Autophagosome–lysosome fusion is independent of V-ATPase-mediated acidification

ARTICLE Received 22 Oct 2014 | Accepted 23 Mar 2015 | Published 11 May 2015 DOI: 10.1038/ncomms8007 OPEN Autophagosome–lysosome fusion is independent of V-ATPase-mediated acidification Caroline Mauvezin1, Péter Nagy2, Gábor Juhász2 & Thomas P. Neufeld1 The ATP-dependent proton pump V-ATPase ensures low intralysosomal pH, which is essential for lysosomal hydrolase activity. Based on studies with the V-ATPase inhibitor BafilomycinA1, lysosomal acidification is also thought to be required for fusion with incoming vesicles from the autophagic and endocytic pathways. Here we show that loss of V-ATPase subunits in the Drosophila fat body causes an accumulation of non-functional lysosomes, leading to a block in autophagic flux. However, V-ATPase-deficient lysosomes remain competent to fuse with autophagosomes and endosomes, resulting in a time-dependent formation of giant autolysosomes. In contrast, BafilomycinA1 prevents autophagosome–lysosome fusion in these cells, and this defect is phenocopied by depletion of the Ca2 þ pump SERCA, a secondary target of this drug. Moreover, activation of SERCA promotes fusion in a BafilomycinA1-sensitive manner. Collectively, our results indicate that lysosomal acidification is not a prerequisite for fusion, and that BafilomycinA1 inhibits fusion independent of its effect on lysosomal pH. 1 Department of Genetics, Cell Biology and Development, University of Minnesota, 6-160 Jackson Hall, 321 Church Street SE, Minneapolis, Minnesota 55455, USA. 2 Department of Anatomy, Cell and Developmental Biology, Eötvös Loránd University, Pazmany s. 1/C. 6.520, Budapest H-1117, Hungary. Correspondence and requests for materials should be addressed to T.P.N. (email: ) NATURE COMMUNICATIONS | 6:7007 | DOI: 10.1038/ncomms8007 | www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. 1 ARTICLE L NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8007 ysosomal degradation and amino-acid recycling are the ultimate steps of the conserved cellular process known as macroautophagy (hereafter autophagy). As dysregulation of autophagy leads to a wide range of pathologies1, a large effort has been made to develop tools for a better understanding and control of the multistep autophagic process2. Autophagy begins with formation of an initiating membrane known as the phagophore, which may develop from multiple sources, including ER–mitochondrial junctions, the ER–Golgi intermediate compartment or plasma membrane, leading to formation of the double-membrane autophagosome3. This vesicle closes to envelop its target, and then converges and fuses with vesicles from the endocytic pathway, thereby forming amphisomes. Ultimately amphisomes and autophagosomes fuse with lysosomes to form autolysosomes (single-membrane vesicles), whose acidic environment leads to activation of the enzymes essential to degrade biological material. The V-ATPases are proton pumps that establish and maintain the acidic environment of lysosomes and other membrane-bound compartments by pumping protons into the lumen, a dynamic process that requires ATP hydrolysis. The V-ATPase is a heteromultimeric enzyme composed of a cytosolic catalytic V1 sector and a membrane-bound V0 proton pore sector. Regulation of the holoenzyme is achieved through reversible binding of the V1–V0 sectors in response to protein kinase A (PKA)-dependent signalling and other cues4,5. Upon formation of a stable V1–V0 complex, ATP hydrolysis drives rotation of a central stalk domain, facilitating the transfer of two protons across the lysosomal membrane for each molecule of ATP hydrolysed (Supplementary Fig. 1a). V-ATPases promote multiple cellular functions in addition to lysosome-mediated degradation, including sorting of cargo in the endosomal and secretory pathways6, proton-coupled transport of ions and solutes and acidification of the pericellular space7. Accordingly, perturbation of V-ATPase function has been linked to a broad spectrum of diseases including lysosomal storage disorders, neurodegeneration, myopathy, bone diseases and cancer8. A better understanding of V-ATPase function, regulation and pharmacology therefore holds promise of leading to improved disease therapies. The V-ATPase inhibitor BafilomycinA1 (BafA1) is a macrolide antibiotic derived from Streptomyces griseus that targets the V0 sector, inhibiting rotation and passage of protons into the lysosomal lumen, thereby reducing vesicle acidification9,10. BafA1 also blocks the fusion between autophagosomes and lysosomes in cultured mammalian cells, but the mechanisms are unknown. These dual properties of BafA1 have led to the view that lysosomal acidification is required for fusion. Although regulation of vesicle fusion by V-ATPase proton pump subunits has been described11, the specific role of V-ATPase in autophagic vesicle fusion is still unknown. Here we use genetic analysis in Drosophila melanogaster to characterize the relationship between fusion and acidification in vivo. In this study, we show that (1) genetic depletion of V-ATPase subunits disrupts lysosomal acidity, blocks autophagic flux and leads to an accumulation of large non-functional autolysosomes; (2) defects in lysosomal acidification do not prevent autophagosome–lysosome fusion; (3) in contrast, BafA1 inhibits autophagosome–lysosome fusion in Drosophila as it does in mammalian cells; and (4) BafA1 potentially targets the Ca2 þ sarco/endoplasmic reticulum Ca2 þ -ATPase (SERCA) pump to inhibit vesicle fusion. These findings thus show that lysosomal acidification and fusion are two separable, independent events. Better understanding of the mode of action of autophagy inhibitors such as BafA1 is essential for further development of such potential therapeutic compounds. 2 Results Loss of V-ATPase disturbs the autophagic pathway. The V-ATPase proton pump is an enzymatic macro-complex composed of eight V1, six V0 and two accessory subunits (Supplementary Fig. 1a). V-ATPase structure and composition are well conserved throughout evolution6,12,13. In Drosophila, there are 13 and 18 partially redundant orthologues for the V1 and the V0 subunits, respectively (Supplementary Fig. 1a and Supplementary Table 1). The powerful genetics of this system offers a convenient model between yeast and mammalian cells systems to study conserved cellular pathways with low genetic redundancy. We used flippase (FLP)-FLP recognition target (FRT)-mediated recombination to generate genetically mosaic tissue in the larval fat body, which functions similarly to mammalian liver and adipose tissue, and efficiently induces autophagy in response to larval starvation. We first examined the role of V-ATPase subunits in autophagy by characterizing the number and size of structures labelled with mCherry-tagged Atg8a (mCh-Atg8a), a specific marker of autophagosomes and autolysosomes2,14. To deplete individually each V-ATPase subunit, we tested commercially available RNA interference (RNAi) lines as well as l (...truncated)