A New Protein Conjugation System in Human
1998; Elsevier BV; Volume: 273; Issue: 51 Linguagem: Inglês
10.1074/jbc.273.51.33889
ISSN1083-351X
AutoresNoboru Mizushima, Hisao Sugita, Tamotsu Yoshimori, Yoshinori Ohsumi,
Tópico(s)Cancer-related gene regulation
ResumoAutophagy is an intracellular process for bulk degradation of cytoplasmic components. We recently found a protein conjugation system essential for autophagy in the yeast,Saccharomyces cerevisiae. The C-terminal glycine of a novel modifier protein, Apg12p, is conjugated to a lysine residue of Apg5p via an isopeptide bond. This conjugation reaction is mediated by Apg7p, a ubiquitin activating enzyme (E1)-like enzyme, and Apg10p, suggesting that it is a ubiquitination-like system (Mizushima, N., Noda, T., Yoshimori, T., Tanaka, Y., Ishii, T., George, M. D., Klionsky, D. J., Ohsumi, M., and Ohsumi, Y. (1998) Nature 395, 395–398). Although autophagy is a ubiquitous process in eukaryotic cells, no molecule involved in autophagy has yet been identified in higher eukaryotes. We reasoned that this conjugation system could be conserved. Here we report cloning and characterization of the human homologue of Apg12 (hApg12). It is a 140-amino acid protein and possesses 27% identity and 48% similarity with the yeast Apg12p, but no apparent homology to ubiquitin. Northern blot analysis showed that its expression was ubiquitous in human tissues. We found that it was covalently attached to another protein. This target protein was identified to be the human Apg5 homologue (hApg5). Mutagenic analyses suggested that this conjugation was formed via an isopeptide bond between the C-terminal glycine of hApg12 and Lys-130 of hApg5. These findings indicate that the Apg12 system is well conserved and may function in autophagy also in human cells. Autophagy is an intracellular process for bulk degradation of cytoplasmic components. We recently found a protein conjugation system essential for autophagy in the yeast,Saccharomyces cerevisiae. The C-terminal glycine of a novel modifier protein, Apg12p, is conjugated to a lysine residue of Apg5p via an isopeptide bond. This conjugation reaction is mediated by Apg7p, a ubiquitin activating enzyme (E1)-like enzyme, and Apg10p, suggesting that it is a ubiquitination-like system (Mizushima, N., Noda, T., Yoshimori, T., Tanaka, Y., Ishii, T., George, M. D., Klionsky, D. J., Ohsumi, M., and Ohsumi, Y. (1998) Nature 395, 395–398). Although autophagy is a ubiquitous process in eukaryotic cells, no molecule involved in autophagy has yet been identified in higher eukaryotes. We reasoned that this conjugation system could be conserved. Here we report cloning and characterization of the human homologue of Apg12 (hApg12). It is a 140-amino acid protein and possesses 27% identity and 48% similarity with the yeast Apg12p, but no apparent homology to ubiquitin. Northern blot analysis showed that its expression was ubiquitous in human tissues. We found that it was covalently attached to another protein. This target protein was identified to be the human Apg5 homologue (hApg5). Mutagenic analyses suggested that this conjugation was formed via an isopeptide bond between the C-terminal glycine of hApg12 and Lys-130 of hApg5. These findings indicate that the Apg12 system is well conserved and may function in autophagy also in human cells. small ubiquitin-related modifier-1 ubiquitin activating enzyme ubiquitin conjugating enzyme ubiquitin cross-reactive protein expressed sequence tag polymerase chain reaction hemagglutinin green fluorescent protein rapid amplification of cDNA ends. Proteins can be modified with a variety of moieties. In particular cases, they are attached by other proteins. Generally, such kinds of post-translational modification alters their metabolic stability, complex formation, or subcellular localization. Ubiquitination is the well known modification system, which is involved in selective protein degradation, endocytosis, etc. (1Hochstrasser M. Annu. Rev. Genet. 1996; 30: 405-439Crossref PubMed Scopus (1444) Google Scholar, 2Hershiko A. Trends Biochem. Sci. 1996; 21: 445-449Abstract Full Text PDF PubMed Scopus (69) Google Scholar, 3Varshavsky A. Trends Biochem. Sci. 1997; 22: 383-387Abstract Full Text PDF PubMed Scopus (511) Google Scholar, 4Weissman A.M. Immunol. Today. 1997; 18: 189-198Abstract Full Text PDF PubMed Scopus (151) Google Scholar). Over the past few years, several other protein conjugation systems for "ubiquitin-related" molecules have been discovered. Conjugation of SUMO-11 to Ran-GAP1 targets the otherwise cytosolic protein to RanBP2, a component of the nuclear pore complex (5Matunis M.J. Coutavas E. Blobel G. J. Cell Biol. 1996; 135: 1457-1470Crossref PubMed Scopus (948) Google Scholar, 6Mahajan R. Delphin C. Guan T. Gerace L. Melchior F. Cell. 1997; 88: 97-107Abstract Full Text Full Text PDF PubMed Scopus (999) Google Scholar, 7Kamitani T. Nguyen H.P. Yeh E.T.H. J. Biol. Chem. 1997; 272: 14001-14004Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar). PML (8Boddy M.N. Howe K. Etkin L.D. Solomon E. Freemont P.S. Oncogene. 1996; 13: 971-982PubMed Google Scholar, 9Sternsdorf T. Jensen K. Will H. J. Cell Biol. 1997; 139: 1621-1634Crossref PubMed Scopus (289) Google Scholar, 10Müller S. Matunis M.J. Dejean A. EMBO J. 1998; 17: 61-70Crossref PubMed Scopus (575) Google Scholar, 11Kamitani T. Nguyen H.P. Kito K. Fukuda-Kamitani T. Yeh E.T.H. J. Biol. Chem. 1998; 273: 3117-3120Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar) and Fas (12Okura T. Gong L. Kamitani T. Wada T. Okura I. Wei C.F. Chang H.M. Yeh E.T.H. J. Immunol. 1996; 157: 4277-4281PubMed Google Scholar) are also modified by SUMO-1. Its yeast homologue, Smt3p, is also attached to other proteins that have not been identified (13Johnson E.S. Schwienhorst I. Dohmen R.J. Blobel G. EMBO J. 1997; 16: 5509-5519Crossref PubMed Scopus (439) Google Scholar). Nedd8 and its yeast homologue Rub1p are conjugated to cullin/Cdc53 (14Osaka F. Kawasaki H. Aida N. Saeki M. Chiba T. Kawashima S. Tanaka K. Kato S. Genes Dev. 1998; 12: 2263-2268Crossref PubMed Scopus (224) Google Scholar, 15Liakopoulos D. Doenges G. Matuschewski K. Jentsch S. EMBO J. 1998; 17: 2208-2214Crossref PubMed Scopus (305) Google Scholar, 16Lammer D. Mathias N. Laplaza J.M. Jiang W. Liu Y. Callis J. Goebl M. Estelle M. Genes Dev. 1998; 12: 914-926Crossref PubMed Scopus (278) Google Scholar). UCRP (17Haas A.L. Ahrens P. Bright P.M. Ankel H. J. Biol. Chem. 1987; 262: 11315-11323Abstract Full Text PDF PubMed Google Scholar) and Fau (18Olvera J. Wool I.G. J. Biol. Chem. 1993; 268: 17967-17974Abstract Full Text PDF PubMed Google Scholar, 19Nakamura M. Xavier R.M. Tsunematsu T. Tanigawa Y. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3463-3467Crossref PubMed Scopus (57) Google Scholar) are also found to be modifier proteins. "Protein conjugation" is getting one of the most important regulatory mechanisms. Recently, we found a quite novel protein conjugation system, which is required for autophagy in yeast (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). Autophagy is a bulk protein degradation process in which cytoplasmic components including organelles are enclosed in double membrane structures termed autophagosomes and delivered to lysosome/vacuole to be degraded (21Seglen P.O. Bohley P. Experientia. 1992; 48: 158-172Crossref PubMed Scopus (368) Google Scholar, 22Dunn Jr., W.A. Trends Cell Biol. 1994; 4: 139-143Abstract Full Text PDF PubMed Scopus (442) Google Scholar). Autophagy is essential for cells to survive during nutrient limitation and may be crucial for cellular remodeling during development and differentiation and removal of obsolete cellular components. This process is ubiquitous in eukaryotes. However, the molecular basis of the autophagic pathway is poorly characterized. We found that autophagy proceeds in the yeast, Saccharomyces cerevisiae, in a manner quite similar to that in animals (23Takeshige K. Baba M. Tsuboi S. Noda T. Ohsumi Y. J. Cell Biol. 1992; 119: 301-311Crossref PubMed Scopus (929) Google Scholar, 24Baba M. Takeshige K. Baba N. Ohsumi Y. J. Cell Biol. 1994; 124: 903-913Crossref PubMed Scopus (395) Google Scholar). It allows us to study the autophagic process at a molecular level. So far we have isolated 14 autophagy-defective (apg) mutants (25Tsukada M. Ohsumi Y. FEBS Lett. 1993; 333: 169-174Crossref PubMed Scopus (1357) Google Scholar). Cloning of APG genes revealed that they were novel genes (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar, 26Kametaka S. Matsuura A. Wada Y. Ohsumi Y. Gene (Amst.). 1996; 178: 139-143Crossref PubMed Scopus (91) Google Scholar, 27Matsuura A. Tsukada M. Wada Y. Ohsumi Y. Gene. 1997; 192: 245-250Crossref PubMed Scopus (376) Google Scholar, 28Funakoshi T. Matsuura A. Noda T. Ohsumi Y. Gene. 1997; 192: 207-213Crossref PubMed Scopus (135) Google Scholar), except APG6, which is allelic toVPS30 involved in vacuolar protein sorting (29Kametaka S. Okano T. Ohsumi M. Ohsumi Y. J. Biol. Chem. 1998; 273: 22284-22291Abstract Full Text Full Text PDF PubMed Scopus (215) Google Scholar). This indicates that little effort has been paid to uncover the molecular mechanism of autophagy. During systematic characterization of Apg proteins, we discovered a new protein conjugation system (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). Apg12p, a 186-amino acid protein, is attached covalently to Apg5p, a 294-amino acid protein. The C-terminal glycine of Apg12p and the Lys-149 of Apg5p is indispensable for the conjugation, suggesting they are conjugated via an isopeptide bond between these two residues. Among Apg proteins, two other factors, Apg7p and Apg10p, are required for the conjugation. Apg7p shows significant homology to the ubiquitin activating enzyme (E1). Apg10p might be a ubiquitin conjugating enzyme (E2)-like protein. Thus, at least four Apg proteins (Apg5p, -7p, -10p, and -12p) function together in the protein conjugation system. Autophagy cannot proceed with a defect in any one of their genes. Although the Apg12p conjugation system is similar to ubiquitination, the amino acid sequence of Apg12p is distinct from that of ubiquitin. Considering that the other modifier proteins discovered are all "ubiquitin-like" (13Johnson E.S. Schwienhorst I. Dohmen R.J. Blobel G. EMBO J. 1997; 16: 5509-5519Crossref PubMed Scopus (439) Google Scholar), Apg12p is the first "ubiquitin-unrelated" modifier. In higher eukaryotes, although autophagy is described morphologically and physiologically (21Seglen P.O. Bohley P. Experientia. 1992; 48: 158-172Crossref PubMed Scopus (368) Google Scholar, 22Dunn Jr., W.A. Trends Cell Biol. 1994; 4: 139-143Abstract Full Text PDF PubMed Scopus (442) Google Scholar), no molecules have been isolated that are directly involved in autophagy. The Apg12 conjugation system would be conserved also in mammals. Here we report the cloning of human Apg12. We also show that it is conjugated to human Apg5 homologue. S. cerevisiae APG12-related genes were searched in the EST data base using the tblastn program. It identified several cDNA fragments (EST183504, EST42556, zo30d07, zx66d05.r1, EST86492, zo17a07.r, etc.) encoding parts of a potential human APG12 homologue. The sequence of 5′-end was obtained using the Marathon cDNA amplification kit (CLONTECH). Briefly, poly(A) mRNA obtained from HeLa cells was reverse-transcribed to cDNA with oligo(dT) primer. The double-stranded cDNA was ligated with the Marathon cDNA adaptor. 5′-RACE (rapid amplification of cDNA ends) was performed with the adaptor primer 1 (AP1) and a gene-specific primer (h12GSP1; 5′-GTTCGTGTTCGCTCTACTGCCCACTT-3′) derived from the ESTs. The PCR products were subcloned and sequenced by automated DNA sequencing (ABI-PRISM model 377). The cDNAs of hApg12 and hApg5 amplified by PCR from total cDNA of HeLa cells were cloned intoEcoRI and SmaI sites of pCI-neo expression vector (Promega), respectively. DNA containing 3× hemagglutinin (HA) or 3× Myc epitope coding sequence was inserted after the first ATG codon of hApg12 (pHA-Apg12 and pMyc-Apg12) and before the stop codon of hApg5 (pApg5-HA). hApg5 was also cloned into pEGFP-C1, a green fluorescent protein (GFP) fusion protein expression vector (CLONTECH) to obtain pGFP-Apg5. Deletion of the C-terminal glycine residue of hApg12 (pHA-Apg12ΔG) and substitution of Lys-130 of hApg5 by arginine (pApg5K130R-HA) were performed by PCR-based mutagenesis and confirmed by DNA sequencing. The cDNA probe of the whole Apg12 coding region was labeled with [α-32P]dCTP using Megaprime labeling kit (Amersham Pharmacia Biotech). Northern blot analysis of human multiple tissue blot (CLONTECH) was performed with the probe according to the manufacturer's instructions. The membrane was stripped and rehybridized with the actin probe (CLONTECH). COS-1 cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal calf serum. For transfection, cells were seeded on 6-well or 35-mm dishes. The next day, cells were transfected with a mixture of 1 μg of plasmid DNA and 3 μl of FuGENE6 (Boehringer Mannheim). For cotransfection, 0.5 μg of each plasmid was used. The transfected cells were harvested 24 h after transfection. The cells were washed once with phosphate-buffered saline and directly lysed in 100 μl of 2× SDS sample buffer containing protease inhibitors and boiled for 5 min. Aliquots (10 μl) of the lysates were separated by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membrane (Millipore). Western blot analysis was performed with mouse monoclonal anti-HA antibody (16B12, BAbCO) and rabbit polyclonal anti-GFP antibody (CLONTECH) and developed by an enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech). A homology search of the human expressed sequence tag (EST) data base identified several cDNA fragments coding regions of a protein that is significantly related to yeast Apg12p (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). However, all of them lack the first methionine codon. To obtain the 5′-end sequence, we performed 5′-RACE method using the gene specific primer (h12GSP1). The PCR amplification yielded a single band of about 250 base pairs. Sequence analysis of the PCR products revealed that two fragments derived from independent PCR contained an ATG codon in the reading frame. One of the PCR fragments reached to nucleotide −19, but no more ATG codon was found. The ATG codon is surrounded by a consensus sequence that is characteristic to the eukaryotic ribosome binding site used for the initiation of translation (30Kozak M. J. Cell Biol. 1991; 115: 887-903Crossref PubMed Scopus (1445) Google Scholar). These suggest that this ATG codon is the translation initiation point (see below). The cDNA encodes a hydrophilic protein, designated as human Apg12 (hApg12), 140 amino acids long with a predicted molecular mass of 15 kDa (Fig. 1). hApg12 is 46 amino acids shorter than the yeast Apg12p and is 27% identical and 48% similar to amino acids 49–186 of the yeast Apg12p (Fig. 1). In particular, the C-terminal half of hApg12 is highly related to the yeast Apg12p. The C-terminal glycine, which is involved in the yeast Apg5p/Apg12p conjugation, is conserved. hApg12, as well as the yeast Apg12p, does not have homology to ubiquitin. The EST sequences of hApg12 were derived from various tissues and cells such as brain, placenta, colon, germinal B cell, and Jurkat T cell. This suggests that hApg12 could be ubiquitously expressed among human tissues. To confirm it, we carried out Northern blot analysis. Four major transcripts of 1.0, 2.4, 3.6, and 5.0 kb were detected in all tissues examined, including heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas, suggesting that hApg12 mRNA is widely expressed (Fig.2). The yeast Apg12p is a novel modifier protein whose C-terminal glycine is conjugated to Apg5p (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). Since the C-terminal regions of hApg12 and the yeast Apg12p are significantly similar, we tested whether hApg12 is also conjugated to other proteins. To detect hApg12, we tagged hApg12 with 3× HA epitopes at its N terminus. COS-1 cells were transfected with pHA-Apg12 and subjected to immunoblot analysis. Most of hApg12 presented as a band at 26 kDa (Fig. 3, lane 2,band a). In addition, we detected a faint band at about 65 kDa (band b). Judging from the predicted molecular mass, the band at 26 kDa represents a monomer of tagged hApg12. The higher band suggests that hApg12 is covalently conjugated to another protein. While we were cloning a human homologue of the yeast Apg5p, it was reported by another group (31Hammond E.M. Brunet C.L. Johnson G.D. Parkhill J. Milner A.E. Brady G. Gregory C.D. Grand R.J.A. FEBS Lett. 1998; 425: 391-395Crossref PubMed Scopus (79) Google Scholar). It is a 276-amino acid-long protein and 26% identical and 46% similar to the yeast Apg5p (26Kametaka S. Matsuura A. Wada Y. Ohsumi Y. Gene (Amst.). 1996; 178: 139-143Crossref PubMed Scopus (91) Google Scholar). To ask whether the higher 65-kDa band (Fig. 3, lane 2, band b) is an hApg12/hApg5 conjugate, we cotransfected COS-1 cells with pHA-Apg12 and pApg5. Overexpression of hApg5 increased the amount of the 65-kDa band (Fig. 3, lane 3), suggesting that it represents an hApg12/hApg5 conjugate. To confirm this, hApg5 was fused with GFP and expressed. The higher band was shifted to about 100 kDa (Fig. 3, lane 4, band c), which was also detected by anti-GFP antibody (Fig. 3, lane 5). Taken together, we concluded that hApg12 is conjugated to hApg5. We next examined whether hApg12 is conjugated to hApg5 in a manner similar to yeast cells. When COS-1 cells were transfected with pHA-Apg12ΔG, in which the C-terminal glycine codon was deleted, the hApg12/hApg5 conjugate was not observed (Fig.4 A, lane 3). This conjugation did not occur even when hApg5 was overexpressed (Fig. 4A,lane 6). These results indicate that C-terminal glycine of hApg12 is required for the conjugation. For analysis of hApg5, we tagged hApg5 with 3× HA epitopes at C terminus, since C-terminal tagging did not affect the function of the yeast Apg5p. On immunoblotting, most of HA-hApg5 presented as a free form of about 37 kDa (Fig. 4 B, lane 2). In addition, another HA-specific band was observed at about 65 kDa (Fig.4 B, lane 2). When untagged hApg12 was overexpressed, intensity of the band was augmented (Fig. 4 B,lane 3). These results indicate that the 65-kDa band represents the hApg12/HA-hApg5 conjugate. The transfected hApg12 generated exactly the same sized band as authentic hApg12 did (Fig.4 B, lanes 2 and 3), supporting the fact that the hApg12 cDNA we cloned contains a whole open reading frame sequence. As expected, overexpression of 3× Myc epitope-tagged hApg12 led to formation of a larger hApg12/hApg5 conjugate (Fig.4 B, lane 4), which was detected with anti-Myc antibody also (data not shown). Replacement of Lys-149 of the yeast Apg5p resulted in complete loss of the conjugate, whereas replacement of the other lysine residues had no effect, suggesting that Lys-149 is the acceptor site for Apg12p conjugation (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). Alignments of the amino acid sequences of the yeast and human Apg5 suggested that Lys-130 of hApg5 is a candidate for acceptor residue. To assess this possibility, we substituted the lysine with arginine (K130R). HA-hApg5K130R was not conjugated by hApg12 (Fig. 4 B, lane 5), even when hApg12 was overexpressed (Fig. 4 B, lane 6). Taken together, these data suggest that hApg12 is conjugated to hApg5 in a manner quite similar to that in yeast; the C-terminal glycine of hApg12 is attached to the central lysine residue of hApg5. We have shown a new human protein conjugation system in which hApg12 is conjugated to hApg5. This is most probably due to an isopeptide bond between the C-terminal glycine of hApg12 and an ε-amino group of Lys-130 of hApg5. These results indicate that the yeast Apg12p conjugation system is well conserved in human. In addition, there are potential Apg12 homologues inArabidopsis thaliana (AA720252) andCaenorhabditis elegans (U32305), suggesting that this system exists in all eukaryotes. In yeast, at least two other proteins, Apg7p (activating enzyme) and Apg10p (possible conjugating enzyme), are required for the conjugation (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). Therefore, these molecules should also have mammalian counterparts. One of the most striking features of the Apg12 system in both yeast and human is a unique character of Apg12 as a "modifier protein." Recently, several "ubiquitin-related" modifiers, such as UCRP (17Haas A.L. Ahrens P. Bright P.M. Ankel H. J. Biol. Chem. 1987; 262: 11315-11323Abstract Full Text PDF PubMed Google Scholar), Fau (18Olvera J. Wool I.G. J. Biol. Chem. 1993; 268: 17967-17974Abstract Full Text PDF PubMed Google Scholar, 19Nakamura M. Xavier R.M. Tsunematsu T. Tanigawa Y. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3463-3467Crossref PubMed Scopus (57) Google Scholar), Smt3p/SUMO-1 (5Matunis M.J. Coutavas E. Blobel G. J. Cell Biol. 1996; 135: 1457-1470Crossref PubMed Scopus (948) Google Scholar, 6Mahajan R. Delphin C. Guan T. Gerace L. Melchior F. Cell. 1997; 88: 97-107Abstract Full Text Full Text PDF PubMed Scopus (999) Google Scholar, 7Kamitani T. Nguyen H.P. Yeh E.T.H. J. Biol. Chem. 1997; 272: 14001-14004Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar, 13Johnson E.S. Schwienhorst I. Dohmen R.J. Blobel G. EMBO J. 1997; 16: 5509-5519Crossref PubMed Scopus (439) Google Scholar), and Rub1p/Nedd8 (14Osaka F. Kawasaki H. Aida N. Saeki M. Chiba T. Kawashima S. Tanaka K. Kato S. Genes Dev. 1998; 12: 2263-2268Crossref PubMed Scopus (224) Google Scholar, 15Liakopoulos D. Doenges G. Matuschewski K. Jentsch S. EMBO J. 1998; 17: 2208-2214Crossref PubMed Scopus (305) Google Scholar, 16Lammer D. Mathias N. Laplaza J.M. Jiang W. Liu Y. Callis J. Goebl M. Estelle M. Genes Dev. 1998; 12: 914-926Crossref PubMed Scopus (278) Google Scholar, 32del Pozo J.C. Timpte C. Tan S. Callis J. Estelle M. Science. 1998; 280: 1760-1763Crossref PubMed Scopus (176) Google Scholar), have been discovered. It has been also uncovered that these modifiers have ubiquitination-like but dedicated conjugation systems (13Johnson E.S. Schwienhorst I. Dohmen R.J. Blobel G. EMBO J. 1997; 16: 5509-5519Crossref PubMed Scopus (439) Google Scholar, 14Osaka F. Kawasaki H. Aida N. Saeki M. Chiba T. Kawashima S. Tanaka K. Kato S. Genes Dev. 1998; 12: 2263-2268Crossref PubMed Scopus (224) Google Scholar, 15Liakopoulos D. Doenges G. Matuschewski K. Jentsch S. EMBO J. 1998; 17: 2208-2214Crossref PubMed Scopus (305) Google Scholar, 16Lammer D. Mathias N. Laplaza J.M. Jiang W. Liu Y. Callis J. Goebl M. Estelle M. Genes Dev. 1998; 12: 914-926Crossref PubMed Scopus (278) Google Scholar,32del Pozo J.C. Timpte C. Tan S. Callis J. Estelle M. Science. 1998; 280: 1760-1763Crossref PubMed Scopus (176) Google Scholar, 33Saitoh H. Pu R. Cavenagh M. Dasso M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3736-3741Crossref PubMed Scopus (163) Google Scholar, 34Desterro J.M.P. Thomson J. Hay R.T. FEBS Lett. 1997; 417: 297-300Crossref PubMed Scopus (302) Google Scholar, 35Gong L. Kamitani T. Fujise K. Caskey L.S. Yeh E.T.H. J. Biol. Chem. 1997; 272: 28198-28201Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 36Schwarz S.E. Matuschewski K. Liakopoulos D. Scheffner M. Jentsch S. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 560-564Crossref PubMed Scopus (187) Google Scholar, 37Lee G.W. Melchior F. Matunis M.J. Mahajan R. Tian Q. Anderson P. J. Biol. Chem. 1998; 273: 6503-6507Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar, 38Johnson E.S. Blobel G. J. Biol. Chem. 1997; 272: 26799-26802Abstract Full Text Full Text PDF PubMed Scopus (404) Google Scholar, 39Narashimhan J. Potter J.L. Haas A.L. J. Biol. Chem. 1996; 271: 324-330Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). Although the Apg12 system is quite similar to the ubiquitin conjugation system, both human and yeast Apg12 are much larger than and have no homology to ubiquitin. However, it is still possible that the three-dimensional structure of Apg12 may resemble that of ubiquitin. All of ubiquitin and its related modifiers are synthesized as precursors and then processed to expose functional C termini that end with Gly-Gly (40Johnson P.R. Hochstrasser M. Trends Cell Biol. 1997; 7: 408-413Abstract Full Text PDF PubMed Scopus (68) Google Scholar) (Fig. 5). In contrast, Apg12 has functional C-terminal "single Gly" instead of "double Gly," and it does not require C-terminal processing for the conjugation (Figs. 1 and 5). Apg12 seems to have a single target protein, Apg5, whereas the other modifiers are conjugated to several proteins, which have not been fully identified. Thus, Apg12 is a quite new type of modifier protein. Our findings may suggest that there are still unidentified modifier proteins that are not necessarily similar to ubiquitin. Hammond et al. (31Hammond E.M. Brunet C.L. Johnson G.D. Parkhill J. Milner A.E. Brady G. Gregory C.D. Grand R.J.A. FEBS Lett. 1998; 425: 391-395Crossref PubMed Scopus (79) Google Scholar) cloned hApg5 as an "apoptosis-specific protein (ASP)." By cross-reaction of polyclonal antibody against c-Jun, they detected somehow modified hApg5 of about 45 kDa. They reported that, otherwise not detected, the 45-kDa band appeared at late stage or as a result of apoptosis, possibly due to post-translational regulation. They failed to detect an unmodified form of hApg5 through the course. Since our experiments shown here are based on transfection, we do not know the natural expression level of hApg5 and hApg12/hApg5 conjugate. However, our results that transfected HA-hApg12 generated the HA-hApg12/hApg5 conjugate in vivo (Fig. 3, lane 2) suggest that hApg5 is constitutively expressed without apoptotic stimuli. Furthermore, the 45-kDa band they detected is much smaller than our 65-kDa band of the conjugate. This discrepancy cannot be attributed to addition of tagged HA peptide (40 amino acids). They may detect degraded or additionally processed forms of hApg5 with anti-c-Jun antibody. Further experiments will be needed to elucidate the role of hApg5 in apoptosis. We do not have direct evidence to show that this system is related to autophagy in human. hApg12 and hApg5 did not complement the autophagy-defective phenotype of yeast strains deficient for Apg12p and Apg5p, respectively (data not shown). However, our results that both the modifier and modified proteins are homologues of yeast Apg proteins strongly suggest that the Apg12 system plays a crucial role in autophagy also in human. Although autophagy is mainly described in hepatocytes and neurons, it is thought to occur in virtually all types of cells (22Dunn Jr., W.A. Trends Cell Biol. 1994; 4: 139-143Abstract Full Text PDF PubMed Scopus (442) Google Scholar). Physiological significance of autophagy in mammals is not clear, but it is probably necessary for normal turnover of cellular components. This is in agreement with the data that mRNA of hApg12 (Fig. 2) and hApg5 (31Hammond E.M. Brunet C.L. Johnson G.D. Parkhill J. Milner A.E. Brady G. Gregory C.D. Grand R.J.A. FEBS Lett. 1998; 425: 391-395Crossref PubMed Scopus (79) Google Scholar) are expressed ubiquitously. Some cDNA clones of hApg12 in EST data base are derived from fetal tissues, suggesting that hApg12 is expressed during fetal stage as well. It remains to be solved at which step in autophagy this modification system works. We speculate that it functions in formation of autophagosome, since (i) accumulation of autophagosomes was not observed in apg5, apg7, apg10, andapg12 mutants, 2M. Baba and Y. Ohsumi, unpublished observation. (ii) most of the yeast Apg12p, Apg5p, and Apg12p/Apg5p conjugate exist in membrane fractions, and (iii) subcellular localization of the yeast Apg12p changes after the conjugation (20Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1249) Google Scholar). On our transfection experiments, about one-half of both overexpressed hApg12 and hApg5 were detected in membrane fractions. 3N. Mizushima, H. Sugita, T. Yoshimori, and Y. Ohsumi, unpublished observation. Therefore, it is expected that the human Apg12 conjugation system is also involved in membrane dynamism. We now are examining their precise intracellular localization and their expression level under various conditions. Finally, generation of mice deficient for this conjugation system will provide important information on the physiological role of the system, moreover that of autophagy in mammals.
Referência(s)