Professional secretory cells produce large amounts of secretory material (hormones, neuropeptides, digestive enzymes, mucin, etc.) and store them in secretory granules (SGs) until a secretagogue elicits their bulk exocytosis. These cells usually produce more secretory material than is released by exocytosis1–3 to provide a sufficient pool of available SGs4. Secretory cells continuously turn over the excess SGs by crinophagy, a specialized form of autophagy to maintain a constant releasable pool of SGs2,5–7. Following this route, abnormal or obsolete SGs may also be subject to crinophagic degradation3,7,8. In addition to degradative crinophagy, SG-lysosome fusions may also contribute to the complex maturation process of SGs and thereby determine their controlled release by exocytosis. During crinophagy, SGs directly fuse with lysosomes that gives rise to degradative crinosomes9.
Easy genetic manipulation and highly conserved molecular mechanisms made Drosophila a powerful in-vivo model for deciphering the molecular regulation of the regulated secretory pathway and crinophagy. Salivary gland cells produce10,11 and secrete12 high amounts of Sgs (Salivary gland secretion)/glue proteins in response to peaks of the molting hormone ecdysone10–12. The released glue is then expelled from the lumen to anchor the metamorphosing prepupae to solid surfaces12. The nascent glue SGs emanate from the TGN 13,14, increase in size by homotypic fusions15,16, and then undergo a complex maturation process during which SGs fuse with lysosomes. This promotes the acidification and profound reorganization of the inner content of SGs3,17–19, preparing them for secretion17,19,20. Excess or abnormal glue can be also degraded by crinophagy, through fusion of non-secreted SGs and lysosomes3,7,8,19,21,22. Taken together, crosstalk and fusion between SGs and the endolysosomal compartment is critical both for SG maturation and crinosome formation, however, the molecular mechanism of these processes is still incompletely understood3,18,19,21–23.
By enabling direct fusion between SGs and lysosomes, crinophagy differs mechanistically from the canonical main autophagic pathway, which mediates the degradation of cytosolic material through autophagosome formation and their subsequent fusion with lysosomes. Accordingly, genes that are required for autophagosome formation proved dispensable to crinophagy3,24,25, while SG-lysosome fusion itself relies on a similar molecular machinery acting in fusions between autophagosomes and lysosomes3,5,21,22. The machinery mediating autophagosome-lysosome fusion is well characterized both in Drosophila and humans by now. Critical components include Rab2, Rab7, and Arl8 small GTPases that also contribute to defining membrane identity21,26–29, homotypic fusion and vacuole protein sorting (HOPS) tethering complex30,31, and a soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex (SNAREpin) that executes the fusion. Based on biochemical properties, functional SNAREpins assemble from three Q- (Qabc) and one R-SNARE domains32. The first discovered SNAREpin that mediates autophagosome-lysosome fusion is composed of Syntaxin 17, Snap29 and Vamp7/830,31,33,34. Recently another R-SNARE: Ykt6 was also discovered to also have a role in the process, either as an R-SNARE potentially substituting for Vamp735 or interacting with Syntaxin 7 and Snap29 to form an alternative SNAREpin36,37. Interestingly, Drosophila crinophagic fusion of glue SGs and lysosomes depends on highly similar machinery, composed of Rab2, Rab7, Arl8, HOPS and a Syntaxin 13, Snap29 and Vamp7 SNAREpin3,21,22. The similarity of the molecular machinery regulating these lysosomal fusions raised the possibility that Ykt6 may also regulate SG-lysosome fusions and crinophagy.
Ykt6 is a highly conserved R-SNARE that consists of an N-terminal longin domain (LD), an R-SNARE domain, and a conserved C-terminal lipidation motif with the amino acid sequence CCAIM. The latter is critical for membrane association38–42 because Ykt6, unlike other R-SNAREs, lacks a canonical transmembrane domain. Moreover, the lipid anchors can hide reversibly in the hydrophobic groove of the protein, which enables Ykt6 to leave membranes and form a cytosolic pool38–41. This way, it can be rapidly incorporated into various intracellular membranes on demand and form a complex with compartment-specific Q-SNAREs to promote vesicle fusion. Membrane-associated Ykt6 regulates the anterograde ER to Golgi43,44, the intra-Golgi45–47, retrograde directed Golgi to ER, and endosome to TGN transports48, and the release of constitutive secretory carriers49 or exosomes50,51 along the secretory pathway. In addition, it also promotes biosynthetic transport to the yeast vacuole and lysosomes in animal cells42,52, endosomal recycling53, and macroautophagic degradation35–37,54,55. However, the role of Ykt6 in SG-lysosome fusion and crinophagy remained unknown.
Here, we show that Ykt6 forms a canonical SNAREpin with Syntaxin 13 and Snap29, which is – similarly to the already known Syntaxin 13, Snap29, Vamp7 SNAREpin – critical for crinophagic degradation. We also demonstrate that Ykt6 localizes to small Lamp1+ (carrier) vesicles and mediates their fusion with SGs, while Vamp7 regulates the fusion of SGs and Arl8 + lysosomes. In summary, we provide evidence that SG maturation preceding exocytosis and crinophagy requires a series of fusions between SGs and two separate lysosome-related vesicle subpopulations, which are governed by different SNAREpins/SNARE complexes.