|
内容記述 |
Autophagy is a major self-degradative process in eukaryotic cells that plays<br />fundamental roles in cellular and organismal homeostasis, and is involved in many<br />physiological and pathological situations. When autophagy is induced, cytoplasmic<br />materials and organelles are sequestered into newly emerging double-membrane<br />vesicles called autophagosomes, and delivered to the lysosome or the vacuole for<br />degradation. <br /> In the past decade, many <i>ATG</i> (<u>a</u>u<u>t</u>opha<u>g</u>y-related) genes have been identified by<br />genetic approaches using the yeast Saccharomyces cerevisiae. Atg8, a ubiquitin-like<br />protein (Ubl), is one of the proteins essential for autophagosome formation. The cysteine protease Atg4 first removes the C-terminal arginine of Atg8 to expose the glycine as the new terminus. This glycine forms a thioester bond with Atg7, an activating enzyme (E1), and transfers to and also forms a thioester bond with Atg3, a conjugation enzyme (E2). Atg8 is eventually conjugated to the amino group in the hydrophilic head of<br />phosphatidylethanolamine (PE). Atg8 is anchored to isolation membrane and<br />autophagosomal membranes probably as this lipid-modified form, and thought to directly participate in the formation of these membranes. Atg4 also catalyzes the deconjugation of Atg8-PE after it has fulfilled its role in autophagosome formation, thus Atg8 is reused. Because the details of this sequential reaction of Atg8 lipidation are unclear, I focus on the mechanism of Atg8 lipidation, in this study. <br /> The Atg8 conjugation system was reconstituted using purified proteins expressed in<br /><i>Escherichia coli</i>and PE-containing liposomes in vitro. First, I successfully capture<br />authentic thioester intermediates, Atg8-Atg7 and Atg8-Atg3, which can not have been<br />detected because of their lability. This allows me to analyze the sequential reaction of<br />Atg8 lipidation. <br /> It was shown that Atg8 could be conjugated with phosphatidylserine (PS) as<br />efficiently as PE in vitro. However, PE was identified as the sole lipid conjugated to the<br />C-terminal glycine of Atg8 in vivo. It suggests that there exists a mechanism that directs<br />Atg8 conjugation preferentially to PE in the cell. In this study, I show that, in contrast to<br />PE conjugation, the PS conjugation of Atg8 is markedly suppressed at physiological<br />(neutral) pH. Then, I show that both of the Atg8-Atg7 and the Atg8-Atg3 intermediates<br />are formed in the presence of PS liposomes as rapidly as in the presence of PE<br />liposomes, and transfer of Atg8 from Atg3 to PS is specifically retarded at neutral pH. <br />Furthermore, the addition of acidic phospholipids to liposomes is also suggested to result in the preferential formation of the Atg8−PE conjugate. I also show that the acidic<br />phospholipids specifically promote the recruitment of the Atg8-Atg7 and the Atg8-Atg3<br />thioester intermediates to the membrane. Furthermore, .it was reported that the<br />Atg12-Atg5 conjugate, which is formed by ubiquitin-like conjugation reaction, is<br />indispensable for Atg8-PE production in vivo, and that recombinant Atg12-Atg5 indeed<br />stimulates Atg8-PE and Atg8-PS production in vitro. <br /> The preferential formation of Atg8-PE can be achieved by combination of neutral<br />pH, acidic phospholipids, and the Atg12-Atg5 conjugate. Furthermore, I show that PS is<br />not essential for autophagosome formation even if Atg8 is conjugated to PS in vivo, <br />because the autophagic activity of cells deficient for the PS synthesis enzyme-deficient<br />(<i>pss1△</i>) cells was normal. However, in vitro, the less efficient but significant production of Atg8-PS was still observed, suggesting that the exclusive formation of Atg8-PE requires precise in vivo settings for these factors and/or other facto(s). Alternatively, this result may imply the production of Atg8-PS in vivo, although its amount should be much less than the PE oonjugate. Previously, PE was detected as the sole lipid conjugated to Atg8 and its mammalian homolog LC3 (microtubule-associated protein light chain 3) in vivo. However, lipidated Atg8 and LC3 were forced to accumulate by mutation or treatment with lysosomal inhibitors under nutrient-replete conditions. Therefore, there also remains an alternative possibility that Atg8-PS is formed in starved cells undergoing autophagy, so, I try to purify lipidated Atg8 from the cells under starvation conditions for detailed analysis of lipids conjugated to Atg8 by LC-MS/MS. <br /> Next, I perform gel filtration chromatography to know the interaction of the<br />components in the Atg8 conjugation reaction in vitro. Atg3 interacts with Atg7 as it shown, and Atg3 also interacts with the Atg8-Atg7 thioester intermediate. Furthermore, I find the<br />Atg8-Atg3 thioester intermediate releases from Atg7. It is reasonable because Atg7 is<br />reused rapidly for the next cycle of the reaction. Furthermore, I show that Atg3 and Atg7<br />are interacted via the disulfide bond between active center cysteines in Atg3 and Atg7. I<br />show the model of the sequential reaction of Atg8 lipidation. |
要旨・審査要旨 (310.7 kB)
