MYO5A-mediated stabilization promotes the acquisition of fusion competence in sealed autophagosomes

TL;DR

This study shows that MYO5A helps autophagosomes mature and fuse with lysosomes, which is crucial for autophagy to function properly.

Contribution

The study identifies MYO5A as a conserved regulator of autophagosome-lysosome fusion through its LIR motifs.

Findings

MYO5A is essential for autophagosome-lysosome fusion in yeast, mammalian cells, and C. elegans.

Loss of MYO5A disrupts the localization of fusion machinery and reduces stationary autophagosomes.

MYO5A binds to autophagosomes via two LIR motifs in its coiled-coil and globular tail domains.

Abstract

Autophagy requires precise regulation of autophagosome-lysosome fusion, yet the molecular details of this process remain incompletely understood. Here, we identify the class V myosin MYO5A as a critical regulator of autophagic flux. The genetic or pharmacological inhibition of MYO5A in Saccharomyces cerevisiae, mammalian cells, or Caenorhabditis elegans blocked autophagic flux by preventing autophagosome-lysosome fusion. MYO5A facilitates the maturation of autophagosomes into fusion-competent intermediates as its loss altered the localization of fusion machinery on autophagosomes and reduced the pool of stationary autophagosomes, a step that proved critical for subsequent fusion with lysosomes. Domain mapping and targeted mutagenesis revealed that two LIR motifs (PAYRVL and QAYIGL) within the coiled-coil and globular tail domains of MYO5A mediate its direct interaction with LC3 on autophagosomes. Live imaging in mammalian cells and C. elegans added support for this role, revealing how MYO5A regulates autophagic flux to ensure fusion. Together, these findings establish MYO5A as a regulator of autophagy and highlight its potential as a target for fine-tuning autophagic flux.

During autophagy, autophagosomes are completely sealed before fusing to lysosomes and releasing their cargo. In this study, MYO5A is shown to reduce directional switching in late stage autophagosomes, enabling maturation into fusion-competent intermediates.

MYO5A is essential for autophagosome–lysosome fusion, with its loss blocking autophagic flux in yeast, mammalian cells, and C. elegans.MYO5A functions during late stages of autophagy, promoting the transition of autophagosomes into fusion-competent intermediates.Loss of MYO5A function mis-localizes core fusion machinery and significantly reduces the stationary autophagosome population required for efficient fusion.MYO5A binds autophagosomes via its two LIR motifs in its coiled-coil and globular tail regions.MYO5A is a conserved regulator ensuring proper autophagic flux.

MYO5A is essential for autophagosome–lysosome fusion, with its loss blocking autophagic flux in yeast, mammalian cells, and C. elegans.

MYO5A functions during late stages of autophagy, promoting the transition of autophagosomes into fusion-competent intermediates.

Loss of MYO5A function mis-localizes core fusion machinery and significantly reduces the stationary autophagosome population required for efficient fusion.

MYO5A binds autophagosomes via its two LIR motifs in its coiled-coil and globular tail regions.

MYO5A is a conserved regulator ensuring proper autophagic flux.

MYO5A interacts with autophagosomes via its LIR motifs and promotes their fusion competence by reducing their directional movement.

Figures

The unconventional myosin Myo2 is required for autophagy in *Saccharomyces cerevisiae.*(A) Schematic of the GFP-Atg8 processing assay for general autophagy. Cells expressing GFP-Atg8 were starved in SD-N medium. Upon autophagosome–vacuole fusion, protease-resistant GFP accumulates in the vacuole and is detected as free GFP by immunoblotting. (B) GFP-Atg8 processing in *atg1Δ* cells as an autophagy-deficient control, where free GFP release is absent. (C–G) GFP-Atg8 processing in WT (*MYO2*) and *ts myo2* alleles (*myo2-12;* p* = 0.0272*, myo2-16;* *p* = 0.0019, 0.0029, p* = 0.0007, *myo2-18;* *

Perturbation of MYO5A inhibits autophagic flux in HeLa cells.(A) Tandem-tagged ptfLC3 (mRFP-GFP-LC3) was expressed in WT and *MYO5A* KO HeLa cells. Cells were either left untreated (GM) or treated with starvation medium (EBSS, 2 h), bafilomycin A1 (BafA1, 100 nM, 2 h), or pentabromopseudilin (PBP, 1 µM, 2 h). Cells were immunostained for MYO5A. Representative fluorescence micrographs are shown. Scale bar: 5 µm; inset: 1 µm. The cell boundary is marked in white and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost corner. (B, C) Quantifica

Lysosomal acidification, positioning, and morphology remain unperturbed upon *MYO5A* knockout.(A) WT and *MYO5A* KO HeLa cells were either untreated or subjected to starvation (EBSS, 2 h), followed by LysoTracker Deep Red (LTDR) staining or antibody labeling against LAMP1 and RAB7. Scale bar, 10 µm. The cell boundary is marked in white. Quantification of mean intensity per cell for (B) RAB7, (C) LTDR-positive, and (D) LAMP1 puncta was performed using the *Measure* plug-in of FIJI ImageJ, and plotted as mean ± SEM from three independent experiments (*N* = 3, *n* = 100 cells). Statistical signif

Loss of MYO5A alters the colocalization dynamics of various stage-specific markers with autophagosomes.(A) WT and *MYO5A* KO HeLa cells were transfected with GFP-LC3B and co-transfected with HA-ATG12, HA-VPS33A, Flag-SNAP29, or Flag-STX17, as indicated (in magenta). WIPI2 and MYO5A were detected by antibody staining. Scale bar, 10 µm. Zoomed insets highlight colocalization events between stage-specific markers and LC3, as well as between MYO5A and stage-specific markers. Scale bar, 1 µm. The cell boundary is marked in white and the white boxes within the image indicate the insets, which are sh

The CC–GTD domain of MYO5A is indispensable for mediating autophagosome–lysosome fusion.(A) Schematic representation of different domains of the MYO5A protein (overexpression constructs): motor region (1–755 aa), IQ region (755–923 aa), coiled-coil domain (923–1227 aa), and globular tail domain (1227–1829 aa). (B) WT HeLa cells were transfected with mCh-LC3B (in green) and different MYO5A constructs, namely full-length GFP-MYO5A (FL), mEmerald-CC–GTD, and mEmerald-M–IQ–CC (as indicated) (in gray). Cells were left untreated (GM) or starved (EBSS, 2 h), followed by LAMP1 staining (in magenta). S

The unconventional myosin Myo2 is required for autophagy in *Saccharomyces cerevisiae.*(A) Schematic of the GFP-Atg8 processing assay for general autophagy. Cells expressing GFP-Atg8 were starved in SD-N medium. Upon autophagosome–vacuole fusion, protease-resistant GFP accumulates in the vacuole and is detected as free GFP by immunoblotting. (B) GFP-Atg8 processing in *atg1Δ* cells as an autophagy-deficient control, where free GFP release is absent. (C–G) GFP-Atg8 processing in WT (*MYO2*) and *ts myo2* alleles (*myo2-12;* p* = 0.0272*, myo2-16;* *p* = 0.0019, 0.0029, p* = 0.0007, *myo2-18;* *

Perturbation of MYO5A inhibits autophagic flux in HeLa cells.(A) Tandem-tagged ptfLC3 (mRFP-GFP-LC3) was expressed in WT and *MYO5A* KO HeLa cells. Cells were either left untreated (GM) or treated with starvation medium (EBSS, 2 h), bafilomycin A1 (BafA1, 100 nM, 2 h), or pentabromopseudilin (PBP, 1 µM, 2 h). Cells were immunostained for MYO5A. Representative fluorescence micrographs are shown. Scale bar: 5 µm; inset: 1 µm. The cell boundary is marked in white and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost corner. (B, C) Quantifica

Lysosomal acidification, positioning, and morphology remain unperturbed upon *MYO5A* knockout.(A) WT and *MYO5A* KO HeLa cells were either untreated or subjected to starvation (EBSS, 2 h), followed by LysoTracker Deep Red (LTDR) staining or antibody labeling against LAMP1 and RAB7. Scale bar, 10 µm. The cell boundary is marked in white. Quantification of mean intensity per cell for (B) RAB7, (C) LTDR-positive, and (D) LAMP1 puncta was performed using the *Measure* plug-in of FIJI ImageJ, and plotted as mean ± SEM from three independent experiments (*N* = 3, *n* = 100 cells). Statistical signif

Loss of MYO5A alters the colocalization dynamics of various stage-specific markers with autophagosomes.(A) WT and *MYO5A* KO HeLa cells were transfected with GFP-LC3B and co-transfected with HA-ATG12, HA-VPS33A, Flag-SNAP29, or Flag-STX17, as indicated (in magenta). WIPI2 and MYO5A were detected by antibody staining. Scale bar, 10 µm. Zoomed insets highlight colocalization events between stage-specific markers and LC3, as well as between MYO5A and stage-specific markers. Scale bar, 1 µm. The cell boundary is marked in white and the white boxes within the image indicate the insets, which are sh

The CC–GTD domain of MYO5A is indispensable for mediating autophagosome–lysosome fusion.(A) Schematic representation of different domains of the MYO5A protein (overexpression constructs): motor region (1–755 aa), IQ region (755–923 aa), coiled-coil domain (923–1227 aa), and globular tail domain (1227–1829 aa). (B) WT HeLa cells were transfected with mCh-LC3B (in green) and different MYO5A constructs, namely full-length GFP-MYO5A (FL), mEmerald-CC–GTD, and mEmerald-M–IQ–CC (as indicated) (in gray). Cells were left untreated (GM) or starved (EBSS, 2 h), followed by LAMP1 staining (in magenta). S

LIR motifs in MYO5A mediate its interaction with autophagosomes and are required for autophagosome–lysosome fusion.(A) Schematic representation of the MYO5A protein. The black boxes indicate the positions of LIR motifs mapped to the coiled-coil and globular tail domains. The positional information of the LIR motifs was analyzed using the iLIR web tool and is depicted below. On the right, the conservation of the LIR motifs across organisms is shown. (B) *MYO5A* KO cells were transfected with GFP-MYO5A, mEmerald-CC–GTD, or mEmerald-tagged P, Q, and DM LIR mutants in the CC–GTD domain of MYO5A, a

Mean directional switches of autophagosomes are altered in the absence of MYO5A.(A) Time-lapse images of WT and *MYO5A* KO cells co-transfected with mCh-LC3 and different MYO5A constructs (as indicated), and stained with LysoTracker Deep Red dye. Scale bar, 1 μm. Arrow marks indicate events of colocalization. (B, C) Quantification of mean directional changes of LC3-positive autophagosomes (one-way ANOVA with Tukey’s post hoc test, p* = 0.0115 and p* = 0.0003 and 0.0002) (B) and lysosomes (C) using the “TrackMate” plug-in of ImageJ. Data were averaged for each cell and plotted as mean ± SEM fro

Loss of *hum-2* results in the accumulation of autophagosomes in the cell body of *C. elegans* TRNs.(A) Schematic of the *C. elegans* PLM neuron. Black boxes indicate the imaged regions. (B) Schematic of the experimental paradigm for neuronal stimulation using blue light. Worms were subjected to 70 V blue-light pulses intermittently (5 s on, 5 s off) for 1 h, followed by a 45 min recovery before imaging for mNG::LGG-1 at different neuronal regions. (C) Representative images of mNG::LGG-1 in the cell body, proximal process, process, branch point, and synapse of PLM neurons in WT, *hum-2(ok596)*

Full text

Introduction

Autophagy is a conserved eukaryotic degradation pathway that maintains cellular homeostasis by eliminating damaged organelles, misfolded proteins, and other surplus cytoplasmic components (Mizushima, 2018). During this process, cargo is sequestered within double-membraned autophagosomes, which subsequently fuse with lysosomes to generate autolysosomes, where degradation and recycling occur (Melia et al, 2020). Over the past two decades, substantial progress has been made in delineating the molecular machinery underlying autophagosome biogenesis. At the phagophore assembly site (PAS), PI3P generated by the class III PI3K complex (VPS34, Beclin1/ATG6, and ATG14) recruits PI3P-binding proteins such as DFCP1 and WIPI2, initiating isolation membrane formation (Dooley et al, 2014; Bozic et al, 2020; Melia et al, 2020; Chang et al, 2021). Membrane expansion is driven by the ATG12–ATG5–ATG16L1 complex and ATG8 family proteins, which conjugate to phosphatidylethanolamine and promote elongation of the phagophore membrane (Romanov et al, 2012). Final closure of the autophagosomal membrane requires the ESCRT machinery, which mediates scission of the double membrane through its polymerization and fission activity (Takahashi et al, 2018; Loi et al, 2019). Following closure, autophagosomes must undergo a critical maturation step before fusion with lysosomes. Fusion is mediated by a conserved membrane trafficking machinery, including the SNARE proteins STX17, SNAP29, and VAMP8, which form a trans-SNARE complex across autophagosome and lysosome membranes (Nakamura and Yoshimori, 2017; Huang et al, 2018; Chen et al, 2021; Koyama-Honda and Mizushima, 2022). The HOPS tethering complex acts as a scaffold that coordinates SNARE assembly, while the small GTPase Rab7 recruits HOPS and effector proteins to the autophagosomal membrane (Jiang et al, 2014; Hegedűs et al, 2016). Additional accessory proteins such as TECPR2 and PLEKHM1 further stabilize this machinery and ensure efficient fusion (Jiang et al, 2014; McEwan et al, 2015; Fraiberg et al, 2021). Thus, autophagosomes, even after closure, require precise molecular loading to become fusion-competent.

Despite these advances, a major gap remains in understanding how autophagosomes are spatially and mechanically positioned to successfully engage lysosomes. Although SNAREs, Rab7, and HOPS define the molecular prerequisites for fusion, the upstream mechanisms that coordinate their localization and recruitment on autophagosomes, and that determine whether autophagosomes move, pause, or dock in proximity to lysosomes, are poorly defined. This represents a critical unresolved step in the autophagy pathway, since the efficiency of degradation ultimately depends on the probability of productive autophagosome–lysosome encounters. Importantly, defects in autophagosome–lysosome fusion are strongly associated with human diseases, including neurodegenerative disorders such as Parkinson’s, Huntington’s, and Alzheimer’s disease, as well as cancer and infectious pathologies, emphasizing the need to resolve the molecular mechanisms that govern this step (Bejarano et al, 2018; Aman et al, 2021; Mizushima, 2018).

The cytoskeleton provides the tracks and forces that regulate autophagosome biogenesis, trafficking, transport and positioning (Aplin et al, 1992; Kast and Dominguez, 2017; Nambiar and Manjithaya, 2024). Microtubules are particularly critical; wherein dynein mediates retrograde transport of autophagosomes toward the lysosome-rich perinuclear region, while kinesins regulate autophagosomal biogenesis by promoting anterograde transport of Atg9 vesicles, as seen with kinesin, UNC-104 in C. elegans (Mackeh et al, 2013; Stavoe et al, 2016; Cason et al, 2022; Cason and Holzbaur, 2023). They also participate in mediating autophagosome–lysosome fusion via the BORC complex by promoting the assembly of fusion proteins on autophagosomes and lysosomes (Jia et al, 2017). Multiple adapters, including JIP1, JIP3/4, HTT-HAP1, and the BORC complex, link autophagosomes to microtubule motors and coordinate trafficking with fusion machinery assembly (Fu and Holzbaur, 2014; Jia et al, 2017; Cason and Holzbaur, 2023). Post-translational modifications and LC3-interacting regions (LIRs) regulate motor–cargo interactions, fine-tuning autophagosome movement along microtubules (Olsvik et al, 2015; Nieto-Torres et al, 2021). These studies have established a central role for microtubule-based transport in autophagosome maturation and fusion.

However, increasing evidence suggests that motility alone is not sufficient to guarantee fusion (Jia et al, 2017). Instead, immobilization or spatial anchoring of vesicles can be equally decisive. For example, detyrosinated microtubules restrict lysosomes to defined domains, thereby enhancing autophagosome–lysosome fusion (Mohan et al, 2019). Myosin-Va has been shown to induce immobilization and clustering of cargoes at the axon initial segment, while actin patches at glutamatergic synapses position dendritic lysosomes to facilitate local degradation (Janssen et al, 2017; Van Bommel et al, 2019). These findings converge on a key principle, i.e., vesicle stalling, rather than continuous movement, can actively promote productive membrane fusion (Janssen et al, 2017; Mohan et al, 2019; Van Bommel et al, 2019). Whether autophagosomes exploit a similar mechanism has remained an open question.

Compared to microtubules, the contribution of the actin cytoskeleton and its associated motors to autophagy is less defined. Actin and actin-binding proteins participate in the early stages of autophagy, particularly in shaping and expanding the phagophore membrane (Coutts and La Thangue, 2015; Kast and Dominguez, 2017; Hu and Mullins, 2019). Unconventional myosins, a large superfamily of actin-based motors, have been implicated in diverse cellular processes including vesicle transport, tethering, exocytosis, and cytoskeletal remodeling (Woolner and Bement, 2009; Fili and Toseland, 2019). A growing body of work suggests their potential involvement in autophagy (Nambiar and Manjithaya, 2024). For example, in Drosophila melanogaster, myosin II regulates starvation-induced cycling of Atg9 vesicles (Tang et al, 2011). MYO1C, a class I myosin, modulates autophagic flux by regulating cholesterol-rich lipid raft trafficking (Brandstaetter et al, 2014). MYO6, a class VI myosin, facilitates autophagosome–lysosome fusion by interacting with endocytic components and also participates in selective autophagy pathways such as mitophagy and xenophagy (Morriswood et al, 2007; Brooks et al, 2017; Kruppa et al, 2018; Hu et al, 2019). Despite these insights, whether unconventional myosins directly regulate the molecular licensing of autophagosomes for fusion, or instead act indirectly through cargo trafficking, remains unresolved.

Building on these observations, a genetic screen conducted in our laboratory using temperature-sensitive (ts) mutants of S. cerevisiae was designed to identify molecular players involved in autophagic flux (Barve et al, 2018b; Singh et al, 2019). The screen focused on proteins involved in vesicular trafficking, for their ability to induce pexophagy, a selective autophagy pathway that degrades excess peroxisomes (Barve et al, 2018b; Barve et al, 2018a; Singh et al, 2019; Klionsky et al, 2021). Among the key findings, we identified the moonlighting roles of two protein complexes, septins and exocyst, and uncovered the involvement of Myo2, the yeast ortholog of the class V myosin MYO5A. Class V myosins represent a particularly interesting family in this context. They are processive, dimeric actin-based motors with a modular architecture comprising an N-terminal motor domain, a lever arm with IQ repeats that bind calmodulin, a coiled-coil region for dimerization, and a globular tail domain that interacts with a variety of cargo adapters (Hammer and Sellers 2012). MYO5A is best known for its roles in melanosome transport, synaptic plasticity, and mRNA trafficking (Yoshimura et al, 2006; Konietzny et al, 2022; Gong et al, 2024; Pan et al, 2024). Its activity is tightly regulated by calcium levels, phosphorylation status, and autoinhibitory conformations relieved by cargo binding (Legesse-Miller et al, 2006; Li et al, 2008; Pranchevicius et al, 2008; Pan et al, 2023). Using three experimental systems, i.e., S. cerevisiae, mammalian cells, and C. elegans, we demonstrate that perturbation of MYO5A impairs autophagic flux by blocking autophagosome–lysosome fusion, without overtly disrupting endo-lysosomal trafficking. Mechanistically, we find that MYO5A promotes the localization of core fusion machinery to autophagosomes and facilitates the formation of a stationary pool of vesicles poised for productive lysosomal encounters. Domain dissection and LIR motif studies revealed that both the coiled-coil and globular tail domains of MYO5A are required for its association with autophagosomal and fusion machinery components, suggesting that cargo recognition and motor coordination cooperate to generate fusion-competent vesicles. These findings reveal a previously unrecognized function for myosin in licensing autophagosomes for fusion, thereby expanding the paradigm of how cytoskeletal elements govern autophagy.

Results

The unconventional myosin, Myo2 is involved in the autophagy pathway in S. cerevisiae

To investigate the role of Myo2 in autophagic flux, we utilized S. cerevisiae temperature-sensitive (ts) mutants carrying point mutations in the COOH-terminal tail domain of Myo2. All ts mutant constructs were sourced from Anthony Bretscher’s laboratory (Schott et al, 1999). The mutations map to the cargo-binding tail region and include substitutions within both the proximal coiled-coil and distal globular subdomains; myo2-16 (M1212T, L1471S, D1497V), myo2-12 (H1373R, Q1441L, D1457V, S1512T), myo2-17 (K1285R, Y1287N, L1436S), and myo2-18 (Y1161F, N1171Y, L1413S, I1453V, I1498M) (Schott et al, 1999, Fig. 1D–G). These sites are distributed across regions of the tail domain known to mediate cargo-adapter interactions (Schott et al, 1999). Autophagic flux was first assessed using the GFP-Atg8 processing assay, in which GFP-tagged Atg8 serves as an autophagosome marker. Upon nutrient starvation (SD-N), autophagosomes labeled with GFP-Atg8 are delivered to the vacuole, where Atg8 is degraded and stable free GFP accumulates (Klionsky et al, 2021, Fig. 1A). In WT cells (MYO2), robust free GFP release was observed at both permissive (PT, 25 °C) and non-permissive temperatures (NPT, 36 °C) (Fig. 1C). As expected, autophagy-deficient cells (atg1Δ) failed to generate free GFP under either condition, validating the specificity of the assay (Fig. 1B). In contrast, Myo2 ts mutant alleles showed a marked reduction in free GFP release at NPT compared to PT, particularly 4 h after the induction of starvation (Fig. 1D–G). To complement immunoblotting, we examined GFP-Atg8 localization by fluorescence microscopy (Fig. 1H–M). Vacuoles were stained with CMAC blue dye, and the percentage of cells displaying punctate versus vacuolar GFP signal was quantified. In WT cells, starvation for 4 h at PT resulted in a strong vacuolar GFP signal, consistent with efficient autophagic flux (Fig. 1H,N,O). By contrast, Myo2 ts mutant alleles showed reduced vacuolar GFP accumulation at NPT, accompanied by an increase in punctate cytoplasmic GFP-Atg8 structures (Fig. 1I–L,N,O). As expected, atg1Δ cells exhibited persistent punctate GFP-Atg8 signal under both PT and NPT, reflecting their inability to form functional autophagosomes (Fig. 1M–O).Figure 1. The unconventional myosin Myo2 is required for autophagy in Saccharomyces cerevisiae.(A) Schematic of the GFP-Atg8 processing assay for general autophagy. Cells expressing GFP-Atg8 were starved in SD-N medium. Upon autophagosome–vacuole fusion, protease-resistant GFP accumulates in the vacuole and is detected as free GFP by immunoblotting. (B) GFP-Atg8 processing in atg1Δ cells as an autophagy-deficient control, where free GFP release is absent. (C–G) GFP-Atg8 processing in WT (MYO2) and ts myo2 alleles (myo2-12; p = 0.0272, myo2-16;* **p = 0.0019, 0.0029, ***p = 0.0007, myo2-18; **p < 0.0001, myo2-17; p < 0.0001) under permissive (PT) and non-permissive (NPT) conditions. Lysates were collected at the indicated time points and analysed by immunoblotting. Graphs show free GFP/GFP-Atg8 ratios (arbitrary units), mean of three independent experiments (error bars: SEM). Statistical significance was determined using one-way ANOVA with Tukey’s post hoc test. (H–M) Representative fluorescence micrographs of GFP-Atg8 localization. Vacuoles were stained with CMAC blue. Scale bar: 5 μm. (N, O) Quantification of cells showing punctate versus vacuolar GFP signal. Data represent mean ± SEM from three independent experiments (n ≥ 100 cells per condition). Statistical significance was determined using one-way ANOVA with Dunn’s post hoc test [**p < 0.01 (0.0018, 0.0054, 0.0017), ****p < 0.0001]. Source data are available online for this figure.

We next asked whether Myo2 function is required for selective autophagy, using a pexophagy assay with Pot1-GFP labeled peroxisomes. Cells were grown in oleate to induce peroxisome proliferation and subsequently shifted to starvation medium (Klionsky et al, 2021, Fig. EV1A). In WT cells, free GFP release was detected under both PT and NPT conditions following starvation, whereas ts mutants (myo2-12, myo2-13, and myo2-16) showed markedly reduced free GFP accumulation at NPT compared to PT (Fig. EV1C–F). As with the GFP-Atg8 assay, atg1Δ cells served as a negative control and failed to release free GFP under either condition (Fig. EV1B). In addition to immunoblotting, we visualized pexophagy at the single-cell level by fluorescence microscopy. Vacuoles were stained with CMAC blue dye, and the percentage of cells showing vacuolar GFP was quantified. In WT cells, GFP accumulation in the vacuole was readily observed at both PT and NPT, after 6 h of starvation, whereas the myo2-13 mutant showed a pronounced reduction in vacuolar GFP signal at NPT (Fig. EV1G,H,J). Consistent with expectations, atg1Δ cells exhibited no detectable vacuolar GFP under either condition (Fig. EV1I,J). These microscopy-based observations further support a requirement for Myo2 in efficient pexophagy.

Together, these findings establish that conditional mutations in the cargo-binding region of the Myo2 tail domain impair both bulk and selective autophagy in yeast. While yeast provides a powerful genetic system to dissect motor protein function, the close spatial arrangement of the pre-autophagosomal structure (PAS) and the vacuole limits resolution of transport and fusion dynamics. To address this limitation, we extended our studies to mammalian cells, where autophagosomes must traverse larger cytoplasmic distances to fuse with lysosomes, thereby enabling a clearer interrogation of MYO5A function in autophagy.

Perturbation of MYO5A inhibits autophagic flux in HeLa cells

To extend the findings from yeast to mammalian systems, we focused on MYO5A, the human ortholog of yeast Myo2 and a member of the class V myosin family. MYO5A function was perturbed by both chemical and genetic approaches in HeLa cells. For chemical inhibition, we used pentabromopseudilin (PBP), a small molecule that binds near the nucleotide-binding site of the MYO5A motor domain (Fedorov et al, 2009; Martin et al, 2009). This interaction interferes with ATP binding, hydrolysis, and ADP release, thereby prolonging the ATPase cycle and limiting MYO5A motility along actin filaments (Fedorov et al, 2009; Martin et al, 2009). In parallel, we generated a MYO5A knockout (KO) cell line and also depleted MYO5A by siRNA (Figs. 2A and EV2A). Autophagic flux was first monitored using the tandem mRFP-GFP-LC3B reporter, which allows discrimination between autophagosomes (yellow puncta, GFP^+ve^; RFP^+ve^) and autolysosomes (red puncta, GFP^–ve^; RFP^+ve^) owing to quenching of GFP fluorescence in acidic lysosomal environments (Kimura et al, 2007; Klionsky et al, 2021). In WT cells, an increase in the number of autolysosomes was observed on starvation (EBSS), consistent with efficient flux (Fig. 2A–C). By contrast, MYO5A KO cells and cells treated with PBP displayed a significant increase in autophagosomes together with a reduction in autolysosomes, indicative of impaired flux (Fig. 2A–C). A similar phenotype was obtained upon siRNA-mediated depletion, which showed elevated GFP fluorescence consistent with autophagosome accumulation (Fig. EV2A,B). To further characterize these defects, we performed live-cell imaging (Fig. 2D,E). In WT cells, mCherry-LC3B-labeled autophagosomes were engaged by lysosomes marked with LAMP1-GFP and LysoTracker Deep Red (Fig. 2D,E). In MYO5A KO cells, autophagosomes displayed markedly reduced association with lysosomes, suggesting inefficient fusion (Fig. 2D,E).Figure 2. Perturbation of MYO5A inhibits autophagic flux in HeLa cells.(A) Tandem-tagged ptfLC3 (mRFP-GFP-LC3) was expressed in WT and MYO5A KO HeLa cells. Cells were either left untreated (GM) or treated with starvation medium (EBSS, 2 h), bafilomycin A1 (BafA1, 100 nM, 2 h), or pentabromopseudilin (PBP, 1 µM, 2 h). Cells were immunostained for MYO5A. Representative fluorescence micrographs are shown. Scale bar: 5 µm; inset: 1 µm. The cell boundary is marked in white and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost corner. (B, C) Quantification of autophagosomes (GFP^+ve^, RFP^+ve^) and autolysosomes (GFP^–ve^, RFP^+ve^) per cell, using the “Analyze Particles” plug-in in FIJI ImageJ. Data represent mean ± SEM from three independent experiments (N  =  3, n  =  75 cells). Statistical significance was assessed by one-way ANOVA with Kruskal–Wallis’s test (2b, ****p  <  0.0001: GM vs. PBP and KO1, ***p  = 0.0005: GM vs. KO2, *p = 0.0418: GM vs. EBSS, *p = 0.0181: GM vs. Baf, 2c, 2b, ****p  <  0.0001: GM vs. BafA1, KO1, and KO2). (D) Time-lapse imaging of WT and MYO5A KO HeLa cells co-expressing mCh-LC3B and LAMP1-GFP, and stained with LysoTracker Deep Red. Scale bar: 1 µm. Arrows indicate autophagosome (mCh-LC3B) - lysosome (LysoTracker^+ve^) fusion events. (E) Quantification of LC3B puncta colocalizing with LysoTracker^+ve^ lysosomes using the “Colocalization” plug-in in ImageJ. Data represent mean ± SEM from three independent experiments (N  =  3, n  =  40 cells). Statistical significance was assessed by Mann–Whitney Student’s t-test (****p  <  0.0001). (F) WT HeLa cells transfected with scrambled (Scr) or MYO5A siRNA were immunostained for p62 and MYO5A. Scale bar: 10 µm; inset: 1 µm. (G) Quantification of p62 puncta area per cell using FIJI ImageJ. Data represent mean ± SEM from three independent experiments (N  =  3, n  =  75 cells). Statistical significance was determined by Mann–Whitney Student’s t-test (****p  <  0.0001). (H) mRFP-GFP-LC3B was expressed in WT, MYO5A KO, or PBP-treated HeLa cells. Cells were treated with BafA1 (100 nM), washed, fixed, and the number of autolysosomes was quantified. Scale bar: 5 µm. (I) Graph shows mean ± SEM from >50 cells per group. Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc test (****p  <  0.0001, WT; 3 h. MYO5A KO, 3 h and PBP, 3 h). (J) Immunoblotting of LC3 and MYO5A in WT and MYO5A KO HeLa cells subjected to GM, EBSS (2 h), or BafA1 (2 h). (K) Quantification of LC3B-II/LC3B-I ratios from three independent experiments (mean ± SEM). Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc test (****p  <  0.0001, *p = 0.0323; MYO5A KO2 BafA1 treated, compared to WT GM). Source data are available online for this figure.

Given that defects in flux often manifest as altered adapter protein turnover, we next examined p62/SQSTM1 (Fig. 2F,G). Immunofluorescence revealed a significant increase in the size and intensity of p62 puncta in MYO5A-depleted cells compared to Scr siRNA controls, pointing to defective clearance. Because autophagosome–lysosome fusion is inherently dynamic and heterogeneous across cells, we synchronized this step using bafilomycin A1 (BafA1), which arrests fusion prior to release (Fig. 2H,I). Upon removal of BafA1, WT cells efficiently generated autolysosomes in a time-dependent manner, whereas MYO5A KO and PBP-treated cells showed a much weaker response, supporting a block in autophagic flux plausibly at the fusion stage (Fig. 2H,I). In addition, on immunoblotting, LC3B-II levels were elevated in KO cells under both basal (GM) and starved (EBSS) conditions (Fig. 2J,K). Importantly, treatment with BafA1 failed to further increase LC3B-II in KO cells, confirming a block in autophagic flux rather than enhanced initiation. LC3B conversion assays following siRNA depletion and PBP treatment produced comparable results, further validating this conclusion. Immunoblotting showed accumulation of p62 in MYO5A knockdown cells compared to controls, reflecting impaired degradation (Fig. EV2C–E).

Taken together, these complementary genetic and chemical approaches consistently show that perturbation of MYO5A function results in autophagosome accumulation, reduced autophagosome–lysosome engagement, increased p62 burden, and elevated LC3B-II levels. These findings point to an essential role for MYO5A in sustaining autophagic flux and cargo degradation. However, whether this defect reflects a block within the autophagy pathway itself or arises from lysosomal dysfunction remains unresolved. To address this, we next examined lysosomal activity and related parameters following MYO5A inhibition.

Lysosomal function and morphology remain unperturbed upon MYO5A knockout

To determine whether the block in autophagic flux upon MYO5A inhibition arises from defects in lysosomes, we examined lysosomal morphology, acidification, and localization. LAMP1 and RAB7 were used as markers for late endosomes and lysosomes, and LysoTracker Deep Red staining was employed to assess lysosomal acidity under basal (GM) and starvation (EBSS) conditions. Fluorescence intensity measurements revealed a significant increase in lysosomal signal during starvation in both WT and MYO5A KO cells, indicative of enhanced lysosomal biogenesis, with no discernible differences between the groups (Fig. 3A–D). Radial distribution analysis further showed perinuclear clustering of lysosomes upon starvation and a more dispersed distribution under basal conditions, consistent with expected physiological responses (Fig. 3E). These findings suggest that MYO5A does not influence lysosomal positioning during nutrient deprivation. To complement these imaging results, we examined lysosomal protein markers by immunoblotting (Fig. 3F). The levels of LAMP1, RAB7, and the lysosomal protease cathepsin D were unchanged between WT and MYO5A KO cells under basal, starvation, or BafA1-treated conditions, reinforcing that lysosomal biogenesis and proteolytic competence were unaffected by MYO5A depletion (Fig. 3G–I). These findings argue against a role for MYO5A in lysosomal maintenance or function. Next, we evaluated lysosomal degradative capacity through endocytic trafficking assays (Fig. 3J). Since epidermal growth factor receptor (EGFR) undergoes ligand-induced internalization and lysosomal degradation, we tracked EGFR turnover following EGF stimulation in MYO5A knockdown or PBP-treated cells (Vats and Manjithaya, 2019). Degradation kinetics were indistinguishable from controls, indicating that receptor trafficking through the endo-lysosomal system remained intact despite MYO5A inhibition (Fig. 3J,K).Figure 3. Lysosomal acidification, positioning, and morphology remain unperturbed upon MYO5A knockout.(A) WT and MYO5A KO HeLa cells were either untreated or subjected to starvation (EBSS, 2 h), followed by LysoTracker Deep Red (LTDR) staining or antibody labeling against LAMP1 and RAB7. Scale bar, 10 µm. The cell boundary is marked in white. Quantification of mean intensity per cell for (B) RAB7, (C) LTDR-positive, and (D) LAMP1 puncta was performed using the Measure plug-in of FIJI ImageJ, and plotted as mean ± SEM from three independent experiments (N  =  3, n  =  100 cells). Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc analysis (3b, p = 0.0024 GM vs. EBSS; WT and p = 0.0063 GM vs. EBSS; KO (3C, D), ****p  <  0.0001). (E) Radial distribution analysis of LAMP1 puncta from the nucleus was quantified using the “Radial Plot” plug-in of FIJI ImageJ, plotted as mean ± SEM from three independent experiments (N  =  3, n  =  70 cells). No significant differences were observed between groups (one-way ANOVA with Kruskal–Wallis’s test). (F) Lysates from WT and MYO5A KO HeLa cells treated with EBSS (2 h) and/or BafA1 (100 nM, 2 h) were immunoblotted with antibodies against Cathepsin D (CTSD), RAB7, LAMP1, and MYO5A. Relative protein levels of (G) CTSD, (H) RAB7, and (I) LAMP1 were quantified from three independent experiments and plotted as mean ± SEM. Statistical significance was assessed by one-way ANOVA with Tukey post hoc analysis relative to WT (GM) control. (J) WT HeLa cells transfected with either Scr or MYO5A siRNA were serum-starved for 3 h and subsequently left untreated or pretreated with PBP before stimulation with EGF (100 ng/ml) for the indicated time points. Lysates were immunoblotted for EGFR. (K) Quantification of EGFR levels from three independent experiments plotted as mean ± SEM. Statistical significance was determined using one-way ANOVA with Tukey’s post hoc analysis. Source data are available online for this figure.

Taken together, these results provide compelling evidence that lysosomes remain structurally and functionally intact in the absence of MYO5A. When considered alongside our earlier findings of autophagosome accumulation and impaired autophagosome–lysosome colocalization, these data suggest that the observed block in autophagic flux originates from the autophagosomal side rather than from lysosomal deficiencies. We reasoned that if lysosomes remain intact, MYO5A must instead regulate aspects of autophagosome competence for fusion. Notably, myosins have been implicated at multiple stages of the autophagy pathway, including autophagosome biogenesis, trafficking, and fusion, making it imperative to dissect the contribution of MYO5A to autophagic flux in a stage-specific manner (Tang et al, 2011; Hu et al, 2019; Feng et al, 2022; Nambiar and Manjithaya, 2024). To this end, we next examined stage-specific markers of the autophagy pathway to pinpoint the precise step at which MYO5A acts during the autophagic flux.

Fusion-incompetent mature autophagosomes are formed in the absence of MYO5A

Autophagy unfolds through a tightly choreographed sequence of events in which specific protein complexes are recruited to autophagosomal membranes in a stage-dependent manner (Melia et al, 2020; Chang et al, 2021). Building on our earlier observations that lysosomal integrity and degradative capacity remain unaltered in the absence of MYO5A, we next sought to determine whether the observed flux defect originates from the autophagosomal side of the pathway. To resolve this, we performed a stage-specific analysis of the autophagic pathway. We first labeled autophagosomes with GFP-LC3B, which marks them from formation through their final fusion with lysosomes, and quantified their colocalization with proteins representing distinct stages of autophagy. In MYO5A KO cells, autophagosomes displayed significantly higher colocalization with the early nucleation factor WIPI2 and the elongation machinery component ATG12 compared to WT cells (Fig. 4A,B). In contrast, their association with key components of the late fusion machinery, including STX17, VPS33A, and SNAP29, was markedly reduced (Fig. 4A,B). These data suggested that while autophagosomes form and elongate normally in the absence of MYO5A, they fail to efficiently engage the fusion complex required for their maturation into autolysosomes. Because defects in autophagosome–lysosome fusion could, in principle, arise from the lysosomal side, we next evaluated whether lysosomes remained capable of accommodating and fusing with incoming cargo. Importantly, the presence of the R-SNARE VAMP8 on lysosomes, a critical determinant of autolysosome formation, was unchanged, as demonstrated by its unaltered colocalization with LAMP1 in both WT and MYO5A KO cells (Fig. 4C,D). These findings reinforced the conclusion that lysosomes retain their fusion competence and degradative potential even in the absence of MYO5A. We then turned to live-cell imaging to capture the dynamics of autophagosome maturation in real time. Autophagosomes in MYO5A KO cells showed higher colocalization with early-stage marker DFCP1, while their colocalization with late-stage fusion components such as VPS33A was reduced (Fig. 4E,F). This imbalance highlighted a bottleneck at the transition from early to late maturation stages, where autophagosomes should normally recruit the fusion machinery. To determine whether these differences reflected altered protein expression or mislocalization, we examined the levels of stage-specific autophagy markers by immunoblotting (Fig. EV3A,B). No changes were detected between WT and MYO5A KO cells, indicating that MYO5A does not regulate the abundance of these proteins (Fig. EV3A,B). Instead, its role plausibly appears to lie in enabling their proper localization to autophagosomes. To directly test this, we immunoprecipitated endogenous LC3B and assessed its association with stage-specific autophagy factors under basal and starvation conditions (Fig. EV3C,D). In MYO5A KO cells, we observed a pronounced reduction in the interaction of LC3B-positive autophagosomes with STX17 and VPS33A, confirming that the localization of fusion machinery is compromised in the absence of MYO5A (Fig. EV3C,D).Figure 4. Loss of MYO5A alters the colocalization dynamics of various stage-specific markers with autophagosomes.(A) WT and MYO5A KO HeLa cells were transfected with GFP-LC3B and co-transfected with HA-ATG12, HA-VPS33A, Flag-SNAP29, or Flag-STX17, as indicated (in magenta). WIPI2 and MYO5A were detected by antibody staining. Scale bar, 10 µm. Zoomed insets highlight colocalization events between stage-specific markers and LC3, as well as between MYO5A and stage-specific markers. Scale bar, 1 µm. The cell boundary is marked in white and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost and leftmost corners, indicating events of colocalization. (B) Quantification of colocalization events between stage-specific markers and LC3B in WT and MYO5A KO cells using the Colocalization plug-in of ImageJ. Data were plotted as mean ± SEM from three independent experiments (N  =  3, n  =  60 cells). Statistical significance was assessed by Mann–Whitney Student’s t-test between the WT and MYO5A KO cells for each marker (****p  <  0.0001, ***p  = 0.0002 for ATG12, ***p  = 0.0008 for SNAP29). (C) WT and MYO5A KO cells were transfected with Flag-VAMP8 and stained for LAMP1. Scale bar, 10 µm. Insets show colocalization events between Flag-VAMP8 and LAMP1. Scale bar, 1 µm. The cell boundary is marked in white, and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost corner, indicating events of colocalization. (D) Quantification of colocalization between LAMP1 and Flag-VAMP8 in WT and MYO5A KO cells. Data represent mean ± SEM from three independent experiments (N  =  3, n  =  60 cells). Statistical significance was assessed by an unpaired Student’s t-test (ns, non significant). (E) Time-lapse images of WT and MYO5A KO HeLa cells transfected with either mCh-LC3B or GFP-LC3B and co-transfected with mCh-DFCP1 or GFP-VPS33A, as indicated. Scale bar, 10 µm. Insets highlight colocalization events between stage-specific markers and LC3. Scale bar, 1 µm. (F) Quantification of colocalization between stage-specific markers and LC3B in WT and MYO5A KO cells using the Colocalization plug-in in ImageJ. Data represent mean ± SEM from three independent experiments (N  =  3, n  =  40 cells). Statistical significance was determined by unpaired Student’s t-test (***p = 0.0001, *p  = 0.0286). (G) WT and MYO5A KO cells were transfected with Flag-VAMP8 and Flag-SNAP29 and either left untreated or subjected to starvation (EBSS, 2 h). Lysates collected 48 h post-transfection were subjected to immunoprecipitation with anti-Flag antibody, followed by immunoblotting with the indicated antibodies. (H) Quantification of the proportion of Flag-SNAP29 and STX17 co-precipitated with Flag-VAMP8 from WT and MYO5A KO cells under basal or starvation conditions. Data were plotted as mean ± SEM from three independent experiments. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc analysis (SNAP29, WT_GM vs. KO_EBSS; *p  = 0.0319, WT_EBSS vs. KO_GM; *p  = 0.0424, WT_EBSS vs. KO_EBSS; *p  = 0.0145; STX17, WT_EBSS vs. KO_EBSS; *p  = 0.0127). (I) Schematic representation of the protease protection assay used to evaluate autophagic cargo entrapment within sealed autophagosomes. p62-tagged cargo inside closed autophagosomes is protected from proteinase K (middle panel), unless Triton X-100 is added (right panel), while untreated lysates serve as a control (left panel). Lysates from HeLa cells transfected with Scr or MYO5A siRNA, or treated with DMSO or PBP, were subjected to combinatorial treatments with proteinase K and Triton X-100 as indicated, followed by immunoblotting with anti-p62. Source data are available online for this figure.

We next asked whether the failure to recruit fusion proteins could be attributed to defective assembly of the SNARE complex itself. To this end, we immunoprecipitated Flag-VAMP8 and examined its interaction with its cognate SNARE partners STX17 and SNAP29 (Fig. 4G,H). Consistent with our earlier findings, MYO5A KO cells exhibited a reduced association of VAMP8 with both STX17 and SNAP29, pointing to impaired assembly of the functional SNARE complex (Fig. 4G,H). Since SNARE recruitment typically occurs only after autophagosome membrane closure, we considered the possibility that defects in autophagosome sealing might underlie the reduced colocalization with late-stage fusion markers. To address this, we performed a protease protection assay using p62/SQSTM1 as a cargo surrogate (Fig. 4I). In sealed autophagosomes, p62 is protected from protease digestion, whereas in unsealed structures it is degraded. Triton X-100 treatment served as a positive control and abolished p62 protection (Fig. 4I). Under basal conditions, only faint p62 bands were detected, reflecting ongoing autophagic flux where autophagosomes rapidly fuse with lysosomes (Fig. 4I). Treatment with BafA1, which blocks autophagosome–lysosome fusion and promotes autophagosome accumulation, led to increased p62 protection, validating the assay (Fig. 4I). Strikingly, perturbation of MYO5A function either by siRNA knockdown or pharmacological inhibition using PBP resulted in accumulation of protease-protected p62. This effect was further enhanced in the presence of BafA1, demonstrating that autophagosomes formed in the absence of MYO5A are fully sealed and accumulate in large numbers (Fig. 4I). These findings excluded the possibility of a closure defect and instead pointed to a downstream block in trafficking or fusion.

Collectively, these findings rule out defects in autophagosome formation, sealing, or lysosomal competence, and instead pinpoint MYO5A as a crucial facilitator of autophagosome–lysosome fusion. Specifically, MYO5A enables the recruitment and assembly of SNARE proteins on sealed autophagosomes, licensing them for productive fusion with lysosomes. Given this critical role at the fusion stage, we next asked how MYO5A itself associates with autophagosomes and whether this interaction underlies its ability to coordinate the fusion machinery.

The CC–GTD domain of the motor protein is indispensable for mediating autophagosome–lysosome fusion

In our earlier experiments, where we immunoprecipitated endogenous LC3B, we detected MYO5A within the pulled-down fraction (Fig. EV3C). This observation suggested that MYO5A associates with autophagosomal membranes. To explore this interaction in greater detail, we next immunoprecipitated MYO5A and examined its association with stage-specific components of the autophagy pathway. We chose markers spanning different stages of autophagic flux: proteins involved in nucleation (FIP200, WIPI2, and ATG9), elongation (ATG5–ATG12–ATG16), the general autophagosome marker LC3B, and late fusion markers (STX17 and LAMP1) (Fig. EV4A). MYO5A consistently co-immunoprecipitated with LC3B, confirming its presence on autophagosomes. Interestingly, MYO5A also interacted with elongation-stage components and with the fusion machinery but not with the nucleation-stage markers (Fig. EV4A). This selective pattern of interaction suggested that MYO5A associates preferentially with autophagosomes at stages beyond initiation, with a bias toward vesicles preparing for fusion. To validate and extend these biochemical observations, we used a proximity ligation assay (PLA), which detects protein–protein interactions within a nanometer-scale range (Fig. EV4B,C). Using MYO5A and LC3B as probes, we observed distinct PLA puncta under basal conditions, which markedly increased during starvation and further accumulated upon lysosomal inhibition (Fig. EV4B,C). Control reactions using either probe alone gave negligible signals, reinforcing the specificity of the observed interaction (Fig. EV4B,C). These findings provided independent support for a close association between MYO5A and autophagosomes, particularly under conditions where flux is upregulated or stalled at the fusion step.

Given these results, we next asked which part of MYO5A mediates this association with autophagosomes. MYO5A is a multi-domain motor protein composed of an N-terminal motor head domain, followed by a neck region with multiple IQ motifs, a long coiled-coil stalk, and a C-terminal globular tail domain (GTD) that is widely implicated in cargo recognition (Hammer and Sellers 2012). To dissect the contribution of these domains, we over-expressed truncated variants of MYO5A: full-length MYO5A (FL, 1–1829 aa; GFP-MYO5A); M-IQ-CC (1–1227 aa; mEmerald-M-IQ-CC), containing the motor domain, IQ repeats, and initial coiled-coil domain (CC1); and CC–GTD (923–1829 aa; mEmerald-CC–GTD), comprising the second and third coiled-coil domains and the GTD (Konietzny et al, 2022) (Fig. 5A). Expression and localization studies revealed differences among these constructs. Full-length MYO5A exhibited a punctate cytoplasmic distribution that became more enriched under starvation conditions (Fig. 5B). Interestingly, the CC–GTD fragment alone displayed a similar punctate localization (Fig. 5B). In contrast, the M–IQ–CC1 fragment was localized towards the periphery, showing little cytosolic enrichment (Fig. 5B). To probe these associations further, we co-expressed the MYO5A fragments with autophagic (mCh-LC3) and lysosomal markers (LAMP1) (Fig. 5B). Both full-length MYO5A and the CC–GTD fragment robustly colocalized with LC3B-positive autophagosomes and LAMP1-positive lysosomes, whereas the M–IQ–CC1 fragment showed minimal overlap with either compartment (Fig. 5B–E). These findings were further substantiated by live-cell imaging, where puncta formed by full-length MYO5A or CC–GTD colocalised with mCh-LC3B-positive vesicles, while M–IQ–CC1 remained largely unassociated (Fig. 5F,G). Taken together, these results strongly implicate the CC–GTD as the critical determinant of the localization of MYO5A to autophagosomes and lysosomes. Having established differential localization patterns, we next asked whether these domains could functionally influence autophagic flux. Overexpression of full-length MYO5A resulted in colocalization of LC3B with LAMP1, indicative of an autolysosomal population (Fig. 5B,C). Expression of the CC–GTD fragment was sufficient to recapitulate this effect, supporting the idea that the tail domain alone can promote autolysosomal formation (Fig. 5B,C). By contrast, expression of the M–IQ–CC1 fragment failed to promote autophagosome–lysosome colocalization events (Fig. 5B,C). At this stage, the evidence suggested that the C-terminal CC–GTD fragment of MYO5A is both necessary and sufficient for its recruitment to autophagic vesicles, and for promoting their subsequent fusion with lysosomes.Figure 5. The CC–GTD domain of MYO5A is indispensable for mediating autophagosome–lysosome fusion.(A) Schematic representation of different domains of the MYO5A protein (overexpression constructs): motor region (1–755 aa), IQ region (755–923 aa), coiled-coil domain (923–1227 aa), and globular tail domain (1227–1829 aa). (B) WT HeLa cells were transfected with mCh-LC3B (in green) and different MYO5A constructs, namely full-length GFP-MYO5A (FL), mEmerald-CC–GTD, and mEmerald-M–IQ–CC (as indicated) (in gray). Cells were left untreated (GM) or starved (EBSS, 2 h), followed by LAMP1 staining (in magenta). Scale bar, 10 µm; zoom insets, 1 µm. (C–E) Graphs represent the percentage colocalization between. The cell boundary is marked in white, and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost corner, indicating events of colocalization. (C) LC3 and LAMP1 (one-way ANOVA with Kruskal–Wallis test; *p = 0.009; ****p < 0.0001), (D) LC3 and MYO5A (Unpaired Student’s t-test; ***p = 0.0004, **p = 0.0087, *p = 0.0208), and (E) LAMP1 and MYO5A (Unpaired Student’s t-test; **p = 0.0049) across different MYO5A constructs in GM and EBSS conditions. Quantification was performed using the “Colocalization” plug-in in ImageJ and plotted as mean ± SEM from three independent experiments (N = 3, 75 cells). (F) Time-lapse images of WT cells co-expressing mCh-LC3B with either GFP-MYO5A, mEmerald-CC–GTD, or mEmerald-M–IQ–CC. Scale bar, 1 µm. (G) The graph shows the percentage colocalization of LC3 with different MYO5A domain constructs, quantified as above from three independent experiments (N = 3, 40 cells). Data plotted as mean ± SEM; statistical significance assessed by Kruskal–Wallis’s test (****p < 0.0001). (H) WT and MYO5A KO cells were transfected with mCh-LC3B and stained for LAMP1. Scale bar, 10 µm; zoom insets, 1 µm. Intensity line profiles show mCh-LC3B and LAMP1 signal overlap. Rescue experiments were performed by expressing different MYO5A constructs in MYO5A KO cells, followed by LAMP1 staining. The cell boundary is marked in white, and the white boxes within the image indicate the insets, which are shown as magnified views in the rightmost corner, indicating events of colocalization. (I) The graph depicts the percentage colocalization of LC3 and LAMP1 in WT, MYO5A KO, and rescue conditions (different MYO5A constructs). Data are plotted as mean ± SEM from three independent experiments (N = 3, 80 cells). Statistical significance was assessed by one-way ANOVA with Kruskal–Wallis post hoc test (WT vs. KO, **p = 0.0037, KO vs. FL, **p = 0.0062, ns non significant). Source data are available online for this figure.

However, whether this reflects a passive localization or an active scaffolding/docking role remained unresolved. To address this, we next tested whether the truncated constructs could rescue the autophagic defects observed in MYO5A KO cells (Fig. 5H). MYO5A KO cells, as expected, displayed reduced autolysosome numbers (Fig. 5H,I). Reintroduction of either full-length MYO5A or the CC–GTD fragment rescued this defect, whereas the M–IQ–CC1 fragment did not (Fig. 5H,I). Complementary PLA assays between LC3B and LAMP1 confirmed that the rescue reflected genuine autophagosome–lysosome fusion events rather than random colocalization (Fig. EV4d). Together, these experiments highlight the critical role of MYO5A, and specifically its CC–GTD region, in bridging autophagosomes with lysosomes to promote their fusion. To further test whether domains of MYO5A contribute to the engagement of the autophagosomes with the fusion machinery, we reintroduced full-length MYO5A or the CC–GTD fragment into MYO5A KO cells and examined the interactions of LC3 with the late fusion regulators STX17 and VPS33A (Fig. EV4E–G). Strikingly, while re-expression of full-length MYO5A restored colocalization of LC3 with both STX17 and VPS33A, expression of the CC–GTD fragment failed to rescue these interactions (Fig. EV4E–G). This indicated that although the CC–GTD is sufficient for targeting MYO5A to autophagic vesicles, the full-length protein is indispensable for coordinating interactions between LC3 and the fusion machinery, ensuring productive autophagosome–lysosome fusion. Thus, our combined genetic, pharmacological, and rescue approaches establish that MYO5A is recruited to autophagic vesicles via its C-terminal CC–GTD domain, which is sufficient for vesicle targeting. However, the presence of the full-length protein is necessary to couple vesicle recruitment with downstream engagement of the autophagosomal SNARE machinery. Given this, we next asked how MYO5A is recruited to autophagosomes at the molecular level. Different motor proteins and their adapters are known to interact with autophagosomes through LC3-interacting region (LIR) motifs (Olsvik et al, 2015; Cheng et al, 2016; Nieto-Torres et al, 2021).

MYO5A engages autophagosomes through conserved LIR motifs in its CC–GTD domain

The LC3-interacting region (LIR) motif is defined by a conserved sequence pattern, “[W/F/Y]xx[L/I/V],” which serves as a recognition site for LC3 proteins on the autophagosomal membrane (Johansen et al, 2017). To determine whether MYO5A harbors such motifs, we analysed its sequence using the iLIR database, focusing on predictions that were also supported by the ANCHOR algorithm (Kalvari et al, 2014). Surprisingly, four of the six predicted conserved LIR motifs corresponded to the coiled-coil domain and two were mapped to the globular tail domain of MYO5A (Fig. 6A). Among these, two motifs located within the CC–GTD region; PAYRVL (referred to as mutant P) and QAYIGL (referred to as mutant Q), were selected for further analysis, as this region, as per our experimental evidence, had previously emerged as critical for autophagosomal association and rescue of the fusion defect observed in MYO5A KO cells (Fig. 6A). To investigate their functional contribution, we generated single LIR mutants by substituting the conserved aromatic and hydrophobic residues with alanine. In addition, a double mutant (referred to as DM) was created by sequentially introducing both substitutions into the same construct, using the CC–GTD fragment as a template. These constructs, along with wild-type full-length GFP-MYO5A and mEmerald-CC–GTD, were expressed in MYO5A KO HeLa cells, together with mCh-LC3, and stained for LAMP1. Colocalization analysis revealed that all LIR mutants showed a significant decrease in overlap with LC3, whereas their colocalization with LAMP1 remained unaffected (Fig. 6B,C). This indicates that disruption of the LIR motifs selectively impairs the ability of MYO5A to associate with autophagosomes without altering its general distribution toward lysosomes. Expression of full-length MYO5A or CC–GTD significantly increased colocalization of mCh-LC3-positive autophagosomes with LAMP1-positive lysosomes, confirming their ability to complement the fusion defect in MYO5A KO cells (Fig. 6B,E). The P mutant retained this ability and behaved similarly to the wild-type CC–GTD (Fig. 6B,E). In contrast, the Q mutant displayed a strong reduction in LC3–LAMP1 colocalization, and the double mutant showed the most severe defect, failing to rescue the fusion phenotype (Fig. 6B,E). These observations suggested that the Q motif plays a more critical role than the P motif, and that combined disruption of both motifs exacerbates the defect, likely by impairing the ability of MYO5A to engage LC3.Figure 6LIR motifs in MYO5A mediate its interaction with autophagosomes and are required for autophagosome–lysosome fusion.(A) Schematic representation of the MYO5A protein. The black boxes indicate the positions of LIR motifs mapped to the coiled-coil and globular tail domains. The positional information of the LIR motifs was analyzed using the iLIR web tool and is depicted below. On the right, the conservation of the LIR motifs across organisms is shown. (B) MYO5A KO cells were transfected with GFP-MYO5A, mEmerald-CC–GTD, or mEmerald-tagged P, Q, and DM LIR mutants in the CC–GTD domain of MYO5A, and co-transfected with mCh-LC3. Zoomed insets highlight colocalization events between mCh-LC3 and MYO5A constructs, as well as between mCh-LC3 and LAMP1. Scale bar, 1 µm. The cell boundary is marked in white. White boxes within the image indicate inset regions, which are shown as magnified views in the rightmost corner to highlight colocalization events. Arrows mark additional points of colocalization. (C–E) Quantification of colocalization events between LC3 and MYO5A (one-way ANOVA with Kruskal–Wallis test; *p = 0.007, CC–GTD vs. Q; ****p < 0.0001) (C), LAMP1 and MYO5A (one-way ANOVA with Tukey’s post hoc test) (D), and LC3 and LAMP1 (E) (one-way ANOVA with Tukey’s post hoc test; *p = 0.0118, ****p < 0.0001) in MYO5A KO cells transfected with different LIR mutant constructs of MYO5A. Analysis was performed using the “Colocalization” plug-in of ImageJ. Data were plotted as mean ± SEM from three independent experiments (N = 3, n = 60 cells). (F) MYO5A KO cells were transfected with mEmerald-CC–GTD or CC–GTD domains harboring different LIR mutants (as indicated). Immunoprecipitation was performed using anti-GFP antibody, and lysates were probed for LC3 (N = 4). (G) Quantification of LC3 levels immunoprecipitated with GFP-tagged CC–GTD or LIR mutant constructs of MYO5A. Data were plotted as mean ± SEM from three independent experiments (N = 4). Statistical significance was assessed by one-way ANOVA with Tukey post hoc test. Source data are available online for this figure.

To probe this interaction biochemically, we performed immunoprecipitation of CC–GTD and its LIR variants, followed by LC3 detection (Fig. 6F,G). Both the Q mutant and the double mutant showed decreased LC3 co-precipitation compared to wild-type CC–GTD, with the double mutant exhibiting the strongest loss of interaction (Fig. 6F,G). Although experimental variability limited statistical significance, the trend was consistent across replicates. Together, these results demonstrate that MYO5A directly engages LC3 through its LIR motifs, and that this interaction is critical for mediating autophagosome–lysosome fusion. While the P motif appears partially dispensable, the Q motif plays a pivotal role, and combined disruption of both sites abolishes productive LC3 binding. This positions the LIR motifs as the molecular interface that allows MYO5A to tether to autophagosomes, thereby coupling its cargo-recognition function with motor activity to promote efficient autophagosome–lysosome fusion. Having established that MYO5A engages autophagosomes through conserved LIR motifs in its CC–GTD domain to facilitate their fusion with lysosomes, we next asked whether MYO5A might also play a role earlier in the pathway, specifically in regulating the dynamics of the autophagosome population.

MYO5A regulates the direction of autophagosomes

Initial colocalization analyses confirmed that MYO5A associates with autophagosomes, an interaction that becomes more pronounced under starvation (Fig. 5B,D). Having established the importance of this association, we next sought to identify the minimal region of MYO5A sufficient to mediate this interaction and, more importantly, to understand how it influences the dynamics of autophagosome–lysosome fusion. Our earlier experiments demonstrated that the CC–GTD domain alone could restore fusion defects in MYO5A KO cells, raising the question of how this domain regulates autophagosome behaviour at a mechanistic level (Fig. 5H,I). To explore this, we employed live-cell imaging assays to track the movement of autophagosomes and lysosomes under both WT and MYO5A KO conditions, as well as in the presence of different MYO5A domains (Fig. 7A). Autophagosomes were visualized using mCh-LC3B, while lysosomes were labeled with LysoTracker Deep Red (Fig. 7A). Quantitative analysis of organelle motility revealed no major differences in mean speed, total distance traveled, or displacement between WT and MYO5A KO cells (Fig. EV5A–c). Furthermore, reintroduction of full-length MYO5A or its CC–GTD or M-IQ-CC domains did not significantly alter these parameters, suggesting that MYO5A does not broadly regulate autophagosome or lysosome velocity (Fig. EV5A– C).Figure 7. Mean directional switches of autophagosomes are altered in the absence of MYO5A.(A) Time-lapse images of WT and MYO5A KO cells co-transfected with mCh-LC3 and different MYO5A constructs (as indicated), and stained with LysoTracker Deep Red dye. Scale bar, 1 μm. Arrow marks indicate events of colocalization. (B,** C**) Quantification of mean directional changes of LC3-positive autophagosomes (one-way ANOVA with Tukey’s post hoc test, *p = 0.0115 and ***p = 0.0003 and 0.0002) (B) and lysosomes (C) using the “TrackMate” plug-in of ImageJ. Data were averaged for each cell and plotted as mean ± SEM from three independent experiments (N = 3, n = 40 cells). Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc test (ns non significant). (D) Time-lapse images of MYO5A KO cells co-transfected with mCh-LC3 and different LIR mutant constructs of MYO5A (as indicated). Scale bar, 1 μm. Arrow marks indicate a lack of colocalization. (E) Quantification of mean directional changes of LC3-positive autophagosomes in MYO5A KO cells transfected with different LIR mutant constructs, analyzed using the “TrackMate” plug-in. Data are plotted as mean ± SEM from three independent experiments (N = 3, n = 40 cells). Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc test (**p = 0.0054, ***p = 0.0009 and ns non significant). Source data are available online for this figure.

However, a striking difference emerged when we analysed the mean directional changes of autophagosomes. In MYO5A KO cells, autophagosomes exhibited markedly higher frequencies of directional switches compared to WT cells (Fig. 7B). This phenotype was rescued by reintroducing either full-length MYO5A or the CC–GTD domain, which restored directional behaviour to near WT levels (Fig. 7B). Interestingly, expression of the M-IQ-CC construct did not rescue this phenotype; instead, it exaggerated directional switching in both WT and knockout cells (Fig. 7B). We speculate that the increased frequency of directional changes in MYO5A KO cells could reduce the likelihood of productive encounters between autophagosomes and lysosomes, thereby contributing to the fusion defects observed in them. In contrast, lysosomal motility parameters remained unaffected across all experimental conditions, consistent with our earlier findings that lysosome dynamics are largely independent of MYO5A activity (Fig. 7C).

To further dissect the molecular determinants of this regulation, we next examined the effects of LIR mutants of MYO5A on autophagosome dynamics. Live-cell tracking of mCh-LC3B revealed that expression of the CC–GTD domain, as well as the P and Q LIR mutants, was able to suppress excessive directional switching of autophagosomes in MYO5A KO cells, restoring their movement patterns to WT levels (Fig. 7D,E). By contrast, the double LIR mutant (DM) failed to rescue this phenotype, and autophagosomes in these cells continued to display increased directional switches (Fig. 7D,E). These findings suggest that while individual LIR motifs may contribute partially to MYO5A–LC3 interactions, their combined disruption severely impairs the ability of MYO5A to stabilize directional track changes of autophagosomes, reinforcing the critical role of LC3 binding in ensuring fusion-competent autophagosome dynamics. While these cellular studies establish the role of MYO5A in autophagic flux and enabling productive fusion with lysosomes, we next asked whether this function is conserved in a physiological setting. To this end, we turned to C. elegans as an in vivo model to test the broader relevance of MYO5A function in maintaining autophagic flux.

MYO5A regulates autophagic flux in vivo in C. elegans neurons

To complement our cellular analyses and validate the physiological relevance of MYO5A function, we next examined its role in an in vivo setting using C. elegans touch receptor neurons (TRNs). The neuronal cells, with its distinct spatial compartmentalization of autophagy, provides an especially sensitive context to dissect flux defects; autophagosomes are predominantly formed at distal neurite tips and are subsequently transported retrogradely towards the soma, where lysosomes are enriched (Maday and Holzbaur, 2016; Cason et al, 2022; Cason and Holzbaur, 2023). Disruption at different stages of autophagy is therefore predicted to produce distinct phenotypes; transport defects would manifest as autophagosome accumulation along the axon, whereas impaired fusion would preferentially lead to build-up at the cell body.

To investigate this, we examined autophagic flux in C. elegans touch receptor neurons (TRNs) of hum-2(ok596), a hypomorphic mutant carrying a 2.1 kb deletion in the hum-2 (C. elegans ortholog of MYO5A) locus. We visualized autophagosomes using the LC3 ortholog reporter mNG::lgg-1. In WT animals, basal autophagy at the L4 stage was characterized by sparse mNG::lgg-1 puncta in distal and proximal regions of the TRN, with occasional puncta observed at the soma (Fig. EV6A–C). In contrast, the dynein mutant dhc-1(js319), which is defective in retrograde transport, displayed robust accumulation of puncta across all compartments; synapse, process, and cell body, confirming its role as a transport-defective control (Fig. EV6A–C). Interestingly, hum-2(ok596) animals showed a distinct distribution pattern: puncta numbers in distal and proximal processes were comparable to WT, yet there was a pronounced accumulation of autophagosomes in the soma (Fig. EV6A–E). This phenotype suggested that hum-2 loss does not disrupt biogenesis or long-range transport, but instead impairs later stages of autophagic flux, most likely at the step of autophagosome–lysosome fusion. We next asked whether this phenotype became more apparent under conditions of elevated autophagic demand. Upon blue-light stimulation, a paradigm that induces autophagy, dhc-1(js319) mutants displayed a global increase in puncta across all neuronal compartments (Fig. 8A–G). hum-2(ok596) animals, however, exhibited a strikingly compartment-specific response; while soma-associated puncta did not increase significantly beyond WT, the proximal neuronal processes showed accumulation of autophagosomes (Fig. 8A–G). This pointed towards a compartmentalized requirement of MYO5A for flux completion. To assess whether these defects were exacerbated with aging, we analysed 7-day-old worms. As expected, basal autophagosomal numbers increased across all genotypes, reflecting age-associated decline in autophagic efficiency (Figs. 8H,I and EV6F–H). Yet hum-2(ok596) worms showed significantly higher accumulation of puncta in proximal processes compared to age-matched WT, highlighting that MYO5A deficiency exacerbates age-associated autophagic decline (Figs. 8H,I and EV6F–H). To rule out gross lysosomal dysfunction as the underlying cause, we examined endo-lysosomal compartments using CTNS-1::mCh as a lysosomal marker (Fig. 8J). The number, distribution, and soma-enrichment of lysosomes were indistinguishable between hum-2 mutants and WT, arguing against lysosomal biogenesis or positioning defects (Fig. 8J). In addition to autophagosomal distribution, we assessed organismal consequences of hum-2 loss. hum-2(ok596) animals exhibited reduced brood size and a significant decline in survival during late adulthood compared to WT animals, suggesting that compromised autophagic flux has physiological consequences at the organismal level (Fig. EV6I–K).Figure 8. Loss of hum-2 results in the accumulation of autophagosomes in the cell body of C. elegans TRNs.(A) Schematic of the C. elegans PLM neuron. Black boxes indicate the imaged regions. (B) Schematic of the experimental paradigm for neuronal stimulation using blue light. Worms were subjected to 70 V blue-light pulses intermittently (5 s on, 5 s off) for 1 h, followed by a 45 min recovery before imaging for mNG::LGG-1 at different neuronal regions. (C) Representative images of mNG::LGG-1 in the cell body, proximal process, process, branch point, and synapse of PLM neurons in WT, hum-2(ok596), and dhc-1(js319) L4 worms after blue-light stimulation. Scale bar, 10 μm. quantification of mNG::LGG-1 puncta in (D) cell body (one-way ANOVA with Kruskal–Wallis’s test; *p = 0.0194), (E) proximal process (one-way ANOVA with Kruskal–Wallis’s test; **p = 0.0085 and ****p < 0.0001), (F) process (one-way ANOVA with Kruskal–Wallis’s test; ****p < 0.0001), and (G) synapse (one-way ANOVA with Kruskal–Wallis’s test; ****p < 0.0001). Graphs represent mean ± SEM from three independent experiments (N = 3, n = 30 worms). (H) representative images of mNG::LGG-1 in the same neuronal regions in WT, hum-2(ok596), and dhc-1(js319) 7-day adult worms. Scale bar, 10 μm. (I) quantification of mNG::LGG-1 puncta in the proximal process, plotted as mean ± SEM from three independent experiments (N = 3, n = 30 worms). Statistical significance was assessed by one-way ANOVA with Kruskal–Wallis’s test; *p = 0.012 and ****p < 0.0001). (J) CTNS-1::mCh in the cell body and proximal process of PLM neurons in WT and hum-2(ok596) L4 worms. Scale bar, 10 μm. Quantification of CTNS-1::mCh distribution length from the cell body using the “Measure” plug-in of ImageJ, plotted as mean ± SEM from three independent experiments (N = 3, n = 30 worms). Statistical significance was assessed by Student’s t-test (ns non significant). (K) Kymographs of mNG::LGG-1 vesicles in the proximal process of PLM neurons were generated using the “Multiple Kymograph” plug-in of ImageJ. Kymographs were annotated and analyzed with the “KymoAnalyzer” plug-in. Arrow marks indicate stationary tracks. (L) Quantification of motion trajectories of mNG::LGG-1 puncta (anterograde, retrograde, stationary) from three independent experiments (N = 3, n = 30 worms). Statistical significance was assessed by Mann–Whitney Student’s t-test (*p = 0.0112). Scale bar, 10 μm (horizontal), 180 s (vertical). Source data are available online for this figure.

Finally, to test whether our live-cell findings on autophagosomal motility extended to an in vivo context, we conducted time-lapse imaging of mNG::lgg-1 puncta in the proximal processes of 7-day-old worms (Fig. 8K,L). Kymograph analysis revealed that the proportions of retrograde and anterograde moving autophagosomes were unchanged in hum-2 mutants compared to WT (Fig. 8K,L). However, the fraction of stationary puncta was significantly reduced in hum-2 mutants, paralleling our earlier observations that MYO5A loss increases directional switching (Fig. 8K,L). This suggests that MYO5A does not control overall motility per se but instead contributes to stabilizing a stationary autophagosomal pool, a property that may be essential for productive fusion with lysosomes. Together, these in vivo findings in C. elegans reinforce the central conclusion that MYO5A is required for efficient autophagic flux. Across yeast, mammalian cells, and C. elegans, MYO5A deficiency consistently leads to autophagosome accumulation, impaired cargo clearance, and a loss of positional stability of autophagosomes. Importantly, in neurons, this manifests as compartment-specific buildup of autophagosomes, with the strongest effects seen in proximal processes and soma, and becomes more severe with age. These data establish MYO5A as a critical regulator of autophagosome–lysosome fusion.

Discussion

Class V myosins are multifunctional motors that couple actin filaments to intracellular cargo transport and organelle positioning (Hammer and Sellers, 2012; Hammer and Wagner, 2013). Among them, MYO5A has been extensively studied for its role in long-range vesicle trafficking, organelle distribution, and local tethering at specialized subcellular domains (Fagarasanu et al, 2009; Donovan and Bretscher, 2012; Pan et al, 2023). Here, we identify a previously unrecognized role for MYO5A in autophagy, i.e., facilitating the transition of autophagosomes from sealed but fusion-incompetent vesicles to fusion-competent organelles. This finding extends the functional repertoire of MYO5A into autophagic flux and highlights a critical stage-specific requirement for actomyosin networks in autophagy.

Autophagy is not a linear process of vesicle movement; it requires timely pauses and stabilization of autophagosomes in order to engage with lysosomes. Our results demonstrate that loss of MYO5A diminishes stationary pools of autophagosomes and leads to the accumulation of sealed but fusion-incompetent vesicles. These observations support a model in which MYO5A acts as a regulatory brake, reducing directional switches and enforcing pausing at the fusion stage. Such a requirement for vesicle stalling has been observed in other contexts, wherein, dendritic lysosomes pause at F-actin patches near glutamatergic synapses; HDAC6-dependent actin remodeling is essential for autophagosome maturation; INPP5E stabilizes actin on lysosomes to ensure fusion competence; and detyrosinated microtubules provide stable tracks that constrain lysosomes to facilitate fusion (Lee et al, 2010; Hasegawa et al, 2016; Mohan et al, 2019; Van Bommel et al, 2019). Our findings thus place MYO5A within a broader paradigm where cytoskeletal regulation of pausing and positioning is indispensable for vesicle fusion.

Mechanistically, our data suggest that MYO5A acts as a stabilizing scaffold that enables autophagosomes to acquire fusion competence. Loss of MYO5A compromised the recruitment of key fusion factors, including VPS33A and STX17, whereas reintroduction of full-length MYO5A restored their localization, pointing to a direct role for MYO5A in organizing the fusion machinery. The presence of multiple LC3-interacting motifs in MYO5A likely provides the molecular basis for this interaction, coupling autophagosome docking with the assembly of tethering and SNARE complexes. Extending these findings to an in vivo context, we observed that MYO5A deficiency in C. elegans neurons led to an accumulation of autophagosomes in the soma, where lysosomes are abundant. This phenotype highlights the physiological relevance of MYO5A in ensuring that autophagosomes not only reach lysosome-rich regions but also undergo productive fusion to maintain autophagic flux. Although our data demonstrate that loss of function of MYO5A disrupts autophagic flux across S. cerevisiae, mammalian cells, and C. elegans, it remains unclear whether the underlying mechanism of inhibition is conserved across these systems. For instance, in yeast, Myo2 has been implicated in the delivery of Atg9 vesicles and in Atg11-dependent cargo transport during selective autophagy (Kumar et al, 2016). However, whether Myo2 directly interacts with Atg8 remains to be determined. It would be worthwhile to test if Myo2–Atg8 interactions occur in yeast, whether these interactions are compromised in temperature-sensitive myo2 mutants, and if they are mediated through LIR motifs analogous to those in MYO5A in mammalian cells. Such analyses would clarify whether LIR-dependent engagement of class V myosins with autophagosomes is an evolutionarily conserved mechanism or a specialized adaptation in higher eukaryotes.

A limitation of our study is that we cannot yet resolve whether the stationary state of autophagosomes imposed by MYO5A is itself required for the loading of fusion machinery, or whether pausing and machinery recruitment occur as independent steps. Although our rescue experiments suggest that reintroduction of full-length MYO5A restores VPS33A and STX17 localization to autophagosomes, the causal relationship between pausing and fusion machinery assembly remains unresolved. Dissecting this temporal sequence upon acute perturbation of MYO5A-actin interactions by employing high-resolution live-imaging methods will provide more mechanistic details. Further investigations are needed to understand the exact process by which MYO5A recruits and stabilizes the fusion complex. Fine mapping of the interaction of MYO5A with the fusion machinery, as well as spatio-temporal resolution of the process, is required to elucidate the sequence of events that precede autophagosome–lysosome fusion. Together, our findings support a model in which MYO5A promotes the transition of sealed autophagosomes into a fusion-competent state by stabilizing their stationary pool and enabling recruitment of the fusion machinery. The presence of multiple LC3-interacting motifs within the CC–GTD domain suggests a direct mechanism for autophagosome recognition, while the motor region of MYO5A may plausibly contribute to tethering events. Future work employing ultrastructural analyses such as transmission electron microscopy (TEM) and super-resolution fluorescence microscopy will be instrumental in validating the spatial organization of MYO5A-associated autophagosomes and further refining the mechanistic framework proposed here. Our study thus, highlights a new layer of regulation in autophagic flux and raises the possibility that dysregulation of MYO5A contributes to diseases marked by impaired autophagosome–lysosome fusion, including neurodegenerative disorders.

Methods

Reagents and tools tableReagent/resourceReference or sourceIdentifier or catalog number Experimental models S. cerevisiae strain details used in the studyBY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::MYO2 [MYO2 in pRS306]This studysAN01BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::myo2-12 [myo2-12 in pRS306]This studysAN02BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::myo2-13 [myo2-13 in pRS306]This studysAN03BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::myo2-14 [myo2-14 in pRS306]This studysAN04BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::myo2-16 [myo2-16 in pRS306]This studysAN05BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::myo2-17 [myo2-17 in pRS306]This studysAN06BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; GFP-Atg8::HIS [pGFP-Atg8 in pRS423]; URA::myo2-18 [myo2-18 in pRS306]This studysAN07BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; POT1::POT1-GFP-HIS; URA::MYO2 [MYO2 in pRS306]This studysAN08BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; POT1::POT1-GFP-HIS; URA::myo2-16 [myo2-16 in pRS306]This studysAN09BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; POT1::POT1-GFP-HIS; URA::myo2-12 [myo2-12 in pRS306]This studysAN10BY4741; his3Δ1 leu2Δ0 ura3Δ0 met15Δ0; POT1::POT1-GFP-HIS; URA::myo2-18 [myo2-18 in pRS306]This studysAN11C. elegans strain details used in the studyN2Koushika Lab, TIFRsAN46 hum-2(ok596) V Koushika Lab, TIFRTT184 dhc-1(js319) I Koushika Lab, TIFRTT1040 tbIs381[mec-4p::ctns-1::mCherry]X (Nadiminti et al, 2024)TT2896 pwSi190[mec-7p::mNeonGreen::lgg-1c::let858];hygR Koushika Lab, TIFRTT3027 dhc-1 (js319) I; pwSi190[mec-7p::mNeonGreen::lgg-1c::let858];hygR This studyTT3374 hum-2 (ok596) V; pwSi190[mec-7p::mNeonGreen::lgg-1c::let858];hygR This studyTT3393 hum-2 (ok596) V; tbIs381[mec-4p::ctns-1::mCherry]X This studyTT3393 Cell lines used in the study HeLa; Cervical adenocarcinoma; Human (tested for mycoplasma)ATCC-CCL-2HSN code; 30019099 Details of plasmid constructs used in the study myo2-12 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN01myo2-13 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN02myo2-14 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN03myo2-16 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN04myo2-17 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN05myo2-18 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN06MYO2 in pRS306Prof. Anthony Bretscher’s Laboratory, Cornell UniversitypAN07GFP-Atg8 in pRS423Cloned in Autophagy Laboratory, JNCASRpAN08ptfLC3#21074 (Addgene), deposited by Tamotsu YoshimoripAN09EGFP-LC3#21074 (Addgene), deposited by Tamotsu YoshimoripAN10mCherry-LC3Dr. Terje Johansen, Norwegian University of Science and Technology – NTNU, NorwaypAN11pCI-neo-HA-hApg12#22950 (Addgene) deposited by Noboru MizushimapAN12FLAG-Stx17#45911 (Addgene), deposited by Noboru MizushimapAN13FLAG-SNAP29#45915 (Addgene), deposited by Noboru MizushimapAN14pEGFP-C1-hApg5#22952 (Addgene), deposited by Noboru MizushimapAN15mCherry-DFCP1#86746 (Addgene), deposited by Do-Hyung KimpAN16LAMP1-GFP#16290 (Addgene), deposited by Ron ValepAN17HA-VPS33ADr. Mahak Sharma, Indian Institute of Science Education and Research (IISER), Mohali, ChandigarhpAN18GFP-VPS33ADr. Mahak Sharma, Indian Institute of Science Education and Research (IISER), Mohali, ChandigarhpAN19GFP-MYO5ADr. Zhiyi Wei, Academy for Advanced Interdisciplinary Studies, Southern University of Science and Technology, Shenzhen, Guangdong, China.pAN20mEmerald-CC–GTDDr. Marina Mikhaylova, ZMNH, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.pAN21mEmerald-M-IQ-CCDr. Marina Mikhaylova, ZMNH, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.pAN22 Reagents and chemicals used in the study Dulbecco’s Modified Eagle Medium (DMEM)Sigma-AldrichD5648Sodium bicarbonateSigma-AldrichS5761Fetal bovine serumGibco10270-106PenstrepGibco15140-122Dulbecco’s Phosphate Buffered Saline (DPBS)Sigma-AldrichD5773Earle’s Balanced Salt Solution (EBSS)Sigma-AldrichE7510Lipofectamine 2000Invitrogen11668-01935 mm Live-cell imaging dishCellvisD35-10-1.5-NECL substrateBio-Rad170-5061siMYO5ASigma-AldrichEHU065861SCR ControlSigma-AldrichSIC001LysoTracker™ Deep RedInvitrogenL12492Duolink PLA kitSigma-AldrichDUO92008Triton X-100Millipore648466Protease inhibitor cocktailSigma-AldrichCOEDTAF-ROPhenylmethylsulfonyl fluoride (PMSF)Sigma-AldrichP7626Dynabeads™ Protein GInvitrogen10004DHuman EGF Recombinant ProteinGibcoPHG0311Proteinase KSigma-Aldrich70663CellTiter-Glo®PromegaG7570TetramisoleSigma-AldrichT1512Bafilomycin A1Sigma-AldrichB1793PentabromopseudilinProf. Hans-Joachim Knolker’s lab Antibodies used in the study MYO5ACell Signaling Technology (CST)#3402ATG9AInvitrogen#MA5-32238LAMP1Cell Signaling Technology (CST)#9091FIP200Cell Signaling Technology (CST)#12436p62Abcamab56416LC3BSigma-AldrichL7543ATG5Cell Signaling Technology (CST)#12994ATG16L1Cell Signaling Technology (CST)#8089FlagSigma-AldrichF1804HASigma-AldrichH9658RAB7ACell Signaling Technology (CST)#9367STX17Sigma-AldrichHPA001204ATG2ACell Signaling Technology (CST)#15011GAPDHInvitrogen#MA5-15738GFPRoche, Sigma-Aldrich11814460001EGFRSanta Cruz Biotechnologysc-03WIPI2Abcamab105459Atto 633 (goat anti-rabbit IgG)Sigma-Aldrich41176Atto 488 (goat anti-rabbit IgG)Sigma-Aldrich41057Atto 550 (goat anti-rabbit IgG)Sigma-Aldrich43328Atto 550 (goat anti-mouse IgG)Sigma-Aldrich43394Atto 633 (goat anti-mouse IgG)Sigma-Aldrich78102Goat anti-mouse HRP antibodyBio-Rad172-1011Goat anti-rabbit HRP antibodyBio-Rad172-1019 Other GraphPad Prism (version 9.5.1)

Yeast strains and media

The yeast strains used in this study were derived from the BY4741 background and included wild-type (WT) and atg1Δ strains, obtained from EUROSCARF. Detailed information about these strains is summarized in the Reagents and tools table. Yeast cells were cultured in various media: YPD medium (consisting of 1% yeast extract, 2% peptone, and 2% dextrose), SD-X drop-out medium (comprising 0.17% yeast nitrogen base, 0.5% ammonium sulfate, 2% dextrose, 0.002% uracil, 0.02% histidine, 0.02% methionine, 0.015% lysine, and 0.01% leucine, where “X” denotes the amino acid excluded for selection), and SD-N starvation medium (containing 0.17% yeast nitrogen base and 2% dextrose). Transformation of strains was performed using the lithium acetate-PEG method as described by Gietz and Woods (2002), with temperature-sensitive (ts) strains incubated overnight at 25 °C in the transformation mix to ensure optimal recovery (Gietz and Woods, 2002). Plasmid constructs of Myo2 temperature-sensitive mutants, kindly provided by Prof. Anthony Bretscher’s Laboratory (Cornell University), were linearized using SpeI restriction digestion before transformation into WT cells (Schott et al, 1999). For strains used in the GFP-Atg8 processing assay, WT cells were initially transformed with myo2 ts mutant plasmid constructs, followed by transformation with GFP-Atg8 plasmid constructs in the pRS423 backbone. WT Pot1-GFP strains, genomically tagged with GFP at the C-terminus of Pot1, were obtained from Prof. Rachubinski (University of Alberta, Canada). For the Pot1-GFP assay, WT Pot1-GFP strains were transformed with myo2 ts mutant plasmids following SpeI digestion. All strain and plasmid details are provided in the Reagents and tools table.

GFP-Atg8 processing assay

GFP-Atg8 processing assay was performed for assessing general autophagy (Klionsky et al, 2021). Cells expressing GFP-Atg8, transformed with the GFP-Atg8 construct in the pRS423 backbone, were cultured in SD-His medium. Log-phase cells were harvested, pelleted, and washed twice with sterile water. Subsequently, the cells were resuspended in starvation medium (SD-N) at a concentration of 3 OD/ml to induce autophagy. The 0-hour timepoint was collected immediately after the washes. The culture was then divided into two tubes: one incubated under permissive temperature (PT, 25 °C) and the other under non-permissive temperature (NPT, 36 °C). Cells were collected at the designated time points, and proteins were precipitated using trichloroacetic acid (TCA). The precipitated samples were analysed by SDS-PAGE, followed by western blotting to assess GFP-Atg8 processing.

Pexophagy assay

Wild-type yeast cells expressing Pot1-GFP were transformed with linearized plasmids digested with SpeI enzyme, as detailed in Reagents and Tools table and Reagents and Tools table. The transformed cells were initially grown in YPD medium. Actively growing Pot1-GFP expressing cells were then transferred to an oleate medium, with a composition of 2.64 mM K_₂_HPO_₄_, 17.36 mM KH_₂_PO_₄_, pH 6.0, 0.1% oleic acid, 0.5% Tween-40, 0.25% yeast extract, and 0.5% peptone at an optical density (A_600_) of 1 to induce peroxisome biogenesis for 14–16 h. After induction, cells were washed twice with sterile water and resuspended in SD-N medium at an A_600_ of 3 to induce pexophagy. The culture was divided into two tubes of permissive temperature (PT, 25 °C) and non-permissive temperature (NPT, 36 °C) for incubation. Samples were collected at specified time points, proteins were precipitated using trichloroacetic acid (TCA), and the processed samples were subjected to western blot analysis.

Western blotting

Cell pellets equivalent to 3 OD A_600_ units, collected for immunoblotting, were resuspended in 400 µl of 12.5% trichloroacetic acid (TCA) by vortexing. The resuspended pellets were stored overnight at −80 °C. The following day, the pellets were thawed on ice and centrifuged at 16,000×g for 10 min. The supernatant was discarded, and the pellets were washed twice with 80% acetone and air-dried. The dried pellets were resuspended in 40 µl of a solution containing 1 N NaOH and 1% SDS, followed by the addition of 10 µl of 5X Laemmli buffer. Samples equivalent to 0.6 OD A_600_ units were loaded onto SDS-PAGE gels and transferred to PVDF membranes using the wet transfer method. Mouse anti-GFP monoclonal antibody (Roche Applied Sciences) was used as the primary antibody for probing GFP at a dilution of 1:3000. Secondary antibody, goat anti-mouse HRP (Bio-Rad) was used at a dilution of 1:10,000.

Cell culture

HeLa cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 3.7 g/L sodium bicarbonate, 10% fetal bovine serum, and 100 units/ml penicillin-streptomycin. The cells were maintained at 37 °C in a humidified incubator with 5% CO₂. Autophagy was induced by washing the cells with Dulbecco’s Phosphate Buffered Saline and incubating them in starvation medium comprising Earle’s balanced salt solution (EBSS) prepared at 1X and supplemented with 2.2 g/L sodium bicarbonate for 2 h or as specified. To inhibit autophagy, cells were treated with Bafilomycin A1 (BafA1) at a working concentration of 100 nM for 2 h. For Pentabromopseudilin (PBP) treatment, cells were incubated for 2 h with a working concentration of 1 µM.

CRISPR-Cas9 MYO5A knockout line

The generation of a MYO5A KO HeLa cell line was accomplished using the CRISPR-Cas9 gene-editing system (Dance, 2015; Adli, 2018). To achieve this, a low passage HeLa cells at 60% confluency were transfected with a customized CRISPR vector plasmid (Sigma-Aldrich, CMV-eCas9-2a-tGFP), which was designed to express the CRISPR-associated protein eCas9 along with a guide RNA (gRNA) sequence. The gRNA, LIR, specifically targeted exon 17 of the MYO5A gene to induce gene disruption. Post-transfection, cells expressing GFP, a marker co-expressed from the plasmid, were identified and isolated as single cells using fluorescence-activated cell sorting (FACS). These sorted single cells were seeded into 96-well plates, each well receiving one cell, to establish clonal cell lines. Clonal lines negative for KO were used as the control cell line. Following colony expansion, the successful knockout of MYO5A was validated through two independent methods: western blotting, which confirmed the absence of the MYO5A protein, and immunofluorescence, which corroborated the loss of MYO5A expression at the cellular level. These analyses collectively ensured the generation of a robust MYO5A KO HeLa cell line for downstream studies.

siRNA-mediated genetic knockdown

To transiently suppress MYO5A expression, small interfering RNA (esiRNA) specific to human MYO5A (siMYO5A, EHU065861) was utilized. A negative control esiRNA (SCR Control, SIC001), which lacks homology to any known human gene, was used to account for non-specific effects. Both esiRNAs were obtained from Sigma-Aldrich. HeLa cells were transfected with these esiRNAs using Lipofectamine™ 2000 transfection reagent following the manufacturer’s protocol. Transfected cells were incubated under standard culture conditions, and samples were collected at 48 h post-transfection to allow for effective knockdown of MYO5A. The collected cells were then processed for downstream analyses, including western blotting to confirm reduced protein expression levels and immunofluorescence to visualize and quantify MYO5A suppression at the cellular level. These experiments validated the efficacy of the esiRNA-mediated knockdown approach for studying the functional consequences of MYO5A depletion.

LIR mutants were generated by substituting the conserved aromatic residue (W/F/Y) and the non-polar aliphatic residue (L/I/V) within the identified motifs with alanine using site-directed mutagenesis, with mEmerald–MYO5A CC–GTD as the template. The primers used were ∆LIR1 FP (5′-GGTGAAATAGCACAAGCAGCGATTGGTGCGAAAGAAACAAACAGGCTC-3′) and ∆LIR1 RP (5′-GAGCCTGTTTGTTTCTTTCGCACCAATCGCTGCTTGTGCTATTTCACC-3′), and ∆LIR2 FP (5′-GCGCCAGGTGCGCCTGCTGCGCGAGTCGCGATGGAACAGCTGACCTC-3′) and ∆LIR2 RP (5′-GAGGTCAGCTGTTCCATCGCGACTCGCGCAGCAGGCGCACCTGGCGC-3′).

∆LIR1 primers were used on the ∆LIR2 mutant construct to generate the double mutant. All constructs were verified by Sanger sequencing.

Immunofluorescence

HeLa cells were seeded onto coverslips placed in a 12-well plate. Transfection was performed the following day when cells reached 60–70% confluency, using Lipofectamine 2000 following the manufacturer’s protocol. Briefly, Lipofectamine 2000 reagent and DNA were diluted separately in 60 µl of OPTI-MEM (for four wells) at a DNA:Lipofectamine ratio of 1:2, with a ratio of 2:5 used for transfection of MYO5A constructs. After a 5-min incubation, the diluted DNA was combined with the diluted Lipofectamine 2000 reagent in a 1:1 ratio and incubated for 20 min to form DNA-Lipofectamine complexes. These complexes were then added to cells at 30 µl per well. After 48 h of transfection, cells were treated as indicated with either EBSS, Bafilomycin A1 (BafA1), or Pentabromopseudilin (PBP) for 2 h, or left untreated. Following treatment, cells were thoroughly washed with DPBS, fixed with 4% paraformaldehyde for 15 min, and permeabilized with 0.25% Triton X-100 for 10 min. Cells were incubated overnight at 4 °C with the appropriate primary antibody, followed by washes with PBS (three washes, 10 min) to remove excess antibody. Coverslips were then incubated with the corresponding fluorescent secondary antibody for 1 h at room temperature. Finally, coverslips were mounted using Vectashield antifade reagent (H-1000/H-1200, Vector Laboratories) following PBS washes (three washes, 10 min). Imaging was performed using a DeltaVision Elite fluorescence microscope (Leica Microsystems) equipped with an Olympus 60X/1.42 Plan ApoN objective and excitation/emission filters for Cy5, FITC, DAPI, and TRITC, utilizing a polychroic Quad filter set.

Live-cell imaging

Live-cell imaging was conducted using cells seeded onto a 35-mm live-cell imaging dish, followed by transfection as described previously. Imaging was performed on a DeltaVision Elite fluorescence microscope (Leica Microsystems) equipped with an inverted Olympus 60X/1.42 NA objective and DAPI, FITC, TRITC, and Cy5 filters. For imaging cells transfected with mCh-LC3, LAMP1-GFP, and stained with Lysotracker Deep Red dye, time-lapse imaging was carried out at 20-s intervals. For live imaging of stage-specific (transfections performed as indicated) markers, time-lapse imaging was performed at 13-s intervals. To analyse the dynamics of autophagosomes and lysosomes in WT and MYO5A KO cells expressing various MYO5A constructs, time-lapse imaging was conducted at 14-s intervals.

Lysotracker staining

Cells were seeded at approximately 50% confluency in appropriate culture dishes. To visualize lysosomal compartments, cells were incubated with LysoTracker™ Deep Red. The dye was used at a working concentration of 100 nM, and the cells were incubated for 20 min at 37 °C. Following staining, the cells were washed with DPBS and proceeded for live-imaging fluorescence microscopy as indicated.

Proximity ligation assay

To detect and quantify protein interactions or modifications, the Duolink® PLA fluorescence protocol was used. Briefly, pretreated samples were blocked with Duolink® Blocking Solution at 37 °C for 1 h. Primary antibodies targeting the proteins of interest were then diluted in Antibody Diluent and incubated with the samples in a humidity chamber. Following washing, PLUS and MINUS PLA probes (specific to the species of the primary antibodies) were added and incubated for 1 h at 37 °C. Subsequently, ligation and amplification steps were performed using reagents prepared according to the manufacturer’s protocol. The ligation solution, containing ligase, was incubated with the samples for 30 min at 37 °C, followed by amplification using polymerase for 100 min under identical conditions. After a series of washes to remove excess reagents, the samples were mounted with Duolink® PLA Mounting Medium containing DAPI. Fluorescence signals were visualized using a DeltaVision Elite fluorescence microscope (Leica Microsystems) equipped with an Olympus 60X/1.42 Plan ApoN objective and excitation/emission filter for TRITC, utilizing a polychroic Quad filter set. Images were further analysed to detect and quantify PLA signals using the “Analyze Particles” plugin of FIJI ImageJ (NIH).

Western blotting

Following the indicated treatments, ~0.8 million cells from each well of a 6-well dish were washed with ice-cold PBS. Cells were lysed in 100 μl of sample buffer containing 10% (w/v) SDS, 10 mM DTT, 20% (v/v) glycerol, 0.2 M Tris-HCl (pH 6.8), and 0.05% (w/v) bromophenol blue. The lysates were collected using a rubber cell scraper, boiled at 99 °C for 10 min, and stored at −80 °C until further use. Immunoblotting was performed using standard protocols. Protein transfer to membranes was carried out via the wet transfer method. MYO5A probing was performed on proteins separated on a 6% SDS-PAGE gel run at 40 V, followed by wet transfer at room temperature. Blots were incubated overnight with the appropriate primary antibodies, as specified. Secondary antibodies, goat anti-mouse (Bio-Rad) or goat anti-rabbit (Bio-Rad) conjugated to HRP, were used at a 1:10,000 dilution. Signal development was performed using the ECL substrate (Bio-Rad), and images were acquired using the auto-capture or series-capture settings of the Syngene G Gel Documentation System (UK) followed by quantification of band intensities using FIJI ImageJ software (NIH).

Immunoprecipitation

After transfection or treatment, whole-cell lysates were prepared using RIPA buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 0.5% Triton X-100; and 1 mM EDTA), supplemented with a protease inhibitor cocktail (Sigma) and 1 mM PMSF (Sigma) to prevent protein degradation. The lysates were homogenized and subsequently centrifuged at 12,000×g for 30 min at 4 °C to separate insoluble debris. The resulting supernatants, containing soluble proteins, were collected for further analysis. For immunoprecipitation, the supernatants were incubated overnight at 4 °C with specific primary antibodies and protein G magnetic dynabeads to capture target proteins. The bead-antibody-protein complexes were washed thrice with ice-cold PBS to remove non-specifically bound proteins. Bound proteins were then eluted by boiling the beads in 1x SDS loading buffer for 10 min at 99 °C. The eluted samples were subjected to western blotting using standard methods.

EGFR trafficking

HeLa cells were seeded into six-well plates at ~60% confluency to ensure optimal conditions for experimental treatments. MYO5A activity was inhibited through two approaches: genetic silencing using siRNA transfection to achieve transient knockdown and pharmacological inhibition via PBP treatment at a concentration of 1 μM for 2 h. On the day of the assay, the cells were gently washed with PBS to remove residual media and subsequently starved in Earle’s balanced salt solution (EBSS) for 3 h. To assess EGFR degradation, the cells were stimulated with a pulse of epidermal growth factor (EGF) at a concentration of 100 ng/ml. Samples were collected at 20-min intervals over a duration of 100 min. The collected samples were then processed for Western blotting.

Protease protection assay

HeLa cells were seeded into six-well plates and subjected to specific treatments, including chemical inhibitors (Bafilomycin A1 and/or PBP) or transfection with either control siRNA (SCR) or MYO5A-specific siRNA. Following treatment, the cells were washed twice with ice-cold PBS, harvested, and homogenized in a buffer containing 20 mM HEPES (pH 7.5), 220 mM mannitol, 70 mM sucrose, and 1 mM EDTA. Homogenization was achieved by passing the cell suspension 45 times through a 26-gauge syringe needle to ensure thorough cell lysis. The homogenates were centrifuged at 500×g for 10 min at 4 °C to separate nuclei and unlysed cells. The resulting supernatants were divided into three aliquots and incubated on ice for 30 min under the following conditions: untreated (control), treated with 100 μg/ml proteinase K, or treated with 100 μg/ml proteinase K and 0.5% Triton X-100 in a total volume of 1 ml. Proteolysis reactions were terminated by adding 1 mM PMSF and incubating the samples on ice for an additional 10 min. Subsequently, proteins were precipitated by adding 250 μl of trichloroacetic acid (TCA) and incubating on ice for 10 min. The precipitated proteins were washed twice with 200 μl of acetone to remove residual TCA. The resulting protein pellets were resuspended in 1x SDS loading buffer and heated at 99 °C for 10 min. The processed samples were then subjected to SDS-PAGE for downstream analysis.

C. elegans maintenance and culturing

C. elegans strains were cultured and maintained at 20 °C on nematode growth media (NGM) plates seeded with E. coli OP50 as the food source. The preparation and use of NGM plates followed standard protocols (Stiernagle, 2006). Worm populations were regularly passaged to fresh plates to prevent overcrowding and maintain healthy cultures.

NGM plate preparation

NGM plates were prepared using standard recipes (Stiernagle, 2006), which included agar, peptone, NaCl, and cholesterol, supplemented with KPO_4_, CaCl_2_ and MgCl_2_ to maintain optimal growth conditions. Plates were poured aseptically and allowed to solidify, followed by overnight bacterial lawn seeding with E. coli OP50 prior to use.

Strain information and imaging

For all experiments, imaging was performed on L3 and L4 larval stage animals or adults aged 7 days post-hatching to ensure uniformity in developmental and aging-related analyses. The strains used in this study, along with their relevant genotypes, are detailed in the Reagents and Tools Table. For imaging experiments, specific developmental stages of C. elegans (as described in individual experiments) were immobilized using 3 mM tetramisole to ensure minimal movement during microscopy. Immobilized worms were carefully mounted on 2–5% agarose pads prepared on glass slides to provide a stable imaging platform. Imaging was conducted using an Olympus IX73 Epifluorescence microscope equipped with a Photometrics Evolve 512 EMCCD camera. The microscope was illuminated with a 120 W X-cite mercury-arc lamp to provide consistent fluorescence excitation. A 100×/1.4 NA DIC oil immersion objective lens was used, offering high resolution with a pixel size of 158 nm per pixel

Brooding assay

To assess reproductive output, ten individual L4-stage C. elegans (N2 and hum-2(ok596) strains) were used per experiment. Each worm was placed on a separate 35 mm NGM plate seeded with E. coli OP50. After 10 h, the worms were transferred to fresh plates to ensure proper separation of embryos laid during each interval. Transfers were performed twice daily to freshly prepared plates to maintain optimal conditions for egg-laying and to prevent mixing of developmental stages. This procedure was repeated for ~7 days, until the animals ceased laying eggs. Embryos laid on each plate were counted immediately after transferring the mother worm to a new plate, providing a detailed record of reproductive output over the course of the assay.

Lifespan assay

To assess lifespan, gravid adult C. elegans of the required genotypes were placed on six separate NGM plates seeded with E. coli OP50, and 30 eggs were allowed to develop per plate. Once the eggs hatched and reached the L4 stage, ten individuals per genotype were transferred to new 60 mm NGM plates seeded with OP50. For each experiment, a total of five plates were prepared per genotype. Worms were monitored daily, and their survival was assessed. During the first week, worms were transferred to fresh plates every other day to prevent overcrowding and maintain optimal conditions. In subsequent weeks, transfers were conducted every 2 days. Worms were observed until all individuals had perished. A worm was scored as dead if it exhibited no movement and failed to respond to a gentle touch with a platinum wire. Lifespan was recorded as the time from hatching to death for each individual worm, and P0 survivors were tracked until no live worms remained.

Image analysis

Image analysis was performed using FIJI ImageJ software (NIH) following standardized procedures for each type of analysis. All experiments and data analyses were conducted under blinded conditions whenever possible. Individuals responsible for designing and performing the experiments were distinct from those conducting data quantification and statistical analysis as much as possible.

Intensity measurement

To measure fluorescence intensity, the entire cell was selected as the region of interest (ROI) for analysis. Background signals in all relevant channels were subtracted uniformly across datasets using the “Process > Subtract Background” function, with a rolling ball radius set to 50 pixels. After background subtraction, the intensity of the signal within the defined ROI for each channel was measured without applying any thresholding. This was achieved using the “Analyze > Measure > Mean Intensity” function in FIJI. These measurements provided accurate representations of signal intensity across experimental conditions.

Particle analysis

The entire cell was selected as the region of interest (ROI) for analysis and added as individual ROIs in the ROI Manager. Background signals were subtracted uniformly across all datasets using the “Process > Subtract Background” function, with the rolling ball radius set to 50 pixels. The number of puncta per cell was determined after thresholding the fluorescence signal in each channel. Thresholding was performed using the “Image > Adjust > Threshold” tool to exclude background signals and isolate distinct puncta for analysis. Background subtraction was carried out as described above. The puncta within the defined ROIs were analyzed using the “Analyze > Analyze Particles” tool. This provided detailed information about the number, area, circularity, and intensity of individual puncta within the ROI. This protocol was followed for all the datasets using a customized automated macro enabling unbiased analysis.

Colocalization analysis

Colocalization between two particles was quantified using the Colocalization plugin in FIJI ImageJ (NIH). The entire cell was selected as the region of interest (ROI) and added to the ROI Manager for analysis. Background signals in each fluorescence channel were subtracted uniformly using the “Process > Subtract Background” function, with a rolling ball radius of 50 pixels. Thresholding was applied to each channel, and the threshold values were input into the Colocalization plugin to define colocalized regions, with a default ratio set to 50 pixels. The resultant colocalization window was auto-thresholded using the “MaxEntropy” algorithm to optimize the detection of overlapping signals. The number of colocalized puncta per cell within the colocalization window was quantified using the “Analyze > Analyze Particles” tool, as described in the puncta quantification section. This tool provided data on the number, area, circularity, and intensity of colocalized puncta.

Radial plot analysis

Radial plots were generated using the “Radial Plot” plugin in FIJI ImageJ (NIH). The entire cell was selected as the region of interest (ROI) and added to the ROI Manager. A line was drawn from the nuclear membrane to the cell periphery to define the radial axis. This line was then evenly divided into concentric circles by the plugin, facilitating the measurement of fluorescence intensities at increasing distances from the nucleus. The intensity at a given distance from the center was calculated as the sum of pixel values along the circumference of each circle. The intensity profiles were plotted as pixel intensity values versus distance from the nucleus, providing a quantitative representation of the spatial distribution of puncta across the cell. The plugin also allowed quantification of puncta intensity per circle, enabling detailed analysis of fluorescence distribution relative to the nucleus.

Tracking analysis

The trajectories of the puncta in live-cell time-lapse videos were analyzed using the TrackMate plugin in FIJI ImageJ (NIH) (Tinevez et al, 2017). The analysis involved tracking individual puncta and quantifying parameters such as velocity, split events, and merge events. Prior to tracking, appropriate thresholding was applied to each channel to enhance particle detection. The log detector algorithm within the TrackMate plugin was used to identify and threshold particles in the videos. The detected tracks were further analyzed to extract parameters, including mean velocity, total distance traveled, net displacement, and mean directional changes of each track. The extracted track information was plotted to visualize and quantify key movement characteristics of the puncta.

Kymograph analysis

All image panels used for time-lapse movie representation and analysis were created using FIJI ImageJ (NIH). Kymographs were generated using the KymoResliceWide v0.5 plugin for FIJI ImageJ (https://github.com/ekatrukha/KymoResliceWide), with the average intensity measured along a polyline of width 3 drawn ~100 μm distal to the PLM cell body. Once generated, the kymographs were manually annotated, and subsequent analysis was performed using the “KymoAnalyzer” plugin in FIJI ImageJ (Neumann et al, 2017). This analysis allowed for the categorization of tracks into anterograde, retrograde, and stationary movements, and provided detailed tracking parameters to assess the dynamics of puncta over time.

Statistical analysis

Data were presented as mean ± SEM, with sample sizes specified in the graphs or figure legends, representing biological replicates. Sample sizes were determined based on established experimental protocols to ensure statistical validity. The normality of the data distribution was assessed using either the D’Agostino-Pearson omnibus normality test or the Shapiro–Wilk test, as implemented in GraphPad Prism (version 9.5.1). Additionally, QQ plots were used to visually confirm the normality and distinguish between parametric and non-parametric distributions. For datasets with a parametric distribution, statistical significance was evaluated using Student’s t-test for two-group comparisons. For non-parametric data, the Mann–Whitney U-test was applied. When comparing multiple groups, one-way ANOVA followed by the Tukey’s multiple comparison test was used for parametric data, while the Kruskal–Wallis test was applied to non-parametric data. A significance threshold (p < 0.05) was used for all statistical tests, and the specific tests performed are indicated in the figure legends.

Ethics approval

All experimental procedures involving biological materials were conducted in full accordance with institutional guidelines and national regulatory standards. The relevant approvals and permits were obtained prior to initiating this study. Work involving recombinant DNA and genetically modified organisms was performed under the Institutional Biosafety Committee (IBSC) approval JNC/IBSC/2022/RM/C-27 at Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Bangalore, India. All necessary safety measures and ethical practices were strictly adhered to throughout the study.

Supplementary information

Peer Review File Source data Fig. 1 Source data Fig. 2 Source data Fig. 3 Source data Fig. 4 Source data Fig. 5 Source data Fig. 6 Source data Fig. 7 Source data Fig. 8 Expanded View Figures