To determine S. pombe genes necessary for ER-phagy, we carried out an imaging-based mutant display screen (Fig 1A). Chemical mutagenesis was utilized to a pressure during which the copy numbers of twenty-two common autophagy genes had been doubled to scale back the prospect of isolating their mutants within the display screen. ER-phagy was monitored by analyzing microscopically the relocalization of the integral ER membrane protein Ost4-GFP to the vacuole upon DTT therapy [26,34,35]. A mutant clone remoted on this display screen was subjected to next-generation sequencing-assisted bulk segregant evaluation [36], which indicated {that a} missense mutation (T17M) in an uncharacterized gene SPBC30D10.09c is a candidate causal mutation (S1A Fig). For causes described under, we named this gene yep1 (for Yop1- and REEP-related protein required for ER-phagy). Yep1 protein belongs to the REEP protein household. This protein household encompasses 2 subfamilies, each of that are current in most metazoan and fungal species (Fig 1B) [37–40]. There are 6 REEP household proteins in people. Amongst them, REEP1-4 proteins belong to 1 subfamily and REEP5-6 proteins belong to the opposite. S. pombe has 2 REEP household proteins, Yep1 and Yop1. Yep1 is the ortholog of human REEP1-4, and Yop1 is the ortholog of human REEP5-6 (Fig 1B). Human REEP1-4 are ER-localizing proteins [38,41,42], Equally, we discovered that Yep1 exhibited an ER localization sample throughout vegetative development (S1B Fig).

Fig 1. Yep1 is required for ER-phagy and nucleophagy.

(A) Schematic of the imaging-based display screen for ER-phagy mutants. Mutagenized clones harboring the Ost4-GFP reporter had been handled with DTT for 16 hours to induce ER-phagy. The phenotype-causing mutation in an ER-phagy-deficient clone was recognized by bulk segregant evaluation. (B) Phylogenetic relationships of REEP household proteins in consultant metazoan and fungal species. A most probability tree was constructed utilizing IQ-TREE and rooted by midpoint rooting. Department labels are the SH-aLRT help values (%) and the UFBoot help values (%) calculated by IQ-TREE. Scale bar signifies 0.3 substitutions per website. (C) Relocalization of the integral ER membrane protein Ost4-CFP to the vacuole after nitrogen hunger (−N) or DTT therapy (+DTT) was abolished in yep1Δ cells. Wild-type (WT), yep1Δ, and atg5Δ cells expressing Ost4-CFP had been examined by microscopy earlier than and after 12-hour hunger or DTT therapy. Cpy1-mCherry is a vacuole lumen marker. Bar, 5 μm. (D–E) Autophagic processing of the ER membrane protein Erg11-GFP was abolished in yep1Δ cells. Cells expressing Erg11-GFP had been collected earlier than and after 24-hour hunger (D) and DTT therapy (E), and whole lysates had been analyzed by immunoblotting utilizing an antibody in opposition to GFP. Submit-immunoblotting staining of the PVDF membrane utilizing Coomassie Sensible Blue (CBB) served because the loading management. The third pattern (yep1Δ Yep1) is a yep1Δ pressure remodeled with an integrating plasmid expressing Yep1 tagged with mCherry. The blot picture is a consultant of triplicate experiments. Quantitation of triplicate experiments is proven under the picture. (F) Autophagic processing of the cortical ER protein Rtn1-GFP was abolished in yep1Δ cells. The blot picture is a consultant of triplicate experiments. Quantitation of triplicate experiments is proven under the picture. (G) Autophagic processing of the nuclear envelope protein Ish1-GFP was abolished in yep1Δ cells. The blot picture is a consultant of triplicate experiments. Quantitation of triplicate experiments is proven under the picture. (H) DTT-induced autophagic processing of the nucleoplasmic protein Pus1-mECtrine was abolished in epr1Δ and yep1Δ cells. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. (I) Nitrogen starvation-induced autophagic processing of the nucleus protein Pus1-mECtrine was diminished in epr1Δ cells and abolished in yep1Δ cells. The blot picture is a consultant of triplicate experiments. Quantitation of triplicate experiments is proven under the picture. (J) Autophagic processing of CFP-Atg8 was largely regular in yep1Δ cells. The blot picture is a consultant of triplicate experiments. Quantitation of triplicate experiments is proven under the picture. Numerical knowledge underlying panels D-J could be present in S1 Information, and uncooked photographs for panels D-J could be present in S1 Uncooked Photos.

https://doi.org/10.1371/journal.pbio.3002372.g001

Threonine 17 in Yep1 is a conserved residue in REEP1-4 subfamily proteins. Thus, we hypothesized that the T17M mutation could compromise the perform of Yep1 and consequently trigger an ER-phagy defect. According to this concept, deletion of yep1 severely diminished the relocalization of Ost4-CFP to the vacuole upon DTT therapy (+DTT) or nitrogen hunger therapy (−N) (Fig 1C). yep1Δ cells additionally exhibited a extreme defect within the autophagic processing of GFP-tagged integral ER membrane protein Erg11 into free GFP (Fig 1D and 1E). Reintroducing Yep1 into yep1Δ cells utterly rescued the defect. These outcomes point out that Yep1 is important for ER-phagy.

As Ost4 and Erg11 localize to each the cortical ER and the nuclear envelope, we additionally examined the autophagic processing of Rtn1-GFP, which localizes solely on the cortical ER, and Ish1-GFP, which is a nuclear envelope protein. The lack of Yep1 abolished DTT- and starvation-induced processing of Rtn1-GFP and Ish1-GFP (Fig 1F and 1G), indicating that Yep1 is required for the autophagy of each subcompartments of the ER.

The remark that the nuclear envelope protein Ish1-GFP is subjected to autophagy prompt to us that the autophagic turnover of nucleoplasmic parts, i.e., nucleophagy, could happen in S. pombe. Certainly, underneath each DTT and hunger therapies, a nucleoplasmic protein Pus1-mECitrine was processed to free mECitrine in an Atg5-dependent method (Fig 1H and 1I). The lack of the ER-phagy receptor Epr1 diminished the processing of Pus1-mECitrine, suggesting that Epr1 additionally acts as a nucleophagy receptor. Deletion of yep1 abolished the processing of Pus1-mECitrine in each DTT- and starvation-treated cells, indicating that Yep1 is important for nucleophagy. According to its function in ER-phagy and nucleophagy, Yep1 was noticed to type puncta at roughly 30% of the websites the place each Atg8 and Epr1 fashioned puncta (S1C and S1D Fig), suggesting that Yep1 participates in ER-phagy and nucleophagy at websites of autophagosome meeting.

In distinction to the extreme ER-phagy and nucleophagy defects of yep1Δ cells, bulk autophagy in yep1Δ cells was largely regular as indicated by the processing of CFP-Atg8 (Fig 1J). As well as, one other readout of bulk autophagy, the processing of fluorescent protein-tagged cytosolic protein Tdh1 (glyceraldehyde-3-phosphate dehydrogenase (GAPDH)) [43], was additionally largely unaffected in yep1Δ cells (S1E Fig). According to the dearth of bulk autophagy defects, transmission electron microscopy (TEM) evaluation confirmed that autophagosome accumulation within the fsc1Δ mutant, which is flawed in autophagosome–vacuole fusion [44], was not notably affected by the deletion of yep1 (S1F Fig). Inspection of the electron micrographs confirmed that in fsc1Δ cells, there are autophagosomes containing a ring-shaped membrane construction, probably of the ER/nuclear envelope origin. We name such autophagosomes double-ring constructions. The extent of double-ring constructions was greater underneath DTT therapy than underneath hunger therapy (S1G Fig). That is inconsistent with the outcomes of the fluorescent protein cleavage assay, probably as a result of the cleavage assay underestimates autophagic flux underneath DTT therapy as a result of an inhibition of vacuolar proteolysis by DTT therapy [35]. The variety of double-ring constructions per cell was markedly decrease in fsc1Δ yep1Δ cells than in fsc1Δ cells (S1F and S1G Fig), suggesting a defect in forming autophagosomes containing ER-phagy and nucleophagy-related cargo membrane constructions.

Essentially the most well-understood sort of proteins necessary for ER-phagy however not bulk autophagy are ER-phagy receptors, whose features depend on their interactions with Atg8. We examined whether or not Yep1 can work together with Atg8 utilizing the Pil1 co-tethering assay [45]. Epr1, however not Yep1, interacted with Atg8 on this assay (S1H Fig), indicating that Yep1 is unlikely to be an ER-phagy receptor.

As the right functioning of the fission yeast ER-phagy receptor Epr1 is dependent upon Ire1, which up-regulates the expression of Epr1 throughout ER stress [26], we examined the likelihood that Yep1 additionally promotes the expression of Epr1. Immunoblotting evaluation confirmed that ER stress-induced up-regulation of the protein stage of Epr1 occurred usually in yep1Δ cells (S1I Fig), ruling out this chance. No interplay between Yep1 and Epr1 was detected utilizing a yeast two-hybrid assay (S1J Fig). Moreover, though the function of Epr1 could be bypassed by fusing a synthetic AIM to the integral ER membrane protein Erg11 [26] (S1K Fig), this fusion didn’t suppress the ER-phagy defect of yep1Δ (S1L Fig), indicating that the principle function of Yep1 shouldn’t be selling the perform of Epr1. According to this concept, growing the extent of Epr1 didn’t suppress yep1Δ (S1L Fig), and growing the extent of Yep1 didn’t suppress epr1Δ (S1K Fig). Collectively, these outcomes recommend that Yep1 and Epr1 play completely different roles in ER-phagy.

Subsequent, we examined whether or not yep1Δ impacts the localization of Epr1 and Atg8 throughout ER-phagy. In each wild-type and yep1Δ cells, Epr1 and Atg8 colocalized at punctate constructions shortly after ER-phagy induction by DTT or hunger therapy (S2A Fig), indicating that Yep1 shouldn’t be important for the preliminary stage of ER-phagy throughout which Epr1 mediates a connection between the ER and Atg8-decorated autophagic membranes.

In our evaluation of nucleophagy phenotypes utilizing strains expressing the nucleoplasmic protein Pus1-mECitrine, we observed that Pus1-mECitrine fashioned cytoplasmic puncta in yep1Δ cells after ER-phagy induction (Fig 2A). We then examined an interior nuclear membrane protein Bqt4 and located that it additionally fashioned cytoplasmic puncta in yep1Δ cells after ER-phagy induction (Fig 2B). No Pus1 or Bqt4 puncta had been noticed in atg5Δ cells and yep1Δ atg5Δ cells (Fig 2A and 2B), suggesting that these puncta are autophagy-related constructions. Beneath hunger therapy, no Pus1 or Bqt4 puncta had been noticed in wild-type cells. Beneath DTT therapy, a smaller variety of Pus1 and Bqt4 puncta had been noticed in wild-type cells, probably as a result of DTT therapy causes a gentle accumulation of autophagosomes within the cytoplasm (see under). We examined whether or not Pus1 puncta and Bqt4 puncta in yep1Δ cells colocalize and located that Pus1 puncta nearly at all times overlapped with Bqt4 puncta and an important majority (>80%) of Bqt4 puncta overlapped with Pus1 puncta (Figs 2C and S2B). Thus, the nucleophagy defect of yep1Δ is accompanied by the cytoplasmic accumulation of nucleus-derived membrane constructions containing interior nuclear membrane parts and nucleoplasmic parts. In yep1Δ cells, lower than 10% of Pus1 puncta and Bqt4 puncta colocalized with Atg8 puncta (S2C and S2D Fig), indicating that the majority nucleus-derived membrane constructions accumulating within the cytoplasm of yep1Δ cells should not related to autophagic membranes embellished with Atg8.

Fig 2. Yep1 is required for the autophagosomal enclosure of ER-phagy and nucleophagy cargos.

(A) Bqt4-mECtrine fashioned cytoplasmic puncta in yep1Δ cells after nitrogen hunger (−N) or DTT therapy. Bar, 5 μm. (B) Pus1-mECtrine fashioned cytoplasmic puncta in yep1Δ cells after nitrogen hunger or DTT therapy. Bar, 5 μm. (C) The colocalization between Bqt4 puncta and Pus1 puncta accumulating in yep1Δ cells. Bar, 5 μm. (D) Electron microscopy evaluation of yep1Δ cells handled with DTT or nitrogen hunger. N, nucleus; V, vacuole; M, mitochondrion. Ring-shaped membrane constructions are denoted by pink arrows. Bar, 1 μm. (E) Electron microscopy evaluation of gold nanoparticle-labeled membrane constructions of the ER/nuclear envelope origin in yep1Δ cells. MTn tagging of Ost4 allowed membrane constructions of the ER/nuclear envelope origin to be labeled by EM-visible gold nanoparticles. N, nucleus; M, mitochondrion. Gold nanoparticle-labeled ring-shaped membrane constructions are denoted by pink arrows. (F and G) Pus1 puncta that amassed in nitrogen hunger–handled yep1Δ cells colocalized with puncta fashioned by ER markers Ost4-mCherry (F) and mCherry-ADEL (G), and these ER marker puncta had been usually related to ER tubules. Pink arrows denote puncta-associated ER tubules. Bar, 2 μm. (H) Percentages of various varieties cytoplasmic Pus1 puncta in photographs acquired within the evaluation proven in G (greater than 200 puncta had been examined). Pus1 puncta had been categorized into 4 varieties primarily based on their places (intracytoplasmic or peripheral) and whether or not the colocalizing mCherry-ADEL punctum was related to ER tubules. (I and J) FRAP evaluation of cytoplasmic Pus1-mCherry puncta (I) and Pus1-colocalizing Ost4-mCherry puncta (J) in yep1Δ cells handled with nitrogen hunger. The photographs had been taken earlier than (pre) and after photobleaching. Photobleached puncta are indicated by purple arrows. Bar, 2 μm. Fluorescence intensities of the puncta had been quantitated and are proven as means ± normal deviations (n = 31 for Pus1-mCherry puncta and n = 30 for Ost4-mCherry puncta). Numerical knowledge underlying panels I and J could be present in S1 Information.

https://doi.org/10.1371/journal.pbio.3002372.g002

To raised perceive the character of the constructions represented by cytoplasmic puncta of Bqt4 and Pus1, we carried out TEM evaluation. This evaluation confirmed that ring-shaped membrane constructions amassed within the cytoplasm of yep1Δ cells however not yep1Δ atg5Δ cells (Figs 2D and S2E and S2F). These constructions differ in location from the ring-shaped autophagosomes that amassed within the fsc1Δ cells (S1D Fig). Autophagosomes accumulating within the fsc1Δ cells are nearly at all times juxtaposed to vacuoles. In distinction, ring-shaped membrane constructions within the cytoplasm of yep1Δ cells had been by no means noticed juxtaposed to vacuoles. The form and dimension of those constructions are just like these of the interior rings within the double-ring constructions noticed within the fsc1Δ cells (S2G Fig). Thus, we proposed that these constructions are ER-phagy/nucleophagy cargos not enclosed inside autophagosomes, they usually embrace constructions represented by the Bqt4 and Pus1 puncta. In DTT-treated, however not starvation-treated wild-type cells, autophagosomes juxtaposed to vacuoles could be noticed (S2E Fig). They possible embrace constructions comparable to the Bqt4 and Pus1 puncta noticed in wild-type cells by fluorescence microscopy.

To additional look at the character of the ring-shaped membrane constructions accumulating within the cytoplasm of yep1Δ cells, we fused a genetically encoded EM tag, MTn, to the integral ER membrane protein Ost4 and examined its distribution in yep1Δ cells utilizing a just lately developed EM expertise [46]. MTn-generated gold nanoparticles had been noticed not solely on the nuclear envelope and the cortical ER but in addition on ring-shaped cytoplasmic membrane constructions resembling the ring-shaped constructions noticed within the TEM evaluation (Fig 2E), confirming that nucleus- and/or cortical-ER-derived constructions with out surrounding autophagic membranes accumulate within the cytoplasm of yep1Δ cells.

To acquire extra corroborating proof on the buildup of ER-phagy/nucleophagy cargos in yep1Δ cells, we employed the degron safety assay [47]. On this assay, an auxin-inducible degron (AID) tag is fused to a protein in order that the protein is degraded when uncovered to the cytosol however not when residing inside a membrane compartment. We used the cytosolic protein Pyk1, a bulk autophagy cargo, to confirm whether or not a protein enclosed throughout the autophagosome is proof against AID-mediated degradation. In untreated cells, Pyk1-AID-mECitrine was utterly degraded upon the addition of the auxin analog 5-adamantyl-IAA (Advert-IAA) [48] (S3A Fig). When autophagy was induced by nitrogen hunger, Pyk1-AID-mECitrine sign in small round-shaped constructions, that are vacuoles labelled by the vacuole lumen marker Cpy1-mCherry, endured in wild-type cells after the addition of Advert-IAA. Equally, persisting Pyk1-AID-mECitrine sign in vacuoles was noticed in nitrogen-starved yep1Δ cells, confirming that bulk autophagy is basically regular in yep1Δ cells. In fsc1Δ cells, that are faulty in autophagosome–vacuole fusion, persisting Pyk1-AID-mECitrine sign in constructions not overlapping with vacuoles was noticed after the addition of Advert-IAA. These constructions are presumably autophagosomes, suggesting that AID-mediated degradation doesn’t occur to proteins enclosed inside autophagosomes. For our evaluation of ER-phagy/nucleophagy cargos accumulating in yep1Δ cells, we selected Epr1 because the degradation goal as it’s a peripheral membrane protein dealing with the cytosol and is concentrated on the websites of ER-phagy/nucleophagy. As anticipated, in untreated wild-type or yep1Δ cells, alerts of Epr1-AID-mECitrine utterly disappeared after the addition of Advert-IAA (S3B Fig). In distinction, in wild-type cells starved for six hours, Epr1-AID-mECitrine relocalized to the vacuole lumen, and the vacuolar sign endured after the addition of Advert-IAA. In yep1Δ cells, Epr1-AID-mECitrine fashioned cytoplasmic puncta, a few of which colocalized with Pus1 cytoplasmic puncta. Upon the addition of Advert-IAA, Epr1-AID-mECitrine puncta utterly disappeared (S3B Fig), indicating that in yep1Δ cells, the outer floor of nucleus-derived membrane constructions accumulating within the cytoplasm is uncovered to the cytosol.

To find out whether or not cortical-ER-derived membrane constructions not enclosed inside autophagosomes additionally amassed in yep1Δ cells, we utilized the degron safety assay to the integral ER membrane protein Rtn1, which localizes to the cortical ER however not the nuclear envelope (S3C Fig). Rtn1-AID-mECitrine sign in untreated wild-type and yep1Δ cells disappeared upon the addition of Advert-IAA. In starvation-treated wild-type cells, Rtn1-AID-mECitrine partially relocalized to the vacuole lumen and the vacuole-localized sign endured upon the addition of Advert-IAA. In starvation-treated yep1Δ cells, Rtn1-AID-mECitrine fashioned cytoplasmic puncta, which disappeared after the addition of Advert-IAA (S3C Fig), suggesting that just like the scenario of nucleophagy cargos, cortical-ER-phagy cargos not enclosed inside autophagosomes additionally amassed in yep1Δ cells.

We subsequent investigated whether or not the ER-phagy/nucleophagy cargo constructions that amassed in yep1Δ cells had been totally separated from the supply compartments. Reside cell imaging confirmed that in starvation-treated yep1Δ cells, cytoplasmic puncta fashioned by the nucleoplasmic protein Pus1-mECitrine colocalized with puncta fashioned by the ER membrane marker Ost4-mCherry and the ER lumen marker mCherry-ADEL (ADEL is an ER retention sign) (Fig 2F and 2G). Notably, these Pus1-positive ER marker puncta had been usually related to cytoplasmic ER tubules or cortical ER (Fig 2F and 2G and 2H), suggesting that the nucleophagy cargo constructions remained hooked up to the nuclear envelope/ER community. Nonetheless, the alerts of Pus1 and the interior nuclear membrane protein Bqt4 had been by no means noticed on ER tubule-like constructions (Fig 2A–2C and 2F and 2G), indicating that the interior contents and the interior membranes of the nucleophagy cargo constructions had been now not steady with the nucleoplasm and the interior nuclear membrane, respectively. Supporting this, fluorescence restoration after photobleaching (FRAP) evaluation confirmed that the fluorescence alerts of photobleached Pus1-mCherry puncta solely recovered barely (Fig 2I). The minor improve in fluorescence alerts can maybe be attributed to the reversible photoswitching of mCherry [49]. In distinction, the fluorescence alerts of Pus1-positive Ost4-mCherry puncta considerably recovered after photobleaching (Figs 2H and 2I and S3F), supporting that the outer membranes of the nucleophagy cargo constructions had been steady with the nuclear envelope/ER community. Utilizing TEM evaluation, we additionally noticed nucleus- and/or cortical-ER-derived cargo constructions that had filamentous membrane protrusions, that are possible ER tubules (S3D Fig).

In fission yeast, 3 ER-shaping proteins, the REEP household protein Yop1, the RTN household protein Rtn1, and the TMEM33 household protein Tts1, localize to tubular ER and act in {a partially} redundant method to take care of tubular ER [50]. We noticed that Yep1 exhibited colocalization with Rtn1, Yop1, and Tts1 (Figs 3A and S4A), suggesting that Yep1 could share comparable features with these proteins.

Within the absence of Rtn1, Yop1, and Tts1, the cortical ER turns into much less reticulate and extra sheet-like, with the frequent look of enormous holes in photographs of the highest or backside aircraft of the cells and prolonged gaps in photographs of the midplane of the cells [50]. We discovered that this alteration of ER morphology could be reversed to a big extent by both introducing again Rtn1 or growing the expression stage of Yep1 (Figs 3B and S4B). Thus, Yep1, when overexpressed, can fulfill the perform of sustaining tubular ER independently of Rtn1, Yop1, and Tts1.

The ER construction aberration attributable to the lack of Rtn1, Yop1, and Tts1 additionally leads not directly to a extreme septum positioning defect, manifesting as long-axis septum, a number of septa, and tilted septum [50]. This phenotype is simpler to attain than the ER morphology phenotype and might reveal the weak phenotypes of single deletion mutants missing Rtn1, Yop1, or Tts1 (Fig 3C). We noticed that yep1Δ brought about a noticeable however even weaker septum positioning defect than rtn1Δ, yop1Δ, or tts1Δ. Mixed yep1Δ with the double and triple deletion of rtn1, yop1, and tts1 invariably resulted in a extra extreme phenotype (Fig 3C). Essentially the most pronounced phenotypic enhancement was noticed when yep1Δ was mixed with the rtn1Δ tts1Δ double deletion. The defect of rtn1Δ tts1Δ yep1Δ could be rescued to the extent of rtn1Δ tts1Δ by reintroducing Yep1 (S4C Fig). Furthermore, growing the expression stage of Yep1 ameliorated the septum positioning defect of rtn1Δ yop1Δ tts1Δ (S4D Fig). These outcomes display that Yep1 shares the membrane-shaping potential of Rtn1, Yop1, and Tts1 and contributes to the upkeep of regular ER construction.

Fig 3. The ER-phagy function of Yep1 depends on its talents to self-interact and form membranes and requires its C-terminal amphipathic helices.

(A) Yep1 colocalized with ER-shaping proteins. Single optical sections targeted on the highest (or backside) of the cells are proven. Photos had been processed by deconvolution to permit higher visualization of the cortical ER community. Bar, 5 μm. (B) The aberrant cortical ER morphology of rtn1Δ yop1Δ tts1Δ cells could be partially rescued by the overexpression of Yep1. A peripheral aircraft and a central aircraft of the identical cells are proven. Pink arrows denote the big holes in photographs of the peripheral planes or the prolonged gaps in photographs of the central planes. Photos had been processed by deconvolution to permit higher visualization of the cortical ER community. Bar, 5 μm. (C) Quantification of the septum abnormality phenotypes (greater than 200 cells with septa had been examined for every pattern). The septum abnormality phenotypes embrace long-axis septum, a number of septa, and tilted septum. Septa had been visualized by calcofluor staining. (D) Yep1 exhibited self-interaction. Yep1-GFP was immunoprecipitated and co-immunoprecipitation of Yep1-mCherry was analyzed by immunoblotting. (E) The outcomes of predicting the constructions of Yep1 oligomers utilizing AlphaFold-Multimer. (F) The AlphaFold-Multimer-predicted Yep1 homodimer construction proven within the cartoon illustration. One subunit is coloured gray. The opposite subunit is coloured to match the diagram in (G), with the transmembrane helices (TMHs) coloured inexperienced, the two C-terminal amphipathic helices (APHs) coloured magenta, and different α-helices coloured mild blue. (G) Topology mannequin of Yep1. The secondary construction is predicated on the anticipated 3D construction proven in (F). The topology is predicated on the TOPCONS membrane protein topology prediction proven in S5C Fig. TMHs are coloured inexperienced. The two C-terminal APHs are coloured magenta. Different α-helices are coloured mild blue. (H) Quantification of the septum abnormality phenotypes of rtn1Δ tts1Δ cells, rtn1Δ tts1Δ yep1Δ cells, and rtn1Δ tts1Δ yep1Δ cells expressing full-length Yep1, Yep1 (1–97), Yep1 (1–113), or Yep1Δ (98–113) (greater than 200 cells with septa had been examined for every pattern). (I) Yep1Δ (98–113) didn’t exhibit self-interaction in a co-immunoprecipitation evaluation. (J) Yep1Δ (98–113) was unable to rescue the ER-phagy defect of yep1Δ. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. (Okay) Inner deletion of each however not both one of many 2 helices in Yep1 (114–150) abolished the ER-phagy perform of Yep1. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. Proteins expressed in yep1Δ had been tagged with mCherry, and their expression ranges had been analyzed by immunoblotting utilizing an antibody in opposition to mCherry. (L) The amphipathic nature of APHs is necessary for the ER-phagy perform of Yep1. Yep1^APH1mut harbors the I117D, F120D, and L124D mutations; Yep1^APH2mut harbors the A137D, V141D, and L148D mutations; Yep1^{APH1mutAPH2mut} harbors the I117D, F120D, L124D, A137D, V141D, and L148D mutations (see helical wheel representations in S6I Fig). The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. Proteins expressed in yep1Δ had been tagged with mCherry, and their expression ranges had been analyzed by immunoblotting utilizing an antibody in opposition to mCherry. Numerical knowledge underlying panels C, H, and J-L could be present in S1 Information, and uncooked photographs for panels D and J-L could be present in S1 Uncooked Photos.

https://doi.org/10.1371/journal.pbio.3002372.g003

We assessed whether or not Rtn1, Yop1, and Tts1 perform in ER-phagy. DTT and starvation-induced processing of Erg11-GFP was solely barely diminished within the rtn1Δ yop1Δ tts1Δ triple deletion mutant (S4E Fig), suggesting that ER-phagy nonetheless happens within the absence of those 3 proteins. As well as, overexpression of Rtn1, Yop1, or Tts1 didn’t alleviate the extreme ER-phagy defect of yep1Δ (S4F Fig). Thus, Rtn1, Yop1, and Tts1 can’t substitute for the important function of Yep1 in ER-phagy.

A number of ER-shaping proteins, together with Yop1, are identified to self-interact [3,4,51–53]. It has been just lately proven that the formation of curved-shape homo-oligomer of ER-shaping proteins is answerable for producing the tubular membrane form [54]. Utilizing the co-immunoprecipitation assay, we discovered that Yep1 can self-interact (Fig 3D). Bimolecular fluorescence complementation (BiFC) confirmed the self-interaction of Yep1 and prompt that Yep1 undergoes self-interaction on the ER membrane (S5A Fig).

We used AlphaFold-Multimer to foretell the constructions of Yep1 homo-oligomers [55] (Figs 3E and S5B). Whatever the variety of Yep1 sequences (from 2 to eight) within the enter, AlphaFold-Multimer solely predicted one sort of oligomeric construction—the construction of the Yep1 dimer, indicating that the dimer is the popular oligomerization state of Yep1. Within the predicted construction of the Yep1 dimer (Fig 3F), inside every Yep1 molecule, there are 3 lengthy α-helices within the N-terminal area. They largely embody the three transmembrane segments predicted by TOPCONS (Figs 3G and S5C). The C-terminal cytoplasmic area incorporates 2 brief α-helices, 2 lengthy α-helices, and a disordered tail [56] (Figs 3G and S5C). Intermolecular contacts primarily contain the primary 2 transmembrane helices (TMHs). The two brief C-terminal helices (amino acids 98–113) additionally contribute to the dimer interface by participating the N-terminal helices of the opposite molecule (Fig 3F).

In the results of the co-immunoprecipitation assay, each Yep1-GFP and Yep1-mCherry appeared as doublets, and the decrease band of Yep1-mCherry was co-immunoprecipitated with Yep1-GFP (Fig 3D). Based mostly on the obvious molecular weights, the higher band and the low band possible correspond to the full-length protein and an N-terminally cleaved protein, respectively. To evaluate the place the cleavage occurred, we expressed 2 N-terminally truncated kinds, Yep1 (35–166)-mCherry and Yep1 (79–166)-mCherry (S5D Fig). Yep1 (35–166)-mCherry appeared as a doublet, whereas Yep1 (79–166)-mCherry appeared as a single band working barely decrease than the decrease doublet band (S5E and S5F Fig), indicating that the cleavage possible occurred shortly upstream of residue 79, between the second and third TMH (S5D Fig). Neither Yep1 (35–166)-mCherry nor Yep1 (79–166)-mCherry was co-immunoprecipitated with Yep1-GFP (S5E and S5F Fig), suggesting that, in step with the anticipated construction of the Yep1 dimer, the N-terminal area of Yep1 is critical for self-interaction. We speculate that the cleavage, which probably occurred throughout protein extraction, didn’t trigger dissociation of the N-terminal area of Yep1 underneath the non-denaturing immunoprecipitation situations and subsequently didn’t have an effect on self-interaction.

Additionally in step with the anticipated construction of the Yep1 dimer, Yep1 (1–97) did not exhibit self-interaction within the co-immunoprecipitation evaluation (S5G Fig), whereas Yep1(1–113) was capable of self-interact (S5H Fig). The septum positioning defect of rtn1Δ tts1Δ yep1Δ was rescued to the extent of rtn1Δ tts1Δ by the expression of Yep1(1–113), however not Yep1(1–97) (Fig 3H), suggesting that the membrane-shaping potential of Yep1 is dependent upon its self-interaction. Inner deletion of residues 98–113 abolished the skills of Yep1 to self-interact and to rescue the septum positioning defect of rtn1Δ tts1Δ yep1Δ (Fig 3H and 3I). Furthermore, Yep1 missing residues 98–113 can now not help ER-phagy (Fig 3J). Collectively, these outcomes recommend that Yep1 self-interaction is necessary for its membrane-shaping potential and indicate that the membrane-shaping potential is necessary for its function in ER-phagy.

Although Yep1 (1–113) can self-interact and possesses the membrane-shaping potential, it can’t help ER-phagy in yep1Δ cells (S6A Fig). In distinction, Yep1 (1–131) and Yep1(1–150), which contained 1 and a couple of extra C-terminal lengthy helices, respectively, can help ER-phagy. Inner deletion of each however not both of these 2 helices abolished the ER-phagy perform of Yep1, suggesting that these 2 lengthy helices play redundant roles for the ER-phagy perform of Yep1 (Fig 3K). According to the outcomes obtained on Yep1 (1–113), the inner deletion mutant Yep1Δ(114–150), which lacks each lengthy helices, and Yep1(1–150) can self-interact and might rescue rtn1Δ tts1Δ yep1Δ to the identical extent as full-length Yep1 (S6B–S6E Fig).

HeliQuest analyses of the C-terminal helices and visible inspection of the AlphaFold-predicted construction indicated that these 2 lengthy helices are amphipathic helices (APHs), whereas the two upstream brief helices don’t exhibit apparent amphipathicity [57] (S6F–S6H Fig). To look at whether or not the amphipathic nature of those 2 APHs is functionally necessary, we substituted 3 hydrophobic amino acids with aspartates in every APH to disrupt their hydrophobic face [58,59] (S6I Fig). Mutating the primary APH considerably weakened, however didn’t abolish, the ER-phagy perform of Yep1 (Fig 3L). Mutating the second APH barely weakened the ER-phagy perform, whereas mutating each APHs rendered Yep1 nonfunctional in ER-phagy (Fig 3L). Collectively, these outcomes display that these APHs are redundantly important for the ER-phagy perform of Yep1.

To know whether or not the structural options of Yep1 are conserved in its homologs, we utilized the identical analyses, together with the inspection of AlphaFold-predicted constructions, TOPCONS prediction of membrane topology, and HeliQuest evaluation of APHs, to a number of different consultant REEP household proteins (Fig 4A and 4B). According to a earlier report [53], our analyses confirmed that REEP1-4 subfamily proteins have 3 TMHs, whereas REEP5-6 subfamily proteins have 4 TMHs. All of them include APHs within the C-terminal cytoplasmic area. REEP5-6 subfamily proteins additionally possess APHs within the N-terminal cytoplasmic area.

Fig 4. REEP1-4 subfamily proteins and Atg40 share the identical ER-phagy perform with Yep1.

(A) Schematics of predicted constructions of consultant members of the REEP household and S. cerevisiae Atg40. The depicted structural components are primarily based on AlphaFold-predicted constructions, TOPCONS prediction of membrane topology, and HeliQuest evaluation of APHs. (B) Topology fashions of consultant REEP household ptoteins and Atg40. Transmembrane helices (TMHs) are coloured inexperienced. C-terminal APHs are coloured magenta. (C) The ER-phagy defect of yep1Δ could be suppressed by expressing any one of many 4 human REEP1-4 subfamily proteins however not by expressing human REEP5 or REEP6. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. Proteins expressed in yep1Δ had been tagged with mCherry, and their expression ranges had been analyzed by immunoblotting utilizing an antibody in opposition to mCherry. (D) S. cerevisiae Atg40 and its PSI–BLAST-detectable sequence homologs in different Saccharomycetaceae species (known as Atg40 proteins right here) are possible divergent REEP1-4 subfamily proteins. Proven on the left is a time-calibrated species tree of consultant Ascomycota species [60]. REEP5-6 subfamily proteins, REEP1-4 subfamily proteins, and Atg40 proteins current in these species are listed on the suitable. All Atg40 proteins and a few REEP1-4 subfamily proteins harbor a C-terminal AIM (see S7B Fig). (E) Saccharomycetaceae genes encoding Atg40 proteins share synteny with non-Saccharomycetaceae genes encoding AIM-containing REEP1-4 subfamily proteins. Syntenic genes are proven as coloured arrows, and nonsyntenic genes are proven as gray arrows. The names of Saccharomyces cerevisiae genes are proven in daring. For the opposite species, the accession numbers of Atg40 proteins and AIM-containing REEP1-4 subfamily proteins are proven on prime of the corresponding genes, that are denoted by inexperienced arrows. (F) Expressing S. cerevisiae Atg40 rescued the ER-phagy defect of yep1Δ in a way impartial of its Atg8-interacting motif (AIM). Atg40^AIMmut harbors the Y242A and M245A mutations. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. Proteins expressed in yep1Δ had been tagged with mCherry, and their expression ranges had been analyzed by immunoblotting utilizing an antibody in opposition to mCherry. (G) Schematics of wild-type and truncated Atg40. (H) Atg40 (1–193) however not Atg40 (1–153) rescued the ER-phagy defect of yep1Δ. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. Proteins expressed in yep1Δ had been tagged with mCherry, and their expression ranges had been analyzed by immunoblotting utilizing an antibody in opposition to mCherry. (I) Fusing an APH-containing fragment from both S. cerevisiae Atg40 or S. pombe Yop1 to Yep1 (1–113) can partially restore its ER-phagy perform. The blot photographs are representatives of triplicate experiments. Quantitation of triplicate experiments is proven under the photographs. Proteins expressed in yep1Δ had been tagged with mCherry, and their expression ranges had been analyzed by immunoblotting utilizing an antibody in opposition to mCherry. Numerical knowledge underlying panels C, F, H, and I could be present in S1 Information, and uncooked photographs for panels C, F, H, and I could be present in S1 Uncooked Photos.

https://doi.org/10.1371/journal.pbio.3002372.g004

In S. pombe, as our knowledge present (S4F Fig), the REEP5-6 subfamily member Yop1 can’t substitute for the ER-phagy perform of the REEP1-4 subfamily member Yep1. We examined whether or not heterologously expressing human REEP household proteins can suppress the ER-phagy defect of yep1Δ. Remarkably, REEP1 and REEP3 totally suppressed yep1Δ, and REEP2 and REEP4 exhibited partial suppression (Fig 4C). In distinction, REEP5 and REEP6 confirmed no suppression (Fig 4C). These outcomes recommend that REEP1-4 subfamily proteins, however not REEP5-6 subfamily proteins, share a conserved ER-phagy perform with S. pombe Yep1.

Although S. cerevisiae doesn’t have an apparent REEP1-4 subfamily member, a identified ER-phagy receptor in S. cerevisiae, Atg40, resembles REEP1-4 subfamily proteins within the quantity and topology of its TMHs [29]. Our evaluation confirmed that it additionally possesses APHs downstream of its 3 TMHs (Figs 4A and 4B and S7A). To know the connection between Atg40 and REEP1-4 subfamily proteins, we surveyed the species distribution of Atg40 homologs and REEP1-4 subfamily proteins within the Ascomycota phylum (Fig 4D). PSI–BLAST-detectable sequence homologs of Atg40 (hereafter known as Atg40 proteins) had been solely present in budding yeast species belonging to the Saccharomycetaceae household. Curiously, these species all lack a REEP1-4 subfamily protein. In distinction, Ascomycota species exterior of the Saccharomycetaceae household all have at the very least 1 REEP1-4 subfamily protein. Remarkably, throughout the subphylum Saccharomycotina (budding yeasts), species not belonging to the Saccharomycetaceae household all possess 1 REEP1-4 subfamily protein harboring a C-terminal AIM, resembling the scenario in Atg40 (Figs 4D and S7B). Moreover, genes encoding these AIM-containing REEP1-4 subfamily proteins share synteny with Saccharomycetaceae genes encoding Atg40 proteins (Fig 4E). These observations point out that Atg40 proteins are divergent orthologs of REEP1-4 subfamily proteins. Supporting this concept, a phylogenetic evaluation confirmed that Atg40 proteins and Ascomycota REEP1-4 subfamily proteins fall into the identical clade, and Ascomycota REEP5-6 subfamily proteins fall right into a sister clade (S7C Fig).

According to the outcomes of the phylogenetic evaluation, heterologous expression of Atg40 in S. pombe rescued the ER-phagy defect of yep1Δ (Fig 4F). In distinction, the S. cerevisiae nucleophagy receptor Atg39 did not rescue yep1Δ. The power of Atg40 to rescue yep1Δ is impartial of its AIM (Fig 4F). Curiously, heterologous expression of Atg40 in S. pombe may rescue the ER-phagy defect of epr1Δ, and this rescue requires its AIM (S7D Fig). Furthermore, Atg40 may even rescue the ER-phagy defect of the yep1Δ epr1Δ double mutant (S7E Fig), suggesting that Atg40 fulfills the mixed roles of Yep1 and Epr1 in ER-phagy.

We examined the function of the APHs in Atg40 by truncation evaluation. Atg40 (1–193), which lacks the C-terminal disordered tail, was nonetheless capable of rescue the ER-phagy defect of yep1Δ (Fig 4G and 4H). In distinction, Atg40(1–153), which lacks the two C-terminal APHs, did not help ER-phagy in yep1Δ cells (Fig 4G and 4H). To deal with the query why REEP1-4 subfamily proteins and Atg40, however not REEP5-6 subfamily proteins, can substitute for Yep1, we changed the C-terminal area of Yep1 with a C-terminal APH-encompassing phase of S. cerevisiae Atg40 or S. pombe Yop1. The ER-phagy defect of yep1Δ could be suppressed to an analogous extent by the two mosaic proteins (Figs 4I and S7F), indicating that the C-terminal APHs of REEP5-6 subfamily proteins are able to supporting ER-phagy. One chance is that the additional TMH and/or the cytoplasmic tail within the N-termini of REEP5-6 subfamily proteins are incompatible with ER-phagy. Supporting this chance, we discovered that including the N-terminal area of Yop1 upstream of its second TMH or solely the primary TMH of Yop1 to the N-terminus of Yep1 disrupted the ER-phagy perform of Yep1 (S7F and S7G Fig). Nonetheless, the presence of the additional N-terminal sequence might not be the one cause why REEP5-6 subfamily proteins can’t help ER-phagy, as eradicating the N-terminal area of Yop1 upstream of its second TMH didn’t render it able to substituting for Yep1 (S7F and S7H Fig).

Fission yeast strains used on this research are listed in S2 Desk, and plasmids used on this research are listed in S3 Desk. The genetic strategies for pressure building and composition of media are as beforehand described [69]. Deletion strains had been generated by PCR-based gene focusing on. The pressure containing 1 extra copy of every of 23 core autophagy genes (atg1, atg2, atg3, atg4, atg5, atg6, atg7, atg8, atg9, atg10, atg11, atg12, atg13, atg14, atg16, atg17, atg18a, atg18b, atg18c, atg101, vps34, vps15, and ctl1) was constructed utilizing the CRISPR-Cas9 system [70]. The plasmids expressing proteins fused with completely different N-terminal or C-terminal tags (GFP, CFP, mECtrine, mCherry, mTurquoise2) underneath exogenous promoters (P41nmt1, Pnmt1, or Padh1) had been constructed using modified pDUAL vectors [71] or modified SIV vectors [72]. The plasmids expressing Erg11-AIM^artwork had been primarily based on modified SIV vectors. AIM^artwork corresponds to three×EEEWEEL [73].

Log part cells had been harvested and resuspended in 0.3 ml of TM buffer (50 mM Tris-maleate). Mutagenesis was induced by including 0.1 ml of two mg/ml N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) to the cell suspension and incubating at room temperature for 60 minutes. The mutagenized cells had been then plated on the EMM medium and incubated at 30°C for 4 days. Small and medium-sized colonies had been chosen, transferred to 96-well deep-well plates, grown for twenty-four hours in EMM liquid medium at 30°C, and handled with 10 mM DTT for 12 hours to induce ER-phagy. A high-content imaging system (Opera LX, PerkinElmer) was used for observing the subcellular localization of Ost4-GFP. Candidate mutants faulty in vacuolar relocalization of Ost4-GFP had been additional evaluated utilizing a DeltaVision PersonalDV system (Utilized Precision). Strains whose phenotypes had been confirmed by reexamination had been backcrossed, and the backcrossed progeny had been analyzed through next-generation sequencing-assisted bulk segregant evaluation [74]. For every backcross, mutations enriched in ER-phagy faulty progeny had been thought-about because the candidate phenotype-causing mutations.

For normal TEM evaluation, 50 OD₆₀₀ models of cells had been harvested after being starved for 12 hours or handled with 10 mM DTT for 12 hours. Cells had been mounted with 1% glutaraldehyde and 4% KMnO4 and dehydrated by way of a graded ethanol collection. The samples had been then embedded in Spurr’s resin [75]. Skinny sections had been examined utilizing an FEI Tecnai G2 Spirit electron microscope geared up with a Gatan 895 4k×4k CCD digital camera. The diameters of the ring-shaped constructions had been decided utilizing the strategy beforehand used for measuring the sizes of autophagosomes or autophagic our bodies [43,76]. P values had been calculated utilizing Welch’s t check. For electron microscopy evaluation using the genetically encoded EM tag MTn, samples had been ready as beforehand described [46]. Briefly, 20 OD600 models of cells expressing Ost4-MTn had been harvested after therapy with 10 mM DTT for 12 hours. Cells had been incubated with 3 mM dithiodipropionic acid (DTDPA) in 0.1 M PIPES at 4°C for half-hour after which handled with 0.5 mg/ml zymolyase 20T and 50 mM DTT in PBS buffer for 10 minutes to take away the cell wall. The zymolyase-treated cells had been permeabilized with 0.05% Triton X-100 for five minutes and processed for gold nanoparticle synthesis. Cells had been combined with 60 mM 2-mercaptoethanol, 0.5 mM HAuCl₄, 50 mM diphenylethylenediamine (DPEN), and 10 μΜ NaBH₄ and subjected to plain high-pressure freezing/freeze-substitution fixation. After resin infiltration, embedding, polymerization, and skinny sectioning, the samples had been used for EM imaging as described above.

To analyze the Atg8-interacting potential, Yep1 and Epr1 had been fused to GFP because the prey, and Atg8 was fused to Pil1-mCherry because the bait [45]. Log-phase cells coexpressing each proteins had been then analyzed by fluorescence microscopy.

Cells cultured in EMM liquid media at 30°C had been used. Previous to remark, cells in EMM liquid media had been handled with 1 μM 5-adamantyl-IAA for 1.5 hours at 30°C [48].

Roughly 100 OD₆₀₀ models of log-phase cells had been harvested. Cells had been combined with 150 μl of lysis buffer (50 mM HEPES–NaOH (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 mM PMSF, 0.05% NP-40, 10% glycerol, 1×Roche protease inhibitor cocktail) and had been lysed by beating with 0.5-mm-diameter glass beads utilizing a FastPrep instrument at a velocity of 6.5 m/s for 3 cycles of 20 seconds every. The lysate was incubated with GFP-Lure agarose beads at 4°C for 3 hours. The beads had been washed twice and proteins had been eluted by boiling in SDS-PAGE loading buffer. Samples had been separated by 10% SDS-PAGE and analyzed by immunoblotting utilizing anti-GFP and anti-mCherry antibodies described above. For the immunoprecipitation coupled with mass spectrometry (IP-MS) experiment utilizing Yep1-GFP as bait, immunoprecipitated samples had been processed and mass spectrometry was carried out as described beforehand [26]. S1 Desk lists ER-localizing proteins (GO:0005783) with a spectral depend of at the very least 30 within the Yep1-GFP IP pattern and greater than 6-fold greater than the spectral depend in a management IP pattern.

To analyze the self-interaction of Yep1, the N-terminal Venus fragment (VN173) and the C-terminal Venus fragment (VC155) had been fused to Yep1, respectively [77]. Log-phase cells coexpressing the VN173 fusing protein and the VC155 fusing protein had been analyzed by fluorescence microscopy.

The constructions of the Yep1 monomer and oligomers had been predicted utilizing AlphaFold-Multimer (model 2.2.0) with default parameters [55]. The construction with the very best confidence rating among the many predicted output was chosen for additional evaluation. The anticipated constructions had been visualized utilizing the Mol* Viewer (model 2.5.0). The transmembrane topology was predicted utilizing the TOPCONS internet server (https://topcons.cbr.su.se/pred/) [56]. The amphipathic nature of helices was analyzed utilizing the HeliQuest internet server (https://heliquest.ipmc.cnrs.fr/cgi-bin/ComputParams.py) [57].