https://orcid.org/0000-0003-0223-7717
Figures
Abstract
Autophagy is essential for cellular homeostasis and function. In neurons, autophagosome biogenesis is temporally and spatially regulated to occur near presynaptic sites, in part via the trafficking of autophagy transmembrane protein ATG-9. The molecules that regulate autophagy by sorting ATG-9 at synapses remain largely unknown. Here, we conduct forward genetic screens at single synapses of C. elegans neurons and identify a role for the long isoform of the active zone protein Clarinet (CLA-1L) in regulating sorting of autophagy protein ATG-9 at synapses, and presynaptic autophagy. We determine that disrupting CLA-1L results in abnormal accumulation of ATG-9 containing vesicles enriched with clathrin. The ATG-9 phenotype in cla-1(L) mutants is not observed for other synaptic vesicle proteins, suggesting distinct mechanisms that regulate sorting of ATG-9-containing vesicles and synaptic vesicles. Through genetic analyses, we uncover the adaptor protein complexes that genetically interact with CLA-1 in ATG-9 sorting. We also determine that CLA-1L extends from the active zone to the periactive zone and genetically interacts with periactive zone proteins in ATG-9 sorting. Our findings reveal novel roles for active zone proteins in the sorting of ATG-9 and in presynaptic autophagy.
Citation: Xuan Z, Yang S, Clark B, Hill SE, Manning L, Colón-Ramos DA (2023) The active zone protein Clarinet regulates synaptic sorting of ATG-9 and presynaptic autophagy. PLoS Biol 21(4): e3002030. https://doi.org/10.1371/journal.pbio.3002030
Academic Editor: Cagla Eroglu, Duke University Medical Center, UNITED STATES
Received: November 14, 2022; Accepted: February 8, 2023; Published: April 13, 2023
Copyright: © 2023 Xuan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: SY was supported by China Scholarship Council-Yale World Scholars Program (no grant number)(https://medicine.yale.edu/bbs/training/initiatives/csc/). ZX, BC, SY, SEH, LM and DACR were supported by NIH R01NS076558, DP1NS111778 (https://www.nih.gov/) and by an HHMI Scholar Award 55108513 (https://www.hhmi.org/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: ADBE, activity-dependent bulk endocytosis; CME, clathrin-mediated endocytosis; EM, electron microscopy; L4, larva 4; SD, standard deviation; SNP, single-nucleotide polymorphism; TGN, trans-Golgi network; WGS, whole-genome sequencing
Introduction
Macroautophagy (herein called autophagy) is a well-conserved cellular degradative pathway, and its disruption in neurons results in axonal degeneration, accumulation of protein aggregates, and cell death [1–4]. Autophagy is spatially and temporally regulated in neurons, and autophagosome biogenesis occurs near presynaptic sites and in response to increased neuronal activity states [5–16]. Directed transport of autophagy proteins to presynaptic sites contributes to local autophagosome biogenesis at synapses [5,9,16–18]. How autophagy proteins are transported, sorted, and locally regulated in neurons to control synaptic autophagy is not well understood.
ATG-9, a transmembrane protein necessary for autophagy, is actively trafficked in vesicles to promote local autophagosome biogenesis [19–24]. In neurons, ATG-9 is transported to presynaptic sites. Like synaptic vesicles, ATG-9-containing vesicles undergo activity-dependent exo-endocytosis regulated by canonical synaptic molecules such as endophilin A and synaptojanin 1 [24]. Activity-dependent exo-endocytosis of ATG-9-containing vesicles supports synaptic autophagy [24]. Yet, beyond the requirement of endophilin A and synaptojanin 1, the mechanisms that sort ATG-9 at presynaptic sites, the relationship of these mechanisms to those that sort canonical synaptic vesicle proteins and the relationship between ATG-9 and the synaptic machinery is not well understood. Knowledge of the molecules required for specific sorting of ATG-9 at synapses is of critical importance to understand the mechanisms that regulate synaptic autophagy.
To better understand the in vivo mechanisms that regulate ATG-9 sorting at synapses and their relationship with synaptic vesicle proteins, we generated transgenic Caenorhabditis elegans (C. elegans) strains that allowed us to simultaneously observe, in single neurons, localization of ATG-9 and synaptic vesicle proteins [24]. We performed unbiased forward genetic screens for mutants in which ATG-9 localization was differentially affected as compared to synaptic vesicle proteins, and identified an allele that affects the long isoform of the active zone protein Clarinet (CLA-1L). Clarinet bears similarity to Drosophila active zone protein Fife and vertebrate active zone proteins RIM, Piccolo, and Bassoon [25–29]. In cla-1(L) mutants ATG-9, but not synaptic vesicle proteins, abnormally accumulate to subsynaptic regions enriched for clathrin. This abnormal ATG-9 phenotype is suppressed by mutants for synaptic vesicle exocytosis, suggesting that the ATG-9 phenotype in cla-1(L) mutants emerges from defects in ATG-9 sorting during exo-endocytosis. Through genetic analyses, we found that mutants of the clathrin-associated adaptor complexes AP-2 and AP180 phenocopy and enhance the ATG-9 phenotypes observed for cla-1(L) mutant, whereas mutants for the AP-1 adaptor complex and the F-BAR protein syndapin 1 suppress the phenotype. We also observed that CLA-1L extends from the exocytic active zone to the endocytic periactive zone and genetically interacts with the periactive zone proteins EHS-1/EPS15 and ITSN-1/intersectin 1 in mediating ATG-9 sorting at presynaptic sites.
Our findings support a model whereby CLA-1L bridges the exocytic active zone regions with the endocytic periactive zone to regulate presynaptic sorting of ATG-9, likely via endosome-mediated sorting. Our findings also suggest that in vivo, ATG-9 containing vesicles represent a distinct subpopulation of synaptic vesicles. Our study uncovers molecules and synaptic machinery specifically involved in ATG-9 sorting and underscores the importance of active zone proteins in regulating local sorting of autophagy proteins and presynaptic autophagy.
Results
The active zone protein Clarinet (CLA-1) regulates ATG-9 sorting at presynaptic sites
To understand the in vivo mechanisms that regulate ATG-9 sorting at synapses and their relationship to the sorting of synaptic vesicle proteins, we simultaneously examined synaptic vesicle proteins and ATG-9 in the AIY interneurons of C. elegans. AIYs are a pair of bilaterally symmetric interneurons that display a stereotyped distribution of presynaptic specializations along their neurites (Fig 1A and 1C) [30,31]. Simultaneous visualization of ATG-9::GFP and the presynaptic marker mCherry::RAB-3 and Synaptogyrin (SNG-1)::BFP indicated that ATG-9 is enriched at presynaptic sites in AIY, consistent with previous studies (Figs 1B–1G and 2A–2D) [16,24].
(A) Schematic of the head of C. elegans, including pharynx (grey region) and the 2 bilaterally symmetric AIY interneurons. The asterisk denotes the cell body. There are 3 distinct segments along the AIY neurite: an asynaptic region proximal to AIY cell body (Zone 1), a large presynaptic region (Zone 2), and a segment with discrete presynaptic clusters at the distal part of the neurite (Zone 3) [30,31]. Presynaptic regions (Zone 2 and Zone 3) are in magenta (AIYL) or violet (AIYR). In axis, A, anterior; P, posterior; L, left; R, right; D, dorsal; V, ventral. (B-D) Distribution of ATG-9::GFP (B) and synaptic vesicle protein (mCherry::RAB-3, pseudo-colored magenta) (C) in the synaptic regions of AIY (merge in D). The dashed box encloses AIY Zone 2. (E-J) Distribution of ATG-9::GFP (E and H) and synaptic vesicle protein (mCherry::RAB-3, pseudo-colored magenta) (F and I) at Zone 2 of AIY (merge in G and J) in wild-type (WT) (E-G) and ola285 mutant (H-J) animals. ATG-9 is evenly distributed in WT but forms subsynaptic foci in ola285 mutants, which are not enriched with RAB-3 (indicated by arrows in H-J). (K) Schematic of the genomic region of cla-1L. The locations of loxP sites and the genetic lesions of the cla-1 alleles examined in this study are indicated. The genetic lesion in allele ola285 (I to N at residue 5753) is shown for both WT and ola285 mutants. The positions of the repetitive region in CLA-1L and the conserved PDZ and C2 domains in all CLA-1 isoforms are also shown in the schematic. (L) Quantification of the index of ATG-9 punctum (ΔF/F; see Methods) at Zone 2 of AIY in wild-type (WT), cla-1(ola285), and cla-1(ok560) mutants. Error bars show standard deviation (SD). “NS” (not significant), *p < 0.05 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. (M) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. Error bars represent 95% confidence interval. “NS” (not significant), ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. (N, O) Endogenous expression of GFP::CLA-1L (WT) (N) and GFP::CLA-1L (I5753N) (O) in the C. elegans nerve ring. (P-S) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (P), floxed cla-1L without Cre (Q), and floxed cla-1L with Cre expressed cell specifically in AIY (R) and cla-1(ok560) (S) animals. Arrows (in R and S) indicate abnormal ATG-9 foci. Scale bar (in B for B-D), 5 μm; (in E for E-J and P-S), 1 μm; (in N for N-O), 10 μm. Data for Fig 1L and 1M can be found in S1 Data.
To identify molecular mechanisms that selectively disrupt ATG-9 sorting at synapses, we performed unbiased forward genetic screens. From our screens, we isolated several alleles that affected ATG-9 expression levels, trafficking to synapses or sorting at synaptic sites (S1 Table). We focused our studies on allele ola285 for 2 reasons: (1) We had previously demonstrated that Synaptojanin 1 is necessary for ATG-9 sorting at presynaptic sites [24], and ola285 allele phenocopies the defects observed for unc-26/synaptojanin 1 mutants (namely, that ATG-9 abnormally localizes to subsynaptic foci); (2) in ola285 allele, the localization of ATG-9 at synapses is differentially affected as compared to the localization of synaptic vesicle proteins, suggesting that ola285 allele specifically affects ATG-9 sorting at synapses.
In wild-type animals, ATG-9 is evenly distributed in the presynaptic-rich region (termed Zone 2; Fig 1E–1G and 1M), whereas in the ola285 mutants, ATG-9 is enriched in subsynaptic foci in about 60% of animals (Fig 1H–1J and 1M). Unlike the ATG-9::GFP foci observed in ola285 mutants, the synaptic vesicle markers SNG-1 (and RAB-3) retained their wild-type phenotype in ola285 mutants in the Zone 2 region of AIY (Figs 1E–1J and 2A–2H). To quantify the expressivity of the phenotype, we calculated an index (Fpeak-Ftrough)/Ftrough) for ATG-9::GFP. The index was calculated from representative micrographs of the ATG-9 phenotype in the indicated genotypes (see Methods). Our quantifications of expressivity revealed a significant redistribution of ATG-9 to subsynaptic foci in ola285 mutants as compared to wild-type animals (Fig 1L).
To identify the genetic lesion corresponding to the ola285 allele, we performed single-nucleotide polymorphism (SNP) mapping coupled with whole-genome sequencing (WGS) [24,32–34]. We identified the genetic lesion of ola285 in the locus of the gene cla-1, which encodes for Clarinet. Clarinet (CLA-1) is an active zone protein that contains PDZ and C2 domains with similarity to vertebrate active zone proteins Piccolo and RIM (Fig 1K; [25–27]). Three lines of evidence support that ola285 is an allele of cla-1. First, ola285 contains a missense mutation in the cla-1 gene that converts Isoleucine (I) to Asparagine (N) at the residue 5753 (I5753N) (Fig 1K). Second, an independent allele of clarinet, cla-1(ok560), phenocopied the ATG-9 localization defects observed for ola285 mutants, both in terms of penetrance and expressivity (Fig 1L and 1M). Third, transheterozygous animals carrying both alleles ola285 and cla-1(ok560) resulted in abnormal ATG-9 localization at synapses, similar to that seen for either ola285 or cla-1(ok560) homozygous mutants (Fig 1M). The inability of cla-1(ok560) to complement the newly isolated allele ola285 supports that they correspond to genetic lesions within the same gene, cla-1. Together, our data indicate that CLA-1, an active zone protein with similarity to Drosophila active zone protein Fife, and vertebrate active zone proteins Piccolo and RIM [25], is required for sorting of the autophagy protein ATG-9 at presynaptic sites.
Clarinet long isoform, CLA-1L, acts cell autonomously to selectively regulate ATG-9 sorting at presynaptic sites
The cla-1 gene encodes 3 isoforms: CLA-1L (long), CLA-1M (medium), and CLA-1S (short) (S1A Fig). The long isoform, which contains a repetitive region predicted to be disordered (Fig 1K), is necessary for synaptic vesicle clustering, whereas the shorter isoforms are required for active zone assembly [25]. The isolated allele cla-1(ola285) (a missense mutation in the coding region of cla-1L), as well as the examined allele cla-1(ok560) (a deletion of the promoter and part of the coding region of cla-1L), only affect CLA-1L, but not CLA-1M or CLA-1S. A null allele affecting all isoforms, cla-1(wy1048), did not display a more severe ATG-9 phenotype than the alleles affecting only CLA-1L (S1A and S1B Fig). These findings suggest that the long isoform of Clarinet (CLA-1L) is necessary for presynaptic sorting of ATG-9.
To better assess the effects of the lesion of allele ola285 in CLA-1 protein product in vivo, we inserted, via CRISPR, a DNA sequence encoding GFP at the 5′-end of the endogenous cla-1 locus. The 5′-end of cla-1 gene corresponds to the N-terminus of CLA-1L (S1A Fig), so the inserted GFP specifically labels CLA-1L (S6B Fig). We observed that CLA-1L expression levels were reduced in ola285 as compared to wild-type animals (Figs 1N–1O and S1C). Based on the loss of function phenotype of ola285 allele, we hypothesize that the missense mutation results in degradation of CLA-1L.
To determine the specific requirement of CLA-1L for ATG-9 sorting at presynaptic sites in AIY, we manipulated the expression of CLA-1L in AIY using a cell-specific knockout strategy [25]. We used a strain in which loxP sites were inserted, via CRISPR, to flank the unique 5′-end gene locus specific to CLA-1L (Figs 1K and S1A). Cell-specific expression of Cre recombinase in AIY, which leads to AIY-specific deletion of the CLA-1L isoform (without affecting CLA-1S and CLA-1M), resulted in the ATG-9 phenotype in AIY (Figs 1R and S1D), which was indistinguishable from that seen for the cla-1 (ok560) allele (Figs 1S and S1D, compare to wild type in Figs 1P, 1Q, and S1D). Together, our data indicate that the allele ola285 affects the long isoform of Clarinet (CLA-1L) and that CLA-1L regulates presynaptic sorting of ATG-9 in a cell-autonomous manner.
ATG-9-containing vesicles cluster at clathrin-rich subsynaptic domains in cla-1(ola285) mutants
In cla-1(ola285) animals, ATG-9 sorting defects phenocopied those observed for unc-26/synaptojanin 1 mutant [24]. However, unlike unc-26/synaptojanin 1, cla-1(ola285) did not exhibit a mutant phenotype for synaptic vesicle proteins in presynaptic-rich Zone 2 (Figs 1E–1J, 2A–2H, and S2A). To better understand the effects of the genetic lesion of cla-1(ola285) on synaptic morphology and synaptic vesicle distribution, we performed transmission electron microscopy (EM) studies. Our ultrastructural analyses in the AIY Zone 2 region revealed that the average length of the active zone is similar between wild-type (2.65 ± 0.23 μm) and cla-1(ola285) mutants (2.45 ± 0.26 μm) (examined in 6 EM reconstructed neurons per genotype; Figs 2I–2M, S3D, and S3E), consistent with fluorescence microscopy observations that CLA-1L is not required for active zone assembly (S3A–S3C Fig; [25]) and suggestive that the defects observed for ATG-9 missorting are not due to general defects in the AIY active zone. Quantification of the presence of synaptic vesicles, dense core vesicles, and endosomal structures in the electron micrographs of cla-1(ola285) and wild-type animals did not reveal major differences that could account for the observed phenotypes of ATG-9 (Figs 2I–2N and S3F-S3H). These observations are consistent with our light microscopy studies on the distribution of synaptic vesicle proteins in cla-1(ola285) mutants (Figs 1E–1G and 2A–2H) and stand in contrast to previous observations in unc-26/synaptojanin 1 mutants, where a general defect in all vesicular structures is observed [24,35]. These findings suggest that the observed phenotype for ATG-9 in cla-1(ola285) mutants is not due to a general problem in synaptic morphology or synaptic vesicle endocytosis.
(A-H) Distribution of SNG-1::BFP (pseudo-colored cyan) (A and E), mCherry::RAB-3 (pseudo-colored magenta) (B and F), and ATG-9::GFP (C and G) at Zone 2 of AIY (merge in D and H) in wild-type (WT) (A-D) and cla-1(ola285) mutant (E-H) animals. While we observe a phenotype for abnormal ATG-9 distribution to subsynaptic foci in cla-1(ola285) mutants (indicated by arrows in G and H), we do not see a similar redistribution for synaptic vesicle proteins SNG-1 and RAB-3. (I, J) Electron microscopy of the Zone 2 region in wild-type (I) and cla-1(ola285) mutant animals (J). Blue lines, outline of AIY Zone 2. Presynaptic dense projections, pointed with arrows in dark blue. “m”, mitochondria. (K, L) Electron micrograph reconstructions of AIY Zone 2 in wild-type (K) and cla-1(ola285) mutant animals (L). The active zones (or dense projections) are highlighted in red. Synaptic vesicles and dense core vesicles are symbolized by yellow and blue spheres, respectively. In axis: A, anterior; P, posterior; L, left; R, right; D, dorsal; V, ventral. (M) Measurement of the length of the active zone (highlighted in red in K and L) in the AIY neurons of 3 wild-type and 3 cla-1(ola285) mutants. Error bars represent standard deviation (SD). “NS” (not significant) by two-tailed Fisher’s exact test. Each dot in the scatter plot represents a single neuron. (N) Quantification of synaptic vesicles in the AIY neurons (AIY-L: AIY on the left side; AIY-R: AIY on the right side) of 1 wild-type and 1 cla-1(ola285) mutant. The differences in AIYR (observed in this figure for cla-1(ola285) mutants, also in S3 for endosomal area in wild type) are consistent with previous findings about asymmetry of AIY neurons [31], including gene expression [114]. Nonetheless, unlike for other examined genotypes by EM (like UNC-26/Synaptojanin) [24], the observed phenotypes do not reveal major differences that could account for the observed light microscopy phenotypes of ATG-9. Error bars represent standard deviation (SD). ***p < 0.001 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single section. Scale bar (in A for A-H), 1 μm; (in I for I and J), 500 nm. Data for Fig 2M and 2N can be found in S1 Data.
To better understand the distribution of ATG-9 in cla-1(ola285) mutants, we performed immuno-EM studies and stained ATG-9::GFP. Although the total number of ATG-9::GFP gold particles displays no significant difference between wild-type and cla-1(ola285) mutant animals (number of gold particles = 503 for wild-type and 599 for cla-1 mutants), the particle distribution displayed differences. In wild-type animals, ATG-9::GFP gold particles are distributed along the Zone 2 synapse (Fig 3A, 3C, and 3E). In cla-1(ola285) mutants, however, we observe ATG-9::GFP gold particles concentrate on subsynaptic regions (Fig 3B, 3D, and 3E). These findings are consistent with our fluorescence microscopy data that ATG-9 localizes to a subsynaptic region in cla-1(ola285) mutants (Fig 1H). Together, our findings indicate that the ATG-9 phenotype in cla-1(ola285) results from differences in the distribution of ATG-9-containing vesicular structures at the synapse.
(A, B) Immunogold electron microscopy at Zone 2 of AIY neurons in wild-type (A) and cla-1(ola285) mutant (B) transgenic animals, with ATG-9::GFP panneuronally expressed and antibodies directed against GFP [24]. Blue line outlines the AIY Zone 2 region; dark blue arrows point at presynaptic dense projections. “m”, mitochondria. Insets at the upper right hand corner correspond to higher magnifications of the regions highlighted with purple squares, with red arrows pointing to a representative immunogold particle detecting ATG-9::GFP in vesicular structures. (C, D) Electron micrograph reconstructions of Zone 2 of AIY and ATG-9::GFP immunogold particles in wild-type (C) and cla-1(ola285) mutant animals (D). Red dots: ATG-9::GFP immunogold particles. In axis: A, anterior; P, posterior; L, left; R, right; D, dorsal; V, ventral. (E) Distribution of ATG-9 immunogold particles density per cross-section in wild-type (blue line and round dots) and cla-1(ola285) mutant animals (orange line and square dots). X axis, Z slices at Zone 2 along the antero-posterior axis. Scale bar (in A for A and B), 500 nm; (in insert of A for inserts of A and B), 100 nm. Data for Fig 3E can be found in S1 Data.
To determine if the abnormal accumulation of ATG-9 in cla-1(ola285) mutants, like that observed for unc-26/synaptojanin 1, results from defects in ATG-9 sorting during exo-endocytosis, we next examined the necessity of synaptic vesicle exocytosis proteins in the ATG-9 phenotype of cla-1(ola285) mutants. We first visualized ATG-9 in putative null alleles unc-13(s69)/Munc13, unc-10(md1117)/RIM, unc-18(e81)/Munc18, and unc-2(e55)/CaV2α1 (voltage-gated calcium channels), all of which encode proteins essential for synaptic vesicle fusion (Fig 4A) [36–43]. Single mutants of unc-13(s69), unc-10(md1117), unc-18(e81)/Munc18, and unc-2(e55) did not disrupt ATG-9 localization (Fig 4D, 4F, 4H, 4J, and 4K). Double mutants of unc-13(s69);cla-1(ola285), unc-10(md1117);cla-1(ola285), unc-18(e81);cla-1(ola285), and unc-2(e55);cla-1(ola285) completely suppressed abnormal ATG-9 localization in cla-1 mutants (Fig 4E, 4G, 4I, 4J, and 4K). These results are consistent with previous findings that ATG-9-positive vesicles undergo exo-endocytosis at presynaptic sites by using the synaptic vesicle cycling machinery [24] and suggest that the ATG-9 phenotype in cla-1(ola285) mutants results from defects in ATG-9 sorting upon ATG-9 exo-endocytosis.
(A) Schematic of the proteins required for the synaptic vesicle cycle and associated with this study (both the names used for C. elegans and vertebrates are listed). (B-I) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (B), cla-1(ola285) (C), unc-13(s69) (D), unc-13(s69);cla-1(ola285) (E), unc-10 (md1117) (F), unc-10(md1117);cla-1(ola285) (G), unc-18(e81) (H), and unc-18(e81);cla-1(ola285) (I) animals. ATG-9 subsynaptic foci are indicated by the arrow (in C). (J) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild type are the same as those in Fig 1M (explained in Methods). Error bars represent 95% confidence interval. ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. (K) Quantification of the index of ATG-9 punctum (ΔF/F) at Zone 2 of AIY in the indicated genotypes. The data used to quantify the index of ATG-9 punctum (ΔF/F) in wild-type and cla-1(ola285) mutants are the same as those in Fig 1L (explained in Methods). Error bars show standard deviation (SD). ****p < 0.0001 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. Scale bar (in B for B-I and J-Q), 1 μm. Data for Fig 4J and 4K can be found in S1 Data.
In unc-26/synaptojanin 1 mutants, ATG-9 abnormally colocalizes with the clathrin heavy chain subunit, CHC-1 [24]. To determine if, in cla-1(ola285) mutants, ATG-9 similarly abnormally colocalizes with the clathrin heavy chain subunit, we examined the colocalization between ATG-9 and CHC-1 in wild-type and cla-1(ola285) mutant animals (Fig 5A–5H). We observed that both ATG-9 and CHC-1 abnormally localized to similar synaptic foci in cla-1(ola285) mutants (Figs 5E–5H and S2B). Together, our findings suggest that in cla-1 mutants, ATG-9-containing vesicles abnormally cluster at clathrin-rich subsynaptic domains.
(A-D) Distribution of ATG-9::GFP (C), BFP::CHC-1 (pseudo-colored cyan) (D), and mCherry::RAB-3 (pseudo-colored magenta) (E) at Zone 2 of AIY (merge in F) in wild-type (WT) animals. (E-H) Distribution of ATG-9::GFP (A), BFP::CHC-1 (pseudo-colored cyan) (B), and mCherry::RAB-3 (pseudo-colored magenta) (C) at Zone 2 of AIY (merge in D) in cla-1(ola285) mutant animals. ATG-9 subsynaptic foci are enriched with CHC-1 in cla-1(ola285) mutants (indicated by arrows in A, B, and D). (I-L) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (E), unc-11(e47)/AP180 (F), dpy-23(e840)/AP2μ (G), and dpy-23(e840);cla-1(ola285) (H) mutant animals. Abnormal ATG-9 subsynaptic foci are indicated by arrows in F-H. (M) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 for the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild-type and cla-1(ola285) mutants are the same as those in Fig 4J (explained in Methods). Error bars represent 95% confidence interval. ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. (N) Quantification of the index of ATG-9 punctum (ΔF/F) at Zone 2 of AIY for the indicated genotypes. The data used to quantify the index of ATG-9 punctum (ΔF/F) in wild-type and cla-1(ola285) mutants are the same as those in Fig 1L (explained in Methods). Error bars show standard deviation (SD). “NS” (not significant), **p < 0.01 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. Scale bar (in A for A-L), 1 μm. Data for Fig 5M and 5N can be found in S1 Data.
The clathrin adaptor complexes, AP-2 and AP180, regulate ATG-9 sorting at presynaptic sites
We next examined the genetic relationship between clathrin adaptor protein complexes and CLA-1L in ATG-9 localization. The AP-2 complex mediates clathrin-mediated endocytosis (CME) of synaptic vesicle proteins [44–49], and it has been implicated in the sorting of ATG-9 during autophagy induction in mammalian nonneuronal cells [50–52]. To determine if the AP-2, and the associated AP180, adaptor complexes were required in presynaptic sorting of ATG-9, we examined ATG-9 localization in the null alleles dpy-23(e840)/AP2μ and unc-11(e47)/AP180. We observed that dpy-23(e840)/AP2μ and unc-11(e47)/AP180 mutants phenocopied cla-1(ola285) mutants in ATG-9 presynaptic sorting defects (Fig 5J, 5K, 5M, and 5N). In addition, the expressivity of the ATG-9 sorting defects was enhanced in dpy-23(e840)/AP2μ;cla-1(ola285) double mutant worms (Fig 5L–5N). These findings suggest shared mechanisms that similarly result in defective ATG-9 sorting when clathrin-associated adaptor complexes, or the active zone protein CLA-1L, are disrupted.
ATG-9 is sorted to endocytic intermediates via SDPN-1/syndapin 1 and the AP-1 adaptor complex
During endocytosis, clathrin-associated adaptor complexes mediate internalization and sorting of cargoes from both the plasma membrane and intracellular endocytic intermediates [53–55]. Since ATG-9 abnormally localizes to subsynaptic foci when disrupting the AP-2 (or the associated AP180) adaptor complexes, we reasoned that the subsynaptic ATG-9-rich foci might represent endocytic intermediates, from which the AP-2 adaptor complex binds to and sort out cargoes. We then sought to identify upstream molecules that mediate the sorting of ATG-9 to the endocytic intermediates. Disrupting those molecules should suppress ATG-9 foci in mutants for CLA-1L or AP-2.
We first examined SDPN-1/syndapin 1, a protein known to play important roles in early stages of membrane invagination during both activity-dependent bulk endocytosis (ADBE) [56,57] and ultrafast endocytosis of synaptic vesicles [58]. We reasoned that if ATG-9-containing vesicles were sorted via SDPN-1-dependent mechanisms, then sdpn-1 mutants would suppress the observed ATG-9 foci for cla-1(ola285) and for mutants of the clathrin-associated adaptor complexes. Consistent with our hypothesis, we observed that ATG-9 localization was not disrupted in sdpn-1(ok1667) single mutants and that the abnormal ATG-9 foci were suppressed in sdpn-1(ok1667);cla-1(ola285) and sdpn-1(ok1667);unc-11(e47)/AP180 double mutant animals (Fig 6A–6H). These findings are consistent with a requirement of SDPN-1 in the sorting of ATG-9 upstream of CLA-1L and clathrin-associated adaptor complexes.
(A-F) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (A), sdpn-1(ok1667) (B), cla-1(ola285) (C), sdpn-1(ok1667);cla-1(ola285) (D), unc-11(e47)/AP180 (E), and sdpn-1(ok1667);unc-11(e47) (F) mutant animals. Abnormal ATG-9 subsynaptic foci are indicated by arrows in C and E. Note that mutations in SDPN-1/syndapin 1 suppress the abnormal ATG-9 phenotypes in cla-1 and unc-11/AP180 mutants. (G) Quantification of the percentage of animals displaying abnormal ATG-9 subsynaptic foci at AIY Zone 2 for the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild-type and cla-1(ola285) mutants are the same as those in Fig 4J; the data used in unc-11(e47) are the same as those in Fig 5M (explained in Methods). Error bars represent 95% confidence interval. “NS” (not significant), ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. (H) Quantification of the index of ATG-9 punctum (ΔF/F) at Zone 2 of AIY for the indicated genotypes. The data used to quantify the index of ATG-9 punctum (ΔF/F) in wild-type and cla-1(ola285) mutants are the same as those in Fig 1L; the data used in unc-11(e47) are the same as those in Fig 5N (explained in Methods). Error bars show standard deviation (SD). “NS” (not significant), *p < 0.05, **p < 0.01, ***p < 0.001 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. Scale bar (in A for A-F), 1 μm. Data for Fig 6G and 6H can be found in S1 Data.
Next, we examined the AP-1 adaptor complex, which acts at presynaptic sites to mediate endosomal sorting of ADBE [54,59,60]. To determine the requirement of the AP-1 adaptor complex in ATG-9 sorting at presynaptic sites, we examined ATG-9 localization in unc-101(m1)/AP1μ1 single mutant animals and in double mutants with cla-1(ola285), dpy-23(e840)/AP2μ, and unc-11(e47)/AP180. We observed that while unc-101(m1)/AP1μ1 single mutant animals did not display detectable phenotypes in ATG-9 localization (Figs 7C, 7I and S4E), unc-101(m1)/AP1μ1 suppressed the ATG-9 phenotype in double mutant animals (Figs 7A–7D, 7G–7I, and S4C-S4E). We confirmed this result by making double mutants of cla-1(ola285) with another allele, unc-101(sy108), and observed it also suppressed the ATG-9 phenotype in cla-1(ola285) (S4E Fig). Furthermore, single-cell expression of the C. elegans cDNA of unc-101(m1)/AP1μ1 in AIY in the unc-101(m1);cla-1(ola285) double mutants reverted the phenotype, indicating that AP-1 acts cell autonomously in AIY to suppress the ATG-9 phenotype in cla-1(L) (Figs 7E, 7I, and S4E). Our findings indicate that, similar to SDPN-1, the AP-1 adaptor complex is required to sort ATG-9 at synapses, likely upstream of CLA-1L and clathrin-associated adaptor complexes.
(A-H) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (A), cla-1(ola285) (B), unc-101(m1)/AP-1μ1 (C), unc-101(m1);cla-1(ola285) (D), unc-101;cla-1 mutants with C. elegans UNC-101/AP-1μ1 cDNA cell specifically expressed in AIY (E), unc-101;cla-1 mutants with mouse AP1μ1 cDNA cell specifically expressed in AIY (F), unc-11(e47)/AP180 (G), and unc-101(m1);unc-11(e47) (H). Abnormal ATG-9 subsynaptic foci are indicated by arrows in B and E-G. (I) Quantification of the index of ATG-9 punctum (ΔF/F) at Zone 2 of AIY for indicated genotypes. The data used to quantify the index of ATG-9 punctum (ΔF/F) in wild-type and cla-1(ola285) mutants are the same as those in Fig 1L; the data used in unc-11(e47) are the same as those in Fig 5N (explained in Methods). Error bars show standard deviation (SD). “NS” (not significant), *p < 0.05 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. Scale bar (in A for A-H), 1 μm. Data for Fig 7I can be found in S1 Data.
The C. elegans UNC-101/AP1μ1 is more similar to the murine AP1μ1 (Query Cover: 100%; Percentage Identity: 74%) than to the murine AP2μ (Query Cover: 98%; Percentage Identity: 41%) (S4F Fig). To examine the conserved role of UNC-101/AP1μ1 in ATG-9 sorting, we expressed murine cDNA of AP1μ1 or AP2μ cell specifically in AIY of the unc-101(m1);cla-1(ola285) double mutants and examined ATG-9 localization. We observed that expression of murine AP1μ1, but not AP2μ cDNA, reverted the suppression seen for unc-101(m1);cla-1(ola285) double mutant animals (Figs 7F, 7I, S4B, and S4E), suggesting that this sorting function is conserved between murine AP1μ1 and C. elegans UNC-101/AP1μ1. Together, our findings are consistent with a model whereby ATG-9 is sorted to vesicular structures via synaptic machinery that includes CLA-1L, and also endosomal sorting proteins AP-1, SDPN-1, AP-2, and AP180 (S5 Fig).
CLA-1L extends to the periactive zone and genetically interacts with endocytic proteins to regulate ATG-9 sorting
Next, we used cell biological approaches to understand the relationship between active zone protein CLA-1L and the endocytic sorting machinery, which localizes primarily to a different subsynaptic domain called the periactive zone [61–63]. Large active zone proteins that bear functional similarity to CLA-1L, such as Piccolo and Bassoon, extend from the active zone subdomains to the periactive zone, a property that has been hypothesized to be important in their roles sorting synaptic components during exo-endocytosis [62,64–70]. CLA-1L is twice the size as Piccolo and Bassoon and contains largely disordered regions that could facilitate its extension from the active zone to the neighboring periactive zones. Consistent with this, in previous studies, we documented that while the C-terminus of CLA-1 localized specifically to the active zone, the unique N-terminus of CLA-1L isoform localized beyond the active zone subdomain [25].
We compared the endogenous C-terminally tagged CLA-1::GFP, or the endogenous N-terminally tagged GFP::CLA-1L (S6A and S6B) [25], with a periactive zone marker APT-4/APA-2/AP-2α [71]. While the C-terminally tagged CLA-1::GFP specifically localizes to small puncta corresponding to the active zone (Fig 8A and 8D), the N-terminally tagged GFP::CLA-1L displays a more distributed presynaptic pattern, extending to other regions of the synaptic bouton beyond the active zone (Fig 8H and 8K). Importantly, we observed that the N-terminally tagged GFP::CLA-1L, but not the C-terminally tagged CLA-1::GFP, colocalizes with the endocytic marker APT-4/APA-2/AP-2α at the periactive zones (Fig 8A–8O). These findings suggest that the long isoform of CLA-1 is anchored, via its C-terminus, to the active zone, but extends to the periactive zone where the endocytic sorting machinery is present.
(A-C) Distribution of endogenous C-terminally tagged CLA-1::GFP (A) and the endocytic zone marker APT-4/APA-2/AP-2α::mCherry (APT-4::mCh, pseudo-colored magenta) (B) in the neurons of the posterior dorsal nerve cord (merge in C) in wild-type animals. Note that APT-4::mCh is expressed in a subset of neurons in the dorsal nerve cord, driven by the punc-129 promoter, while CLA-1::GFP and GFP::CLA-1 are expressed panneuronally (so green puncta can be present where there are no magenta puncta; see Methods). (D-F) Enlarged regions enclosed in dashed boxes in A-C. Endogenous C-terminally tagged CLA-1::GFP (D) localizes to small puncta corresponding to the active zone [25], and different from the pattern observed for periactive zone protein, APT-4::mCh (E, merge in F). Yellow circles are drawn based on the outline of APT-4::mCh puncta in E and are located at the same positions in D-F. (G) Schematic of the localization of the C-terminally tagged CLA-1::GFP, relative to the subsynaptic active and periactive zones. (H-J) Distribution of endogenous N-terminally tagged GFP::CLA-1L (H) and the endocytic zone marker APT-4/APA-2/AP-2α::mCherry (APT-4::mCh, pseudo-colored magenta) (I) in neurons of the posterior dorsal nerve cord (merge in J) in wild-type animals. (K-M) Enlarged regions enclosed in dashed boxes in H-J. Endogenous N-terminally tagged GFP::CLA-1L (K) displays a more distributed synaptic distribution as compared to the C-terminally tagged CLA-1:GFP (compare with A, D, and F; see also [25]) and colocalizes with APT-4::mCh (L, merge in M). White circles are drawn based on the outline of APT-4::mCh puncta in L and are located at the same positions in K-M. (N) Schematic of the localization of the N-terminally tagged GFP::CLA-1L, relative to the subsynaptic active and periactive zones. (O) Pearson correlation coefficient for colocalization between CLA-1::GFP and APT-4::mCh, or between GFP::CLA-1L and APT-4::mCh, both in wild-type animals. Error bars show standard deviation (SD). ****p < 0.0001 by two-tailed unpaired Student t test. Each dot in the scatter plot represents a single animal. (P) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild-type and cla-1(ola285) mutants are the same as those in Fig 4J (explained in Methods). Error bars represent 95% confidence interval. ***p < 0.001, ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. Scale bar (in A for A-C and H-J), 5 μm; (in D for D-F and K-M), 2 μm. Data for Fig 8O and 8P can be found in S1 Data.
Based on the localization of CLA-1L to these presynaptic subdomains, and the ATG-9 phenotypes observed in cla-1 and endocytic mutants, we hypothesized the existence of genetic interactions between CLA-1L and endocytic proteins at the periactive zone. To test this hypothesis, we examined the localization of ATG-9 in the double mutants of cla-1(ola285) with genes encoding periactive zone endocytic proteins, ehs-1(ok146)/EPS15 or itsn-1(ok268)/intersectin 1. We focused our analyses on the endocytic regulators EHS-1/EPS15 and ITSN-1/intersectin 1 because of their hypothesized roles in coupling synaptic vesicle exocytosis at the active zone, and endocytosis at the periactive zone [61,62,72–79]. We observed that in null alleles of ehs-1(ok146)/EPS15 and itsn-1(ok268)/intersectin 1, 30% of worms displayed abnormal ATG-9 foci (compared to 60% in cla-1(L) mutants) (Figs 8P, S7C, and S7E). Interestingly, we observed that ehs-1(ok146);cla-1(ola285) and itsn-1(ok268);cla-1(ola285) enhanced the ATG-9 phenotype as compared to any of the single mutants (S7A–S7G Fig), both in terms of penetrance (Fig 8P) and expressivity (S7G Fig). Our findings uncover a cooperative genetic relationship between CLA-1L and the EHS-1-ITSN-1 endocytic scaffolding complex, suggesting that the active zone protein CLA-1L acts in pathways that are partially redundant to the EHS-1-ITSN-1 complex in linking the active zone and periactive zone regions to regulate ATG-9 sorting at presynapses.
Disrupted ATG-9 sorting in cla-1(ola285) mutants is associated with a deficit in activity-induced autophagosome formation
ATG-9 is important for autophagosome biogenesis at presynaptic sites [16,24]. To examine how autophagosome formation is affected in cla-1(L) mutants, we measured the average number of LGG-1/Atg8/GABARAP puncta (an autophagosomal marker) in the AIY neurites [12,16,24] (Fig 9A–9C). Previously, we observed the average number of LGG-1 puncta increased in AIY when the wild-type animals were cultivated at 25°C [12,24], a condition known to increase the activity state of the AIY neurons [80]. Worms with impaired exocytosis (in unc-13 mutants) or impaired autophagy (in atg-9 mutants) failed to display increased LGG-1 synaptic puncta [12,24]. We found that, unlike wild-type animals, the average number of LGG-1 puncta did not increase in cla-1(L) mutants (alleles ola285 and ok560) in response to cultivation temperatures that increase the activity state of the neuron (Figs 9D and S8A). These findings indicate that activity-induced autophagosome formation at synapses is impaired in cla-1(L) mutants.
(A-C) Confocal micrographs of GFP::LGG-1 (A) and cytoplasmic mCherry (cyto::mCh) (pseudo-colored magenta, B) in AIY (merge in C). Inset is the enlarged region enclosed in dashed box to show one LGG-1 punctum in AIY synaptic Zone 2. (D) Quantification of the average number of LGG-1 puncta in the AIY neurites at 20°C and at 25°C for 4 h in wild-type (WT) and cla-1(ola285) mutants. (Note: The activity state of the thermotaxis interneurons AIY is reported to increase when animals are cultivated at 25°C for 4 h, compared to animals at 20°C [80,115,116]. Error bars represent 95% confidence interval. “NS” (not significant), ***p < 0.001 by Kruskal–Wallis test with Dunn’s multiple comparisons test. The number on the bars indicates the number of animals scored. (E-H) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (E), epg-9(bp320) (F), cla-1(ola285) (G), and epg-9(bp320); cla-1(ola285) (H) mutant animals. Arrows (in F-H) indicate abnormal ATG-9 foci. (I) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. Error bars represent 95% confidence interval. “NS” (not significant), ***p < 0.001, ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bar indicates the number of animals scored. Scale bar (in A for A-C), 5 μm; (in inset of A for inset of A-C), 2 μm; (in E for E-H), 1 μm. Data for Fig 9D and 9I can be found in S1 Data.
Previously, we found that in mutants that affect early stages of autophagy such as epg-9(bp320), ATG-9 accumulated into subsynaptic foci, which colocalized with the clathrin heavy chain CHC-1 [24]. To determine a potential cross-talk between CLA-1L-mediated ATG-9 endocytosis and autophagy, we generated epg-9(bp320);cla-1(ola285) double mutant animals. The ATG-9 phenotype is enhanced in the double mutants, compared to single mutants (Fig 9E–9I). Our findings are consistent with a model whereby disrupted ATG-9 sorting in cla-1(L) mutants contributes to deficits in activity-induced autophagosome formation at synapses.
Discussion
The active zone protein Clarinet (CLA-1L) regulates ATG-9 sorting at synapses and presynaptic autophagy. Autophagy, a conserved cellular degradative pathway, is temporally and spatially regulated in neurons to occur at synaptic compartments and in response to increased synaptic activity states [6,8–12,15,16,81,82]. How synaptic autophagy and synaptic activity states are coordinated in neurons is not well understood. In previous studies, we had determined that in C. elegans synapses and in vivo, ATG-9 is trafficked to synapses and that disruption of ATG-9 trafficking or its sorting results in disruptions to activity-dependent autophagosome biogenesis at synapses [16,24].
The synaptic machinery that sorts ATG-9 at presynaptic sites to regulate local autophagy remain largely unknown. In this study, by performing unbiased forward genetic screens, we uncover an unexpected role for the active zone protein Clarinet in synaptic sorting of ATG-9 and in activity-dependent autophagosome formation at synapses. We note that we did not detect differences in baseline autophagy in cla-1 mutants, although we do observe defects in ATG-9 sorting under those conditions. Both ATG-9 sorting and autophagosome formation represent processes that are in flux, which increases upon neuronal activation. Our data suggest that the defects in ATG-9 sorting in cla-1 mutants is not able to meet the needs for increased autophagy flux when neuronal activity increases.
The mechanisms of CLA-1L regulation of synaptic autophagy are likely distinct from those observed for other active zone proteins such as Bassoon. Previous studies in primary hippocampal neurons had demonstrated that active zone proteins such as Bassoon and Piccolo play important roles in synaptic protein homeostasis [83], in part via the regulation of presynaptic autophagy [84–86]. Bassoon negatively regulates presynaptic autophagy by serving as a scaffold for Atg5 [86], and the E3 ubiquitin ligase Parkin [84,85]. We find that instead of inhibiting autophagy, CLA-1L is required for activity-dependent synaptic autophagy, likely by sorting ATG-9 at synapses. Despite the mechanistic differences, together, these studies support the concept that in neurons, active zone proteins play important roles in regulating activity-dependent synaptic autophagy.
The CLA-1 long isoform, CLA-1L, extends from the active zone to the periactive zone and is specifically required for ATG-9 sorting by genetically interacting with proteins involved in endocytosis and sorting of synaptic cargo. Unlike the shorter isoforms, CLA-1L is not required for active zone assembly or synapse formation [25]. Our ultrastructural studies similarly demonstrate that morphological features of the synapse (including active zone length, synaptic vesicles, dense core vesicles, and endosomes) are largely unaffected in the cla-1(ola285) mutants, which specifically affect the CLA-1L isoform (Figs 2, S1G, and S1H). These findings suggest that the ATG-9 phenotype in cla-1(L) does not result from general defects in AIY synaptic morphology, or synaptic vesicle endocytosis. CLA-1L has been proposed to be functionally similar to Piccolo and Bassoon in its roles at the active zone [25]. In vertebrate synapses, Piccolo and Bassoon extend from the active zone region to periactive zones, and this architecture has been proposed to couple exocytosis at the active zone, and protein sorting during endocytosis at the periactive zone [62,64–70]. CLA-1L is an 8922 amino acid protein, twice the size of Bassoon (3942 amino acids) and Piccolo (4969 amino acids). It is anchored to the active zone via its C-terminus [25], with a large disordered N-terminal domain extending to the periactive zone, where endocytic processes occur. Consistent with these observations on CLA-1L size and position at the synapses, we uncover genetic interactions between CLA-1L and periactive zone proteins EHS-1/EPS15 or ITSN-1/intersectin 1, which have been suggested as linkers between active zone exocytosis and periactive zone endocytosis [61,62,72–79]. We propose that the specific requirement of CLA-1L in sorting ATG-9 at synapses is mediated via its capacity to extend across presynaptic subdomains, from the exocytic active zone to the endocytic periactive zone.
CLA-1L selectively regulates the sorting of ATG-9 by genetically interacting with clathrin-associated adaptor complexes such as AP-1, AP-2, and AP180. Our findings are consistent with studies in nonneuronal cells, which demonstrated that the AP-1 and AP-2 complexes mediate ATG-9 trafficking between the plasma membrane, the trans-Golgi network (TGN), the recycling endosomes, and the growing autophagosomes [50,52,87,88]. It is noteworthy that in both cla-1 and Synaptojanin/unc-26 mutants, sorting defects of ATG-9 result in ATG-9-containing vesicles abnormally colocalizing onto subsynaptic foci. Clusters of ATG-9-containing vesicular structures have also been observed in vertebrate cells defective for autophagy [89], suggesting that clustering of ATG-9 vesicles, such as those seen for cla-1 mutants, might be a hallmark of defective autophagy. Consistent with this interpretation, we observe that mutations in early autophagy protein EPG-9 also result in abnormal accumulation of ATG-9 in synaptic foci, which were enhanced by cla-1(L) mutants, further underscoring the relationship between ATG-9 sorting at the synapse and autophagy.
Sorting of ATG-9 at synapses is genetically separable from the sorting of synaptic vesicle proteins. Similar to synaptic vesicles, ATG-9 vesicles are transported to synapses via the canonical synaptic vesicle kinesin UNC-104/KIF-1A and undergo exo-endocytosis in an UNC-13/Munc13- and UNC-26/Synaptojanin-dependent manner. However, we hypothesize that ATG-9-containing vesicles probably represent a distinct subpopulation of vesicles at the synapse, for 3 reasons: First, only a small amount of ATG-9 was found in the synaptic vesicle fractions via mass spectrometry [90–92], suggesting that ATG-9 does not localize to all vesicles at synapses. Second, our cell biological studies reveal that while ATG-9 localizes to presynaptic sites, it only partially colocalizes with synaptic vesicle proteins [24] (Fig 2A–2D), consistent with the existence of distinct vesicle pools. Third, in both fibroblasts and nerve terminals, vertebrate ATG9A does not coassemble into synaptophysin-positive vesicle condensates, consistent with it undergoing differential sorting relative to synaptic vesicle proteins [93]. While further biochemical studies on the composition of ATG-9-containing vesicles are required to better understand the relationship between ATG-9 vesicles and synaptic vesicles in neurons, our cell biological studies suggest that they belong to distinct subpopulation of vesicles at the synapse. Moreover, while we do not fully understand the biochemical interactions of CLA-1L resulting in ATG-9 sorting, our in vivo genetic studies suggest a model of how the synaptic machinery could cooperate with the autophagy pathway in regulating local synaptic autophagy. We favor a model in which AP-1 adaptor complex and the F-BAR protein syndapin I (SDPN-1) mediate trafficking of ATG-9 to a transient sorting station from which AP2-AP180 complexes facilitate clathrin-mediated ATG-9 vesicle budding (Fig 10).
At Zone 2 of AIY, both synaptic vesicles and ATG-9 vesicles undergo exo-endocytosis [24]. Our data are consistent with ATG-9 undergoing distinct sorting pathways and displaying distinct phenotypes than those seen for synaptic vesicle proteins. We propose a model whereby ATG-9 is sorted by the adaptor complex AP1 to intracellular endocytic intermediates (symbolized here by “ATG-9 vesicle cluster”). CLA-1L, together with presynaptic endocytic proteins that reside in the periactive zone, such as EHS-1 and ITSN-1, as well as the adaptor complexes such as AP-2 and AP180, are necessary for sorting of ATG-9 from endocytic intermediates. Mutations in the active zone gene cla-1L result in abnormal accumulation of ATG-9 into endocytic intermediates and defects in activity-dependent autophagosome formation.
Protein turnover (e.g., via autophagy) is correlated with activity states of synapses and is necessary for optimal synaptic function [94,95]. It is interesting that in cla-1(L) mutants, only activity-induced autophagy, but not baseline autophagy, is affected in the AIY interneurons. We speculate that other molecules play a redundant role in supporting baseline autophagy. Proteins like CLA-1L could be necessary to boost synaptic autophagy for the degradation of damaged synaptic components under high activity states and via sorting of ATG-9. We hypothesize that the sorting could potentiate ATG-9 role as a lipid scramblase in the nucleation of the growth of the autophagosome isolation membrane [96–99], perhaps by the activation of the scramblase activity via exo-endocytosis, and the subsequent transport of lipids necessary for autophagosome biogenesis [100,101]. In summary, we propose a model whereby active zone proteins, like CLA-1L, which bridge the exocytic active zone with the endocytic periactive zone, could regulate ATG-9 sorting to modulate this activity-dependent presynaptic autophagy.
Materials and methods
C. elegans strains and genetics
C. elegans Bristol strain worms were raised on NGM plates at 20°C using OP50 Escherichia coli as a food source [102]. Larva 4 (L4) stage hermaphrodites were examined. For a full list of strains used in the study, please see S2 Table.
Molecular biology and transgenic lines
Expression clones were made in the pSM vector [103]. The transgenic strains (0.5 to 50 ng/μl) were generated using standard injection techniques and coinjected with markers Punc122::GFP (15 to 30 ng/μl), Punc122::dsRed (15 to 30 ng/μl), or Podr-1::rfp (15 to 30 ng/μl). UNC-101, mouse AP1 mu1, and mouse AP2 mu isoform1 were PCR amplified from C. elegans and mouse cDNA (PolyATtract mRNA Isolation Systems, Promega and ProtoScript First Strand cDNA Synthesis Kit, NEB). The plasmid sequences with annotations are included in S1–S5 Plasmids files and can be viewed via ApE (https://jorgensen.biology.utah.edu/wayned/ape/), a free, multiplatform application for visualizing, designing, and presenting biologically relevant DNA sequences.
Forward genetic screen, SNP mapping, and whole-genome sequencing (WGS)
Cla-1(ola285) was isolated from a visual forward genetic screen designed to identify mutants with abnormal localization of ATG-9::GFP in the interneuron AIY. OlaIs34 animals, expressing ATG-9::GFP and mCherry::RAB-3 in AIY interneurons, were mutagenized with ethyl methanesulfonate (EMS) as described previously [102], and their F2 progenies were visually screened under A Leica DM 5000 B compound microscope with an oil objective of HCX PL APO 63×/1.40–0.60 for abnormal ATG-9::GFP localization. SNP mapping [32] was used to map the lesion corresponding to the ola285 allele to a 6.6- to 11.3-Mbp region on chromosome IV. WGS was performed at the Yale Center for Genome Analysis (YCGA) and analyzed on www.usegalaxy.org using “Cloudmap Unmapped Mutant workflow (w/ subtraction of other strains)” as described [33,34]. The ola285 allele was sequenced by using Sanger sequencing (Genewiz), and the genetic lesion confirmed as a single T-to-A nucleotide substitution in Exon 15 of cla-1L that results in a missense mutation I5753N. Complementation tests were performed by generating ola285/cla-1(ok560) trans-heterozygotes. The ola285 allele failed to complement cla-1(ok560).
N-terminal endogenous labeling of CLA-1L
To endogenously tag CLA-1L at the N-terminus in cla-1(ola285) mutants, a CRISPR protocol [25,104] was used to create cla-1(ola506[gfp::cla-1L]);cla-1(ola285).
Cell autonomy of CLA-1L
A CRISPR protocol [105,106] was used to create cla-1 (ola324), in which 2 loxP sites were inserted into 2 introns of cla-1L [25]. Cell-specific removal of CLA-1L in AIY was achieved with a plasmid driving the expression of Cre cDNA under the AIY-specific mod-1 promoter fragment [107].
Cell autonomy and cell-specific rescues
The ATG-9 phenotype in unc-101(m1);cla-1(ola285) was suppressed by cell specifically expressing the C. elegans cDNA of unc-101(m1)/AP1μ1C and the murine cDNA of AP1μ1. Cell-specific expression was achieved using AIY-specific ttx-3G promoter fragment [108]. Cell-specific expression of the murine cDNA of AP2μ in AIY did not suppress the ATG-9 phenotype in unc-101(m1);cla-1(ola285).
Confocal imaging
Hermaphrodite animals of larval stage 4 (L4) were anesthetized using 10 mM levamisole (Sigma-Aldrich) or 50 mM muscimol in M9 buffer, mounted on 2% to 5% agar pads and imaged by using a 60× CFI Plan Apochromat VC, NA 1.4, oil objective (Nikon) on an UltraView VoX spinning-disc confocal microscope (PerkinElmer). Images were processed with Volocity software. Maximum-intensity projections presented in the figs were generated using Fiji (NIH) for all the confocal images. Quantifications were performed on maximal projections of raw data. Note that the ATG-9 subsynaptic foci in cla-1(ola285) were imaged with a different (lower exposure) confocal setting from the wild-type control (higher exposure), to avoid saturating the signal in cla-1 (ola285) animals and at the same time to maximize ATG-9 signal (below saturation) in wild-type animals. The Index of ATG-9 punctum at Zone 2 of AIY for each genotype was consistent across different confocal settings (S1E–S1I Fig).
Electron microscopy
High-pressure freezing, freeze substitution, and sectioning were all performed as previously described in [24,109–111]. AIY Zone 2 was identified based on the anatomical landmarks described in the original C. elegans connectome and recent Zone 2 reconstruction [24,31].
For immuno-EM, sections of 50 nm were collected on nickel slot grids covered with Formvar (EMS). Grids were incubated at 20°C on 50 μl droplets of 0.05 M glycine PBS for 5 min, 1% BSA and 1% CWFS gelatin in PBS for 20 min, anti-GFP rabbit polyclonal (1:20 in 0.3% BSA and 0.3% CWFS gelatin in PBS, ab6556 Abcam) overnight at 4°C and then 60 min at 20°C, 6 PBS washes over 30 min, Protein A Gold conjugated to 10 nm gold (1:75 in 0.3% BSA and 0.3% CWFS gelatin in PBS, University Medical Center Utrecht) for 60 min, 6 PBS washes over 30 min, 2% glutaraldehyde in PBS for 5 min, 3 water washes for 10 s. After drying, grids were post-stained in 2% uranyl acetate for 4 min, and lead citrate for 1 min.
Images were acquired on TALOS L120 (Thermo Fisher) equipped with a Ceta 4k × 4k CMOS camera. For serial sections, images were aligned in z using the TrakEM2 plugin in FIJI [112].
Quantification and statistical analyses
Quantifications of penetrance and expressivity.
A Leica DM500B compound fluorescent microscope was used to visualize and screen the worms in the indicated genetic backgrounds. Penetrance was scored blindly as either wild type (distributed throughout the Zone 2 synaptic region) or mutant (localized into subsynaptic foci) phenotypes for ATG-9 or SNG-1 at Zone 2 of AIY. Mutant phenotype was defined as one or more subsynaptic foci of ATG-9::GFP or SNG-1::GFP at Zone 2 of AIY. For each genotype, at least 40 animals were scored. The Y axis of graphs is named as “% of animals with ATG-9 (or SNG-1) puncta at Zone 2 of AIY.” Statistics for penetrance quantification was determined using two-tailed Fisher’s exact test. Error bars represent 95% of confidence intervals.
In addition, we validated our scoring results by quantifying the expressivity of the phenotypes. Specifically, confocal micrographs of around 15 representative worms for each genotype were acquired using a spinning-disc confocal microscope (PerkinElmer) within the dynamic range of fluorescence (avoiding saturated pixels). Fluorescence values for each AIY Zone 2 were obtained after background subtraction by drawing a freehand line using Fiji along the long axis of Zone 2. The fluorescence peak values and trough values were acquired via the Profile Plot function. Subsynaptic enrichment index was then calculated as (Fpeak-Ftrough)/Ftrough. The Y axes of the graphs are named as “Index of ATG-9 (or SNG-1) punctum at Zone 2 of AIY.” To compare penetrance across different genotypes presented in different graphs of the study, quantifications of wild type from Fig 1M and cla-1(ola285) mutant from Fig 4J were included as references in the following figures: Figs 1M, 4J, 5M, 6G, 8P, S1B, S1G, and S4E. To compare expressivity across different genotypes presented in different graphs of the study, quantifications of wild-type and cla-1(ola285) mutant from Fig 1L were included as references in the following figures: Figs 1L, 4K, 5N, 6H, 7I, S1L, and S7G (we clearly stated this in the figure legends). Similarly, quantifications of percentage and expressivity for unc-11(47) from Fig 5M and 5N were included as references for Figs 5M, 6G, and S4E and Figs 5N, 6H, and 7I respectively.
Mean intensity of SYD-2 at the presynaptic Zone 2 of AIY neurons.
To measure the level of SYD-2 at presynaptic regions, we obtained the fluorescent value using Fiji as indicated above. All settings for the confocal microscope and camera were kept identical to compare the intensity of SYD-2 between the wild-type and cla-1(ok560) mutants. The same ROI was drawn for all micrographs analyzed. The mean fluorescent value of SYD-2 was measured by Fiji.
Mean intensity of CLA-1L at the nerve ring region.
To measure the level of CLA-1L at the nerve ring region, we drew the same ROI in all micrographs analyzed and the mean fluorescent value of CLA-1L was measured by Fiji. All settings for the confocal microscope and camera were kept identical to compare the intensity of CLA-1L between the wild-type and cla-1(ola285) mutants.
Colocalization analysis and analyses of endogenous CLA-1 localization at the presynaptic regions.
The Coloc2 plugin of Fiji was used to measure the Pearson correlation coefficient for colocalization between ATG-9::GFP and CHC-1::BFP, or ATG-9::GFP and SNG-1::BFP, both in cla-1(ola285) mutants. ROI was drawn to include the entire Zone 2 in all Z-stacks. The Coloc2 plugin of Fiji was also used to measure the Pearson correlation coefficient for colocalization between APT-4::mCh and CLA-1::GFP or APT-4::mCh and GFP::CLA-1L.
To be able to rigorously examine the relationship of CLA-1 to subsynaptic domains, we chose a neuronal system in which active zone and periactive zone positions had been well characterized both by EM and light microscopy: the posterior dorsal nerve cord [31,71]. Note that APT-4::mCh is expressed in a subset of neurons in the dorsal nerve cord, driven by the punc-129 promoter, while CLA-1::GFP and GFP::CLA-1 are expressed panneuronally (so green puncta can be present where there are no magenta puncta; Fig 8D–8N and Methods). Therefore, ROI was drawn based on the outline of APT-4::mCh puncta for both CLA-1::GFP and GFP::CLA-1L to compare the colocalization between CLA-1::GFP or GFP::CLA-1L with APT-4::mCh in the neurons that express APT-4::mCh.
Activity-dependent autophagy.
To measure the synaptic autophagosomes in AIY, olaIs35 animals expressing eGFP::LGG-1 and cytoplasmic mCherry in AIY interneurons in the wild-type, cla-1(ola285), and cla-1(ok560) mutant backgrounds were raised in 20°C and then shifted to 25°C for 4 h as described [24] to alter the activity state of the AIY interneuron [80]. LGG-1 puncta in the neurite of AIY was scored under a Leica DM 1048 5000B compound microscope as described [12].
Quantification of transmission EM and immuno-EM.
Quantifications were performed using the TrakEM2 plugin in FIJI. Plasma membranes, endosomes, and dense projections areas in AIY were manually traced as area lists. Synaptic vesicles, dense core vesicles, and immunogold particles were manually marked as ball objects. The layers were scaled in the z-dimension (40 nm for transmission EM, 50 nm for immuno-EM) before making 3D models. Active zone length was calculated by counting the number of continuous sections, which had a dense projection present in AIY and then multiplying by 40 nm (morphology section thickness). The number of synaptic vesicles or dense core vesicles per cross section was calculated by taking the number of marked synaptic vesicles or dense core vesicles within AIY’s cell membrane per z-layer. The endosomes area was calculated by measuring the total marked endosomes area in the sections that possess endosomes. The relationship between the area measurements and the volume is the thickness of the section, which is 40 nm. Because this number is constant for all sections, the volume is directly proportional to all our area measurements. The Analyze Particles tool was used in FIJI to count the immunogold particles, which labelled the ATG-9::GFP. AIY’s cell membrane was manually traced and then a threshold set to only show the darkest 3% on the image. The Analyze Particles tool was set to detect particles 40 to 120 nm2 with a circularity of 0.8 to 1.0. The ATG-9 density (immunogold particles per area) was calculated by dividing the number of ATG-9 immunogold particles by the area of AIY neuron in the corresponding sections.
Statistical analyses.
Statistical analyses were conducted with Prism 7 software and reported in the figure legends. Briefly, two-tailed Fisher’s exact test was used to determine statistical significance of categorical data such as the penetrance of ATG-9 phenotype. For continuous data, such as Index of ATG-9 punctum, two-tailed unpaired Student t test or ordinary one-way ANOVA with Tukey’s post hoc analysis was used. For LGG-1 scoring, since the number of LGG-1 puncta for each condition did not pass D’Agostino and Pearson normality test, the nonparametric analysis Kruskal–Wallis with Dunn’s multiple comparisons test was used. The error bars represent either 95% confidence interval or standard deviation (SD), as indicated in the figure legends, along with the p-values.
Protein sequence alignment of the AP adaptor complex μ subunit
The sequence of C. elegans unc-101/AP-1μ1 (K11D2.3) was acquired from WormBase (https://wormbase.org). The sequences of murine AP-1μ1 (NP_031482.1) and murine AP-2μ (NP_001289899.1) were acquired from NCBI (https://www.ncbi.nlm.nih.gov/protein/?term=). Sequence alignment of worm UNC-101/AP1μ1, mouse AP1μ1, and mouse AP2μ was generated using Tcoffee in Jalview 2.10.5 [113].
Supporting information
S1 Data. Source data for graphs in this paper.
https://doi.org/10.1371/journal.pbio.3002030.s001
(XLSX)
S1 Table. Annotated list of alleles from forward genetic screen.
List of all alleles identified in 3 semiclonal forward genetic screens and categorized by phenotypic class. We focused our study on the ola285 allele due to its phenotype (affects ATG-9 localization but does not affect synaptic vesicle protein localization in Zone 2 synapses) and the higher penetrance and expressivity of its phenotype (see Fig 1).
https://doi.org/10.1371/journal.pbio.3002030.s002
(DOCX)
S1 Fig. Examination of the active zone protein SYD-2 in cla-1(L) mutants and of ATG-9 distribution in cla-1(wy1048) null allele.
(A) Schematic of cla-1 gene, with different protein isoforms. The ok560 allele specifically affects the long protein isoform, while wy1048 allele affects all CLA-1 protein isoforms. (B) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild-type are the same as those in Fig 1M (explained in Methods). Error bars represent 95% confidence interval. ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. (C) Mean intensity of GFP::CLA-1L (WT) and GFP::CLA-1L (I5753N) in the worm nerve ring. Error bars show standard deviation (SD). ****p < 0.0001 by two-tailed unpaired Student t test. Each dot in the scatter plot represents a single animal. (D) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild-type are the same as those in Fig 1M (explained in Methods). Error bars represent 95% confidence interval. ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bar indicates the number of animals scored. (E-H) Distribution of ATG-9::GFP at Zone 2 of AIY in wild type (WT) (E), WT with lower exposure setting (F), cla-1(ola285) (G), and cla-1(ola285) with lower exposure setting (H). Arrows (in G and H) indicate abnormal ATG-9 foci. Note that for this study, the ATG-9 subsynaptic foci in cla-1(ola285) were imaged with a different (lower exposure) confocal setting from the wild-type control (higher exposure), to avoid saturating the signal in cla-1 (ola285) animals and at the same time to maximize ATG-9 signal (below saturation) in wild-type animals. (I) Quantification of the index of ATG-9 punctum (ΔF/F) at Zone 2 of AIY for indicated conditions. The data used to quantify the index of ATG-9 punctum (ΔF/F) in wild-type and cla-1(ola285) mutants are the same as those in Fig 1L (explained in Methods). Error bars show standard deviation (SD). “NS” (not significant), **p < 0.01 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. Note that the index was consistent for ATG-9 in wild-type animals with different imaging confocal settings and was smaller than that in cla-1 (ola285) animals. Scale bar (in E for E-H), 1 μm. Data for S1B, S1C, S1D and S1I Fig can be found in S1 Data.
https://doi.org/10.1371/journal.pbio.3002030.s004
(EPS)
S2 Fig. ATG-9 colocalizes better with clathrin than with the integral synaptic vesicle membrane protein SNG-1.
(A) Quantification of the index of SNG-1::GFP punctum (ΔF/F) at Zone 2 of AIY in wild-type (WT) and cla-1(ola285) mutants. Error bars show standard deviation (SD). “NS” (not significant) by two-tailed unpaired Student t test. Each dot in the scatter plot represents a single animal. (B) Pearson correlation coefficient for colocalization between ATG-9::GFP and SNG-1::BFP, or between ATG-9::GFP and BFP::CHC-1, both in cla-1(ola285) mutants. Error bars show standard deviation (SD). ****p < 0.0001 by two-tailed unpaired Student t test. Each dot in the scatter plot represents a single animal. Data for S2A and S2B Fig can be found in S1 Data.
https://doi.org/10.1371/journal.pbio.3002030.s005
(EPS)
S3 Fig. Dense core vesicles and endosomal structures in the electron micrographs of wild-type and cla-1(ola285) mutants did not reveal major differences.
(A, B) Distribution of SYD-2::GFP at the synaptic Zone 2 and Zone 3 regions of AIY in wild-type (A) and cla-1(ok560) (B) animals. The dashed boxes highlight the presynaptic Zone 2 of AIY examined in this study. (C) Mean intensity of SYD-2 at AIY Zone 2 in wild-type (WT) and cla-1(ok560) mutants. Error bars show standard deviation (SD). “NS” (not significant) by two-tailed unpaired Student t test. Each dot in the scatter plot represents a single animal. (D, E) Electron microscopy of the Zone 2 region in wild-type (D) and cla-1(ola285) mutant animals (E). Same as Fig 2I and 2J, without annotations. (F) Quantification of dense core vesicles in the AIY neurons (AIYL: AIY on the left side; AIYR: AIY on the right side) of 1 wild-type and 1 cla-1(ola285) mutant. Error bars represent standard deviation (SD). ****p < 0.0001 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single section. (G) Measurement of endosome area in the AIY neurons (AIYL: AIY on the left side; AIYR: AIY on the right side) of 2 wild-type and 1 cla-1(ola285) mutant. Error bars represent standard deviation (SD). *p < 0.05 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single section. (H) Total number of endosomes in the AIY neurons of wild-type and cla-1(ola285) mutants. Error bars represent standard deviation (SD). “NS” (not significant) by two-tailed unpaired Student t test. n = 4 for wild-type and n = 2 for cla-1(ola285) mutants. Scale bar (in A for A and B), 5 μm; (in D for D and E), 500 nm. Data for S3C, S3F, S3G and S3H Fig can be found in S1 Data.
https://doi.org/10.1371/journal.pbio.3002030.s006
(EPS)
S4 Fig. The AP-1 and AP-2 adaptor complexes mediate presynaptic trafficking of ATG-9.
(A-D) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (A), unc-101(m1);cla-1(ola285) mutants with mouse AP2μ cDNA cell specifically expressed in AIY (B), dpy-23(e840)/AP2μ (C), and unc-101(m1);dpy-23(e840) (D) mutant animals. ATG-9 subsynaptic foci are indicated by the arrow (in C). (E) Quantification of the percentage of animals displaying ATG-9 subsynaptic foci at AIY Zone 2 in the indicated genotypes. The data used to quantify the percentage of animals displaying ATG-9 subsynaptic foci in wild-type and cla-1(ola285) mutants are the same as those in Fig 4J; the data used in unc-11(e47) are the same as those in Fig 5M (explained in Methods). Error bars represent 95% confidence interval. “NS” (not significant), *p < 0.05, **p < 0.01, ****p < 0.0001 by two-tailed Fisher’s exact test. The number on the bars indicates the number of animals scored. Black asterisks indicate comparison between each group with the wild-type control. Blue asterisks indicate comparison between two specific groups (highlighted with brackets). (F) Sequence alignment of C. elegans UNC-101/AP1μ1, mouse AP1μ1, and mouse AP2μ, generated using Tcoffee in Jalview (Waterhouse and colleagues [113]). Although both AP-1 μ1 and AP-2μ share similarity, C. elegans UNC-101/AP1μ1 is more similar to the mouse AP1μ1 (Query Cover: 100%; Percentage Identity: 74%) than the mouse AP2μ (Query Cover: 98%; Percentage Identity: 41%). Scale bar (in A for A-D), 1 μm. Data for S4E Fig can be found in S1 Data.
https://doi.org/10.1371/journal.pbio.3002030.s007
(EPS)
S5 Fig. Cartoon diagrams of the genetic interactions, and model, in this study.
(A) Mutants for CLA-1L, AP-2, and AP180 adaptor complexes display similar ATG-9 phenotypes at synapses and are necessary for resolving ATG-9-containing foci (clathrin-rich endocytic intermediates). Note that the pink-filled symbols represent the abnormal ATG-9 foci in mutants for CLA-1L, AP-2, and AP180 adaptor complexes. (B) Cartoon diagram of the genetic interactions, and model displaying the genetic interactions of CLA-1L with SDPN-1/syndapin 1 and the AP-1 adaptor complex, which suppress the ATG-9 phenotypes observed for cla-1(ola285), AP-2, or AP180 mutants, consistent with roles for SDPN-1/syndapin 1 and the AP-1 upstream of the formation of the abnormal ATG-9 foci (similar to what was observed for exocytosis mutants in Fig 3, which also suppressed ATG-9 phenotypes in cla-1(ola285) mutants).
https://doi.org/10.1371/journal.pbio.3002030.s008
(EPS)
S6 Fig. Schematics of the strategies for endogenously tagging CLA-1 at C-terminus or N-terminus via CRISPR.
(A) Schematics of the strategy for endogenously tagging CLA-1 at C-terminus via CRISPR. FLP-on-GFP (let-858 3′ UTR flanked by FRT sites followed by GFP) was inserted at the common C-terminus of CLA-1 (shared by CLA-1L, CLA-1M, and CLA-1S protein isoforms; see Xuan and colleagues [25]). FLPase driven by the Prab-3 promoter is expressed panneuronally to induce expression of CLA-1::GFP in an endogenous manner (see Fig 8). (B) Schematics of endogenously tagging CLA-1L at N-terminus via CRISPR (Xuan and colleagues [25]). GFP was inserted at the unique N-terminus of CLA-1L (see Fig 8).
https://doi.org/10.1371/journal.pbio.3002030.s009
(EPS)
S7 Fig. CLA-1L genetically interacts with the endocytic proteins at the periactive zone to regulate ATG-9 trafficking.
(A-F) Distribution of ATG-9::GFP at Zone 2 of AIY in wild-type (WT) (A), cla-1(ola285) (B), ehs-1(ok146) (C), ehs-1(ok146);cla-1(ola285) (D), itsn-1(ok268) (E), and itsn-1(ok268);cla-1(ola285) (F) mutant animals. (G) Quantification of the index of ATG-9 punctum (ΔF/F) at Zone 2 of AIY in the indicated genotypes. The data used to quantify the index of ATG-9 punctum (ΔF/F) in wild-type and cla-1(ola285) mutants are the same as those in Fig 1L (explained in Methods). Error bars show standard deviation (SD). “NS” (not significant), *p < 0.05 by ordinary one-way ANOVA with Tukey’s multiple comparisons test. Each dot in the scatter plot represents a single animal. Scale bar (in A for A-F), 1 μm. Data for S7G Fig can be found in S1 Data.
https://doi.org/10.1371/journal.pbio.3002030.s010
(EPS)
S8 Fig. Disrupted ATG-9 trafficking in cla-1(ok560) mutants is associated with a deficit in activity-induced autophagosome formation.
(A) Quantification of the average number of LGG-1 puncta in the AIY neurites at 20°C and at 25°C for 4 h in wild-type (WT) and cla-1(ok560) mutants. Error bars represent 95% confidence interval. “NS” (not significant), ***p < 0.001 by Kruskal–Wallis test with Dunn’s multiple comparisons test. The number on the bars indicates the number of animals scored. Data for S8A Fig can be found in S1 Data.
https://doi.org/10.1371/journal.pbio.3002030.s011
(EPS)
S1 Plasmids. Please use ApE to open (https://jorgensen.biology.utah.edu/wayned/ape/).
https://doi.org/10.1371/journal.pbio.3002030.s012
(STR)
S2 Plasmids. Please use ApE to open (https://jorgensen.biology.utah.edu/wayned/ape/).
https://doi.org/10.1371/journal.pbio.3002030.s013
(STR)
S3 Plasmids. Please use ApE to open (https://jorgensen.biology.utah.edu/wayned/ape/).
https://doi.org/10.1371/journal.pbio.3002030.s014
(STR)
S4 Plasmids. Please use ApE to open (https://jorgensen.biology.utah.edu/wayned/ape/).
https://doi.org/10.1371/journal.pbio.3002030.s015
(STR)
S5 Plasmids. Please use ApE to open (https://jorgensen.biology.utah.edu/wayned/ape/).
https://doi.org/10.1371/journal.pbio.3002030.s016
(STR)
Acknowledgments
We thank Jihong Bai (Basic Sciences Division, Fred Hutch) and Kang Shen (Department of Biology, Stanford University) for providing strains and constructs. We thank Lin Shao (Department of Neuroscience, Yale University) for assistance with image quantification and statistics. We thank Josh Hawk for providing mouse cDNA. We thank Center for Cellular and Molecular Imaging, Electron Microscopy Facility, the Neuroscience EM core at Yale Medical School for assistance with the work presented here, and Morven Graham, Leslie Gunther-Cummins, Yumei Wu, Irina Kolotuev, Leslie Gunther-Cummins, David Hall, Maike Kittelmann, and Szi-chieh Yu for advice on immunoelectron microscopy experiments. We thank current and past members of the Colón-Ramos lab for help, advice, and insightful comments on the project. We also thank Andrea Stavoe, Ian Gonzalez, Mia Dawn, Peri Kurshan, Janet Richmond, and Pietro De Camilli for assistance and comments on the project. We thank the Caenorhabditis Genetics Center (funded by NIH Office of Research Infrastructure Programs P40 OD010440) for C. elegans strains.
References
- 1. Liang Y, Sigrist S. Autophagy and proteostasis in the control of synapse aging and disease. Curr Opin Neurobiol. 2018;48:113–121. Epub 2017/12/25. pmid:29274917.
- 2. Menzies FM, Fleming A, Caricasole A, Bento CF, Andrews SP, Ashkenazi A, et al. Autophagy and Neurodegeneration: Pathogenic Mechanisms and Therapeutic Opportunities. Neuron. 2017;93(5):1015–1034. Epub 2017/03/11. pmid:28279350.
- 3. Wang Y, Song M, Song F. Neuronal autophagy and axon degeneration. Cell Mol Life Sci. 2018;75(13):2389–2406. Epub 2018/04/21. pmid:29675785.
- 4. Luningschror P, Sendtner M. Autophagy in the presynaptic compartment. Curr Opin Neurobiol. 2018;51:80–85. Epub 2018/03/20. pmid:29549710.
- 5. Vijayan V, Verstreken P. Autophagy in the presynaptic compartment in health and disease. J Cell Biol. 2017;216(7):1895–1906. Epub 2017/05/19. pmid:28515275; PubMed Central PMCID: PMC5496617.
- 6. Soykan T, Haucke V, Kuijpers M. Mechanism of synaptic protein turnover and its regulation by neuronal activity. Curr Opin Neurobiol. 2021;69:76–83. Epub 2021/03/22. pmid:33744822.
- 7. Bunge MB. Fine structure of nerve fibers and growth cones of isolated sympathetic neurons in culture. J Cell Biol. 1973;56(3):713–735. pmid:4347207; PubMed Central PMCID: PMC2108939.
- 8. Soukup SF, Kuenen S, Vanhauwaert R, Manetsberger J, Hernandez-Diaz S, Swerts J, et al. A LRRK2-Dependent EndophilinA Phosphoswitch Is Critical for Macroautophagy at Presynaptic Terminals. Neuron. 2016;92(4):829–844. Epub 2016/10/11. pmid:27720484.
- 9. Maday S, Wallace KE, Holzbaur EL. Autophagosomes initiate distally and mature during transport toward the cell soma in primary neurons. J Cell Biol. 2012;196(4):407–417. pmid:22331844; PubMed Central PMCID: PMC3283992.
- 10. Katsumata K, Nishiyama J, Inoue T, Mizushima N, Takeda J, Yuzaki M. Dynein- and activity-dependent retrograde transport of autophagosomes in neuronal axons. Autophagy. 2010;6(3):378–385. pmid:20150763.
- 11. Shehata M, Matsumura H, Okubo-Suzuki R, Ohkawa N, Inokuchi K. Neuronal Stimulation Induces Autophagy in Hippocampal Neurons That Is Involved in AMPA Receptor Degradation after Chemical Long-Term Depression. J Neurosci. 2012;32(30):10413–10422. WOS:000307125300029. pmid:22836274
- 12. Hill SE, Kauffman KJ, Krout M, Richmond JE, Melia TJ, Colon-Ramos DA. Maturation and Clearance of Autophagosomes in Neurons Depends on a Specific Cysteine Protease Isoform, ATG-4.2. Dev Cell. 2019;49(2):251–266.e8. WOS:000465117900009. pmid:30880001
- 13. Maday S, Holzbaur EL. Compartment-Specific Regulation of Autophagy in Primary Neurons. J Neurosci. 2016;36(22):5933–5945. Epub 2016/06/03. pmid:27251616; PubMed Central PMCID: PMC4887563.
- 14. Wang T, Martin S, Papadopulos A, Harper C, Mavlyutov T, Niranjan D, et al. Control of autophagosome axonal retrograde flux by presynaptic activity unveiled using botulinum neurotoxin type-A. J Neurochem. 2015;134:165. WOS:000360206300390. pmid:25878289
- 15. Kulkarni VV, Anand A, Herr JB, Miranda C, Vogel MC, Maday S. Synaptic activity controls autophagic vacuole motility and function in dendrites. J Cell Biol. 2021;220(6). WOS:000648748300001. pmid:33783472
- 16. Stavoe AKH, Hill SE, Hall DH, Colon-Ramos DA. KIF1A/UNC-104 Transports ATG-9 to Regulate Neurodevelopment and Autophagy at Synapses. Dev Cell. 2016;38(2):171–185. WOS:000380825000011. pmid:27396362
- 17. Soukup SF, Verstreken P. EndoA/Endophilin-A creates docking stations for autophagic proteins at synapses. Autophagy. 2017;13(5):971–972. Epub 2017/03/11. pmid:28282269; PubMed Central PMCID: PMC5446079.
- 18. Maday S, Holzbaur EL. Autophagosome biogenesis in primary neurons follows an ordered and spatially regulated pathway. Dev Cell. 2014;30(1):71–85. pmid:25026034; PubMed Central PMCID: PMC4109719.
- 19. Karanasios E, Walker SA, Okkenhaug H, Manifava M, Hummel E, Zimmermann H, et al. Autophagy initiation by ULK complex assembly on ER tubulovesicular regions marked by ATG9 vesicles. Nat Commun. 2016;7. WOS:000381526400001. pmid:27510922
- 20. Reggiori F, Tucker KA, Stromhaug PE, Klionsky DJ. The Atg1-Atg13 complex regulates Atg9 and Atg23 retrieval transport from the pre-autophagosomal structure. Dev Cell. 2004;6(1):79–90. WOS:000222442700009. pmid:14723849
- 21. Reggiori F, Shintani T, Nair U, Klionsky DJ. Atg9 cycles between mitochondria and the pre-autophagosomal structure in yeasts. Autophagy. 2005;1(2):101–109. WOS:000238564000006. pmid:16874040
- 22. Sawa-Makarska J, Baumann V, Coudevylle N, von Bulow S, Nogellova V, Abert C, et al. Reconstitution of autophagosome nucleation defines Atg9 vesicles as seeds for membrane formation. Science. 2020;369(6508). Epub 2020/09/05. pmid:32883836.
- 23. Sekito T, Kawamata T, Ichikawa R, Suzuki K, Ohsumi Y. Atg17 recruits Atg9 to organize the pre-autophagosomal structure. Genes Cells. 2009;14(5):525–538. Epub 2009/04/18. pmid:19371383.
- 24. Yang S, Park D, Manning L, Hill SE, Cao M, Xuan Z, et al. Presynaptic autophagy is coupled to the synaptic vesicle cycle via ATG-9. Neuron. 2022;110(5):824–840 e10. Epub 2022/01/24. pmid:35065714; PubMed Central PMCID: PMC9017068.
- 25. Xuan Z, Manning L, Nelson J, Richmond JE, Colon-Ramos DA, Shen K, et al. Clarinet (CLA-1), a novel active zone protein required for synaptic vesicle clustering and release. Elife. 2017;6. Epub 2017/11/22. pmid:29160205; PubMed Central PMCID: PMC5728718.
- 26. Mullen G, Frisby D, Mathews E, Mc Manus J, Grundahl K, A. D, et al. The C. elegans RIM-Piccolo homolog CLA-1 is required for normal synaptic transmission. International Worm Meeting. 2017.
- 27. Rand J, McManus J, Grundahl K, Frisby D, Fields S, Duke A. The rip-1 gene encodes a set of presynaptic proteins similar to vertebrate RIM and Piccolo. International Worm Meeting. 2003.
- 28. Bruckner JJ, Gratz SJ, Slind JK, Geske RR, Cummings AM, Galindo SE, et al. Fife, a Drosophila Piccolo-RIM homolog, promotes active zone organization and neurotransmitter release. J Neurosci. 2012;32(48):17048–17058. Epub 2012/12/01. pmid:23197698; PubMed Central PMCID: PMC3524967.
- 29. Wang Y, Okamoto M, Schmitz F, Hofmann K, Sudhof TC. Rim is a putative Rab3 effector in regulating synaptic-vesicle fusion. Nature. 1997;388(6642):593–598. Epub 1997/08/07. pmid:9252191.
- 30. Colon-Ramos DA, Margeta MA, Shen K. Glia promote local synaptogenesis through UNC-6 (netrin) signaling in C-elegans. Science. 2007;318(5847):103–106. WOS:000249915400051. pmid:17916735
- 31. White JG, Southgate E, Thomson JN, Brenner S. The structure of the nervous system of the nematode Caenorhabditis elegans. Philos Trans R Soc Lond B Biol Sci. 1986;314(1165):1–340. Epub 1986/11/12. pmid:22462104.
- 32. Davis MW, Hammarlund M, Harrach T, Hullett P, Olsen S, Jorgensen EM. Rapid single nucleotide polymorphism mapping in C-elegans. BMC Genomics. 2005;6. Artn 118 WOS:000232371800001.
- 33. Doitsidou M, Poole RJ, Sarin S, Bigelow H, Hobert O. C. elegans Mutant Identification with a One-Step Whole-Genome-Sequencing and SNP Mapping Strategy. PLoS ONE. 2010;5(11). ARTN e15435 WOS:000283920000025. pmid:21079745
- 34. Minevich G, Park DS, Blankenberg D, Poole RJ, Hobert O. CloudMap: a cloud-based pipeline for analysis of mutant genome sequences. Genetics. 2012;192(4):1249–1269. Epub 2012/10/12. pmid:23051646; PubMed Central PMCID: PMC3512137.
- 35. Harris TW, Hartwieg E, Horvitz HR, Jorgensen EM. Mutations in synaptojanin disrupt synaptic vesicle recycling. J Cell Biol. 2000;150(3):589–600. Epub 2000/08/10. pmid:10931870; PubMed Central PMCID: PMC2175188.
- 36. Richmond JE, Davis WS, Jorgensen EM. UNC-13 is required for synaptic vesicle fusion in C-elegans. Nat Neuroscience. 1999;2(11):959–964. WOS:000083883300011.
- 37. Koushika SP, Richmond JE, Hadwiger G, Weimer RM, Jorgensen EM, Nonet ML. A post-docking role for active zone protein Rim. Nat Neurosci. 2001;4(10):997–1005. Epub 2001/09/18. pmid:11559854; PubMed Central PMCID: PMC2585766.
- 38. Hata Y, Slaughter CA, Sudhof TC. Synaptic vesicle fusion complex contains unc-18 homologue bound to syntaxin. Nature. 1993;366(6453):347–351. pmid:8247129.
- 39. Sassa T, Harada S, Ogawa H, Rand JB, Maruyama IN, Hosono R. Regulation of the UNC-18-Caenorhabditis elegans syntaxin complex by UNC-13. J Neurosci. 1999;19(12):4772–4777. Epub 1999/06/15. pmid:10366611; PubMed Central PMCID: PMC6782667.
- 40. Schafer WR, Kenyon CJ. A Calcium-Channel Homolog Required for Adaptation to Dopamine and Serotonin in Caenorhabditis-Elegans. Nature. 1995;375(6526):73–78. WOS:A1995QW60400058. pmid:7723846
- 41. Gracheva EO, Hadwiger G, Nonet ML, Richmond JE. Direct interactions between C. elegans RAB-3 and Rim provide a mechanism to target vesicles to the presynaptic density. Neurosci Lett. 2008;444(2):137–142. WOS:000260516300006. pmid:18721860
- 42. Han Y, Kaeser PS, Sudhof TC, Schneggenburger R. RIM determines Ca(2)+ channel density and vesicle docking at the presynaptic active zone. Neuron. 2011;69(2):304–316. Epub 2011/01/26. pmid:21262468; PubMed Central PMCID: PMC3259453.
- 43. Kaeser PS, Deng LB, Wang Y, Dulubova I, Liu XR, Rizo J, et al. RIM Proteins Tether Ca2+ Channels to Presynaptic Active Zones via a Direct PDZ-Domain Interaction. Cell. 2011;144(2):282–295. WOS:000286459900012. pmid:21241895
- 44. Maycox PR, Link E, Reetz A, Morris SA, Jahn R. Clathrin-coated vesicles in nervous tissue are involved primarily in synaptic vesicle recycling. J Cell Biol. 1992;118(6):1379–1388. Epub 1992/09/01. pmid:1325974; PubMed Central PMCID: PMC2289614.
- 45. Royle SJ, Lagnado L. Endocytosis at the synaptic terminal. J Physiol-London. 2003;553(2):345–355. WOS:000187279700003. pmid:12963793
- 46. Takei K, Mundigl O, Daniell L, DeCamilli P. The synaptic vesicle cycle: A single vesicle budding step involving clathrin and dynamin. J Cell Biol. 1996;133(6):1237–1250. WOS:A1996UT40700008. pmid:8682861
- 47. Keen JH, Beck KA. Identification of the clathrin-binding domain of assembly protein AP-2. Biochem Biophys Res Commun. 1989;158(1):17–23. Epub 1989/01/16. pmid:2912448.
- 48. Traub LM. Sorting it out: AP-2 and alternate clathrin adaptors in endocytic cargo selection. J Cell Biol. 2003;163(2):203–208. WOS:000186331100001. pmid:14581447
- 49. Hirst J, Robinson MS. Clathrin and adaptors. BBA-Mol Cell Res. 1998;1404(1–2):173–193. WOS:000075714500015. pmid:9714795
- 50. Puri C, Renna M, Bento CF, Moreau K, Rubinsztein DC. Diverse autophagosome membrane sources coalesce in recycling endosomes. Cell. 2013;154(6):1285–1299. pmid:24034251; PubMed Central PMCID: PMC3791395.
- 51. Popovic D, Dikic I. TBC1D5 and the AP2 complex regulate ATG9 trafficking and initiation of autophagy. EMBO Rep. 2014;15(4):392–401. pmid:24603492; PubMed Central PMCID: PMC3989670.
- 52. Imai K, Hao F, Fujita N, Tsuji Y, Oe Y, Araki Y, et al. Atg9A trafficking through the recycling endosomes is required for autophagosome formation. J Cell Sci. 2016;129(20):3781–3791. WOS:000389595200009. pmid:27587839
- 53. McMahon HT, Boucrot E. Molecular mechanism and physiological functions of clathrin-mediated endocytosis. Nat Rev Mol Cell Biol. 2011;12(8):517–533. Epub 2011/07/23. pmid:21779028.
- 54. Cheung G, Cousin MA. Adaptor protein complexes 1 and 3 are essential for generation of synaptic vesicles from activity-dependent bulk endosomes. J Neurosci. 2012;32(17):6014–6023. Epub 2012/04/28. pmid:22539861; PubMed Central PMCID: PMC3348540.
- 55. Kononenko NL, Puchkov D, Classen GA, Walter AM, Pechstein A, Sawade L, et al. Clathrin/AP-2 mediate synaptic vesicle reformation from endosome-like vacuoles but are not essential for membrane retrieval at central synapses. Neuron. 2014;82(5):981–988. Epub 2014/06/09. pmid:24908483.
- 56. Cheung G, Cousin MA. Synaptic vesicle generation from activity-dependent bulk endosomes requires a dephosphorylation-dependent dynamin-syndapin interaction. J Neurochem. 2019;151(5):570–583. WOS:000490703100001. pmid:31479508
- 57. Clayton EL, Cousin MA. The molecular physiology of activity-dependent bulk endocytosis of synaptic vesicles. J Neurochem. 2009;111(4):901–914. Epub 2009/09/22. pmid:19765184; PubMed Central PMCID: PMC2871311.
- 58. Imoto Y, Raychaudhuri S, Ma Y, Fenske P, Sandoval E, Itoh K, et al. Dynamin is primed at endocytic sites for ultrafast endocytosis. Neuron. 2022;110(17):2815–2835 e13. Epub 2022/07/10. pmid:35809574; PubMed Central PMCID: PMC9464723.
- 59. Glyvuk N, Tsytsyura Y, Geumann C, D’Hooge R, Huve J, Kratzke M, et al. AP-1/sigma1B-adaptin mediates endosomal synaptic vesicle recycling, learning and memory. EMBO J. 2010;29(8):1318–1330. Epub 2010/03/06. pmid:20203623; PubMed Central PMCID: PMC2868567.
- 60. Park J, Cho OY, Kim JA, Chang S. Endosome-mediated endocytic mechanism replenishes the majority of synaptic vesicles at mature CNS synapses in an activity-dependent manner. Sci Rep. 2016;6. WOS:000381560200001. pmid:27534442
- 61. Haucke V, Neher E, Sigrist SJ. Protein scaffolds in the coupling of synaptic exocytosis and endocytosis. Nat Rev Neurosci. 2011;12(3):127–138. WOS:000287474900001. pmid:21304549
- 62. Maritzen T, Haucke V. Coupling of exocytosis and endocytosis at the presynaptic active zone. Neurosci Res. 2018;127:45–52. WOS:000425857300006. pmid:29221907
- 63. Lou XL. Sensing Exocytosis and Triggering Endocytosis at Synapses: Synaptic Vesicle Exocytosis-Endocytosis Coupling. Front Cell Neurosci. 2018;12. WOS:000427422300001. pmid:29593500
- 64. Mukherjee K, Yang XF, Gerber SH, Kwon HB, Ho A, Castillo PE, et al. Piccolo and bassoon maintain synaptic vesicle clustering without directly participating in vesicle exocytosis. Proc Natl Acad Sci U S A. 2010;107(14):6504–6509. WOS:000276374400068. pmid:20332206
- 65. Fenster SD, Chung WJ, Zhai R, Cases-Langhoff C, Voss B, Garner AM, et al. Piccolo, a presynaptic zinc finger protein structurally related to bassoon. Neuron. 2000;25(1):203–214. WOS:000085043200022. pmid:10707984
- 66. Gundelfinger ED, Reissner C, Garner CC. Role of Bassoon and Piccolo in Assembly and Molecular Organization of the Active Zone. Front Synaptic Neurosci. 2015;7:19. Epub 2016/01/23. pmid:26793095; PubMed Central PMCID: PMC4709825.
- 67. Dieck ST, Sanmarti-Vila L, Langnaese K, Richter K, Kindler S, Soyke A, et al. Bassoon, a novel zinc-finger CAG/glutamine-repeat protein selectively localized at the active zone of presynaptic nerve terminals. J Cell Biol. 1998;142(2):499–509. WOS:000075111300016. pmid:9679147
- 68. CasesLanghoff C, Voss B, Garner AM, Appeltauer U, Takei K, Kindler S, et al. Piccolo, a novel 420 kDa protein associated with the presynaptic cytomatrix. Eur J Cell Biol. 1996;69(3):214–223. WOS:A1996UA00600002. pmid:8900486
- 69. Dani A, Huang B, Bergan J, Dulac C, Zhuang XW. Superresolution Imaging of Chemical Synapses in the Brain. Neuron. 2010;68(5):843–856. WOS:000285664900007. pmid:21144999
- 70. Fenster SD, Kessels MM, Qualmann B, Chung WJ, Nash J, Gundelfinger ED, et al. Interactions between Piccolo and the actin/dynamin-binding protein Abp1 link vesicle endocytosis to presynaptic active zones. J Biol Chem. 2003;278(22):20268–20277. WOS:000183078000089. pmid:12654920
- 71. Bai JH, Hu ZT, Dittman JS, Pym ECG, Kaplan JM. Endophilin Functions as a Membrane-Bending Molecule and Is Delivered to Endocytic Zones by Exocytosis. Cell. 2010;143(3):430–441. WOS:000283603900016. pmid:21029864
- 72. Koh TW, Korolchuk VI, Wairkar YP, Jiao W, Evergren E, Pan H, et al. Eps15 and Dap160 control synaptic vesicle membrane retrieval and synapse development. J Cell Biol. 2007;178(2):309–322. Epub 2007/07/11. pmid:17620409; PubMed Central PMCID: PMC2064449.
- 73. Pechstein A, Bacetic J, Vahedi-Faridi A, Gromova K, Sundborger A, Tomlin N, et al. Regulation of synaptic vesicle recycling by complex formation between intersectin 1 and the clathrin adaptor complex AP2. P Natl Acad Sci USA. 2010;107(9):4206–4211. WOS:000275131100048. pmid:20160082
- 74. Koh TW, Verstreken P, Bellen HJ. Dap160/intersectin acts as a stabilizing scaffold required for synaptic development and vesicle endocytosis. Neuron. 2004;43(2):193–205. WOS:000222905400009. pmid:15260956
- 75. Marie B, Sweeney ST, Poskanzer KE, Roos J, Kelly RB, Davis GW. Dap160/intersectin scaffolds the periactive zone to achieve high-fidelity endocytosis and normal synaptic growth. Neuron. 2004;43(2):207–219. Epub 2004/07/21. pmid:15260957.
- 76. Gubar O, Morderer D, Tsyba L, Croise P, Houy S, Ory S, et al. Intersectin: The Crossroad between Vesicle Exocytosis and Endocytosis. Front Endocrinol (Lausanne). 2013;4:109. Epub 2013/08/30. pmid:23986746; PubMed Central PMCID: PMC3753573.
- 77. Wang W, Bouhours M, Gracheva EO, Liao EH, Xu KL, Sengar AS, et al. ITSN-1 controls vesicle recycling at the neuromuscular junction and functions in parallel with DAB-1. Traffic. 2008;9(5):742–754. WOS:000254793000011. pmid:18298590
- 78. Gowrisankaran S, Houy S, Del Castillo JGP, Steubler V, Gelker M, Kroll J, et al. Endophilin-A coordinates priming and fusion of neurosecretory vesicles via intersectin. Nat Commun. 2020;11(1):1266. Epub 2020/03/11. pmid:32152276; PubMed Central PMCID: PMC7062783.
- 79. Mishra R, Sengul GF, Candiello E, Schu P. Synaptic AP2 CCV life cycle regulation by the Eps15, ITSN1, Sgip1/AP2, synaptojanin1 interactome. Sci Rep. 2021;11(1):8007. Epub 2021/04/15. pmid:33850201; PubMed Central PMCID: PMC8044098.
- 80. Hawk JD, Calvo AC, Liu P, Almoril-Porras A, Aljobeh A, Torruella-Suarez ML, et al. Integration of Plasticity Mechanisms within a Single Sensory Neuron of C. elegans Actuates a Memory. Neuron. 2018;97(2):356–367.e4. WOS:000422692000013. pmid:29307713
- 81. Hill SE, Colon-Ramos DA. The Journey of the Synaptic Autophagosome: A Cell Biological Perspective. Neuron. 2020;105(6):961–973. Epub 2020/03/20. pmid:32191859.
- 82. Stavoe AKH, Holzbaur ELF. Autophagy in Neurons. Annu Rev Cell Dev Bi. 2019;35:477–500. WOS:000490122300020. pmid:31340124
- 83. Waites CL, Leal-Ortiz SA, Okerlund N, Dalke H, Fejtova A, Altrock WD, et al. Bassoon and Piccolo maintain synapse integrity by regulating protein ubiquitination and degradation. EMBO J. 2013;32(7):954–969. Epub 2013/02/14. pmid:23403927; PubMed Central PMCID: PMC3616282.
- 84. Hoffmann-Conaway S, Brockmann MM, Schneider K, Annamneedi A, Rahman KA, Bruns C, et al. Parkin contributes to synaptic vesicle autophagy in Bassoon-deficient mice. Elife. 2020;9. Epub 2020/05/05. pmid:32364493; PubMed Central PMCID: PMC7224700.
- 85. Montenegro-Venegas C, Annamneedi A, Hoffmann-Conaway S, Gundelfinger ED, Garner CC. BSN (bassoon) and PRKN/parkin in concert control presynaptic vesicle autophagy. Autophagy. 2020;16(9):1732–1733. WOS:000557617100001. pmid:32718208
- 86. Okerlund ND, Schneider K, Leal-Ortiz S, Montenegro-Venegas C, Kim SA, Garner LC, et al. Bassoon Controls Presynaptic Autophagy through Atg5. Neuron. 2017;93(4):897–913 e7. Epub 2017/02/24. pmid:28231469.
- 87. Zhou C, Ma K, Gao R, Mu C, Chen L, Liu Q, et al. Regulation of mATG9 trafficking by Src- and ULK1-mediated phosphorylation in basal and starvation-induced autophagy. Cell Res. 2017;27(2):184–201. Epub 2016/12/10. pmid:27934868; PubMed Central PMCID: PMC5339848.
- 88. Guo YJ, Chang CM, Huang R, Liu B, Bao L, Liu W. AP1 is essential for generation of autophagosomes from the trans-Golgi network. J Cell Sci. 2012;125(7):1706–1715. WOS:000303911300011. pmid:22328508
- 89. Olivas TJ, Wu Y, Yu S, Luan L, Choi P, Nag S, et al. ATG9 vesicles comprise the seed membrane of mammalian autophagosomes. bioRxiv. 2022:2022.08.16.504143.
- 90. Boyken J, Gronborg M, Riedel D, Urlaub H, Jahn R, Chua JJ. Molecular profiling of synaptic vesicle docking sites reveals novel proteins but few differences between glutamatergic and GABAergic synapses. Neuron. 2013;78(2):285–297. Epub 2013/04/30. pmid:23622064.
- 91. Chantranupong L, Saulnier JL, Wang W, Jones DR, Pacold ME, Sabatini BL. Rapid purification and metabolomic profiling of synaptic vesicles from mammalian brain. Elife. 2020;9. Epub 2020/10/13. pmid:33043885; PubMed Central PMCID: PMC7575323.
- 92. Taoufiq Z, Ninov M, Villar-Briones A, Wang HY, Sasaki T, Roy MC, et al. Hidden proteome of synaptic vesicles in the mammalian brain. Proc Natl Acad Sci U S A. 2020;117(52):33586–33596. Epub 2020/12/31. pmid:33376223; PubMed Central PMCID: PMC7776996.
- 93. Park D, Wu Y, Wang X, Gowrishankar S, Baublis A, De Camilli P. Synaptic vesicle proteins and ATG9A self-organize in distinct vesicle phases within synapsin condensates. Nat Commun. 2023;14(1):455. Epub 2023/01/29. pmid:36709207; PubMed Central PMCID: PMCPMC9884207
- 94. Bingol B, Sheng M. Deconstruction for reconstruction: the role of proteolysis in neural plasticity and disease. Neuron. 2011;69(1):22–32. Epub 2011/01/12. pmid:21220096.
- 95. Heo S, Diering GH, Na CH, Nirujogi RS, Bachman JL, Pandey A, et al. Identification of long-lived synaptic proteins by proteomic analysis of synaptosome protein turnover. Proc Natl Acad Sci U S A. 2018;115(16):E3827–E3836. Epub 2018/04/04. pmid:29610302; PubMed Central PMCID: PMC5910858.
- 96. Matoba K, Kotani T, Tsutsumi A, Tsuji T, Mori T, Noshiro D, et al. Atg9 is a lipid scramblase that mediates autophagosomal membrane expansion. Nat Struct Mol Biol. 2020. Epub 2020/10/28. pmid:33106658.
- 97. Maeda S, Yamamoto H, Kinch LN, Garza CM, Takahashi S, Otomo C, et al. Structure, lipid scrambling activity and role in autophagosome formation of ATG9A. Nat Struct Mol Biol. 2020. Epub 2020/10/28. pmid:33106659.
- 98. Matoba K, Noda NN. Secret of Atg9: lipid scramblase activity drives de novo autophagosome biogenesis. Cell Death Differ. 2020. Epub 2020/11/13. pmid:33177618.
- 99. Ghanbarpour A, Valverde DP, Melia TJ, Reinisch KM. A model for a partnership of lipid transfer proteins and scramblases in membrane expansion and organelle biogenesis. P Natl Acad Sci USA. 2021;118(16). ARTN e2101562118 WOS:000642462300008.
- 100. Yamamoto H, Kakuta S, Watanabe TM, Kitamura A, Sekito T, Kondo-Kakuta C, et al. Atg9 vesicles are an important membrane source during early steps of autophagosome formation. J Cell Biol. 2012;198(2):219–233. WOS:000306686100010. pmid:22826123
- 101. Feng YC, Klionsky DJ. Autophagic membrane delivery through ATG9. Cell Res. 2017;27(2):161–162. WOS:000394187900001. pmid:28072404
- 102. Brenner S. The genetics of Caenorhabditis elegans. Genetics. 1974;77(1):71–94. pmid:4366476; PubMed Central PMCID: PMC1213120.
- 103. Shen K, Bargmann CI. The immunoglobulin superfamily protein SYG-1 determines the location of specific synapses in C. elegans. Cell. 2003;112(5):619–630. Epub 2003/03/12. pmid:12628183.
- 104. Dickinson DJ, Pani AM, Heppert JK, Higgins CD, Goldstein B. Streamlined Genome Engineering with a Self-Excising Drug Selection Cassette. Genetics. 2015;200(4):1035–1049. Epub 2015/06/06. pmid:26044593; PubMed Central PMCID: PMC4574250.
- 105. Paix A, Wang YM, Smith HE, Lee CYS, Calidas D, Lu T, et al. Scalable and Versatile Genome Editing Using Linear DNAs with Microhomology to Cas9 Sites in Caenorhabditis elegans. Genetics. 2014;198(4):1347–1356. WOS:000346059300001. pmid:25249454
- 106. Arribere JA, Bell RT, Fu BXH, Artiles KL, Hartman PS, Fire AZ. Efficient Marker-Free Recovery of Custom Genetic Modifications with CRISPR/Cas9 in Caenorhabditis elegans. Genetics. 2014;198(3):837–846. WOS:000344373300003. pmid:25161212
- 107. Flavell SW, Pokala N, Macosko EZ, Albrecht DR, Larsch J, Bargmann CI. Serotonin and the Neuropeptide PDF Initiate and Extend Opposing Behavioral States in C. elegans. Cell. 2013;154(5):1023–1035. WOS:000323767300016. pmid:23972393
- 108. Wenick AS, Hobert O. Genomic cis-regulatory architecture and trans-acting regulators of a single interneuron-specific gene battery in C-elegans. Dev Cell. 2004;6(6):757–770. WOS:000222443200009. pmid:15177025
- 109. Rostaing P, Weimer RM, Jorgensen EM, Triller A, Bessereau JL. Preservation of immunoreactivity and fine structure of adult C. elegans tissues using high-pressure freezing. J Histochem Cytochem. 2004;52(1):1–12. Epub 2003/12/23. pmid:14688212.
- 110. Manning L, Richmond J. High-Pressure Freeze and Freeze Substitution Electron Microscopy in C. elegans. Methods Mol Biol. 2015;1327:121–140. Epub 2015/10/02. pmid:26423972.
- 111. Kolotuev I. Positional correlative anatomy of invertebrate model organisms increases efficiency of TEM data production. Microsc Microanal. 2014;20(5):1392–1403. Epub 2014/09/03. pmid:25180638.
- 112. Cardona A, Saalfeld S, Schindelin J, Arganda-Carreras I, Preibisch S, Longair M, et al. TrakEM2 Software for Neural Circuit Reconstruction. PLoS ONE. 2012;7(6). WOS:000305652700008. pmid:22723842
- 113. Waterhouse AM, Procter JB, Martin DM, Clamp M, Barton GJ. Jalview Version 2—a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009;25(9):1189–1191. Epub 2009/01/20. pmid:19151095; PubMed Central PMCID: PMC2672624.
- 114. Hobert O. Development of left/right asymmetry in the Caenorhabditis elegans nervous system: from zygote to postmitotic neuron. Genesis. 2014;52(6):528–543. Epub 2014/02/11. pmid:24510690.
- 115. Clark DA, Biron D, Sengupta P, Samuel ADT. The AFD sensory neurons encode multiple functions underlying thermotactic behavior in Caenorhabditis elegans. J Neurosci. 2006;26(28):7444–7451. WOS:000238987700015. pmid:16837592
- 116. Biron D, Shibuya M, Gabel C, Wasserman SM, Clark DA, Brown A, et al. A diacylglycerol kinase modulates long-term thermotactic behavioral plasticity in C. elegans. Nat Neurosci. 2006;9(12):1499–1505. Epub 2006/11/07. pmid:17086178.