Multiple overlapping SNARE complexes drive endosome maturation in Drosophila nephrocytes

Dávid Hargitai; Márton Molnár; András Rubics; Iván Bodor; Dóra Baukál; Anikó Nagy; Virág Balogh; Zsófia Simon-Vecsei; Gábor Juhász; Péter Lőrincz

doi:10.1038/s41467-026-77479-z

. 2026 Sep 3;17:10512. doi: 10.1038/s41467-026-77479-z

Multiple overlapping SNARE complexes drive endosome maturation in Drosophila nephrocytes

Dávid Hargitai ^1,^2,^#, Márton Molnár ^1,^2,^#, András Rubics ¹, Iván Bodor ^1,², Dóra Baukál ¹, Anikó Nagy ^1,², Virág Balogh ^1,², Zsófia Simon-Vecsei ^1,², Gábor Juhász ^1,³, Péter Lőrincz ^1,^2,^4,^✉

PMCID: PMC13639004 PMID: 42834050

Abstract

Endosomal maturation determines whether internalized cargo is recycled or degraded, yet the logic governing early endosomal fusion remains incompletely defined. Although commonly depicted as a linear Rab5-to-Rab7 transition mediated by a single SNARE pathway, we show that early endosome maturation is driven by multiple parallel, non-interchangeable SNARE complexes. Using Drosophila nephrocytes as a tractable in vivo model, we define a Syx12L–Snap29–Ykt6 complex that mediates homotypic early endosomal fusion and identify two related assemblies–Syx7L–Snap29–Ykt6 and Syx7L–Snap29–Vamp7–that promote later endosomal and lysosomal fusion with distinct Rab requirements. When the canonical Syx12L pathway is disrupted, alternative complexes remain active and generate aberrant endolysosomal swirls which also form in the absence of Rab5 or early endosomal tethers. Thus, Rab5-independent membrane progression toward lysosomes persists but is insufficient to produce fully functional degradative organelles. Together, our findings reveal a network of parallel fusion pathways underlying endosome maturation.

Subject terms: SNARE, Membrane fusion, Endosomes, Lysosomes, Endocytosis

Hargitai, Molnar et al. show that parallel, non-interchangeable SNARE complexes drive endosome maturation in Drosophila nephrocytes, revealing alternative routes that allow membrane progression toward lysosomes.

Introduction

Eukaryotic cells internalize transmembrane receptors and extracellular material through endocytosis. Following uptake, cargo is either recycled back to the plasma membrane or delivered to lysosomes for degradation. Early (sorting) endosomes constitute the central decision-making hub of this pathway. These dynamic tubulo-vesicular compartments mature through coordinated membrane remodeling and fusion events that are tightly regulated by phosphatidylinositol-phosphates and small GTPases, including Rab proteins. In their GTP-bound state, Rab proteins recruit effector molecules that control vesicle targeting, tethering, and fusion. Membrane fusion itself is executed by SNARE (Soluble N-ethylmaleimide–sensitive factor Attachment protein REceptor) proteins, whose assembly is promoted by the SNARE chaperone SM-proteins and tethering factors that bridge opposing membranes and stabilize trans-SNARE complexes^1–5.

In addition to large coiled-coil tethering proteins such as Early Endosomal Antigen 1 (EEA1) and Rabenosyn-5, endosome maturation relies primarily on two related multisubunit tethering complexes. The HOPS complex promotes fusion of lysosomes with Rab7- and/or Rab2-positive late endosomes, autophagosomes, secretory granules, and Golgi-derived carriers^6–18. Its counterpart, CORVET (Class C core vacuole/endosome tethering), mediates tethering and fusion of Rab5-positive early endosomal membranes^9,19–27. HOPS and CORVET share a conserved class C core composed of Vps11, Vps16, Vps18, and Vps33, but differ in their Rab-interacting subunits: CORVET contains the Rab5-binding proteins Vps8 and Vps3 (or Vps39-2/ Tgfbrap1/Trap1 in mammals), whereas HOPS incorporates the Rab7 adaptor Vps41 and the Rab2-binding subunit Vps39 in yeast^{9,21,22,27–29}. In Drosophila, CORVET is further simplified into a smaller variant termed miniCORVET, composed of Vps8 and only three class C subunits (Dor/Vps18, Car/Vps33A, and Vps16A), rendering Vps11 specific to HOPS in flies^30,31.

SNARE complexes generally consist of three or four SNARE proteins, each contributing one or two SNARE motifs. Within the resulting four-helix bundle, the motifs fall into four classes - Qa, Qb, Qc, and R³². Functional SNARE complexes obey the QabcR rule, incorporating one motif from each class, and therefore usually consist of four different proteins (32–35). An important exception is provided by Qbc-SNAREs, which contain two SNARE motifs; in these cases, a single Qa- and a single R-SNARE are sufficient to form a functional complex³³. By the late 2000s, a canonical set of SNARE complexes had been proposed to mediate the major fusion steps of endosomal maturation in yeast and mammalian cells. In mammals, the STX12 (Qa)–VTI1A (Qb)–STX6 (Qc)–VAMP4 (R) complex was shown to drive homotypic early endosome fusion³⁴. As endosomes mature, late endosomes form and undergo homotypic fusion or heterotypic fusion with early endosomes via the STX7 (Qa)–VTI1b (Qb)–STX8 (Qc)–VAMP8 (R) complex^35–37. Subsequent fusion of late endosomes with lysosomes is mediated by the closely related STX7 (Qa)–VTI1b (Qb)–STX8 (Qc)–VAMP7 (R) complex, which also supports homotypic lysosome fusion^38,39.

More recent work, however, has substantially revised this linear view. As summarized by Dingjan and colleagues, multiple SNARE complexes can often catalyze the same fusion step, with their relative contributions varying across cell types and physiological contexts⁴⁰. Proteins such as VTI1a and VTI1b, previously considered essential and functionally distinct, can be partially redundant or even dispensable in specific settings^41–44. Similar context-dependent redundancy has been proposed for several other SNAREs^40,45–48, underscoring the plasticity and complexity of endosomal fusion in mammalian cells.

In contrast to mammals, SNARE redundancy is markedly lower in Drosophila, which encodes roughly half as many SNARE proteins while retaining clear homologs for most mammalian counterparts^49–51. In flies, Syx7 (Qa)^52,53, Snap29 (Qbc)⁵⁴, Syb (R)⁵⁵, and nSyb (R)⁵⁵ have been implicated in early endocytic steps, whereas Syx13 (Qa)^14,56,57, Syx16 (Qa)⁵⁸, Syx17 (Qa)^59,60, Syx18 (Qa)⁶¹, Sec20 (Qb)⁶¹, Snap29 (Qbc)^54,59,60, Ykt6 (R)⁶², and Vamp7 (R)^14,59,60 have been linked to lysosomal fusions and autophagy. Additional roles in recycling and retrograde transport have been described for Syx1 (Qa)⁶³, Syx4 (Qa)⁶³, Syx16 (Qa)⁶⁴, Ykt6 (R)⁶⁵, and Syb (R)^55,64,66. Importantly, we noticed a notable and still unexplained difference in the literature between mammals and Drosophila: Syntaxin7 is consistently described as a late endosomal/lysosomal SNARE in mammalian cells^38,67, yet in flies it has been reported to function primarily at early endosomes^52,53.

Despite the identification of numerous SNAREs associated with endosome maturation, only a limited number of functional SNARE complexes have been rigorously defined. Many studies have focused on individual SNAREs or small subsets of the family, often without biochemical validation of complex formation, leaving key mechanistic questions unresolved. Increasing evidence from multiple model systems indicates that several SNARE assemblies can act in parallel during endosomal maturation, complicating efforts to define linear pathways and obscuring how vesicle fate is determined within this highly interconnected system.

Here, we take advantage of the reduced SNARE redundancy of Drosophila to perform a comprehensive, system-level analysis of SNARE function in the endolysosomal pathway. Using Drosophila nephrocytes as a well-established model of endocytosis^11,30,31, we identify multiple SNARE complexes that act in parallel during the early stages of endosome maturation. Our results resolve long-standing ambiguities regarding the functional assignment of Syx7-family Qa-SNAREs in flies, leading us to propose a revised nomenclature that reflects both evolutionary history and functional specialization. Based on genetic, biochemical, and ultrastructural analyses, we present an updated model of early endosomal fusion that incorporates pathway branching, differential Rab5/Rab7 dependence, and regulated partial redundancy among SNARE complexes. Together, our findings establish how overlapping SNARE machineries coordinate vesicle fate decisions within the endolysosomal system.

Results

Syx7 (Qa), Snap29 (Qbc) and Ykt6 (R) are required for early endosomal fusion in nephrocytes

To systematically assess the contribution of SNARE proteins to endosomal maturation, we performed an RNAi-based screen by targeting 23 of the 25 Drosophila SNARE genes one by one using 77 independent, commercially available RNAi constructs. We compared all phenotypes to a control (luciferase RNAi); Vti1b was excluded due to the lack of available strains, and Syx17 was omitted because we previously showed that its depletion blocks endosome maturation indirectly via a secondary “tethering lock” effect⁶⁸. As a primary readout, we quantified the size of Rab7–positive endosomes, which sensitively reflects defects along the endocytic pathway (Fig. 1A shows statistics for representative RNAi constructs, and representative micrographs are shown in Supplementary Figs. 1, 2). Impaired early endosomal fusion typically results in reduced late endosome size (Fig. 1A–E)³⁰, whereas defects in lysosomal fusion, degradation, or recycling lead to enlarged late endosomes (Fig. 1A)^{11,30,31,69–72}. SNAREs were considered hits when at least two independent RNAi lines produced comparable phenotypes with the same effect on Rab7+ endosome size. Based on increased late endosome size, we identified Syx1 (Supplementary Fig. 1B), Syx13 (Supplementary Fig. 1F), Syx16 (Supplementary Fig. 1G), Syx18 (Supplementary Fig. 1H), Membrin (Supplementary Fig. 1J), Sec20 (Supplementary Fig. 1K), Syb (Supplementary Fig. 2E) and Vamp7 (Supplementary Fig. 2G) as SNAREs acting in later stages of endosome maturation and/or recycling in Drosophila. Strikingly, silencing of only three SNAREs - Syx7 (Qa) (Fig. 1C and Supplementary Fig. 1E), Snap29 (Qbc) (Fig. 1D and Supplementary Fig. 2C), and Ykt6 (R) (Fig. 1E and Supplementary Fig. 2H)—consistently resulted in reduced average Rab7-positive endosome size, indicative of defects at earlier stages Notably, the Qa–Qbc–R identities of Syx7, Snap29 and Ykt6 make them a plausible candidate set for a single functional SNARE complex. While a single RNAi construct against Use1 (Qc) (Supplementary Fig. 1P) also caused a mild endosome size reduction, the lack of independent supporting evidence led us to exclude it from further analysis.

To characterize Syx7, Snap29 and Ykt6 RNAi phenotypes in more detail, we analyzed early endosomes and lysosomes using Rab5-GFP (Fig. 1B–G) and anti-Lamp1 (Fig. 1H–M) staining, respectively, in combination with anti-Rab7 labeling. In all three knockdowns, Rab5-positive early endosomes were significantly smaller (Fig. 1F), supporting a role for Syx7, Snap29, and Ykt6 in early endosomal fusions. Snap29 depletion caused the most pronounced reduction in early endosome size. Moreover, colocalization between Rab5-GFP and anti-Rab7, quantified using the Threshold Overlap Score (TOS) method (Fig. 1G), was increased in all three conditions, indicating impaired endosomal maturation. According to the original description of the method, positive TOS values (0.1 to 1) indicate colocalization, values close to zero ( − 0.1 to 0.1) indicate random overlap, whereas negative values (−1 to −0.1) indicate segregation of the two signals⁷³. Late endosomes were also affected in all three knockdowns, although with distinct phenotypic signatures. Snap29 silencing resulted in uniformly small Rab7-positive compartments (Fig. 1D), mirroring the early endosome phenotype. In contrast, Syx7 knockdown (Fig. 1C) reduced average late endosome size but additionally produced a subset of unusually large Rab5-GFP negative, Rab7-positive compartments with surprisingly thick Rab7 signal–a phenotype not observed in any other single SNARE depletion. Ykt6 silencing (Fig. 1E) generated smaller-than-control Rab7-positive endosomes with markedly increased Rab5-GFP colocalization.

Lysosomes were similarly affected: anti-Lamp1–positive compartments were smaller in all three knockdowns than in the control (Fig. 1H). In Syx7-depleted cells (Fig. 1I), Lamp1 strongly colocalized with Rab7 on the above-mentioned large aberrant endolysosomal structures. In contrast, Snap29 (Fig. 1J) silencing did not increase Lamp1–Rab7 colocalization relative to controls, whereas Ykt6 depletion (Fig. 1K) resulted in a population of Rab7/Lamp1 double-positive endolysosomes.

Silencing of Snap29, Syx7, or Ykt6 impairs endocytic uptake

Our morphological analyses suggested that Snap29, Syx7, and Ykt6 function at early stages of endosome maturation. To determine whether these defects also compromise endocytic activity, we performed tracer uptake assays in nephrocytes. As an initial screen, we monitored short-term (5 min) uptake of FITC–Avidin following RNAi-mediated silencing of individual SNARE genes using constructs selected based on the results of our late endosome size (anti-Rab7) screen (Fig. 2A). While depletion of Syx1 (Fig. 2A and Supplementary Fig. 3B), Syx5 (Fig. 2A and Supplementary Fig. 3D), or Use1 (Fig. 2A and Supplementary Fig. 3P) caused a moderate reduction in tracer internalization, silencing of Syx7 (Fig. 2A and Supplementary Fig. 3E), Snap29 (Fig. 2A, Supplementary Fig. 3S), or Ykt6 (Fig. 2A and Supplementary Fig. 3X) completely abolished detectable FITC–Avidin uptake (Fig. 2A). This result indicated a severe disruption of endocytosis, in line with our endosome size and colocalization analyses. Garland nephrocytes internalize material through deep plasma membrane invaginations termed lacunae, which are lined by a slit diaphragm that is structurally and functionally analogous to that of mammalian podocytes. This filtration barrier is essential for nephrocyte detoxification and hemolymph sieving, and it has a size exclusion limit close to that of FITC–Avidin (66 kDa). Previous studies have shown that impaired trafficking of slit diaphragm components can lead to “clogging” at the lacunar entrance, reducing the effective exclusion limit and thereby blocking uptake of larger tracers (Fig. 2B)⁷⁴.

To test whether such clogging contributes to the observed phenotype, we compared uptake of FITC–Avidin and a much smaller fluorescent dextran (10 kDa) following 5- and 30-min incubations. In control nephrocytes, both tracers were efficiently internalized, with intracellular signal increasing over time (Fig. 2C, D, L). In contrast, Syx7- (Fig. 2E, F, K) or Snap29-depleted cells (Fig. 2G, H, K) failed to internalize FITC–Avidin, while dextran uptake was reduced relative to controls (Fig. 2F, H, K) and did not increase with prolonged incubation (Fig. 2M, N, P). These findings are consistent with clogging of the slit diaphragm. However, if clogging were the sole defect, intracellular dextran would still be expected to accumulate gradually over time, albeit more slowly. Instead, dextran levels remained essentially unchanged between the 5- and 30-min incubations, suggesting that internalized dextran was continuously removed from the cell at a rate similar to its uptake. This observation raised the possibility that impaired endosome maturation promotes accelerated recycling of internalized membrane and cargo back to the plasma membrane. We therefore tested whether impaired endosome maturation promotes accelerated recycling to the plasma membrane. Nephrocytes were incubated with both tracers for 30 min, followed by a 10-min chase in tracer-free medium (Fig. 2Q–U). In control cells, tracer levels remained unchanged (Fig. 2Q, U). In contrast, Syx7- (Fig. 2R, U) or Snap29-depleted cells (Fig. 2S, U) were almost completely devoid of dextran after the chase, supporting the notion of rapid recycling in these conditions.

Ykt6 depletion produced a distinct phenotype. After 5 min (Fig. 2I–K), uptake resembled that observed upon Syx7 or Snap29 silencing (Fig. 2F, H). However, after 30 min, FITC–Avidin was clearly internalized, and dextran levels increased compared to the short incubation (Fig. 2I, O, P). This indicates that loss of Ykt6 does not disrupt slit diaphragm sieving but instead reduces overall endocytic efficiency. Consistently, tracer retention following the 10 min chase (Fig. 2I, T, U) was comparable to that after continuous incubation (Fig. 2O), indicating that Ykt6 silencing does not trigger rapid recycling. To further corroborate these findings, we performed a silver nitrate uptake assay by rearing larvae on silver nitrate–containing food, in which silver accumulates and precipitates in the lysosomes of nephrocytes. In all three SNARE knockdowns, but most profoundly in Syx7 and Snap29, silver accumulation was reduced relative to controls (Supplementary Fig. 3Y) and mirrored the altered lysosomal patterns observed by anti-Lamp1 staining (Fig. 1H–K).

Together, these uptake assays demonstrate that Syx7, Snap29, and Ykt6 are required for efficient early endocytic function, although the loss of each SNARE affects the pathway in a distinct manner. To determine whether these proteins act together within the same molecular machinery, we next examined their localization and physical interactions.

Snap29, Syx7, and Ykt6 localize to Rab5-positive endosomes and form a SNARE complex

To determine the subcellular localization of the three candidate SNAREs, we analyzed garland nephrocytes expressing Rab5–GFP (Fig. 3A–C) or stained for Rab7 (Fig. 3D–F). Syx7 (Fig. 3A, D, H) and Snap29 (Fig. 3B, E, H, I) were detected using specific antibodies, while Ykt6 localization was assessed using a Ykt6–HA transgene (Fig. 3C, F, H, I). All three SNAREs showed strong colocalization with Rab5-positive early endosomes (Fig. 3A-C) and minimal overlap with Rab7-positive compartments (Fig. 3D–F), as quantified by Threshold Overlap Score (TOS) (Fig. 3G). Notably, in control cells the TOS values between Rab5–GFP and Rab7 were close to zero similar to the SNARE and Rab7 pairs (Fig. 1G), suggesting that Syx7, Snap29, and Ykt6 still be transiently present on Rab5/Rab7 double-positive endosomes and are removed when Rab7 conversion occurs.

We next asked whether these SNAREs reside on the same vesicles. Double immunostaining for Syx7 and Snap29 (Fig. 3H), as well as for Snap29 and Ykt6–HA (Fig. 3I), revealed substantial colocalization in both cases (Fig. 3G). The slightly lower Threshold Overlap Score (TOS) observed for Snap29–Ykt6 (Fig. 3G) is compatible with a model in which Qa/Qbc precomplexes assemble before R-SNARE engagement, although it may also reflect additional Snap29-independent functions of Ykt6 on endosomal membranes. The strong functional similarity at early endosomes and their overlapping localization suggested that Syx7, Snap29, and Ykt6 act together as a SNARE complex. To test this, we performed biochemical interaction assays. In GST pull-down experiments (Fig. 3J), bacterially expressed GST–Syx7 SNARE motif efficiently bound full-length MBP–Snap29. In contrast, MBP–Ykt6 bound to GST–Syx7 only in the presence of MBP–Snap29, indicating that Syx7 engages R-SNAREs via the Qbc-SNARE Snap29. Because previous work has implicated Vamp7 as an alternative endosomal R-SNARE in Snap29-containing complexes^14,59,75, we tested whether the Syx7–Snap29 precomplex could also engage Vamp7. Indeed, the closely related MBP–Vamp7 SNARE motif showed the same Snap29-dependent binding behavior as Ykt6, whereas control GST and MBP proteins showed negligible binding.

These experiments demonstrate three key points: (i) Syx7, Snap29, and Ykt6 can assemble into a SNARE complex without intrinsic biochemical constraints at the SNARE-motif level; (ii) the Qa-SNARE Syx7 engages R-SNAREs only through the Qbc-SNARE Snap29; and (iii) the Syx7–Snap29 Q-SNARE precomplex can interact with more than one type of R-SNARE, including both Ykt6 and Vamp7.

Because in vitro pull-down assays do not capture the full regulatory context of SNARE assembly⁷⁶, we next tested complex formation in cultured fly cells. Using Drosophila S2R+ cells expressing tagged proteins, co-immunoprecipitation experiments (Fig. 3K) showed that Syx7–FLAG precipitated both Ykt6–HA and Snap29–HA. Importantly, Syx7–FLAG also co-precipitated endogenous Snap29. This interaction was further confirmed by co-immunoprecipitation from lysates expressing only Syx7–HA (Fig. 3L), which likewise precipitated endogenous Snap29.

We next asked whether depletion of individual SNAREs altered the localization of the remaining endosomal SNAREs (Supplementary Fig. 4A–R). As expected, RNAi against Syx12L, Snap29, or Ykt6 efficiently reduced the level of the corresponding target protein (Supplementary Fig. 4B, F, I, L, P, R) compared with controls (Supplementary Fig. 4A, G, M). In contrast, depletion of one SNARE did not grossly affect the localization or apparent abundance of the remaining components. Syx12L remained associated with Rab5-positive early endosomes in Snap29- or Ykt6-depleted cells (Supplementary Fig. 4C–F), and Ykt6-HA likewise retained its early endosomal localization following depletion of Syx12L or Snap29 (Supplementary Fig. 4N–Q). Similarly, Snap29 remained predominantly associated with early endosomes in Syx12L- or Ykt6-depleted cells (Supplementary Fig. 4H–K), although its overlap with Rab5-positive structures was reduced compared with controls. These observations indicate that depletion of individual SNAREs does not cause widespread mislocalization or destabilization of the remaining endosomal SNARE machinery.

Together, our localization and biochemical data demonstrate that Syx7, Snap29, and Ykt6 form a functional SNARE complex acting at Rab5 positive early endosomes.

Drosophila Syx7/Avl (CG5081) is functionally analogous to mammalian STX12, whereas Syx13 (CG11278) corresponds to STX7

Comparison of our findings with the mammalian literature revealed an apparent inconsistency: in mammalian cells, STX7 functions at late endosomes and lysosomes, whereas in Drosophila Syx7 localizes to early endosomes and is required for early endosomal fusion^52,77. Given the generally conserved roles of SNARE proteins, we sought to resolve this discrepancy. In silico analyses of SNARE evolution indicate that the endosomal Syntaxin7 family comprised two paralogs (Syx7′a and Syx7′b) in the last metazoan common ancestor (LMCA; Fig. 4A)⁴⁹. Drosophila retained both genes: Avl/Syx7 (derived from Syx7′a) and Syx13/Syx20 (derived from Syx7′b). In vertebrates, Syx7′a underwent duplication, giving rise to STX7 and STX12/STX13, while Syx7′b (named TSNARE1 or STX20 in human) was disrupted by insertion of a Harbinger transposon⁷⁸, resulting in transformation of its SNARE function into a regulatory one⁷⁹.

Fig. 4 — A Schematic depicting the evolutionary relationships of Syntaxin7 family members in humans and Drosophila. LECA: Last Eukaryotic Common Ancestor, LMCA: Last Metazoan Common Ancestor. C, D Syx13-HA localizes to Rab7⁺ late endosomes (C) and Lamp1⁺ lysosomes (D) in nephrocytes. Only minimal but detectable colocalization is observed with Rab5⁺ early endosomes (B, yellow arrowheads). Blue: nuclei (DAPI). Rab5-GFP⁺ (B), Rab7⁺ (C) and Lamp1⁺ (D) structures are outlined by continuous orange lines in the black-and-white insets. Scale bars: 10 μm. E–U Rescue experiments using human STX7 and STX12 transgenes in Drosophila nephrocytes. E–M, O–T) Blue: nuclei (DAPI). Cell outlines are indicated by magenta lines. Scale bars: 10 μm. E–G Rab7⁺ endosomes appear normal in control (E), hSTX7-expressing (F), and hSTX12-expressing (G) cells. H–J Rab7⁺ endosomes are significantly smaller in Syx7 RNAi cells (H); this phenotype is not rescued by hSTX7 (I) but is partially rescued by hSTX12 (J). K–M Rab7⁺ endosomes are enlarged in Syx13 RNAi cells (K); this phenotype is not rescued by hSTX12 (M) but is partially rescued by hSTX7 (L). N Quantification of Rab7⁺ endosome size shown in (E–M). n = 100 endosomes randomly selected from 10 cells per genotype, derived from 5 independent animals (biological replicates). Experiments were independently repeated on two different days with similar results. Statistical significance was assessed using a Kruskal–Wallis test followed by Dunn’s multiple-comparisons test. Asterisks (*) indicate comparisons with control (Luciferase RNAi), orange hashes (#) indicate comparisons with Syx7 RNAi, and cyan hashes (#) indicate comparisons with Syx13 RNAi. ****P < 0.0001; ####P < 0.0001. Error bars represent SD. Exact P values are as follows. Syx7 dataset, compared with control: hSTX7-HA, P > 0.9999; hSTX12-HA, P > 0.9999; Syx7 RNAi, P < 0.0001; hSTX7-HA + Syx7 RNAi, P < 0.0001; hSTX12-HA + Syx7 RNAi, P > 0.9999. Compared with Syx7 RNAi: control, P < 0.0001; hSTX7-HA, P < 0.0001; hSTX12-HA, P < 0.0001; hSTX7-HA + Syx7 RNAi, P = 0.4201; hSTX12-HA + Syx7 RNAi, P < 0.0001. Syx13 dataset, compared with control: hSTX7-HA, P > 0.9999; hSTX12-HA, P > 0.9999; Syx13 RNAi, P < 0.0001; hSTX7-HA + Syx13 RNAi, P > 0.9999; hSTX12-HA + Syx13 RNAi, P < 0.0001. Compared with Syx13 RNAi: control, P < 0.0001; hSTX7-HA, P < 0.0001; hSTX12-HA, P < 0.0001; hSTX7-HA + Syx13 RNAi, P < 0.0001; hSTX12-HA + Syx13 RNAi, P > 0.9999. O–T Avidin uptake and transport to endosomes proceed normally in control (O), hSTX7 expressing (P), and hSTX12 expressing (Q) cells. In Syx7 RNAi cells (R), avidin uptake is nearly abolished; this defect is not rescued by hSTX7 (S) but is partially rescued by hSTX12 (T). Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. U Quantification of Avidin uptake based on tracer coverage of the cell (%) shown in (O–T). n = 15 cells per genotype, derived from 6 independent animals (biological replicates). Experiments were independently repeated on two different days with similar results. Statistical significance was assessed using ordinary one-way ANOVA followed by Tukey’s multiple-comparisons test. Asterisks (*) indicate comparisons with control (Luciferase RNAi), and orange hashes (#) indicate comparisons with Syx7 RNAi. ****P < 0.0001; ####P < 0.0001. Error bars represent SD. Exact P values are as follows. Compared with control: hSTX7-HA, P > 0.9999; hSTX12-HA, P > 0.9999; Syx7 RNAi, P < 0.0001; hSTX7-HA + Syx7 RNAi, P < 0.0001; hSTX12-HA + Syx7 RNAi, P < 0.0001. Compared with Syx7 RNAi: control, P < 0.0001; hSTX7-HA, P < 0.0001; hSTX12-HA, P < 0.0001; hSTX7-HA + Syx7 RNAi, P = 0.8644; hSTX12-HA + Syx7 RNAi, P < 0.0001. V Schematic of the updated nomenclature of Drosophila Syntaxin 7 family members: Syx7 is renamed Syx12L and Syx13 is renamed Syx7L.

Because of the similarity in nomenclature, one might infer that Drosophila Syx7 corresponds to human STX7; similarly, at least one study inferred that Drosophila Syx13 corresponds to STX12/13⁵⁷. However, phylogenetic analysis indicates a different relationship: Drosophila Syx13 represents the ancestral Syx7′b lineage whose SNARE function was disrupted in mammals^78,79, whereas Drosophila Syx7 corresponds to the Syx7′a lineage that duplicated in vertebrates^49,50. In support of this assignment, Syx7 localizes predominantly to Rab5-positive early endosomes in Drosophila, whereas Syx13–HA is enriched on Rab7-positive late endosomes and Lamp1-positive lysosomes, with only minimal overlap with Rab5, indicating a primarily late endosomal function (Fig. 4B–D). To directly test functional equivalence, we expressed human STX7–HA or STX12–HA in Syx7- or Syx13-depleted nephrocytes (Fig. 4E–N). Expression of either human SNARE in control cells had no significant effect on late endosome size (Fig. 4E–G). In Syx7-silenced cells, human STX12–HA (Fig. 4J)—but not STX7–HA (Fig. 4I)— rescued the endosomal phenotype (Fig. 4H, N). Conversely, in Syx13-silenced cells (Fig. 4K), STX7–HA (Fig. 4L)—but not STX12–HA (Fig. 4M)—restored normal late endosome size (Fig. 4N). Functional equivalence was further supported by uptake assays: the near-complete loss of FITC–Avidin internalization caused by Syx7 silencing (Fig. 4O, R, U) was rescued by STX12–HA (Fig. 4Q, T, U) but not by STX7–HA (Fig. 4P, S, U).

To determine whether this functional conservation extends beyond the Syntaxin family, we next tested whether the remaining components of the canonical early endosomal SNARE complex could likewise be functionally replaced by their human orthologues. Expression of human SNAP29 or human YKT6 efficiently rescued the reduced Rab7-positive endosome size caused by depletion of the corresponding Drosophila proteins (Supplementary Fig. 4S–W), indicating that the Qbc- and R-SNARE components of the canonical early endosomal fusion machinery are also functionally conserved between flies and mammals.

Together, our evolutionary analysis, cross-species rescue experiments and the existing literature indicate that functional orthology between Drosophila and mammalian Syntaxin7 family members has shifted through paralogous compensation. In Drosophila, Syx7 retains early endosomal function, while Syx13 acts at late endosomes; in mammals, these roles are partitioned between STX12 and STX7, respectively, most likely because the ancestral Syx7′b lineage mostly lost its SNARE function following Harbinger transposon insertion^78,79. Since the current nomenclature of Drosophila Syx7 and Syx13 reflects neither their evolutionary history nor their functional equivalence to mammalian proteins, we propose a revised naming scheme (Fig. 4V). We suggest renaming Drosophila Syx7 as Syntaxin12-like (Syx12L) and Syx13 as Syntaxin7-like (Syx7L), where “-like” denotes functional analogy while allowing for species-specific divergence. Hereafter, we adopt this revised nomenclature throughout the manuscript.

Disruption of early endosome fusion enhances recycling in a Ykt6- and Exocyst-dependent manner

Fluorescence microscopy revealed clear but previously unexplained phenotypic differences among knockdowns of the three SNAREs. To resolve these differences at early stages of endosome maturation, we examined the nephrocyte periphery by electron microscopy using tannic acid impregnation to visualize lacunae and clathrin-coated pits (Fig. 5A–D). Control nephrocytes (Fig. 5A) displayed well-developed lacunae with abundant clathrin-coated pits and large endosomes. In contrast, Syx12L silencing (Fig. 5B) caused a marked deepening of lacunae accompanied by expanded luminal spaces and a pronounced reduction in endosome size. Snap29 depletion (Fig. 5C) also increased lacunar depth, and endosomes were small. Notably, Snap29-deficient lacunae frequently contained intracellular cargo, consistent with our previous finding that Snap29 loss promotes autophagosome secretion in nephrocytes (Supplementary Fig. 5Y)⁶⁸. In sharp contrast, Ykt6 silencing (Fig. 5D) led to a drastic shallowing of lacunae, while clathrin-coated pits and medium-sized endosomes were preserved, which fits with the result of our uptake experiments.

Fig. 5 — A–D Transmission electron micrographs of the peripheral regions of tannic acid (TA)–impregnated nephrocytes. TA fills the lacunar system and appears as dark electron-dense material in the lacunar lumen (colored orange in false-color images). Scale bars: 1 μm. m, mitochondria; α, late endosomes (colored green). White arrowheads indicate TA-filled clathrin-coated pits; black arrowheads indicate clathrin-coated and uncoated early endocytic vesicles and small early endosomes. A) Lacunar depth is normal in control cells, with large late endosomes present. B, C Lacunar depth is markedly increased in Syx12L. B and Snap29 (C) RNAi cells. These cells lack normally sized endosomes and instead contain numerous small endosomes. Note that lacunae are dilated in Syx12L RNAi cells (B), whereas the lacunar lumen contains secreted autophagosomes in Snap29 RNAi cells (C; higher magnification in Supplementary Fig. 3Y). D Ykt6 RNAi cells lack deep lacunae, and endosomes are smaller than in controls but larger than in Syx12L (B) or Snap29 (C) RNAi cells. E Schematic illustrating the relationship between lacunar depth and endosomal recycling versus degradation. CUBAM: cubilin-amnionless complex. F–K Channel diffusion assay using fluorescent BSA shows increased lacunar channel depth in Syx12L (G) and Snap29 (H) RNAi cells compared with controls (F). In contrast, lacunar channels are shallower in Ykt6 RNAi cells (I), similar to Sec5 single (J) and Sec5/Snap29 double RNAi (K) cells. Note the intense internal BSA signal in Syx12L cells (G), corresponding to lacunar dilations seen in (B). Blue channel shows nuclei (DAPI); cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. L Quantification of lacunar channel depth from (F–K and Supplementary Fig. 4 A1–A6). The dotted line indicates the mean lacunar depth in control cells. n = 15 cells per genotype, derived from 4 independent animals (biological replicates). Experiments were independently repeated on two different days with similar results. Statistical significance was assessed using ordinary one-way ANOVA followed by Tukey’s multiple-comparisons test. Asterisks (*) indicate comparisons with control (Luciferase RNAi), and hashes (#) indicate comparisons with Snap29 RNAi. ****P < 0.0001; ####P < 0.0001; **P < 0.01. Error bars represent SD. Exact P values are as follows. Compared with control: P < 0.0001 for all genotypes except Vps8 RNAi, P = 0.005, and Syx7L RNAi, P > 0.9999. Compared with Snap29 RNAi: P < 0.0001 for all genotypes except Syx12L RNAi, P = 0.9243. The dotted line indicates the mean lacunar depth in control cells.

The ultrastructural differences in lacunar depth pointed to altered recycling dynamics. In nephrocytes, endocytosis occurs primarily at the lacunar membrane, and membrane homeostasis depends on the balance between endocytosis-driven membrane removal and recycling-mediated membrane return (Fig. 5E). Accordingly, enhanced recycling would be expected to deepen lacunae, whereas impaired recycling would decrease their depth. To quantify recycling activity and strengthen our EM data, we used a fluorescent channel diffusion assay in prefixed cells to label the lacunar lumen (Fig. 5F–L and Supplementary Fig. 5A). Consistent with the EM data, Syx12L (Fig. 5G) or Snap29 (Fig. 5H) silencing significantly increased average channel depth, albeit with distinct morphologies. In contrast, Ykt6 depletion (Fig. 5I) reduced channel depth, with channels absent from large portions of the cell surface.

Because early endosomal fusion is impaired upon Syx12L or Snap29 depletion, yet based on our uptake experiments endocytosis remains rapid, we hypothesized that fusion-defective endocytic vesicles are rerouted into the recycling pathway. Supporting this idea, silencing of the early endosomal tethering factors Rbsn5 (Fig. 5L, Supplementary Fig. 5A1) or Vps8 (Fig. 5L and Supplementary Fig. 5A2) similarly increased channel depth, whereas depletion of Vps11 (Fig. 5L and Supplementary Fig. 5A3) or Vps41 (Fig. 5L and Supplementary Fig. 5A4) subunits of late endosomal HOPS tethering complex resulted in shallower channels. In turn, silencing of late endosomal Syx7L (Fig. 5L and Supplementary Fig. 5A5) resulted in channel depth similar to control. Thus, enhanced recycling is specifically triggered by disruption of early—but not late—endosomal fusions. Ykt6 depletion (Fig. 5I, L), however, produced the opposite phenotype, suggesting an additional role in recycling. Consistent with this, Ykt6 silencing was epistatic to Snap29 depletion (Fig. 5L and Supplementary Fig. 5A6) with respect to channel depth, mirroring the behavior of the Sec5 subunit of the recycling tether Exocyst⁸⁰ (Fig. 5J–L). This suggests that Ykt6 plays an important role in membrane recycling.

To further assess the balance of endocytosis and recycling, we studied a physiologically critical cargo class of nephrocytes, the slit diaphragm components. In both Drosophila garland nephrocytes and mammalian podocytes, the filtration function depends on the correct assembly and plasma membrane localization of the slit diaphragm^81,82. In control nephrocytes, heat fixation followed by immunostaining for slit diaphragm proteins Pyd (ZO-1) and Sns (Nephrin) revealed a characteristic fingerprint-like, homogeneous cortical pattern (Supplementary Fig. 5B), while medial sections showed predominant localization at the plasma membrane (Supplementary Fig. 5L). At the ultrastructural level, slit diaphragms appeared as electron-dense structures at the opening of lacunae (Supplementary Fig. 5C, M).

Syx12L depletion disrupted this organization (Supplementary Fig. 5D, J, K), resulting in reduced cortical colocalization of Pyd and Sns, decreased slit diaphragm density, and extensive surface regions lacking organized sieving membranes, confirmed by electron microscopy (Supplementary Fig. 5E). In medial sections (Supplementary Fig. 5N, T), slit diaphragm proteins were detected deeper within the cell, corresponding to multiple ectopic slit diaphragms observed by electron microscopy (Supplementary Fig. 5O). Snap29 depletion (Supplementary Fig. 5F, G, J, K) did not reduce overall cortical slit diaphragm density or component colocalization, but led to the formation of deeper, ectopic slit diaphragms (Supplementary Fig. 5P, Q, T), likely reflecting aberrant recycling and altered lacunar morphology. These ectopic structures observed both in Syx12L and Snap29 silencing provide a plausible explanation for the ‘clogged’ lacunae and the consequent block in FITC–Avidin uptake (Fig. 2). In contrast, Ykt6 depletion (Supplementary Fig. 5H–K) caused a pronounced reduction in cortical slit diaphragm density accompanied with the loss of the ‘fingerprint’ pattern and the marked decrease of the colocalization between Pyd and Sns, indicating defective slit diaphragm assembly and trafficking. Although most slit diaphragm proteins remained associated with the plasma membrane (Supplementary Fig. 5R, T), electron microscopy revealed only sparse slit diaphragms at the openings of abnormally shallow lacunae (Supplementary Fig. 5S).

Consistent with these findings, the recycling cargo Cubilin⁸³ showed reduced peripheral localization in all three SNARE knockdowns (Supplementary Fig. 5U–X). In Syx12L-depleted cells (Supplementary Fig. 5V), Cubilin accumulated in a pattern similar to the expanded lacunae, whereas in Snap29-depleted cells (Supplementary Fig. 5W) it retained a lacuna-associated distribution. In contrast, Cubilin predominantly localized to the intracellular region, most probably to endosomes upon Ykt6 depletion (Supplementary Fig. 5X), consistent with impaired recycling and explaining the reduced endocytic uptake observed under these conditions (Fig. 2).

Early endosome fusion bypass leads to aberrant endolysosomal ‘Swirl’ formation

To assess later stages of endolysosomal maturation, we analyzed deeper regions of tannic acid–impregnated nephrocytes (Fig. 6A–D) and compared them with Lamp1 immunostaining (Fig. 1H–K). Control cells (Fig. 6A) contained multiple large endosomes, endolysosomes and smaller, denser lysosomes. In contrast, Snap29-depleted cells (Fig. 6C) contained numerous small, electron-dense perinuclear lysosomes together with small endosomes distributed throughout the cytoplasm.. Ykt6 depletion (Fig. 6D) produced smaller than control but otherwise mature endolysosomes, indicating partial progression of maturation. In Syx12L-silenced cells (Fig. 6B and Supplementary Fig. 6A), small endosomes and lysosomes comparable in size to those observed upon Snap29 depletion were still present, but in addition, strikingly large abnormal structures emerged. These consisted of multiple concentric membrane layers, often enclosing cytosol or smaller vesicles (Fig. 6B) and lipid droplets (Supplementary Fig. 6A). Super-resolution structured illumination microscopy (SIM) of Rab7-positive swirls revealed a remarkably similar organization (Fig. 6E), strongly supporting the conclusion that the large Rab7- and Lamp1 positive structures identified by fluorescent microscopy (Fig. 1C, I) correspond to the membrane swirls observed by transmission electron microscopy. Similar structures were also observed upon silencing of other early endosomal proteins, such as Rab5 (Fig. 6F), Rbsn5 (Fig. 6G and Supplementary Fig. 6B), or Vps8 (Fig. 6H), as well as in Vps8 mutant nephrocytes (Supplementary Fig. 6C–E). These structures lacked hallmarks of intraluminal membrane reorganization common for multilamellar late endosomes and instead appeared to arise through vesicle elongation and coiling, based on observation of membrane ends (Supplementary Fig. 6C’) and enveloped cytoplasm. We therefore refer to them as endolysosomal swirls. Using a Rab2-pHluorin reporter, we found that swirls were Rab2-positive (Fig. 6I, J) but largely lacked the clathrin light chain - GFP marker, ER marker Calnexin, the trans-Golgi marker hGALNT2-YFP, the PI(3)P reporter FYVE-GFP, the PI(4,5)P₂ reporter PH-mCherry, and the autophagy markers Atg8a and p62 (Fig. 7Q and Supplementary Fig. 6F–N, respectively). Together, these observations suggest that the swirls represent aberrant late endolysosomal compartments rather than early endosomal, Golgi-derived, or autophagic structures. Likewise, a cytosolic GFP reporter did not accumulate within the swirls (Fig. 7Q, Supplementary Fig. 6M), arguing against nonspecific trapping of soluble cytoplasmic proteins. Notably, swirl-like structures were absent in Rab2-pHluorin expressing Snap29 (Fig. 6K) or Ykt6-depleted cells (Fig. 6L), suggesting the possibility that their formation may require these SNAREs.

The formation of multilayered endolysosomal swirls following impairment of early endosomal fusion prompted us to characterize these structures in greater detail (Fig. 7). To determine whether endocytosed cargo accumulates within the swirls, we first followed the fate of lumenal tracers. Despite prolonged ex vivo incubation (50 min), fluorescent dextran did not accumulate within Rab7-positive swirls (Fig. 7A, B, Q). Likewise, although lysosomes in control nephrocytes contained prominent electron-dense silver deposits following silver nitrate feeding, the swirls remained devoid of detectable silver (Fig. 7C, D), indicating that soluble endocytosed cargo does not efficiently reach or accumulate within these structures.

In contrast, membrane-associated markers behaved differently. Incubation with the plasma membrane dye FM1-43 (Fig. 7E, F, Q), together with expression of the plasma membrane-associated reporters mCD8-RFP, CAAX-mCherry, and human transferrin receptor-GFP (hTfR-GFP; Fig. 7G–N, Q), showed that endocytosed membranes readily accumulated within the swirls. Whereas both FM1-43, mCD8-RFP and CAAX-mCherry labeled these structures, hTfR-GFP was detected only in a subset of swirls (Fig. 7N, Q). Because transferrin receptor is predominantly recycled following endocytosis in mammalian cells, this observation suggests that membrane cargos destined for recycling may be removed before reaching, or alternatively from, the swirls. Consistent with this interpretation, the recycling endosome marker Rab11-YFP also localized to Rab7-positive swirls (Fig. 7O–Q). Together, these findings indicate that the swirls represent aberrant endolysosomal compartments that accumulate membrane-associated cargo while largely excluding soluble lumenal cargo, suggesting differential sorting of membrane and lumenal components following disruption of canonical early endosomal fusion.

Two additional SNARE complexes complement the Syx12L–Snap29–Ykt6 pathway in early endosome maturation

Because partial endosome maturation still occurs in Syx12L- or Ykt6-depleted nephrocytes, we performed epistasis analyses to define the relationships among the three identified SNAREs (Fig. 8). Simultaneous silencing of Syx12L and Ykt6 (Fig. 8D) abolished the formation of Rab7-positive endolysosomal swirls (Fig. 8A, G) and reduced the size of both Rab5- and Rab7-positive endosomes compared with Ykt6 single RNAi (Fig. 8B, H and Supplementary Fig. 5H, I). This supports the existence of a common Syx12L–Snap29–Ykt6 pathway, while also indicating that Syx12L and Ykt6 can independently promote endosomal maturation toward distinct endpoints. Consistently, combined silencing of Snap29 with either Syx12L (Fig. 8E) or Ykt6 (Fig. 8F) reduced Rab5- and Rab7-positive endosome size relative to the respective single knockdowns (Fig. 8G, H and Supplementary Fig. 5H, I). Snap29 RNAi also prevented swirl formation in Syx12L-depleted cells (Fig. 8E, G). These results place Snap29 as a central component shared by all early endosomal maturation routes in Drosophila.

Fig. 8 — A–F Double RNAi of Syx12L with Ykt6 (D) further reduces the size of small endosomes compared with Syx12L (A) or Ykt6 (B) single RNAi, similar to Syx12L (E) or Ykt6 (F) double RNAi in a Snap29 RNAi background (C). Notably, Snap29 and Ykt6 RNAi eliminate the membrane swirls characteristic of Syx12L RNAi cells (A, D, F), consistent with Syx12L- and Ykt6-containing SNARE complexes acting in parallel. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. G Quantification of Rab7⁺ endosome diameter shown in (A, D, F, I and P). n = 100 endosomes randomly selected from 10 cells per genotype, derived from 5 independent animals (biological replicates). Experiments were independently repeated on two different days with similar results. Statistical significance was assessed using a Kruskal–Wallis test followed by Dunn’s multiple-comparisons test, comparing each double-RNAi dataset with Syx12L RNAi. ****P < 0.0001; ***P < 0.001; **P < 0.01. Error bars represent SD. Exact P values are as follows: Syx12L RNAi + Snap29 RNAi, P < 0.0001; Syx12L RNAi + Ykt6 RNAi, P = 0.0001; Syx12L RNAi + Syx7L RNAi, P = 0.0049; Syx12L RNAi + Vamp7 RNAi, P > 0.9999. H Quantification of Rab7⁺ endosome diameter shown in (B, E, F, J and Q). n = 100 endosomes randomly selected from 10 cells per genotype, derived from 5 independent animals (biological replicates). Experiments were independently repeated on two different days with similar results. Statistical significance was assessed using a Kruskal–Wallis test followed by Dunn’s multiple-comparisons test, comparing each double-RNAi dataset with Ykt6 RNAi. ****P < 0.0001; *P < 0.05. Error bars represent SD. Exact P values are as follows: P < 0.0001 for all double-RNAi genotypes except Ykt6 RNAi + Syx7L RNAi, P = 0.039. I, J Double RNAi of Ykt6 with Vamp7 results in smaller endosomes compared with either Vamp7 (I) or Ykt6 (B) single RNAi, consistent with parallel routes of action. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. K Co-immunoprecipitation from Drosophila S2R⁺ cells shows that Syx12L–3×FLAG precipitates Snap29–3×HA and Vamp7–3×HA. L–O) Vamp7–GFP predominantly localizes to Rab7⁺ late endosomes (L) and Lamp1⁺ lysosomes (M) in nephrocytes. Only minimal but detectable colocalization is observed with Rab5⁺ early endosomes and with Syx12L (L, M; yellow arrowheads). Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Syx12L⁺ (L), Rab5⁺ (M), Rab7⁺ (N) and Lamp1⁺ (O) structures are outlined by continuous orange lines in the black-and-white insets Scale bars: 10 μm. P, Q Double RNAi of the Qa-SNAREs Syx12L and Syx7L (Q) results in smaller endosomes compared with single RNAi (A and P) and eliminates membrane swirls, consistent with parallel routes of action. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. R Double RNAi of Syx12L with Vamp7 reduces swirl size but does not completely eliminate them, consistent with the action of a parallel Syx7L–Snap29–Ykt6 complex. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. S Double RNAi of Syx7L with Ykt6 results in slightly larger late endosomes compared with Ykt6 single RNAi (B), but smaller compared with Syx7L single RNAi (S), consistent with the presence of a parallel Syx12L–Snap29–Vamp7 complex. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. T Double RNAi of Syx7L with Vamp7 results in larger early endosomes compared with either Vamp7 or Syx7L single RNAi (R, S), consistent with the presence of a parallel Syx12L–Snap29–Ykt6 complex. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Scale bars: 10 μm. U Vamp7–GFP colocalizes with Syx7L–HA in nephrocytes. Blue: nuclei (DAPI). Cell outlines are indicated by dashed cyan lines. Syx7L-HA⁺ structures are outlined by continuous orange lines in the black-and-white insets Scale bars: 10 μm. V Schematic illustrating the composition of the four endosomal Syx12L- and Syx7L-based SNARE complexes in flies, based on the current work and previous studies^14,56.

In the absence of Ykt6, Syx12L and Snap29 must therefore act with an alternative R-SNARE. Our in vitro pull-down experiments indicated that Syx12L–Snap29 can associate with Vamp7 (Fig. 3J), which prompted us to test this interaction in vivo. Vamp7 silencing caused enlargement of Rab7-positive endosomes (Fig. 8I and Supplementary Fig. 5I), consistent with our initial RNAi screen (Fig. 1A and Supplementary Fig. 2G, I). Importantly, combined Vamp7 and Ykt6 silencing (Fig. 8J) reduced Rab5- and Rab7-positive endosome size compared with Ykt6 RNAi alone (Fig. 8B, H and Supplementary Fig. 5I). Co-immunoprecipitation from S2 cells confirmed that FLAG–Syx12L forms a complex with both Snap29 and Vamp7 (Fig. 8K). Although Vamp7–GFP localized predominantly to Rab7-positive late endosomes and lysosomal layers, overlapping vesicles with Syx12L and Rab5 were also detected (Fig. 8L–O). Together, these data support a model in which, in the absence of Ykt6, a Syx12L–Snap29–Vamp7 complex can drive early endosomal maturation toward smaller endolysosomes.

Conversely, when Syx12L is absent, Snap29 and Ykt6 promote endosomes to mature into endolysosomal swirls, implying the involvement of an alternative Qa-SNARE. Given the close evolutionary relationship between Syx7L and Syx12L, and the documented ability of Syx7L/STX7 to form complexes with Snap29 and Ykt6 during crinophagy in flies and autophagy in mammalian cells^45,56, we tested Syx7L as a candidate. Syx7L localized mainly to Rab7-positive late endosomes and lysosomal layers (Fig. 4C, D), with minimal but detectable overlap with Rab5-positive early endosomes (Fig. 4B), and its silencing resulted in enlarged late endosomes (Fig. 8P), consistent with our initial screen (Supplementary Fig. 1F). Notably, simultaneous depletion of Syx12L and Syx7L (Fig. 8Q) abolished swirl formation. This indicates that, in the absence of Syx12L, a Syx7L–Snap29–Ykt6 complex drives the formation of endolysosomal swirls.

Previous work has shown that Syx7L–Snap29–Vamp7 complexes can also form in Drosophila¹⁴ We found that Syx7L–HA colocalizes with Vamp7–GFP on endosome/lysosome-like vesicles in nephrocytes (Fig. 8U). To assess their role within the identified pathways, we analyzed Syx12L–Vamp7 and Ykt6–Syx7L double knockdowns. Syx12L–Vamp7 double RNAi (Fig. 8R) reduced swirl size but allowed the formation of larger Rab5- and Rab7-positive vesicles than Syx12L RNAi alone (Fig. 8G, R; Supplementary Fig. 7A), indicating that the Syx7L–Snap29–Vamp7 complex functions downstream of the Syx7L–Snap29–Ykt6 complex during swirl formation. In contrast, Ykt6–Syx7L double RNAi (Fig. 8S) increased Rab5-positive endosome size compared with Ykt6 single RNAi (Fig. 8B, H and Supplementary Fig. 7B), suggesting a maturation defect. This implies that the Syx7L–Snap29–Vamp7 complex functions downstream of the Syx12L–Snap29–Vamp7 complex. Consistently, Syx7L–Vamp7 double RNAi cells (Fig. 8T) contained significantly larger early endosomes than either single RNAi condition (Fig. 8B, I and Supplementary Fig. 7C, D), whereas late endosomes showed a similar trend that did not reach statistical significance. Together, these data indicate that alternative Syx7L/Vamp7-containing complexes primarily act on early endosomes in parallel to the Syx12L–Snap29–Ykt6 pathway.

Collectively, our epistasis, localization, and interaction analyses identify three partially redundant yet mechanistically distinct SNARE-driven pathways operating during early endosome maturation, as well as a late endosomal pathway (Fig. 8V). The Syx12L–Snap29–Ykt6 pathway supports the earliest stages of maturation. In Ykt6-deficient cells, a Syx12L–Snap29–Vamp7 complex mediates limited progression toward small endolysosomes. Conversely, in Syx12L-deficient cells, a Syx7L–Snap29–Ykt6 complex drives maturation toward endolysosomal swirls. All three pathways converge downstream on a Syx7L–Snap29–Vamp7 complex to support further endosomal maturation.

Distinct Rab5 and Rab7 requirements define the Ykt6- and Syx12L-independent maturation pathways

The existence of Syx12L- and Ykt6-independent pathways prompted us to ask whether these routes differ not only in their SNARE composition but also in their Rab GTPase requirements. Because Rab5 and Rab7 define key stages of endosomal maturation, we analyzed how perturbing each pathway affected their functional involvement.

To assess Rab5 involvement, we first depleted Rab5 and observed that swirl-like structures still formed (Fig. 6E), indicating that their formation is Rab5-independent. We next expressed constitutively active Rab5 (Rab5-CA), which remains GTP-bound and enhances Rab5-dependent fusion events (Fig. 9A–J). In control nephrocytes, Rab5-CA expression (Fig. 9F, U) increased the size of Rab5- and Rab7-positive endosomes, including Rab5/Rab7 double-positive structures. However, in Snap29-depleted cells (Fig. 9G, V), Rab5-CA failed to enlarge early endosomes, indicating that all Rab5-dependent fusion pathways require Snap29. In Syx12L-depleted cells (Fig. 9H, W), Rab5-CA did not increase swirl size, although Rab5-CA remained associated with swirl membranes, further supporting their endolysosomal identity and indicating that Syx12L-independent pathways operate largely independently of Rab5. In contrast, Rab5-CA markedly increased the size of Rab5- and Rab7-positive endosomes in Ykt6-depleted cells (Fig. 9I, X), demonstrating that the Ykt6-independent pathway is Rab5-dependent.

We next assessed Rab7 involvement (Fig. 9K–O). In control cells, Rab7 depletion did not change early endosome size, as expected (Fig. 9K). However, in all three SNARE knockdowns in a Rab7 RNAi background (Fig. 9L–O), early endosomes were smaller than in Rab7 RNAi alone, indicating that both Syx12L- and Ykt6-dependent pathways can engage Rab7. Notably, early endosomes were further reduced in size in Ykt6-depleted cells when Rab7 was also silenced (Fig. 9D, N, X). Similarly, when constitutively active Rab5 (Rab5-CA) was expressed in combination with Rab7 silencing (Fig. 9P–T), enlarged Rab5-positive endosomes formed only in Ykt6-depleted cells (Fig. 9S), and swirls were not observed even when Syx12L was silenced (Fig. 9R). These results indicate that the Syx12L-independent pathway and swirl formation require Rab7, whereas the Ykt6-independent pathway can proceed, at least partially, without Rab7.

Together, these experiments demonstrate that the three early endosomal maturation pathways differ not only in their SNARE composition but also in their Rab GTPase requirements. This strongly argues that these pathways are not simple compensatory mechanisms but represent distinct, regulated routes of early endosome maturation, likely engaging different tethering factors and regulatory inputs.

Discussion

Endosomal maturation has traditionally been viewed as a largely linear process, driven by sequential Rab GTPase conversion and the ordered deployment of tethering factors and SNARE complexes. While this framework has been invaluable, accumulating evidence - particularly from mammalian systems - has suggested that multiple fusion machineries may operate at the same trafficking step. However, extensive redundancy in these systems has made it difficult to disentangle core mechanisms from compensatory effects. By exploiting the reduced SNARE repertoire of Drosophila nephrocytes^49–51, our study reveals that early endosome maturation is executed by multiple, parallel SNARE-dependent pathways, each with distinct molecular requirements, functional consequences, and morphological outputs. Our data indicate that early endosomal fusion in Drosophila does not rely on a single indispensable SNARE complex. Instead, at least three partially redundant SNARE assemblies promote maturation beyond the early endosomal stage (Fig. 9). These pathways converge downstream but differ markedly in their reliance on specific SNAREs, Rab GTPases, and recycling dynamics. Importantly, our data suggest that these pathways are not simply interchangeable backups; rather, they are all required for shaping endosomal architecture, recycling flux, and cargo handling.

The existence of parallel fusion routes explains why early endosomal maturation can proceed, albeit aberrantly, upon loss of individual fusion components, while simultaneously producing strikingly distinct phenotypes depending on which early-acting SNARE is disrupted. These findings support a model that differs from a strictly hierarchical SNARE cascade and instead support a model in which SNARE composition itself is likely to be an important determinant of endosomal fate.

Among the SNAREs analyzed, Snap29 occupies a unique position. Genetic, morphological, and functional analyses demonstrate that all identified early endosomal fusion routes depend on Snap29. Loss of Snap29 abolishes compensatory fusion events, disrupts maturation at multiple stages, and produces phenotypes that cannot be bypassed through activation of alternative pathways. These findings support a model in which Snap29 occupies a central position in early endosome maturation and its progression toward the late endosome–lysosome system in Drosophila.

This central role is consistent with extensive prior evidence implicating Snap29 across lysosome-related pathways. Snap29 is required for autophagosome–lysosome fusion through interactions with Syx17 and Vamp7, and for secretory granule maturation and crinophagy via complexes containing Syx7L (previously termed Syx13) together with Vamp7 or Ykt6^{14,16,45,54,56,59,60,62,75,84}. In addition, Snap29 mutations cause defects in eye pigment granule formation, a lysosome-related organelle in flies^54,85. Consistent with this view, Snap29-deficient cells contain lysosomes that remain small and fail to undergo homotypic or heterotypic fusion.

This system-wide role is consistent with Snap29’s identity as a TM-domain–lacking Qbc-SNARE, which enables its availability to participate in multiple complexes because it is soluble and not membrane-bound. Beyond its biochemical versatility, our data indicate that Snap29 acts as a central integrator of endolysosomal fusion, coordinating parallel pathways rather than serving a single step. This framework offers a mechanistic explanation for the severe endolysosomal defects observed upon SNAP29 loss in humans, such as in CEDNIK syndrome, where Snap29 dysfunction cannot be compensated by alternative SNARE assemblies^86,87.

A notable outcome of our analysis is the clarification of the roles of Syntaxin7 family members in Drosophila. In mammalian cells, STX7 is generally regarded as a late endosomal/lysosomal Qa-SNARE^35–38,45, whereas STX12 has been described to function at early endosomes³⁴. In contrast, in Drosophila, Syx12L (previously termed Syx7) was originally assigned an early endosomal role^52,53, while Syx7L (previously termed Syx13) has been implicated in a specialized late endolysosomal fusion pathway - secretory granule–lysosome fusion (crinophagy) - suggesting a primary function at later stages of the pathway^14,56.

Our functional, genetic, and localization data demonstrate that Drosophila Syx12L (formerly Syx7) operates predominantly at early endosomes, where it mediates fusion events preceding late endosomal maturation. In contrast, late endosomal fusion is mediated by Syx7L-containing complexes (formerly Syx13), supporting a reassignment of functional orthology between Drosophila and mammalian Syntaxin family members. Although the apparent mismatch between mammalian and Drosophila Syntaxin7 family assignments has persisted in the literature⁵⁷, it has not been explicitly addressed. Based on our findings and previous evolutionary analyses^49,50, we propose that Syntaxin7 family proteins in vertebrates underwent paralogous compensation following gene duplication, likely associated with functional inactivation of the ancestral Syx7′b lineage due to insertion of a Harbinger transposon⁷⁸. In contrast, Drosophila Syx12L and Syx7L retained their ancestral functions, corresponding functionally to human STX12 and STX7, respectively. The ability of the human orthologues STX12, STX7, SNAP29, and YKT6 to rescue the corresponding Drosophila phenotypes further highlights the remarkable functional conservation of the canonical early endosomal SNARE machinery despite evolutionary diversification of the Syntaxin7 family.

One limitation of the present study is that all functional analyses were performed in Drosophila nephrocytes, a highly specialized endocytic cell type characterized by exceptionally high endocytic flux and the presence of lacunar channels. Consequently, the relative contribution of the individual SNARE pathways identified here may differ in other cell types that internalize cargo through distinct mechanisms, such as epithelial cells or professional phagocytes. Nevertheless, the core SNARE proteins examined in this study are broadly expressed and evolutionarily conserved, suggesting that the existence of multiple partially overlapping maturation pathways may represent a general organizational principle of the endolysosomal system, even if their relative importance varies between tissues.

One of the most striking insights from this study is the tight coupling between early endosomal fusion and endosomal recycling. Disruption of early endosome fusion–either by SNARE or tethering factor depletion–does not simply stall maturation but instead reroutes membrane traffic toward recycling pathways. This is manifested in nephrocytes by elongated lacunar channels.

These observations suggest that early endosome fusion acts as a rate-limiting checkpoint that determines whether internalized membrane proceeds toward degradation or is rapidly returned to the plasma membrane. When fusion is impaired, our data suggest that endocytic vesicles are preferentially shunted into recycling routes, preserving membrane flux but altering compartment morphology and cargo distribution. Based on our nephrocyte data and independent studies in wing cells⁶⁵, Ykt6 is required for endosomal recycling. These findings suggest that Ykt6 plays a particularly prominent role in endocytic processes, acting not only in endosomal fusion but also in restraining excessive recycling, thereby maintaining balance between degradative and recycling pathways.

Our epistasis and Rab5 activation experiments further reveal that the identified SNARE pathways differ fundamentally in their Rab requirements. While the Ykt6-independent pathway (Syx12L–Snap29–Vamp7 complex) relies on Rab5 activity, the Syx12L-independent route (Ykt6–Snap29–Syx7L) can bypass Rab5 and proceed in its absence. Rab7 dependence is likewise pathway-specific. Certain maturation routes retain fusion competence even when Rab7 function is compromised. These findings prompt a refinement of simplified models of Rab5-to-Rab7 conversion as an obligatory step in endosome maturation. Instead, they support a model in which SNARE identity can partially uncouple fusion from canonical Rab logic, allowing maturation to proceed through alternative routes (Fig. 10).

Fig. 10 — A In control cells, the Syx12L–SNAP29–Ykt6 complex mediates the earliest Rab5-dependent fusions between early endocytic vesicles and early endosomes. Subsequent maturation involves Rab5-dependent as well asRab5-independent fusion events and incudes both Rab7-independent and Rab7-dependent steps driven by Syx12L–SNAP29–Vamp7 and Syx7L–SNAP29–Ykt6 complexes, respectively. Fusions of late endosomes or lysosomes are executed primarily by the Syx7L–SNAP29–Vamp7 complex. Recycling endosomes are fused with the plasma membrane (PM) by a yet unidentified Ykt6-containing complex. Balanced recycling versus degradation maintains slit diaphragm (SD) integrity and lacunar depth. B Simplified schematic of the SNARE network highlighting SNAP29 as a central shared component. C In SNAP29 knockdown cells, none of the endosomal/lysosomal SNARE complexes are functional. As a result, endosomes and lysosomes remain small and unfused. A large fraction of endosomes is redirected into the recycling pathway, where a yet unidentified Ykt6-containing SNARE complex promotes their fusion with the PM, leading to lacunar deepening and clogging of SDs. D In Syx12L knockdown cells, only Syx7L-containing complexes remain functional. Early endosomes fail to undergo homotypic fusion and their fusion with more mature compartments is repressed. Maturation is not completely blocked: Syx7L–SNAP29–Ykt6 drives their conversion into small endolysosome-like structures, and Syx7L–SNAP29–Vamp7 mediates their homotypic fusion into endolysosomal “swirls.” In parallel, many unfused early endosomes are redirected into the recycling pathway, where Ykt6-dependent fusion with the PM deepens lacunae and disrupts SD architecture. E) In Ykt6 knockdown cells, homotypic early endosome fusion is lost due to inactivation of the Syx12L–SNAP29–Ykt6 complex, and fusion of early endosomes with more mature compartments is also impaired. Maturation is partially maintained through Vamp7-containing complexes, which generate smaller endolysosomes. Because Ykt6 is also required for recycling, its absence diverts most endosomal cargo toward degradation, leading to progressive loss of lacunae due to continuous uptake of their components.

The formation of multilayered endolysosomal swirls is a distinctive morphological consequence of impaired early endosomal fusion. Our ultrastructural, genetic, and marker analyses are consistent with these structures arising through elongation and coiling of endolysosomal membranes, rather than through intraluminal vesicle accumulation likely driven by unbalanced fusion events and membrane remodeling. The accumulation of multiple plasma membrane markers, including FM1-43, mCD8-RFP, and CAAX-mCherry within these structures, their dependence on specific SNARE combinations, and their absence when both Syx12L and Syx7L are depleted together suggest that they arise through a defined alternative maturation route rather than representing a purely nonspecific consequence of membrane trafficking defects. Consistent with this interpretation, the swirls preferentially accumulated endocytosed membrane-associated cargo and the recycling endosome marker Rab11, whereas soluble lumenal cargo failed to accumulate, indicating that these aberrant compartments retain selective membrane trafficking properties rather than representing passive membrane aggregates.

Our data further indicate that lysosome formation is not strictly contingent on canonical Rab5-dependent early endosomal fusion. Aberrant lysosomal “swirls” arise not only upon loss of the Syx12L-Snap29-Ykt6 pathway, but also when core early endosomal tethers (miniCORVET and Rbsn-5) or Rab5 itself are depleted. Thus, even in the absence of Rab5-driven homotypic early endosome fusion, membrane flow toward lysosome-like compartments still occurs. These findings suggest a refinement of the prevailing view that Rab5 is absolutely required for all steps of endosomal maturation. In mouse hepatocytes, Zeiger and colleagues showed that Rab5 depletion eliminates most of the endolysosomal pathway and that Rab5 is essential for establishing early endosome identity⁸⁸. Consistent with these observations, our data show that Rab5-dependent fusion is mandatory for generating fully functional degradative lysosomes, but not for all membrane progression into the late endolysosomal pathway. Instead, Rab5-independent SNARE routes appear capable of promoting membrane progression toward lysosomal compartments, albeit in a dysregulated and inefficient manner. This framework may explain why Rab5 mutant Drosophila fat cells form aberrant autolysosome-like structures during starvation⁸⁹ and clarifies how lysosomal identity can be partially acquired even in the absence of proper early endosomal maturation.

Together, our findings support a revised model of early endosome maturation in which multiple SNARE-dependent pathways operate in parallel, non-interchangeable ways, each capable of driving fusion to a limited extent and each associated with distinct Rab usage, recycling behavior, and morphological outcomes. Rather than serving as simple backups, these pathways act together, providing the endosomal system with flexibility, robustness, and regulatory capacity. The extensive functional conservation observed between Drosophila and human SNARE proteins suggests that this organizational principle is likely to extend beyond nephrocytes and may represent a general feature of metazoan endosomal maturation. Such an organization allows cells to adapt endosomal trafficking to physiological demands, maintain membrane homeostasis, and buffer against perturbations.

A longstanding question in membrane trafficking is whether SNAREs themselves act as targeting signals that confer vesicle identity^76,90–92. Recent biochemical and structural studies have challenged this view, suggesting that beyond the QabcR rule of complex formation⁹², SNARE pairing is often permissive, and that the surrounding molecular context plays a dominant role in specifying productive fusion events^{76,91,93–95}. Given that Qa, Qb, Qc, and R SNARE families have undergone hundreds of millions of years of independent evolution^49,50,96, the conservation of such interaction promiscuity argues for its functional importance. Our study provides an in vivo example of this principle: early endosome maturation in Drosophila relies on partially overlapping SNARE complexes that differ by only a single Qa or R component, yet these drive distinct fusion outcomes within a robust network. Rather than acting as strict determinants of vesicle identity, SNAREs therefore appear to function as permissive fusion modules that expand the range of possible trafficking routes. In this view, vesicle identity is not encoded by a single SNARE, but emerges from the combinatorial assembly of multiple factors, allowing trafficking pathways to remain flexible while preserving overall system robustness.

In conclusion, our study demonstrates that early endosome maturation is not governed by a single linear SNARE cascade but by a network of partially overlapping fusion machineries (Fig. 10). By defining the composition, regulation, and consequences of these pathways, we provide a framework for understanding how endosomal fate decisions are encoded at the molecular level.

Methods

Molecular cloning

For pull-down experiments, DNA constructs encoding Drosophila SNAREs were generated by PCR using primers listed in Supplementary Data 1. PCR products were cloned into the BamHI–XhoI sites of the pETMBP vector, except for the Syx7 SNARE domain, which was cloned into the BamHI–NotI sites of pETARA⁹⁷. Inserts were assembled using the NEBuilder HiFi DNA Assembly Cloning Kit (New England Biolabs, Massachusetts, USA). DH5α competent E. coli cells (New England Biolabs, Massachusetts, USA) were transformed and plated on LB–ampicillin agar plates. After overnight growth at 37 °C, single colonies were used to inoculate 4 ml LB cultures, and plasmids were purified using the GeneJet Plasmid Miniprep Kit (Thermo Fisher Scientific, Massachusetts, USA).

For immunoprecipitation experiments, the coding region of Syx7 was amplified from Drosophila cDNA (primers in Supplementary Data 1). To generate N-terminally 3×FLAG-tagged Syntaxin7, the insert was cloned into the NotI–Acc65I sites of pUAST-attB-Flag-dTSC2 WT (Addgene #111807). For N-terminally 3×HA-tagged Syntaxin7, the insert was cloned into the Acc65I–XbaI sites of pUAST-3xHA-V5-RpL10Ab (Addgene #125223). All constructs were verified by Sanger sequencing (Microsynth AG, Switzerland). UAS-VAMP7-3xHA, UAS-SNAP29-3xHA, pMT-GAL4, and UAS-Ykt6-3xHA constructs were described previously^59,62.

Protein production and purification

BL21(DE3) E. coli cells were transformed with pulldown constructs and grown at 37 °C to OD600 = 0.4. Protein expression was induced with 500 µM IPTG and cultures were incubated overnight at 18 °C. Cells were harvested by centrifugation at 3315 g for 5 min (Beckman JLA-9.1000 rotor, Beckman Coulter Life Sciences, Indianapolis, IN, USA), resuspended in lysis buffer (50 mM Na2HPO4 pH 8.0, 300 mM NaCl, 20 mM imidazole, 0.1% Triton X-100, 2 mM β-mercaptoethanol, 10 mg/ml lysozyme), and lysed by freeze–thaw cycles followed by sonication. Lysates were clarified at 48 400 g for 30 min at 4 °C. Supernatants were incubated with Ni Sepharose Excel resin (Cytiva, Chicago, IL, USA) for 1 h at 4 °C with rotation. Bound proteins were washed sequentially and eluted in elution buffer (20 mM Tris-HCl pH 8.0, 200 mM NaCl, 400 mM imidazole, 10% glycerol, 2 mM β-mercaptoethanol). Eluates were concentrated if needed (Amicon Ultra-15, 10 kDa MWCO, Millipore, Massachusetts, USA), supplemented with 2 mM TCEP, aliquoted, and stored at −80 °C.

GST pull-down

GST pull-down assays were performed using Glutathione Sepharose High-Performance resin (Cytiva, Chicago, IL, USA). GST-Syx7-SNARE-6×His was bound to resin for 1 h at 4 °C. After washing, prey proteins were added and incubated for 30 min at 4 °C. Resin was washed three times and bound proteins were eluted in Laemmli buffer, boiled, and analyzed by SDS-PAGE followed by Coomassie Brilliant Blue staining.

S2R⁺ cell culture, transfection and immunoprecipitation

S2R⁺ cells (DGRC #150, RRID:CVCL Z831) were maintained at 26 °C in Insect XPress medium (Lonza) with 10% FBS and 1% Penicillin–Streptomycin. For IP experiments, 7.5 × 10⁶ cells were transfected using the calcium phosphate method. Cells were co-transfected with pMT-GAL4 and protein expression was induced with 500 µM CuSO₄ 24 h post-transfection. Cells were lysed in Triton X-100 buffer and cleared at 20 000 g. Inputs were reserved, and lysates were incubated with anti-HA agarose (MBL-561-8) or anti-FLAG M2 agarose (Millipore A2220) for 3 h at 4 °C. Beads were washed and eluted in Laemmli buffer. Proteins were separated by SDS-PAGE and transferred to PVDF membranes using the Bio-Rad Trans-Blot Turbo system.

Antibodies

Details of antibodies are listed in Supplementary Data 1. Anti-Cubilin, anti-Lamp1, anti-Sns and anti-Syx12L (previously Avl/Syx7) were kindly provided by Joaquim Culi⁹⁸, Andreas Jenny⁹⁹, Michael Krahn¹⁰⁰ and David Bilder⁵², respectively. Anti-Snap29 and anti-GFP were already in our hands⁶²; all other antibodies were obtained from commercial suppliers.

Fly work and treatments

Flies were maintained at 25 °C on standard minimal food. Wandering third-instar (L3) larvae of both sexes were used for all experiments, except for those involving the X-linked Ykt6-HA transgene, for which only female progeny were analyzed to ensure consistent expression levels. Sources and details of fly stocks are listed in Supplementary Data 1. UAS-GFP-VAMP7, UAS-pHluorinSE-Rab2, and UAS-Syx13-HA were kindly provided by Amy Kiger, Ole Kjaerulff, and Fen-Biao Gao, respectively^57,70,101. UAS-Ykt6.HA and vps8[1] were already in our hands^30,62, and all other lines were obtained from stock centers. Detailed genotypes of experimental animals are listed in Supplementary Data 2. FlyBase was used for gene and stock annotation¹⁰².

Silver nitrate uptake was performed as described previously³⁰. Briefly, adult flies were allowed to lay eggs on food containing 0.005% AgNO₃ (20 g yeast in 35 ml of 1.5% agar). Garland nephrocytes were dissected in PBS and fixed in 4% formaldehyde in PBS for 30 min at room temperature (RT). Images were acquired using a Plan-Neofluar 40×/0.75 NA objective on an AxioImager Z1 microscope with an AxioCam ICc camera (Zeiss). Before imaging, samples were exposed to UV light for 10–15 s using an HBO 100 lamp and a DAPI filter set to enhance contrast of intracellular silver inclusions³⁰.

Immunohistochemistry

Nephrocytes from wandering L3 larvae were dissected in PBS and fixed in 4% formaldehyde in PBS for 45 min at RT. Samples were washed three times for 10 min and permeabilized in PBS containing 0.1% Triton X-100 (PBTX) for 15 min³⁰.

For slit diaphragm visualization, samples were heat-fixed¹⁰⁰. Nephrocytes were dissected in ice-cold M3 medium, transferred to boiling 0.03% Triton X-100 in PBS, and incubated for 22 s in boiling water. Samples were then processed in 0.03% PBTX.

Fixed samples were blocked in 5% fetal bovine serum (FBS) in PBTX for 30 min at RT and incubated with primary antibodies overnight at 4 °C. After washing, samples were incubated in PBTX containing 4% NaCl for 15 min to reduce nonspecific binding, followed by secondary antibody incubation for 3 h at RT. Samples were mounted in DAPI-containing VECTASHIELD.

Channel diffusion assay

Nephrocytes were fixed in ice-cold 4% formaldehyde in PBS for 5 min to stop endocytosis. Samples were incubated with Alexa Fluor 647–BSA in M3 medium for 15 min at RT, followed by a second fixation for 45 min. Samples were washed and mounted in DAPI-containing VECTASHIELD (Vector)¹⁰³.

Uptake assay

Nephrocytes were dissected in ice-cold M3 medium and incubated with FITC-avidin (1:100) and, where indicated, fluorescent dextran (10 kDa; 1:20) for 5 min (early endosomes), 30 or 50 min (late endosomes/endolysosomes) at room temperature. For tracer clearance assays, samples were chased in tracer-free M3 medium for 10 min following the 30 min pulse before fixation. For plasma membrane uptake experiments dissected nephrocytes were incubated with the membrane dye FM1-43FX (10 µM in M3 medium) for 50 min. Samples were then fixed in 4% formaldehyde for 45 min, washed, and mounted Where indicated, fixed samples were subsequently processed for immunostaining as described above.

Fluorescence and SIM imaging

Images were acquired at RT using an AxioImager.M2 microscope with an ApoTome2 unit, a Plan-Apochromat 63×/1.40 oil objective, and an Orca Flash 4.0 LT sCMOS camera. Imaging settings were identical for all comparable experiments. Images were processed using Zeiss Efficient Navigation 2 and assembled in Adobe Photoshop CS4. Super-resolution structured illumination microscopy (SIM) images were acquired using a Multi-SIM system (Beijing NanoInsights-tech Co., Ltd., Beijing, China) installed on a Nikon Ti2-E microscope equipped with a CFI SR HP Apochromat TIRF 100× oil-immersion objective. Images were acquired using VSIM v2.3 software (Beijing NanoInsights-tech Co., Ltd.) and processed using Fiji/ImageJ.

Electron microscopy

Nephrocytes were fixed overnight at 4 °C in 3.2% paraformaldehyde, 1% glutaraldehyde, 1% sucrose, and 0.028% CaCl₂ in 0.1 M sodium cacodylate buffer (pH 7.4). Samples were post-fixed in 0.5% osmium tetroxide for 1 h and half-saturated uranyl acetate for 30 min, dehydrated in graded ethanol, and embedded in Durcupan. For the silver uptake experiments, larvae were fed silver nitrate-containing food as described above. Dissected nephrocytes were exposed to UV light for 10–15 s using a Vilber Lourmat UV transilluminator to enhance the contrast of intracellular silver deposits before fixation and were subsequently processed for TEM as described above. For tannic acid impregnation¹⁰⁴, samples were treated after osmification with 1% tannic acid in 0.1 M cacodylate buffer (pH 7.4) for 30 min at room temperature, and uranyl acetate staining was omitted. Ultrathin sections were stained with Reynolds’ lead citrate and examined on a JEM-1011 transmission electron microscope at 80 kV. Sections from the silver uptake experiments were examined without additional staining.

Quantification and statistical analysis, illustrations

Fluorescent structures were quantified from original, unmodified single optical sections using Fiji/ImageJ. Signal thresholds for each channel were set consistently by the same person across all samples. Cells were randomly selected for analysis, and only cells with nuclei present in the focal plane were included to ensure sampling of both perinuclear and peripheral regions, except for measurements of slit diaphragm surface density and colocalization. Each experiment was performed at least twice on different days.

Endosome size measurements were obtained for Rab5-GFP/YFP, Rab5-CA-YFP, anti-Rab7, and anti-Lamp1–positive compartments by measuring vesicle diameters using the Line tool in ImageJ. For each genotype, 3–6 animals were analyzed, and 10 endosomes were measured per cell from 10 cells, yielding n = 100 endosomes per condition.

Pairwise colocalization between reporter proteins and antibody stainings was quantified using the Threshold Overlap Method. Both channels were thresholded identically, and the area fraction of each cell covered by the fluorescent signal was measured. Overlapping signal was determined using the Image Calculator (AND function). Threshold overlap scores (TOS) were calculated according to Sheng et al⁷³. from 12 cells per genotype and condition (n = 12), using samples from 4–7 animals.

RNAi efficiency was assessed using antibodies against endogenous Syx12L and Snap29 or by quantifying the anti-HA immunofluorescence signal of ectopically expressed Ykt6-HA. The mean gray value of the fluorescence signal was measured for each nephrocyte and normalized to the mean gray value of an adjacent control tissue region (proventriculus) of identical size and shape to account for between-sample variability. For comparison between control and RNAi-expressing cells, the mean normalized signal of the control group was set to 100. Three to five animals were analyzed for each genotype, and fluorescence was quantified in 12 nephrocytes per genotype (n = 12).

Fluorescent tracer uptake (FITC–avidin and dextran) was quantified by measuring the area fraction of each cell covered by tracer signal using uniform threshold settings. Cell outlines were determined by enhancing autofluorescence. For each genotype and condition, 4–7 animals were analyzed and fluorescence was quantified in 13–15 cells (n = 13–15).

Lacunar depth (BSA–Alexa Fluor 647) was estimated by outlining the total cell area and the area defined by the lacuna endpoints. The difference in radii of the corresponding calculated circles from the area was used as an approximation of lacunar depth. Measurements were obtained from 15 cells per genotype from 4–6 animals (n = 15).

Surface density of slit diaphragm components (Pyd and Sns) was quantified from cortical sections of heat-fixed, antibody-stained nephrocytes. Using the Line tool in ImageJ, a line (6–15 μm, depending on cell size) was drawn perpendicular to slit diaphragm stripes. Peak numbers were extracted from Plot Profile outputs and normalized to a 10 μm line length. Measurements were obtained from 12 cells per genotype using samples from 4–7 animals (n = 12).

The maximum signal depth of the slit diaphragm components Pyd and Sns was measured in medial optical sections by drawing a line perpendicular to the cell surface and recording the maximum depth of the detected fluorescence signal. For each genotype, 12 nephrocytes from 4–7 animals were analyzed (n = 12).

To assess the enrichment of fluorescent reporters and immunostained proteins within the multilamellar swirls, we used a method previously described⁶⁸. Briefly, the fluorescence intensity ratio between the swirl and the surrounding cytoplasm was determined and compared with that measured in Syx12L (formerly Syx7) RNAi nephrocytes expressing cytosolic enhanced green fluorescent protein (eGFP). The boundaries of the swirls were defined using anti-Rab7 immunostaining, Rab7-YFP, or anti-Lamp1 immunostaining, depending on the experiment. The mean gray value of the unaltered fluorescence signal was measured within each swirl and in an adjacent cytoplasmic region of identical size and shape. The swirl-to-cytoplasm fluorescence intensity ratio was then calculated for each swirl. For each marker or reporter, three to five animals were analyzed, and 15 swirls were quantified (n = 15).

Statistical analyses were performed using GraphPad Prism. Normality was assessed using the D’Agostino–Pearson test. For non-normally distributed datasets, Mann-Whitney test (if two datasets were compared) or Kruskal–Wallis tests with Dunn’s multiple comparisons were used. For normally distributed datasets, Student’s T-test (if two datasets were compared) or one-way ANOVA with appropriate post hoc tests was applied. P value thresholds were defined as * or #, p < 0.05; ** or ##, p < 0.01; *** or ###, p < 0.001; **** or ####, p < 0.0001. Figure legends specify reference groups for all comparisons.

Dot plots display individual data points with mean ± standard deviation. Full statistical details (including n numbers) are provided in Supplementary Data 3.

Microscopy images were processed and figures assembled using Adobe Photoshop CS4. Graphs and plots were generated using GraphPad Prism, and schematic illustrations were created using Inkscape 1.4.4.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

Supplementary Information^{(2.7MB, pdf)}

41467_2026_77479_MOESM2_ESM.pdf^{(51.3KB, pdf)}

Description of Additional Supplementary Files

Supplementary Data 1^{(52.5KB, docx)}

Supplementary Data 2^{(32.3KB, docx)}

Supplementary Data 3^{(72.5KB, docx)}

Reporting Summary^{(2.2MB, pdf)}

Transparent Peer Review file^{(214KB, pdf)}

Source data

Source Data^{(27.6MB, xlsx)}

Acknowledgements

We thank S. Pálfia, M. Truszka, and I. Répássy for technical assistance and colleagues listed in Materials and Methods for providing reagents. We thank Prof. Dong Li, founder of Beijing NanoInsights-tech Co., Ltd., for kindly providing access to a demonstration Multi-SIM system used for acquiring the super-resolution SIM images presented in this study.

Author contributions

D.H.: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Validation, Visualization and Writing–original draft, Writing–review & editing. M.M.: Conceptualization, Investigation, Methodology, Validation, Visualization and Writing–original draft, Writing – review & editing. A.R.: Investigation, Methodology and Writing–original draft. I.B.: Investigation. D.B.: Investigation. A.N.: Investigation, Writing–review & editing. V.B.: Formal analysis and Investigation. Zs. S.-V.: Investigation and Methodology. G.J.: Funding acquisition, Methodology and Resources, Writing–review & editing. P.L.: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization and Writing–original draft, Writing–review & editing.

Peer review

Peer review information

Nature Communications thanks the anonymous, reviewers for their contribution to the peer review of this work. A peer review file is available.

Funding

This work was funded by the following organizations: Hungarian Academy of Sciences (Magyar Tudományos Akadémia): LP2022-13 to P.L., LP2023-6 to G.J.; National Research, Development, and Innovation Office of Hungary (Nemzeti Kutatási, Fejlesztési és Innovációs Hivatal): FK138851 to P.L. and K146634 to G.J.; Eötvös Loránd University Excellence Fund: EKA 2022/045-P101 to P.L.; DKÖP-23 Doctoral Excellence Program of the Ministry for Culture and Innovation from the source of the National Research, Development and Innovation Fund of Hungary): DKÖP-2023-ELTE-13 to D.H.. D.H. is the recipient of the Joseph Cours Scholarship (ELTE Eötvös Loránd University, Budapest, Hungary).

Data availability

The data generated in this study are provided within the paper, Supplementary Information, and Source Data file. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Dávid Hargitai, Márton Molnár.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-77479-z.

References

1.Sudhof, T. C. & Rothman, J. E. Membrane fusion: grappling with SNARE and SM proteins. Sci. (N. Y.)323, 474–477 10.1126/science.1161748 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Kummel, D. & Ungermann, C. Principles of membrane tethering and fusion in endosome and lysosome biogenesis. Curr. Opin. Cell Biol.29, 61–66, 10.1016/j.ceb.2014.04.007 (2014). [DOI] [PubMed] [Google Scholar]
3.Hong, W. & Lev, S. Tethering the assembly of SNARE complexes. Trends Cell Biol.24, 35–43 10.1016/j.tcb.2013.09.006 (2014). [DOI] [PubMed] [Google Scholar]
4.Wickner, W. & Schekman, R. Membrane fusion. Nat. Struct. Mol. Biol.15, 658–664 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Zick, M. & Wickner, W. T. A distinct tethering step is vital for vacuole membrane fusion. eLife3, e03251 10.7554/eLife.03251 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.McEwan, D. G. et al. PLEKHM1 regulates autophagosome-lysosome fusion through HOPS complex and LC3/GABARAP proteins. Mol. cell57, 39–54 10.1016/j.molcel.2014.11.006 (2015). [DOI] [PubMed] [Google Scholar]
7.Pankiv, S. et al. FYCO1 is a Rab7 effector that binds to LC3 and PI3P to mediate microtubule plus end-directed vesicle transport. J. Cell Biol.188, 253–269 10.1083/jcb.200907015 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.van der Kant, R. et al. Late endosomal transport and tethering are coupled processes controlled by RILP and the cholesterol sensor ORP1L. J. Cell Sci.126, 3462–3474 10.1242/jcs.129270 (2013). [DOI] [PubMed] [Google Scholar]
9.Balderhaar, H. J. & Ungermann, C. CORVET and HOPS tethering complexes–coordinators of endosome and lysosome fusion. J. Cell Sci.126, 1307–1316 (2013). [DOI] [PubMed] [Google Scholar]
10.Solinger, J. A. & Spang, A. Tethering complexes in the endocytic pathway: CORVET and HOPS. FEBS J.280, 2743–2757 (2013). [DOI] [PubMed] [Google Scholar]
11.Lorincz, P. et al. Rab2 promotes autophagic and endocytic lysosomal degradation. J. Cell Biol.216, 1937–1947 10.1083/jcb.201611027 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Fujita, N. et al. Genetic screen in Drosophila muscle identifies autophagy-mediated T-tubule remodeling and a Rab2 role in autophagy. eLife6, e23367 10.7554/eLife.23367 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Kajiho, H. et al. RAB2A controls MT1-MMP endocytic and E-cadherin polarized Golgi trafficking to promote invasive breast cancer programs. EMBO Rep.17, 1061–1080 10.15252/embr.201642032 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Csizmadia, T. et al. Molecular mechanisms of developmentally programmed crinophagy in Drosophila. J. Cell Biol.217, 361–374 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Wang, Z. et al. The Vici Syndrome Protein EPG5 Is a Rab7 Effector that Determines the Fusion Specificity of Autophagosomes with Late Endosomes/Lysosomes. Mol. cell63, 781–795 10.1016/j.molcel.2016.08.021 (2016). [DOI] [PubMed] [Google Scholar]
16.Jiang, P. et al. The HOPS complex mediates autophagosome-lysosome fusion through interaction with syntaxin 17. Mol. Biol. cell25, 1327–1337 10.1091/mbc.E13-08-0447 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Angers, C. G. & Merz, A. J. HOPS interacts with Apl5 at the vacuole membrane and is required for consumption of AP-3 transport vesicles. Mol. Biol. cell20, 4563–4574 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Pols, M. S., ten Brink, C., Gosavi, P., Oorschot, V. & Klumperman, J. The HOPS proteins hVps41 and hVps39 are required for homotypic and heterotypic late endosome fusion. Traffic (Cph., Den.)14, 219–232 10.1111/tra.12027 (2013). [DOI] [PubMed] [Google Scholar]
19.Balderhaar, H. J. et al. The CORVET complex promotes tethering and fusion of Rab5/Vps21-positive membranes. Proc. Natl. Acad. Sci. USA110, 3823–3828 10.1073/pnas.1221785110 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Plemel, R. L. et al. Subunit organization and Rab interactions of Vps-C protein complexes that control endolysosomal membrane traffic. Mol. Biol. cell22, 1353–1363 10.1091/mbc.E10-03-0260 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Brocker, C. et al. Molecular architecture of the multisubunit homotypic fusion and vacuole protein sorting (HOPS) tethering complex. Proc. Natl. Acad. Sci. USA109 1991–1996, 10.1073/pnas.1117797109 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Lürick, A., Kümmel, D. & Ungermann, C. Multisubunit tethers in membrane fusion. Curr. Biol.28, R417–R420 (2018). [DOI] [PubMed] [Google Scholar]
23.Seals, D. F., Eitzen, G., Margolis, N., Wickner, W. T. & Price, A. A Ypt/Rab effector complex containing the Sec1 homolog Vps33p is required for homotypic vacuole fusion. Proc. Natl. Acad. Sci. USA97, 9402–9407 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Wurmser, A. E., Sato, T. K. & Emr, S. D. New component of the vacuolar class C-Vps complex couples nucleotide exchange on the Ypt7 GTPase to SNARE-dependent docking and fusion. J. Cell Biol.151, 551–562 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Peplowska, K., Markgraf, D. F., Ostrowicz, C. W., Bange, G. & Ungermann, C. The CORVET tethering complex interacts with the yeast Rab5 homolog Vps21 and is involved in endo-lysosomal biogenesis. Dev. Cell12, 739–750 (2007). [DOI] [PubMed] [Google Scholar]
26.Rieder, S. E. & Emr, S. D. A novel RING finger protein complex essential for a late step in protein transport to the yeast vacuole. Mol. Biol. cell8, 2307–2327 (1997). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.van der Kant, R. et al. Characterization of the mammalian CORVET and HOPS complexes and their modular restructuring for endosome specificity. J. Biol. Chem 10.1074/jbc.M115.688440 (2015). [DOI] [PMC free article] [PubMed]
28.Perini, E. D., Schaefer, R., Stoter, M., Kalaidzidis, Y. & Zerial, M. Mammalian CORVET is required for fusion and conversion of distinct early endosome subpopulations. Traffic (Cph., Den.)15, 1366–1389 10.1111/tra.12232 (2014). [DOI] [PubMed] [Google Scholar]
29.Lachmann, J., Glaubke, E., Moore, P. S. & Ungermann, C. The Vps39-like TRAP1 is an effector of Rab5 and likely the missing Vps3 subunit of human CORVET. Cell. Logist.4 e970840, 10.4161/21592780.2014.970840 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Lorincz, P. et al. MiniCORVET is a Vps8-containing early endosomal tether in Drosophila. eLife5, e14226 10.7554/eLife.14226 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Lorincz, P. et al. Vps8 overexpression inhibits HOPS-dependent trafficking routes by outcompeting Vps41/Lt. eLife8, e45631 10.7554/eLife.45631 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Fasshauer, D., Sutton, R. B., Brunger, A. T. & Jahn, R. Conserved structural features of the synaptic fusion complex: SNARE proteins reclassified as Q-and R-SNAREs. Proc. Natl. Acad. Sci.95, 15781–15786 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Jahn, R. & Scheller, R. H. SNAREs-engines for membrane fusion. Nat. Rev. Mol. cell Biol.7, 631–643 10.1038/nrm2002 (2006). [DOI] [PubMed] [Google Scholar]
34.Brandhorst, D. et al. Homotypic fusion of early endosomes: SNAREs do not determine fusion specificity. Proc. Natl. Acad. Sci.103, 2701–2706 10.1073/pnas.0511138103 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Advani, R. J. et al. Seven novel mammalian SNARE proteins localize to distinct membrane compartments*. J. Biol. Chem.273, 10317–10324 10.1074/jbc.273.17.10317 (1998). [DOI] [PubMed] [Google Scholar]
36.Antonin, W., Holroyd, C., Tikkanen, R., Höning, S. & Jahn, R. The R-SNARE Endobrevin/VAMP-8 mediates homotypic fusion of early endosomes and late endosomes. Mol. Biol. cell11, 3289–3298 10.1091/mbc.11.10.3289 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Wong, S. H., Xu, Y., Zhang, T. & Hong, W. Syntaxin 7, a novel syntaxin member associated with the early endosomal compartment*. J. Biol. Chem.273, 375–380 10.1074/jbc.273.1.375 (1998). [DOI] [PubMed] [Google Scholar]
38.Pryor, P. R. et al. Combinatorial SNARE complexes with VAMP7 or VAMP8 define different late endocytic fusion events. EMBO Rep.5, 590–595 10.1038/sj.embor.7400150 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Advani, R. J. et al. Vamp-7 Mediates Vesicular Transport from Endosomes to Lysosomes. J. Cell Biol.146, 765–776 10.1083/jcb.146.4.765 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Dingjan, I. et al. Endosomal and Phagosomal SNAREs. Physiol. Rev.98, 1465–1492, 10.1152/physrev.00037.2017 (2018). [DOI] [PubMed] [Google Scholar]
41.Atlashkin, V. et al. Deletion of the SNARE vti1b in mice results in the loss of a single SNARE partner, syntaxin 8. Mol. Cell. Biol.23, 5198–5207 10.1128/mcb.23.15.5198-5207.2003 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Bollmann, C. et al. Primary neurons lacking the SNAREs vti1a and vti1b show altered neuronal development. Neural Dev.17, 12 10.1186/s13064-022-00168-2 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Kunwar, A. J. et al. Lack of the endosomal SNAREs vti1a and vti1b led to significant impairments in neuronal development. Proc. Natl. Acad. Sci.108, 2575–2580 10.1073/pnas.1013891108 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Music, A., et al The SNARE protein Vti1b is recruited to the sites of BCR activation but is redundant for antigen internalisation, processing and presentation. Front. Cell. Dev. Biol.10 10.3389/fcell.2022.987148 (2022). [DOI] [PMC free article] [PubMed]
45.Matsui, T. et al. Autophagosomal YKT6 is required for fusion with lysosomes independently of syntaxin 17. J. cell Biol.217, 2633–2645 10.1083/jcb.201712058 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Jian, F. et al. The STX17-SNAP47-VAMP7/VAMP8 complex is the default SNARE complex mediating autophagosome–lysosome fusion. Cell Res.34, 151–168 (2024). [DOI] [PMC free article] [PubMed]
47.Davis, L. J. et al. Organelle tethering, pore formation and SNARE compensation in the late endocytic pathway. J. Cell Sci.134 10.1242/jcs.255463 (2021). [DOI] [PMC free article] [PubMed]
48.Fujiwara, T. et al. Analysis of knock-out mice to determine the role of HPC-1/Syntaxin 1A in expressing synaptic plasticity. J. Neurosci.26, 5767 10.1523/JNEUROSCI.0289-06.2006 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Kloepper, T. H., Kienle, C. N. & Fasshauer, D. An elaborate classification of SNARE proteins sheds light on the conservation of the eukaryotic endomembrane system. Mol. Biol. cell18, 3463–3471 10.1091/mbc.e07-03-0193 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Kloepper, T. H., Kienle, C. N. & Fasshauer, D. SNAREing the basis of multicellularity: consequences of protein family expansion during evolution. Mol. Biol. Evol.25, 2055–2068 10.1093/molbev/msn151 (2008). [DOI] [PubMed] [Google Scholar]
51.Littleton, J. T. A genomic analysis of membrane trafficking and neurotransmitter release in Drosophila. J. Cell Biol.150, F77–F82 10.1083/jcb.150.2.F77 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Lu, H. & Bilder, D. Endocytic control of epithelial polarity and proliferation in Drosophila. Nat. cell Biol.7 1232–1239, 10.1038/ncb1324 (2005). [DOI] [PubMed] [Google Scholar]
53.Morrison, H. A. et al. Regulation of early endosomal entry by the Drosophila tumor suppressors Rabenosyn and Vps45. Mol. Biol. cell19, 4167–4176 10.1091/mbc.E08-07-0716 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Morelli, E. et al. Multiple functions of the SNARE protein Snap29 in autophagy, endocytic, and exocytic trafficking during epithelial formation in Drosophila. Autophagy0 10.4161/15548627.2014.981913 (2014). [DOI] [PMC free article] [PubMed]
55.Yamazaki, Y. et al. Goliath family E3 ligases regulate the recycling endosome pathway via VAMP3 ubiquitylation. EMBO J.32, 524–537 10.1038/emboj.2013.1 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Szenci, G., Glatz, G., Takáts, S. & Juhász, G. The Ykt6–Snap29–Syx13 SNARE complex promotes crinophagy via secretory granule fusion with Lamp1 carrier vesicles. Sci. Rep.14, 3200 10.1038/s41598-024-53607-x (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Lu, Y., Zhang, Z., Sun, D., Sweeney, S. T. & Gao, F. B. Syntaxin 13, a genetic modifier of mutant CHMP2B in frontotemporal dementia, is required for autophagosome maturation. Mol. cell52, 264–271 10.1016/j.molcel.2013.08.041 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Akbar, M. A., Ray, S. & Kramer, H. The SM protein Car/Vps33A regulates SNARE-mediated trafficking to lysosomes and lysosome-related organelles. Mol. Biol. cell20, 1705–1714 10.1091/mbc.E08-03-0282 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Takats, S. et al. Autophagosomal Syntaxin17-dependent lysosomal degradation maintains neuronal function in Drosophila. J. Cell Biol.201, 531–539 10.1083/jcb.201211160 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Takats, S. et al. Interaction of the HOPS complex with Syntaxin 17 mediates autophagosome clearance in Drosophila. Mol. Biol. cell25, 1338–1354 10.1091/mbc.E13-08-0449 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Lakatos, Z. et al. Sec20 is required for autophagic and endocytic degradation independent of Golgi-ER retrograde transport. Cells8 10.3390/cells8080768 (2019). [DOI] [PMC free article] [PubMed]
62.Takats, S. et al. Non-canonical role of the SNARE protein Ykt6 in autophagosome-lysosome fusion. PLoS Genet.14, e1007359 10.1371/journal.pgen.1007359 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Zhou, L. et al. Convergence of secretory, endosomal, and autophagic routes in trans-Golgi-associated lysosomes. J. Cell Biol.222 10.1083/jcb.202203045 (2023). [DOI] [PMC free article] [PubMed]
64.Csizmadia, T. et al. Developmental program-independent secretory granule degradation in larval salivary gland cells of Drosophila. Traffic (Cph., Den.)23, 568–586 10.1111/tra.12871 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Linnemannstöns, K. et al. Ykt6-dependent endosomal recycling is required for Wnt secretion in the Drosophila wing epithelium. Dev. (Camb., Engl.)147 dev185421 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Fujii, S. et al. Recycling endosomes attach to the trans-side of Golgi stacks in Drosophila and mammalian cells. J. Cell Sci.133 10.1242/jcs.236935 (2020). [DOI] [PubMed]
67.Ward, D. M., Pevsner, J., Scullion, M. A., Vaughn, M. & Kaplan, J. Syntaxin 7 and VAMP-7 are soluble N-ethylmaleimide-sensitive factor attachment protein receptors required for late endosome-lysosome and homotypic lysosome fusion in alveolar macrophages. Mol. Biol. cell11, 2327–2333 10.1091/mbc.11.7.2327 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Hargitai, D. et al. HOPS-dependent vesicle tethering lock inhibits endolysosomal fusions and autophagosome secretion upon the loss of Syntaxin17. Sci. Adv.11, eadu9605 10.1126/sciadv.adu9605 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Boda, A. et al. Drosophila Arl8 is a general positive regulator of lysosomal fusion events. Biochim. Biophys. Acta (BBA)-Mol. Cell Res.1866, 533–544 (2019). [DOI] [PubMed] [Google Scholar]
70.Lund, V. K., Madsen, K. L. & Kjaerulff, O. Drosophila Rab2 controls endosome-lysosome fusion and LAMP delivery to late endosomes. Autophagy14, 1520–1542 10.1080/15548627.2018.1458170 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Fu, Y. et al. Comprehensive functional analysis of Rab GTPases in Drosophila nephrocytes. Cell Tissue Res368, 615–627 10.1007/s00441-017-2575-2 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Maruzs, T. et al. Interaction of the sorting nexin 25 homologue Snazarus with Rab11 balances endocytic and secretory transport and maintains the ultrafiltration diaphragm in nephrocytes. Mol. Biol. cell34, ar87 10.1091/mbc.E22-09-0421 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Sheng, H., Stauffer, W. & Lim, H. N. Systematic and general method for quantifying localization in microscopy images. Biol. open5, 1882–1893 10.1242/bio.019893 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Lang, K. et al. Selective endocytosis controls slit diaphragm maintenance and dynamics in Drosophila nephrocytes. eLife11, e79037 10.7554/eLife.79037 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Itakura, E., Kishi-Itakura, C. & Mizushima, N. The hairpin-type tail-anchored SNARE syntaxin 17 targets to autophagosomes for fusion with endosomes/lysosomes. Cell151, 1256–1269 10.1016/j.cell.2012.11.001 (2012). [DOI] [PubMed] [Google Scholar]
76.Koike, S. & Jahn, R. SNAREs define targeting specificity of trafficking vesicles by combinatorial interaction with tethering factors. Nat. Commun.10, 1608 10.1038/s41467-019-09617-9 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Vaccari, T., Lu, H., Kanwar, R., Fortini, M. E. & Bilder, D. Endosomal entry regulates Notch receptor activation in Drosophila melanogaster. J. Cell Biol.180, 755–762 10.1083/jcb.200708127 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Smith, J. J., Sumiyama, K. & Amemiya, C. T. A living fossil in the genome of a living fossil: Harbinger transposons in the coelacanth genome. Mol. Biol. Evol.29, 985–993 10.1093/molbev/msr267 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Plooster, M. et al. Schizophrenia-linked protein tSNARE1 regulates endosomal trafficking in cortical neurons. J. Neurosci.: J. Soc. Neurosci.41, 9466–9481 10.1523/jneurosci.0556-21.2021 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
80.Wen, P., Zhang, F., Fu, Y., Zhu, J. Y. & Han, Z. Exocyst Genes Are Essential for Recycling Membrane Proteins and Maintaining Slit Diaphragm in Drosophila Nephrocytes. J. Am. Soc. Nephrology: JASN31, 1024–1034 10.1681/asn.2019060591 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Weavers, H. et al. The insect nephrocyte is a podocyte-like cell with a filtration slit diaphragm. Nature457, 322–326 10.1038/nature07526 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Zhuang, S. et al. Sns and Kirre, the Drosophila orthologs of Nephrin and Neph1, direct adhesion, fusion and formation of a slit diaphragm-like structure in insect nephrocytes. Dev. (Camb., Engl.)136, 2335–2344 10.1242/dev.031609 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Zhang, F., Zhao, Y., Chao, Y., Muir, K. & Han, Z. Cubilin and amnionless mediate protein reabsorption in Drosophila nephrocytes. J. Am. Soc. Nephrology: JASN24, 209–216 10.1681/asn.2012080795 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Zheng, D. et al. Human YKT6 forms priming complex with STX17 and SNAP29 to facilitate autophagosome-lysosome fusion. Cell Rep.43, 113760 (2024). [DOI] [PubMed]
85.Dean, D. M., Codd, L. E., Constanza, R. & Segel, X. M. purpleoid1, a classic Drosophila eye color mutation, is an allele of the t-SNARE-encoding gene SNAP29. microPubl. Biol.2025 10.17912/MICROPUB.BIOLOGY.001563 (2025). [DOI] [PMC free article] [PubMed]
86.Fuchs-Telem, D. et al. CEDNIK syndrome results from loss-of-function mutations in SNAP29. Br. J. Dermatol.164, 610–616, 10.1111/j.1365-2133.2010.10133.x (2011). [DOI] [PubMed] [Google Scholar]
87.Sprecher, E. et al. A mutation in SNAP29, coding for a SNARE protein involved in intracellular trafficking, causes a novel neurocutaneous syndrome characterized by cerebral dysgenesis, neuropathy, ichthyosis, and palmoplantar keratoderma. Am. J. Hum. Genet.77, 242–251 10.1086/432556 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Zeigerer, A. et al. Rab5 is necessary for the biogenesis of the endolysosomal system in vivo. Nature485, 465–470 10.1038/nature11133 (2012). [DOI] [PubMed] [Google Scholar]
89.Hegedus, K. et al. The Ccz1-Mon1-Rab7 module and Rab5 control distinct steps of autophagy. Mol. Biol. cell27, 3132–3142 10.1091/mbc.E16-03-0205 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
90.Koike, S. & Jahn, R. SNARE proteins: zip codes in vesicle targeting? Biochemical J.479, 273–288 10.1042/bcj20210719 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Koike, S. & Jahn, R. Rab GTPases and phosphoinositides fine-tune SNAREs dependent targeting specificity of intracellular vesicle traffic. Nat. Commun.15, 2508 10.1038/s41467-024-46678-x (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Yadav, D. et al. A look beyond the QR code of SNARE proteins. Protein Sci.: a Publ. Protein Soc.33, e5158. 10.1002/pro.5158 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
93.Schwartz, M. L. et al. Sec17 (alpha-SNAP) and an SM-tethering complex regulate the outcome of SNARE zippering in vitro and in vivo. eLife6 10.7554/eLife.27396 (2017). [DOI] [PMC free article] [PubMed]
94.Orr, A., Song, H., Rusin, S. F., Kettenbach, A. N. & Wickner, W. HOPS catalyzes the interdependent assembly of each vacuolar SNARE into a SNARE complex. Mol. Biol. cell28, 975–983 10.1091/mbc.E16-10-0743 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
95.Song, H., Orr, A. S., Lee, M., Harner, M. E. & Wickner, W. T. HOPS recognizes each SNARE, assembling ternary trans-complexes for rapid fusion upon engagement with the 4th SNARE. eLife9, e53559 10.7554/eLife.53559 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
96.Koumandou, V. L. et al. Molecular paleontology and complexity in the last eukaryotic common ancestor. Crit. Rev. Biochem. Mol. Biol.48, 373–396 10.3109/10409238.2013.821444 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
97.Glatz, G., Gogl, G., Alexa, A. & Remenyi, A. Structural mechanism for the specific assembly and activation of the extracellular signal regulated kinase 5 (ERK5) module. J. Biol. Chem.288, 8596–8609 10.1074/jbc.M113.452235 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
98.Atienza-Manuel, A., Castillo-Mancho, V., De Renzis, S., Culi, J. & Ruiz-Gómez, M. Endocytosis mediated by an atypical CUBAM complex modulates slit diaphragm dynamics in nephrocytes. Development (Cambridge, England)148 10.1242/dev.199894 (2021). [DOI] [PMC free article] [PubMed]
99.Chaudhry, N. et al. Lamp1 mediates lipid transport, but is dispensable for autophagy in Drosophila. Autophagy18, 2443–2458 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
100.Hochapfel, F. et al. Distinct functions of Crumbs regulating slit diaphragms and endocytosis in Drosophila nephrocytes. Cell. Mol. Life Sci.74, 4573–4586 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
101.Jean, S., Cox, S., Nassari, S. & Kiger, A. A. Starvation-induced MTMR13 and RAB21 activity regulates VAMP8 to promote autophagosome–lysosome fusion. EMBO Rep.16, 297–311 10.15252/embr.201439464 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
102.Öztürk-Çolak, A. et al. FlyBase: updates to the Drosophila genes and genomes database. Genetics227, iyad211 10.1093/genetics/iyad211 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
103.Milosavljevic, J. et al. Nephrotic syndrome gene TBC1D8B is required for endosomal maturation and nephrin endocytosis in drosophila. J. Am. Soc. Nephrol.33, 2174–2193 10.1681/asn.2022030275 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
104.Kosaka, T. & Ikeda, K. Reversible blockage of membrane retrieval and endocytosis in the garland cell of the temperature-sensitive mutant of Drosophila melanogaster, shibirets1. J. Cell Biol.97, 499–507 (1983). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information^{(2.7MB, pdf)}

41467_2026_77479_MOESM2_ESM.pdf^{(51.3KB, pdf)}

Description of Additional Supplementary Files

Supplementary Data 1^{(52.5KB, docx)}

Supplementary Data 2^{(32.3KB, docx)}

Supplementary Data 3^{(72.5KB, docx)}

Reporting Summary^{(2.2MB, pdf)}

Transparent Peer Review file^{(214KB, pdf)}

Source Data^{(27.6MB, xlsx)}

Data Availability Statement

The data generated in this study are provided within the paper, Supplementary Information, and Source Data file. Source data are provided with this paper.

[CR1] 1.Sudhof, T. C. & Rothman, J. E. Membrane fusion: grappling with SNARE and SM proteins. Sci. (N. Y.)323, 474–477 10.1126/science.1161748 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Kummel, D. & Ungermann, C. Principles of membrane tethering and fusion in endosome and lysosome biogenesis. Curr. Opin. Cell Biol.29, 61–66, 10.1016/j.ceb.2014.04.007 (2014). [DOI] [PubMed] [Google Scholar]

[CR3] 3.Hong, W. & Lev, S. Tethering the assembly of SNARE complexes. Trends Cell Biol.24, 35–43 10.1016/j.tcb.2013.09.006 (2014). [DOI] [PubMed] [Google Scholar]

[CR4] 4.Wickner, W. & Schekman, R. Membrane fusion. Nat. Struct. Mol. Biol.15, 658–664 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Zick, M. & Wickner, W. T. A distinct tethering step is vital for vacuole membrane fusion. eLife3, e03251 10.7554/eLife.03251 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR6] 6.McEwan, D. G. et al. PLEKHM1 regulates autophagosome-lysosome fusion through HOPS complex and LC3/GABARAP proteins. Mol. cell57, 39–54 10.1016/j.molcel.2014.11.006 (2015). [DOI] [PubMed] [Google Scholar]

[CR7] 7.Pankiv, S. et al. FYCO1 is a Rab7 effector that binds to LC3 and PI3P to mediate microtubule plus end-directed vesicle transport. J. Cell Biol.188, 253–269 10.1083/jcb.200907015 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.van der Kant, R. et al. Late endosomal transport and tethering are coupled processes controlled by RILP and the cholesterol sensor ORP1L. J. Cell Sci.126, 3462–3474 10.1242/jcs.129270 (2013). [DOI] [PubMed] [Google Scholar]

[CR9] 9.Balderhaar, H. J. & Ungermann, C. CORVET and HOPS tethering complexes–coordinators of endosome and lysosome fusion. J. Cell Sci.126, 1307–1316 (2013). [DOI] [PubMed] [Google Scholar]

[CR10] 10.Solinger, J. A. & Spang, A. Tethering complexes in the endocytic pathway: CORVET and HOPS. FEBS J.280, 2743–2757 (2013). [DOI] [PubMed] [Google Scholar]

[CR11] 11.Lorincz, P. et al. Rab2 promotes autophagic and endocytic lysosomal degradation. J. Cell Biol.216, 1937–1947 10.1083/jcb.201611027 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Fujita, N. et al. Genetic screen in Drosophila muscle identifies autophagy-mediated T-tubule remodeling and a Rab2 role in autophagy. eLife6, e23367 10.7554/eLife.23367 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Kajiho, H. et al. RAB2A controls MT1-MMP endocytic and E-cadherin polarized Golgi trafficking to promote invasive breast cancer programs. EMBO Rep.17, 1061–1080 10.15252/embr.201642032 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Csizmadia, T. et al. Molecular mechanisms of developmentally programmed crinophagy in Drosophila. J. Cell Biol.217, 361–374 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Wang, Z. et al. The Vici Syndrome Protein EPG5 Is a Rab7 Effector that Determines the Fusion Specificity of Autophagosomes with Late Endosomes/Lysosomes. Mol. cell63, 781–795 10.1016/j.molcel.2016.08.021 (2016). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Jiang, P. et al. The HOPS complex mediates autophagosome-lysosome fusion through interaction with syntaxin 17. Mol. Biol. cell25, 1327–1337 10.1091/mbc.E13-08-0447 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR17] 17.Angers, C. G. & Merz, A. J. HOPS interacts with Apl5 at the vacuole membrane and is required for consumption of AP-3 transport vesicles. Mol. Biol. cell20, 4563–4574 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Pols, M. S., ten Brink, C., Gosavi, P., Oorschot, V. & Klumperman, J. The HOPS proteins hVps41 and hVps39 are required for homotypic and heterotypic late endosome fusion. Traffic (Cph., Den.)14, 219–232 10.1111/tra.12027 (2013). [DOI] [PubMed] [Google Scholar]

[CR19] 19.Balderhaar, H. J. et al. The CORVET complex promotes tethering and fusion of Rab5/Vps21-positive membranes. Proc. Natl. Acad. Sci. USA110, 3823–3828 10.1073/pnas.1221785110 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Plemel, R. L. et al. Subunit organization and Rab interactions of Vps-C protein complexes that control endolysosomal membrane traffic. Mol. Biol. cell22, 1353–1363 10.1091/mbc.E10-03-0260 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Brocker, C. et al. Molecular architecture of the multisubunit homotypic fusion and vacuole protein sorting (HOPS) tethering complex. Proc. Natl. Acad. Sci. USA109 1991–1996, 10.1073/pnas.1117797109 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Lürick, A., Kümmel, D. & Ungermann, C. Multisubunit tethers in membrane fusion. Curr. Biol.28, R417–R420 (2018). [DOI] [PubMed] [Google Scholar]

[CR23] 23.Seals, D. F., Eitzen, G., Margolis, N., Wickner, W. T. & Price, A. A Ypt/Rab effector complex containing the Sec1 homolog Vps33p is required for homotypic vacuole fusion. Proc. Natl. Acad. Sci. USA97, 9402–9407 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Wurmser, A. E., Sato, T. K. & Emr, S. D. New component of the vacuolar class C-Vps complex couples nucleotide exchange on the Ypt7 GTPase to SNARE-dependent docking and fusion. J. Cell Biol.151, 551–562 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Peplowska, K., Markgraf, D. F., Ostrowicz, C. W., Bange, G. & Ungermann, C. The CORVET tethering complex interacts with the yeast Rab5 homolog Vps21 and is involved in endo-lysosomal biogenesis. Dev. Cell12, 739–750 (2007). [DOI] [PubMed] [Google Scholar]

[CR26] 26.Rieder, S. E. & Emr, S. D. A novel RING finger protein complex essential for a late step in protein transport to the yeast vacuole. Mol. Biol. cell8, 2307–2327 (1997). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.van der Kant, R. et al. Characterization of the mammalian CORVET and HOPS complexes and their modular restructuring for endosome specificity. J. Biol. Chem 10.1074/jbc.M115.688440 (2015). [DOI] [PMC free article] [PubMed]

[CR28] 28.Perini, E. D., Schaefer, R., Stoter, M., Kalaidzidis, Y. & Zerial, M. Mammalian CORVET is required for fusion and conversion of distinct early endosome subpopulations. Traffic (Cph., Den.)15, 1366–1389 10.1111/tra.12232 (2014). [DOI] [PubMed] [Google Scholar]

[CR29] 29.Lachmann, J., Glaubke, E., Moore, P. S. & Ungermann, C. The Vps39-like TRAP1 is an effector of Rab5 and likely the missing Vps3 subunit of human CORVET. Cell. Logist.4 e970840, 10.4161/21592780.2014.970840 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Lorincz, P. et al. MiniCORVET is a Vps8-containing early endosomal tether in Drosophila. eLife5, e14226 10.7554/eLife.14226 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Lorincz, P. et al. Vps8 overexpression inhibits HOPS-dependent trafficking routes by outcompeting Vps41/Lt. eLife8, e45631 10.7554/eLife.45631 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Fasshauer, D., Sutton, R. B., Brunger, A. T. & Jahn, R. Conserved structural features of the synaptic fusion complex: SNARE proteins reclassified as Q-and R-SNAREs. Proc. Natl. Acad. Sci.95, 15781–15786 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Jahn, R. & Scheller, R. H. SNAREs-engines for membrane fusion. Nat. Rev. Mol. cell Biol.7, 631–643 10.1038/nrm2002 (2006). [DOI] [PubMed] [Google Scholar]

[CR34] 34.Brandhorst, D. et al. Homotypic fusion of early endosomes: SNAREs do not determine fusion specificity. Proc. Natl. Acad. Sci.103, 2701–2706 10.1073/pnas.0511138103 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Advani, R. J. et al. Seven novel mammalian SNARE proteins localize to distinct membrane compartments*. J. Biol. Chem.273, 10317–10324 10.1074/jbc.273.17.10317 (1998). [DOI] [PubMed] [Google Scholar]

[CR36] 36.Antonin, W., Holroyd, C., Tikkanen, R., Höning, S. & Jahn, R. The R-SNARE Endobrevin/VAMP-8 mediates homotypic fusion of early endosomes and late endosomes. Mol. Biol. cell11, 3289–3298 10.1091/mbc.11.10.3289 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Wong, S. H., Xu, Y., Zhang, T. & Hong, W. Syntaxin 7, a novel syntaxin member associated with the early endosomal compartment*. J. Biol. Chem.273, 375–380 10.1074/jbc.273.1.375 (1998). [DOI] [PubMed] [Google Scholar]

[CR38] 38.Pryor, P. R. et al. Combinatorial SNARE complexes with VAMP7 or VAMP8 define different late endocytic fusion events. EMBO Rep.5, 590–595 10.1038/sj.embor.7400150 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Advani, R. J. et al. Vamp-7 Mediates Vesicular Transport from Endosomes to Lysosomes. J. Cell Biol.146, 765–776 10.1083/jcb.146.4.765 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Dingjan, I. et al. Endosomal and Phagosomal SNAREs. Physiol. Rev.98, 1465–1492, 10.1152/physrev.00037.2017 (2018). [DOI] [PubMed] [Google Scholar]

[CR41] 41.Atlashkin, V. et al. Deletion of the SNARE vti1b in mice results in the loss of a single SNARE partner, syntaxin 8. Mol. Cell. Biol.23, 5198–5207 10.1128/mcb.23.15.5198-5207.2003 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Bollmann, C. et al. Primary neurons lacking the SNAREs vti1a and vti1b show altered neuronal development. Neural Dev.17, 12 10.1186/s13064-022-00168-2 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Kunwar, A. J. et al. Lack of the endosomal SNAREs vti1a and vti1b led to significant impairments in neuronal development. Proc. Natl. Acad. Sci.108, 2575–2580 10.1073/pnas.1013891108 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Music, A., et al The SNARE protein Vti1b is recruited to the sites of BCR activation but is redundant for antigen internalisation, processing and presentation. Front. Cell. Dev. Biol.10 10.3389/fcell.2022.987148 (2022). [DOI] [PMC free article] [PubMed]

[CR45] 45.Matsui, T. et al. Autophagosomal YKT6 is required for fusion with lysosomes independently of syntaxin 17. J. cell Biol.217, 2633–2645 10.1083/jcb.201712058 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Jian, F. et al. The STX17-SNAP47-VAMP7/VAMP8 complex is the default SNARE complex mediating autophagosome–lysosome fusion. Cell Res.34, 151–168 (2024). [DOI] [PMC free article] [PubMed]

[CR47] 47.Davis, L. J. et al. Organelle tethering, pore formation and SNARE compensation in the late endocytic pathway. J. Cell Sci.134 10.1242/jcs.255463 (2021). [DOI] [PMC free article] [PubMed]

[CR48] 48.Fujiwara, T. et al. Analysis of knock-out mice to determine the role of HPC-1/Syntaxin 1A in expressing synaptic plasticity. J. Neurosci.26, 5767 10.1523/JNEUROSCI.0289-06.2006 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR49] 49.Kloepper, T. H., Kienle, C. N. & Fasshauer, D. An elaborate classification of SNARE proteins sheds light on the conservation of the eukaryotic endomembrane system. Mol. Biol. cell18, 3463–3471 10.1091/mbc.e07-03-0193 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Kloepper, T. H., Kienle, C. N. & Fasshauer, D. SNAREing the basis of multicellularity: consequences of protein family expansion during evolution. Mol. Biol. Evol.25, 2055–2068 10.1093/molbev/msn151 (2008). [DOI] [PubMed] [Google Scholar]

[CR51] 51.Littleton, J. T. A genomic analysis of membrane trafficking and neurotransmitter release in Drosophila. J. Cell Biol.150, F77–F82 10.1083/jcb.150.2.F77 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Lu, H. & Bilder, D. Endocytic control of epithelial polarity and proliferation in Drosophila. Nat. cell Biol.7 1232–1239, 10.1038/ncb1324 (2005). [DOI] [PubMed] [Google Scholar]

[CR53] 53.Morrison, H. A. et al. Regulation of early endosomal entry by the Drosophila tumor suppressors Rabenosyn and Vps45. Mol. Biol. cell19, 4167–4176 10.1091/mbc.E08-07-0716 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Morelli, E. et al. Multiple functions of the SNARE protein Snap29 in autophagy, endocytic, and exocytic trafficking during epithelial formation in Drosophila. Autophagy0 10.4161/15548627.2014.981913 (2014). [DOI] [PMC free article] [PubMed]

[CR55] 55.Yamazaki, Y. et al. Goliath family E3 ligases regulate the recycling endosome pathway via VAMP3 ubiquitylation. EMBO J.32, 524–537 10.1038/emboj.2013.1 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR56] 56.Szenci, G., Glatz, G., Takáts, S. & Juhász, G. The Ykt6–Snap29–Syx13 SNARE complex promotes crinophagy via secretory granule fusion with Lamp1 carrier vesicles. Sci. Rep.14, 3200 10.1038/s41598-024-53607-x (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Lu, Y., Zhang, Z., Sun, D., Sweeney, S. T. & Gao, F. B. Syntaxin 13, a genetic modifier of mutant CHMP2B in frontotemporal dementia, is required for autophagosome maturation. Mol. cell52, 264–271 10.1016/j.molcel.2013.08.041 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Akbar, M. A., Ray, S. & Kramer, H. The SM protein Car/Vps33A regulates SNARE-mediated trafficking to lysosomes and lysosome-related organelles. Mol. Biol. cell20, 1705–1714 10.1091/mbc.E08-03-0282 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Takats, S. et al. Autophagosomal Syntaxin17-dependent lysosomal degradation maintains neuronal function in Drosophila. J. Cell Biol.201, 531–539 10.1083/jcb.201211160 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR60] 60.Takats, S. et al. Interaction of the HOPS complex with Syntaxin 17 mediates autophagosome clearance in Drosophila. Mol. Biol. cell25, 1338–1354 10.1091/mbc.E13-08-0449 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Lakatos, Z. et al. Sec20 is required for autophagic and endocytic degradation independent of Golgi-ER retrograde transport. Cells8 10.3390/cells8080768 (2019). [DOI] [PMC free article] [PubMed]

[CR62] 62.Takats, S. et al. Non-canonical role of the SNARE protein Ykt6 in autophagosome-lysosome fusion. PLoS Genet.14, e1007359 10.1371/journal.pgen.1007359 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR63] 63.Zhou, L. et al. Convergence of secretory, endosomal, and autophagic routes in trans-Golgi-associated lysosomes. J. Cell Biol.222 10.1083/jcb.202203045 (2023). [DOI] [PMC free article] [PubMed]

[CR64] 64.Csizmadia, T. et al. Developmental program-independent secretory granule degradation in larval salivary gland cells of Drosophila. Traffic (Cph., Den.)23, 568–586 10.1111/tra.12871 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Linnemannstöns, K. et al. Ykt6-dependent endosomal recycling is required for Wnt secretion in the Drosophila wing epithelium. Dev. (Camb., Engl.)147 dev185421 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR66] 66.Fujii, S. et al. Recycling endosomes attach to the trans-side of Golgi stacks in Drosophila and mammalian cells. J. Cell Sci.133 10.1242/jcs.236935 (2020). [DOI] [PubMed]

[CR67] 67.Ward, D. M., Pevsner, J., Scullion, M. A., Vaughn, M. & Kaplan, J. Syntaxin 7 and VAMP-7 are soluble N-ethylmaleimide-sensitive factor attachment protein receptors required for late endosome-lysosome and homotypic lysosome fusion in alveolar macrophages. Mol. Biol. cell11, 2327–2333 10.1091/mbc.11.7.2327 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR68] 68.Hargitai, D. et al. HOPS-dependent vesicle tethering lock inhibits endolysosomal fusions and autophagosome secretion upon the loss of Syntaxin17. Sci. Adv.11, eadu9605 10.1126/sciadv.adu9605 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR69] 69.Boda, A. et al. Drosophila Arl8 is a general positive regulator of lysosomal fusion events. Biochim. Biophys. Acta (BBA)-Mol. Cell Res.1866, 533–544 (2019). [DOI] [PubMed] [Google Scholar]

[CR70] 70.Lund, V. K., Madsen, K. L. & Kjaerulff, O. Drosophila Rab2 controls endosome-lysosome fusion and LAMP delivery to late endosomes. Autophagy14, 1520–1542 10.1080/15548627.2018.1458170 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR71] 71.Fu, Y. et al. Comprehensive functional analysis of Rab GTPases in Drosophila nephrocytes. Cell Tissue Res368, 615–627 10.1007/s00441-017-2575-2 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR72] 72.Maruzs, T. et al. Interaction of the sorting nexin 25 homologue Snazarus with Rab11 balances endocytic and secretory transport and maintains the ultrafiltration diaphragm in nephrocytes. Mol. Biol. cell34, ar87 10.1091/mbc.E22-09-0421 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR73] 73.Sheng, H., Stauffer, W. & Lim, H. N. Systematic and general method for quantifying localization in microscopy images. Biol. open5, 1882–1893 10.1242/bio.019893 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR74] 74.Lang, K. et al. Selective endocytosis controls slit diaphragm maintenance and dynamics in Drosophila nephrocytes. eLife11, e79037 10.7554/eLife.79037 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR75] 75.Itakura, E., Kishi-Itakura, C. & Mizushima, N. The hairpin-type tail-anchored SNARE syntaxin 17 targets to autophagosomes for fusion with endosomes/lysosomes. Cell151, 1256–1269 10.1016/j.cell.2012.11.001 (2012). [DOI] [PubMed] [Google Scholar]

[CR76] 76.Koike, S. & Jahn, R. SNAREs define targeting specificity of trafficking vesicles by combinatorial interaction with tethering factors. Nat. Commun.10, 1608 10.1038/s41467-019-09617-9 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Vaccari, T., Lu, H., Kanwar, R., Fortini, M. E. & Bilder, D. Endosomal entry regulates Notch receptor activation in Drosophila melanogaster. J. Cell Biol.180, 755–762 10.1083/jcb.200708127 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR78] 78.Smith, J. J., Sumiyama, K. & Amemiya, C. T. A living fossil in the genome of a living fossil: Harbinger transposons in the coelacanth genome. Mol. Biol. Evol.29, 985–993 10.1093/molbev/msr267 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR79] 79.Plooster, M. et al. Schizophrenia-linked protein tSNARE1 regulates endosomal trafficking in cortical neurons. J. Neurosci.: J. Soc. Neurosci.41, 9466–9481 10.1523/jneurosci.0556-21.2021 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR80] 80.Wen, P., Zhang, F., Fu, Y., Zhu, J. Y. & Han, Z. Exocyst Genes Are Essential for Recycling Membrane Proteins and Maintaining Slit Diaphragm in Drosophila Nephrocytes. J. Am. Soc. Nephrology: JASN31, 1024–1034 10.1681/asn.2019060591 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR81] 81.Weavers, H. et al. The insect nephrocyte is a podocyte-like cell with a filtration slit diaphragm. Nature457, 322–326 10.1038/nature07526 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR82] 82.Zhuang, S. et al. Sns and Kirre, the Drosophila orthologs of Nephrin and Neph1, direct adhesion, fusion and formation of a slit diaphragm-like structure in insect nephrocytes. Dev. (Camb., Engl.)136, 2335–2344 10.1242/dev.031609 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR83] 83.Zhang, F., Zhao, Y., Chao, Y., Muir, K. & Han, Z. Cubilin and amnionless mediate protein reabsorption in Drosophila nephrocytes. J. Am. Soc. Nephrology: JASN24, 209–216 10.1681/asn.2012080795 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.Zheng, D. et al. Human YKT6 forms priming complex with STX17 and SNAP29 to facilitate autophagosome-lysosome fusion. Cell Rep.43, 113760 (2024). [DOI] [PubMed]

[CR85] 85.Dean, D. M., Codd, L. E., Constanza, R. & Segel, X. M. purpleoid1, a classic Drosophila eye color mutation, is an allele of the t-SNARE-encoding gene SNAP29. microPubl. Biol.2025 10.17912/MICROPUB.BIOLOGY.001563 (2025). [DOI] [PMC free article] [PubMed]

[CR86] 86.Fuchs-Telem, D. et al. CEDNIK syndrome results from loss-of-function mutations in SNAP29. Br. J. Dermatol.164, 610–616, 10.1111/j.1365-2133.2010.10133.x (2011). [DOI] [PubMed] [Google Scholar]

[CR87] 87.Sprecher, E. et al. A mutation in SNAP29, coding for a SNARE protein involved in intracellular trafficking, causes a novel neurocutaneous syndrome characterized by cerebral dysgenesis, neuropathy, ichthyosis, and palmoplantar keratoderma. Am. J. Hum. Genet.77, 242–251 10.1086/432556 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR88] 88.Zeigerer, A. et al. Rab5 is necessary for the biogenesis of the endolysosomal system in vivo. Nature485, 465–470 10.1038/nature11133 (2012). [DOI] [PubMed] [Google Scholar]

[CR89] 89.Hegedus, K. et al. The Ccz1-Mon1-Rab7 module and Rab5 control distinct steps of autophagy. Mol. Biol. cell27, 3132–3142 10.1091/mbc.E16-03-0205 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR90] 90.Koike, S. & Jahn, R. SNARE proteins: zip codes in vesicle targeting? Biochemical J.479, 273–288 10.1042/bcj20210719 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR91] 91.Koike, S. & Jahn, R. Rab GTPases and phosphoinositides fine-tune SNAREs dependent targeting specificity of intracellular vesicle traffic. Nat. Commun.15, 2508 10.1038/s41467-024-46678-x (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR92] 92.Yadav, D. et al. A look beyond the QR code of SNARE proteins. Protein Sci.: a Publ. Protein Soc.33, e5158. 10.1002/pro.5158 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR93] 93.Schwartz, M. L. et al. Sec17 (alpha-SNAP) and an SM-tethering complex regulate the outcome of SNARE zippering in vitro and in vivo. eLife6 10.7554/eLife.27396 (2017). [DOI] [PMC free article] [PubMed]

[CR94] 94.Orr, A., Song, H., Rusin, S. F., Kettenbach, A. N. & Wickner, W. HOPS catalyzes the interdependent assembly of each vacuolar SNARE into a SNARE complex. Mol. Biol. cell28, 975–983 10.1091/mbc.E16-10-0743 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR95] 95.Song, H., Orr, A. S., Lee, M., Harner, M. E. & Wickner, W. T. HOPS recognizes each SNARE, assembling ternary trans-complexes for rapid fusion upon engagement with the 4th SNARE. eLife9, e53559 10.7554/eLife.53559 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR96] 96.Koumandou, V. L. et al. Molecular paleontology and complexity in the last eukaryotic common ancestor. Crit. Rev. Biochem. Mol. Biol.48, 373–396 10.3109/10409238.2013.821444 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR97] 97.Glatz, G., Gogl, G., Alexa, A. & Remenyi, A. Structural mechanism for the specific assembly and activation of the extracellular signal regulated kinase 5 (ERK5) module. J. Biol. Chem.288, 8596–8609 10.1074/jbc.M113.452235 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR98] 98.Atienza-Manuel, A., Castillo-Mancho, V., De Renzis, S., Culi, J. & Ruiz-Gómez, M. Endocytosis mediated by an atypical CUBAM complex modulates slit diaphragm dynamics in nephrocytes. Development (Cambridge, England)148 10.1242/dev.199894 (2021). [DOI] [PMC free article] [PubMed]

[CR99] 99.Chaudhry, N. et al. Lamp1 mediates lipid transport, but is dispensable for autophagy in Drosophila. Autophagy18, 2443–2458 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR100] 100.Hochapfel, F. et al. Distinct functions of Crumbs regulating slit diaphragms and endocytosis in Drosophila nephrocytes. Cell. Mol. Life Sci.74, 4573–4586 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR101] 101.Jean, S., Cox, S., Nassari, S. & Kiger, A. A. Starvation-induced MTMR13 and RAB21 activity regulates VAMP8 to promote autophagosome–lysosome fusion. EMBO Rep.16, 297–311 10.15252/embr.201439464 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR102] 102.Öztürk-Çolak, A. et al. FlyBase: updates to the Drosophila genes and genomes database. Genetics227, iyad211 10.1093/genetics/iyad211 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR103] 103.Milosavljevic, J. et al. Nephrotic syndrome gene TBC1D8B is required for endosomal maturation and nephrin endocytosis in drosophila. J. Am. Soc. Nephrol.33, 2174–2193 10.1681/asn.2022030275 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR104] 104.Kosaka, T. & Ikeda, K. Reversible blockage of membrane retrieval and endocytosis in the garland cell of the temperature-sensitive mutant of Drosophila melanogaster, shibirets1. J. Cell Biol.97, 499–507 (1983). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Multiple overlapping SNARE complexes drive endosome maturation in Drosophila nephrocytes

Dávid Hargitai

Márton Molnár

András Rubics

Iván Bodor

Dóra Baukál

Anikó Nagy

Virág Balogh

Zsófia Simon-Vecsei

Gábor Juhász

Péter Lőrincz

Abstract

Introduction

Results

Syx7 (Qa), Snap29 (Qbc) and Ykt6 (R) are required for early endosomal fusion in nephrocytes

Fig. 1. RNAi screen identifies Syx7, Snap29, and Ykt6 as key executors of early endosomal fusion.

Silencing of Snap29, Syx7, or Ykt6 impairs endocytic uptake

Fig. 2. Silencing of Syx7, Snap29, or Ykt6 impairs endocytic uptake.

Snap29, Syx7, and Ykt6 localize to Rab5-positive endosomes and form a SNARE complex

Fig. 3. Syx7, Snap29, and Ykt6 localize to early endosomes and form a SNARE complex.

Drosophila Syx7/Avl (CG5081) is functionally analogous to mammalian STX12, whereas Syx13 (CG11278) corresponds to STX7

Fig. 4. Functional reassignment of Syntaxin7 family members in Drosophila.

Disruption of early endosome fusion enhances recycling in a Ykt6- and Exocyst-dependent manner

Fig. 5. Lacunar depth is increased in Syx12L and Snap29 RNAi cells but is decreased in Ykt6 depleted cells.

Early endosome fusion bypass leads to aberrant endolysosomal ‘Swirl’ formation

Fig. 6. Endosome/lysosome maturation defects and formation of aberrant endo-lysosomal “swirls” upon bypass of early endosomal fusion.

Fig. 7. Aberrant endolysosomal swirls accumulate membrane-associated but not lumenal cargo.

Two additional SNARE complexes complement the Syx12L–Snap29–Ykt6 pathway in early endosome maturation

Fig. 8. Parallel Syx12L- and Syx7L-based SNARE complexes drive endosome maturation.

Distinct Rab5 and Rab7 requirements define the Ykt6- and Syx12L-independent maturation pathways

Fig. 9. Rab5 activation and Rab7 depletion reveal distinct Rab dependencies of Syx12L-, Snap29-, and Ykt6-dependent pathways.

Discussion

Fig. 10. Proposed model of endosome maturation driven by Syx12L (previously Syx7) - and Syx7L (previously Syx13) - containing SNARE complexes in Drosophila nephrocytes.

Methods

Molecular cloning

Protein production and purification

GST pull-down

S2R+ cell culture, transfection and immunoprecipitation

Antibodies

Fly work and treatments

Immunohistochemistry

Channel diffusion assay

Uptake assay

Fluorescence and SIM imaging

Electron microscopy

Quantification and statistical analysis, illustrations

Reporting summary

Supplementary information

Source data

Acknowledgements

Author contributions

Peer review

Peer review information

Funding

Data availability

Competing interests

Footnotes

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

S2R⁺ cell culture, transfection and immunoprecipitation