Mamoru Miyasaka1,2, Tetsuo Mioka1, Takuma Kishimoto1, Eriko Itoh1, Kazuma Tanaka1. 1. Division of Molecular Interaction, Institute for Genetic Medicine, Hokkaido University Graduate School of Life Science, Sapporo, Hokkaido, Japan. 2. Department of Gastroenterological Surgery II, Hokkaido University Faculty of Medicine, Sapporo, Hokkaido, Japan.
Abstract
In eukaryotic cells, phospholipid flippases translocate phospholipids from the exoplasmic to the cytoplasmic leaflet of the lipid bilayer. Budding yeast contains five flippases, of which Cdc50p-Drs2p and Neo1p are primarily involved in membrane trafficking in endosomes and Golgi membranes. The ANY1/CFS1 gene was identified as a suppressor of growth defects in the neo1Δ and cdc50Δ mutants. Cfs1p is a membrane protein of the PQ-loop family and is localized to endosomal/Golgi membranes, but its relationship to phospholipid asymmetry remains unknown. The neo1Δ cfs1Δ mutant appears to function normally in membrane trafficking but may function abnormally in the regulation of phospholipid asymmetry. To identify a gene that is functionally relevant to NEO1 and CFS1, we isolated a mutation that is synthetically lethal with neo1Δ cfs1Δ and identified ERD1. Erd1p is a Golgi membrane protein that is involved in the transport of phosphate (Pi) from the Golgi lumen to the cytoplasm. The Neo1p-depleted cfs1Δ erd1Δ mutant accumulated plasma membrane proteins in the Golgi, perhaps due to a lack of phosphatidylinositol 4-phosphate. The Neo1p-depleted cfs1Δ erd1Δ mutant also exhibited abnormal structure of the endoplasmic reticulum (ER) and induced an unfolded protein response, likely due to defects in the retrieval pathway from the cis-Golgi region to the ER. Genetic analyses suggest that accumulation of Pi in the Golgi lumen is responsible for defects in Golgi functions in the Neo1p-depleted cfs1Δ erd1Δ mutant. Thus, the luminal ionic environment is functionally relevant to phospholipid asymmetry. Our results suggest that flippase-mediated phospholipid redistribution and luminal Pi concentration coordinately regulate Golgi membrane functions.
In eukaryotic cells, phospholipid flippases translocate phospholipids from the exoplasmic to the cytoplasmic leaflet of the lipid bilayer. Budding yeast contains five flippases, of which Cdc50p-Drs2p and Neo1p are primarily involved in membrane trafficking in endosomes and Golgi membranes. The ANY1/CFS1 gene was identified as a suppressor of growth defects in the neo1Δ and cdc50Δ mutants. Cfs1p is a membrane protein of the PQ-loop family and is localized to endosomal/Golgi membranes, but its relationship to phospholipid asymmetry remains unknown. The neo1Δ cfs1Δ mutant appears to function normally in membrane trafficking but may function abnormally in the regulation of phospholipid asymmetry. To identify a gene that is functionally relevant to NEO1 and CFS1, we isolated a mutation that is synthetically lethal with neo1Δ cfs1Δ and identified ERD1. Erd1p is a Golgi membrane protein that is involved in the transport of phosphate (Pi) from the Golgi lumen to the cytoplasm. The Neo1p-depleted cfs1Δ erd1Δ mutant accumulated plasma membrane proteins in the Golgi, perhaps due to a lack of phosphatidylinositol 4-phosphate. The Neo1p-depleted cfs1Δ erd1Δ mutant also exhibited abnormal structure of the endoplasmic reticulum (ER) and induced an unfolded protein response, likely due to defects in the retrieval pathway from the cis-Golgi region to the ER. Genetic analyses suggest that accumulation of Pi in the Golgi lumen is responsible for defects in Golgi functions in the Neo1p-depleted cfs1Δ erd1Δ mutant. Thus, the luminal ionic environment is functionally relevant to phospholipid asymmetry. Our results suggest that flippase-mediated phospholipid redistribution and luminal Pi concentration coordinately regulate Golgi membrane functions.
The lipid composition of the plasma membrane (PM) in eukaryotic cells differs between the inner (cytoplasmic) leaflet and outer (exoplasmic) leaflet. Generally, phosphatidylcholine (PC) and sphingolipids distribute in the outer leaflet, whereas phosphatidylserine (PS) and phosphatidylethanolamine (PE) distribute in the inner leaflet [1]. This asymmetrical distribution of phospholipids across the bilayer is conserved in yeast as well as higher forms of life. Changes in phospholipid asymmetry involve various cellular functions. For instance, exposure of PS to the outer leaflet of the PM signals the onset of blood coagulation and apoptosis and triggers phagocytosis [2], [3]. Cellular polarity is also induced due to local transbilayer changes in PE and PS levels in the PM [4], [5], [6].P4-ATPases, termed flippases, are essential to maintain an asymmetric PM structure. These membrane proteins control the distribution of bilayer lipids by flipping phospholipids across the lipid bilayer from the outer to the inner leaflet [7]. Select flippases are localized to endosomes and Golgi membranes to regulate membrane trafficking that occurs between the membranes [8], [9]. In the budding yeastSaccharomyces cerevisiae, five flippases can be found: Drs2p, Dnf1p, Dnf2p, Dnf3p, and Neo1p [10]. These flippases, except for Neo1p, form complexes with a member of the CDC50 protein family [11]. Drs2p, Dnf1p/Dnf2p, and Dnf3p form complexes with Cdc50p, Lem3p and Crf1p, respectively. Further, these five flippases functionally overlap, but NEO1 is an essential gene on its own [12]. The Cdc50p-Drs2p complex is localized to endosomes and the trans-Golgi network (TGN), and its functions in membrane trafficking have been extensively characterized. It has been proposed that phospholipid flipping by the Drs2p-Cdc50p complex promotes the formation of transport vesicles, including the clathrin-coated vesicle, but its mechanism remains unclarified [8], [9], [13]. Neo1p is also localized at endosomes and Golgi membranes to regulate membrane trafficking. Neo1p is required for retrograde transport from the cis-Golgi to the endoplasmic reticulum (ER) [14]. Neo1p is also involved in endocytosis and the vacuolar sorting pathway [15]. Flippase activity of Neo1p has not yet been demonstrated, but the neo1 mutant expresses loss of PE asymmetry in the PM [16]. In addition, NEO1 was isolated as a multicopy suppressor of the cdc50Δ mutation [11], and Neo1p has similar functions to the Cdc50p-Drs2p complex in terms of the endocytic recycling pathway [17], which suggests that Neo1p is also a flippase.ANY1/CFS1 was identified as a suppressor mutation of growth defects in the neo1Δ and cdc50Δ mutants [18], [19]. Cfs1p belongs to the PQ-loop family, which has seven-helix membrane topology, and partially colocalizes with Drs2p and Neo1p. The cfs1Δ mutation completely suppressed the lethality of the neo1Δ mutant, and the neo1Δ cfs1Δ mutant did not exhibit any defects in membrane trafficking. However, the neo1Δ cfs1Δ mutant may not be equivalent to the wild type in its Golgi functions, otherwise a set of these two genes including the essential NEO1 gene would be dispensable. Thus, a number of genes that are functionally relevant to phospholipid asymmetry may function to alleviate defects in the endosomal/Golgi membrane function in the neo1Δ cfs1Δ mutant. To determine such genes, we performed synthetic lethal screening with the neo1Δ cfs1Δ mutations, and the ERD1 gene was identified. Erd1p is involved in the transport of phosphate (Pi) from the luminal space of the Golgi membrane to the cytoplasm [20]. Our results suggest that phospholipid asymmetry and luminal Pi concentration coordinately regulate functions of the Golgi membrane.
Materials and methods
Media and genetic methods
Chemicals were purchased from Wako Pure Chemicals Industry (Osaka, Japan), unless otherwise stated. Standard genetic manipulations of the yeast were performed according to a previously described method [21]. Yeast transformations were performed using the lithium acetate method [22], [23]. Yeast strains were then cultured in YPDA-rich medium [1% yeast extract (Difco Laboratories, Detroit, MI), 2% Bacto-peptone (Difco), 2% glucose, and 0.01% adenine]. Strains that carry plasmids were placed in a synthetic medium [SD: 0.67% yeastnitrogen base (YNB) without amino acids (Difco) and 2% glucose] that contain required nutritional supplements [24]. The SDA medium was a SD medium that contains 0.5% casamino acid (Difco). To induce the GAL1 promoter, 3% galactose and 0.2% sucrose, instead of glucose, were used as carbon sources (YPGA and SG-Leu media). When expression of the GAL1 promoter was attenuated, 2% raffinose and either 0.01% or 0.1% galactose were used as carbon sources. A Pi-depleted SD medium was prepared with YNB without amino acids and phosphate (ForMedium Ltd., Hunstanton, UK). To adjust the Pi concentration in the SD and YPDA medium at pH 6.6, 1 M K2HPO4 and 1 M KH2PO4 were added at a ratio of 38.1: 61.9.
Yeast strains and plasmids
The yeast strains used in this study are listed in the “Supplemental Material” section (S1 Table). PCR-based procedures were used to construct gene deletions and gene fusions with the GAL1 promoter, green fluorescent protein (GFP), mCherry, and monomeric red fluorescent protein 1 (mRFP1) [25], [26]. Standard molecular biological techniques were used for plasmid construction, PCR amplification, and DNA sequencing [27]. Escherichia coli strain XL1-Blue was used to construct and amplify the plasmids. Gene deletions of CFS1, ERD1, BST1, EMP24, ESP1, ERP1, ERP2, RER1, VAN1, MNN10, MNN2, MNN5, MNN4, MNN6, and MNN1 in the YEF 473 genetic background [28] were performed as follows. Regions with the KanMX6 disruption marker and flanking sequences were PCR-amplified using genomic DNA that were derived from the knockout strain in the BY4741 background [29] as a template. The amplified DNA fragments were introduced into the appropriate strains, and G418-resistant transformants were selected. To construct KanMX6::P-GFP-PIK1 strains, the TRP1::P-GFP-PIK1 allele was replaced with the KanMX6::P-GFP-PIK1 fragment by the PCR-based allele replacement method [30]. All constructs that were produced by the PCR-based procedure were verified by colony PCR amplification to confirm that the replacement or insertion occurred at the expected locus. Sequences of PCR primers are available on request.The plasmids of this study are listed in the “Supplemental Material” section (S2 Table). To construct pRS316-mCherry-evt-2 PH (pKT2205), the mCherry-evt-2 PH fragment from pmCherry-Evc2-C2 (a gift from T. Taguchi, Tohoku University) was cloned into the NheI-BamHI gap of pRS316-P-T. To construct pRS306-2× UPRE-GFP (pKT2196), the 2× UPRE [31] and GFP fragments were inserted into the HindIII–KpnI gap of pRS306-P-T. These pRS306-based plasmids were linearized and integrated into the URA3 locus. pRS416-P-GFP-PHO87 (pKT2203) and pRS416-P-GFP-PHO90 (pKT2204) were constructed by inserting the coding regions of PHO87 and PHO90, respectively, into the BamH1–Sal1 gap of pRS416-P-GFP-PEP12-T (pKT1487). Multicopy plasmids that carry ERD1, PHO87, PHO90, ERS1, SAR1, or YIP1 were constructed by ligating DNA fragments, which were PCR-amplified using the genomic DNA of wild type cells (YKT38), into the YEplac181 vector plasmid. Schemes that detail the construction of plasmids and DNA sequences of nucleotide primers are available on request.
Isolation of mutants that are synthetically lethal with the neo1Δ cfs1Δ mutation
Mutants that are synthetically lethal with neo1Δ cfs1Δ were isolated according to a previously described procedure [32]. From 11,100 colonies that were screened, three single recessive mutations were identified by genetic analyses, and the corresponding wild type genes were cloned. These genes encoded DRS2, CDC50, and ERD1. Mutations in DRS2/CDC50 were shown to be synthetically lethal with neo1Δ cfs1Δ [18]. Likewise, the present study confirms that the erd1Δ mutation is synthetically lethal with neo1Δ cfs1Δ (Fig 1A).
Fig 1
Synthetic lethality and membrane trafficking defects in the neo1Δ cfs1Δ erd1Δ mutant.
(A) The erd1Δ mutation exhibited synthetic lethality with the neo1Δ cfs1Δ mutations. Tetrad dissection of the neo1Δ/NEO1 cfs1Δ/cfs1Δ ERD1/erd1Δ diploid. Tetrad genotypes (TT, tetra type; PD, parental ditype; NPD, nonparental ditype) are indicated, and the identities of the triple mutant are shown in parentheses. (B) The growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were grown to the early log phase in YPDA for approximately 12 h to deplete Neo1p, washed, and adjusted to a concentration of 1.0 × 107 cells/mL. Next, 4-μl drops of 4-fold serial dilutions were spotted onto YPGA (galactose) and YPDA (glucose) plates and then incubated at 30°C for 1.5 d. The strains used were wild type (WT) (YKT38), P-NEO1 (YKT2134), P-NEO1 cfs1Δ (YKT2135), P-NEO1 cfs1Δ erd1Δ (YKT2136), cfs1Δ erd1Δ (YKT2137), and erd1Δ (YKT2138). (C) Intracellular accumulation of GFP-Snc1p-pm in the Neo1p-depleted cfs1Δ erd1Δ mutant. Upper panels: Strains that express GFP-Snc1p-pm were grown to the exponential phase in YPDA medium at 30°C for ~20 h except for P-NEO1 and P-NEO1 erd1Δ (12 h) to deplete Neo1p, followed by observation using a fluorescence microscope. The strains used were WT (YKT2139), P-NEO1 (YKT2140), P-NEO1 cfs1Δ (YKT2141), P-NEO1 cfs1Δ erd1Δ (YKT2142), cfs1Δ erd1Δ (YKT2143), erd1Δ (YKT2144), and P-NEO1 erd1Δ (YKT2225), all of which carry GFP-SNC1-pm. Lower panel: The percentage of cells with intracellularly accumulated GFP-Snc1p-pm was determined (n = 200) and is shown as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). (D) Intracellular accumulation of Pdr5p-GFP and GFP-Sso1p in the Neo1p-depleted cfs1Δ erd1Δ mutant. Strains were cultured and observed as in (C), except that GFP-SSO1 strains were cultured in SDA-Ura. The strains used were WT (YKT38 and YKT2145), P-NEO1 (YKT2134 and YKT2146), P-NEO1 cfs1Δ (YKT2135 and YKT2147), and P-NEO1 cfs1Δ erd1Δ (YKT2148 and YKT2149), all of which carry pRS416-GFP-SSO1 (pKT1476) and PDR5-GFP, respectively. Bar, 5 μm. DIC, differential interference contrast.
Synthetic lethality and membrane trafficking defects in the neo1Δ cfs1Δ erd1Δ mutant.
(A) The erd1Δ mutation exhibited synthetic lethality with the neo1Δ cfs1Δ mutations. Tetrad dissection of the neo1Δ/NEO1 cfs1Δ/cfs1Δ ERD1/erd1Δ diploid. Tetrad genotypes (TT, tetra type; PD, parental ditype; NPD, nonparental ditype) are indicated, and the identities of the triple mutant are shown in parentheses. (B) The growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were grown to the early log phase in YPDA for approximately 12 h to deplete Neo1p, washed, and adjusted to a concentration of 1.0 × 107 cells/mL. Next, 4-μl drops of 4-fold serial dilutions were spotted onto YPGA (galactose) and YPDA (glucose) plates and then incubated at 30°C for 1.5 d. The strains used were wild type (WT) (YKT38), P-NEO1 (YKT2134), P-NEO1 cfs1Δ (YKT2135), P-NEO1 cfs1Δ erd1Δ (YKT2136), cfs1Δ erd1Δ (YKT2137), and erd1Δ (YKT2138). (C) Intracellular accumulation of GFP-Snc1p-pm in the Neo1p-depleted cfs1Δ erd1Δ mutant. Upper panels: Strains that express GFP-Snc1p-pm were grown to the exponential phase in YPDA medium at 30°C for ~20 h except for P-NEO1 and P-NEO1erd1Δ (12 h) to deplete Neo1p, followed by observation using a fluorescence microscope. The strains used were WT (YKT2139), P-NEO1 (YKT2140), P-NEO1 cfs1Δ (YKT2141), P-NEO1 cfs1Δ erd1Δ (YKT2142), cfs1Δ erd1Δ (YKT2143), erd1Δ (YKT2144), and P-NEO1erd1Δ (YKT2225), all of which carry GFP-SNC1-pm. Lower panel: The percentage of cells with intracellularly accumulated GFP-Snc1p-pm was determined (n = 200) and is shown as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). (D) Intracellular accumulation of Pdr5p-GFP and GFP-Sso1p in the Neo1p-depleted cfs1Δ erd1Δ mutant. Strains were cultured and observed as in (C), except that GFP-SSO1 strains were cultured in SDA-Ura. The strains used were WT (YKT38 and YKT2145), P-NEO1 (YKT2134 and YKT2146), P-NEO1 cfs1Δ (YKT2135 and YKT2147), and P-NEO1 cfs1Δ erd1Δ (YKT2148 and YKT2149), all of which carry pRS416-GFP-SSO1 (pKT1476) and PDR5-GFP, respectively. Bar, 5 μm. DIC, differential interference contrast.
Isolation of multicopy suppressors of the neo1Δ cfs1Δ erd1Δ mutant
The neo1Δ cfs1Δ erd1Δ strain (YKT2173) with the YCplac33-NEO1 plasmid (pKT1470) was transformed using a yeast genomic DNA library that was constructed with the multicopy plasmid YEp13 [33]. Transformants were spread onto SD-Leu plates and incubated at 30°C for 3 days. These plates were replica-plated onto SD-Leu containing 0.1% 5-fluoroorotic acid (5-FOA) plates to select clones that had lost URA3-containing YCplac33-NEO1 [34]. Approximately 3 × 105 transformants were screened, and 190 clones were obtained, of which 61 were further analyzed. 18 clones were found to contain NEO1 and ERD1 by colony PCR and were thus eliminated. Plasmids were then recovered from the remaining 43 clones, and restriction enzyme digestion of the plasmids indicated that they originated from 13 independent clones. These 13 plasmids were reintroduced into the original mutant to confirm the suppression. Finally, they were grouped into five different genomic regions by DNA sequencing. Fragment subcloning revealed that PHO87, PHO90, ERS1, SAR1 and YIP1 were responsible for the suppression.
Microscopic observations
Cells were observed using a Nikon ECLIPSE E800 microscope (Nikon Instec, Tokyo, Japan) equipped with an HB-10103AF superhigh-pressure mercury lamp and a 1.4 numerical aperture 100 × Plan Apo oil immersion objective lens (Nikon Instec) with appropriate fluorescence filter sets (Nikon Instec) or differential interference contrast optics. Images were acquired using a cooled digital charge-coupled device camera (C4742-95-12NR; Hamamatsu Photonics, Hamamatsu, Japan) and AQUACOSMOS software (Hamamatsu Photonics). GFP-, mRFP1-, or mCherry-tagged proteins were observed in living cells, which were grown to early to mid-logarithmic phase, harvested, and resuspended in SD medium. Cells were immediately observed using a GFP bandpass (for GFP) or G2-A (for mRFP1 and mCherry) filter set. Observations were compiled from examining at least 200 cells. For statistical analysis, the observation of 200 cells was repeated five times for each strain.
Results
Isolation of erd1Δ as a mutation that is synthetically lethal with neo1Δ cfs1Δ mutations
We previously showed that the cfs1Δ mutation suppressed the lethality of the neo1Δ mutant. The cfs1Δ neo1Δ mutant grew normally, and the cfs1Δ mutation efficiently suppressed the membrane trafficking defects of the Neo1p-depleted cells [18]. The neo1 mutations caused abnormal phospholipid asymmetry; the neo1 mutants exposed PE and PS in the PM [16]. The any1Δ/cfs1Δ mutation completely suppressed PE and PS exposure in the neo1-2 temperature-sensitive mutant [19], whereas it did not suppress the exposure in the neo1Δ mutant [35], which suggests that Cfs1p partially antagonizes the flippase action of Neo1p. To isolate a gene that is functionally relevant to NEO1 and CFS1, we performed synthetic lethal screening using the neo1Δ cfs1Δ mutant. Consequently, a mutation in ERD1 was isolated (Fig 1A). ERD1 was identified as a gene that encodes a membrane protein required for the retention of proteins that are localized to the endoplasmic reticulum (ER) and was subsequently shown to be required for glycosylation of some proteins in the Golgi apparatus [36], [37]. A recent study suggested that Erd1p transports Pi from the lumen of the Golgi apparatus to the cytoplasm and recycles the Pi byproducts of glycosylation reactions [20]. To phenotypically analyze the neo1 cfs1 erd1 mutant, we constructed the P-NEO1 cfs1Δ erd1Δ mutant, in which the expression of NEO1 is controlled by the glucose-repressible GAL1 promoter. The erd1Δ mutation exhibited synthetic growth defects with the Neo1p-depleted cfs1Δ mutation, but not with the cfs1Δ mutation (Fig 1B). We examined whether other mutations that are involved in ER retention exhibited synthetic lethality with the Neo1p-depleted cfs1Δ mutation, but none did (S1A Fig). We also examined mutations in Golgi glycosylation, but they also were not synthetically lethal with Neo1p-depleted cfs1Δ (S1B Fig). These results suggest that a defect specific to erd1Δ causes synthetic lethality with Neo1p-depleted cfs1Δ.We examined membrane trafficking in the Neo1p-depleted cfs1Δ erd1Δ mutant using GFP-Snc1p-pm as a marker protein. Snc1p-pm is a PM-localized mutant v-SNARE with point mutations that inhibit endocytosis [38]. Neo1p-depletion intracellularly accumulated GFP-Snc1p-pm, and the cfs1Δ mutation suppressed this phenotype, as previously reported [18] (Fig 1C). The Neo1p-depleted cfs1Δ erd1Δ mutant accumulated GFP-Snc1p-pm like Neo1p-depleted cells (Fig 1C). We also examined the localization of Pdr5p and Sso1p, which are two GFP-fused PM proteins. Pdr5p is an ATP-binding cassette transporter [39],while Sso1p is a t-SNARE that is involved in the fusion of secretory vesicles with the PM [40]. Similar to GFP-Snc1p-pm, Neo1p-depletion accumulated Pdr5p-GFP and GFP-Sso1p in ~77% and 94% of the cells, respectively, (n = 200 cells) (Fig 1D). The Neo1p-depleted cfs1Δ mutant also accumulated Pdr5p-GFP and GFP-Sso1p to a certain extent (~16% and 28%, respectively, n = 200 cells). In contrast, the Neo1p-depleted cfs1Δ erd1Δ mutant accumulated Pdr5p-GFP and GFP-Sso1p to a large extent (~45% and 84%, respectively, n = 200 cells), although the accumulation of Pdr5p-GFP was relatively low compared to GFP-Snc1p-pm and GFP-Sso1p. These results suggest that the Neo1p-depleted cfs1Δ erd1Δ mutant exhibits major defects in membrane trafficking pathways.
PM proteins accumulate in the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant
Neo1p is localized to the TGN, cis- and medial-Golgi, and endosomal compartments [14], [15], [41]. Cfs1p is partially localized to the TGN [18], whereas the localization of Erd1p has not yet been determined. Erd1p-GFP, which was expressed from its own promoter, was barely detectable, and thus Erd1p-GFP was overexpressed under the control of the GAL1 promoter. We compared the localization of Erd1p-GFP to that of Mnn9p-mRFP1, which is a cis-Golgi marker [42], and Sec7p-mRFP1, which is a TGN marker [43]. Erd1p-GFP colocalized with Mnn9p-mRFP1, rather than with Sec7p-mRFP1 (Fig 2A); 44% of the Erd1p-GFP dots were colocalized with Mnn9p-mRFP1 dots (n = 221), and 43% of the Mnn9p-mRFP1 dots were co-localized with Erd1p-GFP dots (n = 225). In contrast, 9% of the Erd1p-GFP dots were colocalized with Sec7p-mRFP1 dots (n = 316), and 12% of the Sec7p-mRFP1 dots were colocalized with Erd1p-GFP dots (n = 221). These results suggest that Erd1p is mainly localized to the cis-Golgi.
Fig 2
Localization of GFP-Snc1p-pm in the Neo1p-depleted cfs1Δ erd1Δ mutant.
(A) Colocalization of Erd1p-GFP with Mnn9p-mRFP1. Cells were cultured in YPGA at 30°C for 7 h, followed by microscopic observation. The strains used were P-ERD1-GFP MNN9-mRFP1 (YKT2150) and P-ERD1-GFP SEC7-mRFP1 (YKT2151). (B) Localization of GFP-Snc1p-pm and Mnn9p-mCherry in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were wild type (WT) (YKT2152), P-NEO1 (YKT2153), and P-NEO1 cfs1Δ erd1Δ (YKT2154), all of which carry GFP-SNC1-pm and MNN9-mCherry. (C) Localization of GFP-Snc1p-pm and Sec7p-mRFP1 in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2155), P-NEO1 (YKT2156), and P-NEO1 cfs1Δ erd1Δ (YKT2157), all of which carry GFP-SNC1-pm and SEC7-mRFP1. Bar, 5 μm. DIC, differential interference contrast.
Localization of GFP-Snc1p-pm in the Neo1p-depleted cfs1Δ erd1Δ mutant.
(A) Colocalization of Erd1p-GFP with Mnn9p-mRFP1. Cells were cultured in YPGA at 30°C for 7 h, followed by microscopic observation. The strains used were P-ERD1-GFP MNN9-mRFP1 (YKT2150) and P-ERD1-GFP SEC7-mRFP1 (YKT2151). (B) Localization of GFP-Snc1p-pm and Mnn9p-mCherry in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were wild type (WT) (YKT2152), P-NEO1 (YKT2153), and P-NEO1 cfs1Δ erd1Δ (YKT2154), all of which carry GFP-SNC1-pm and MNN9-mCherry. (C) Localization of GFP-Snc1p-pm and Sec7p-mRFP1 in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2155), P-NEO1 (YKT2156), and P-NEO1 cfs1Δ erd1Δ (YKT2157), all of which carry GFP-SNC1-pm and SEC7-mRFP1. Bar, 5 μm. DIC, differential interference contrast.The localization patterns of Neo1p, Cfs1p, and Erd1p suggest that the Neo1p-depleted cfs1Δ erd1Δ mutant has defective membrane trafficking through the Golgi apparatus. We examined whether GFP-Snc1p-pm was accumulated in the cis-Golgi or the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant. In both the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants, 25% (n = 224) and 26% (n = 237) of GFP-Snc1p-pm-positive structures were colocalized with Mnn9p-mCherry structures, respectively (Fig 2B). In contrast, 87% (n = 200) and 70% (n = 215) of GFP-Snc1p-pm structures were colocalized with Sec7p-mRFP1 structures in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants, respectively (Fig 2C). These results suggest that Snc1p-pm mainly accumulates in the TGN of the Neo1p-depleted cfs1Δ erd1Δ mutant.
Abnormal distribution of phospholipids in the TGN of the Neo1p-depleted cfs1Δ erd1Δ mutant
We next examined the distribution of phospholipids in the Neo1p-depleted cfs1Δ erd1Δ mutant. The C2 domain of lactadherin (Lact-C2) and the pleckstrin homology (PH) domain of evectin-2 (evt-2 PH) were used as probes for PS in the cytoplasmic leaflet of the PM and organelle membranes [44], [45]. GFP-Lact-C2 normally localizes only to the PM, but it localizes to the TGN membranes in flippase mutants, which suggests that PS is exposed to the cytoplasmic leaflet of the TGN [46]. In the P-NEO1 mutant, 67% (n = 207) and 71% (n = 208) of the GFP-Snc1p-pm structures were colocalized with mRFP1-Lact-C2 and mCherry-evt-2 PH structures, respectively (Fig 3A and 3B). Similarly, in the Neo1p-depleted cfs1Δ erd1Δ mutant, 51% (n = 255) and 61% (n = 206) of GFP-Snc1p-pm structures were colocalized with mRFP1-Lact-C2 and mCherry-evt-2 PH structures, respectively. These results suggest that PS is exposed to the cytoplasmic leaflet of the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant.
Fig 3
The cytoplasmic leaflet of the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant contains PS and decreased level of PI4P.
(A) and (B) PS is exposed to the cytoplasmic leaflet of the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant. (A) Localization of mRFP1-Lact-C2. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were wild type (WT) (YKT2139), P-NEO1 (YKT2158), and P-NEO1 cfs1Δ erd1Δ (YKT2142), all of which carry GFP-SNC1-pm and pRS416-mRFP1-Lact-C2 (pKT1755). (B) Localization of mCherry-evt-2 PH. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2139), P-NEO1 (YKT2158), and P-NEO1 cfs1Δ erd1Δ (YKT2142), all of which carry GFP-SNC1-pm and pRS316-mCherry-evt-2 PH (pKT2205). (C) and (D) PI4P was absent from the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant. (C) Localization of Osh2p-PH-GFP. Left panels: Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2159), P-NEO1 (YKT2160), P-NEO1 cfs1Δ (YKT2161), P-NEO1 cfs1Δ erd1Δ (YKT2162), cfs1Δ erd1Δ (YKT2163), and erd1Δ (YKT2164), all of which carry OSH2-PH-GFP. Right panel: The percentage of cells with Osh2p-PH-GFP only at the PM was determined (n = 200) and is expressed as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). (D) Localization of Osh2p-PH-GFP and Sec7p-mRFP1. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2165), P-NEO1 (YKT2166), P-NEO1 cfs1Δ (YKT2167), and P-NEO1 cfs1Δ erd1Δ (YKT2168), all of which carry OSH2-PH-GFP and SEC7-mRFP1. (E) Colocalization of GFP-Pik1p and Sec7p-mRFP1 in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2169), P-NEO1 (YKT2170), P-NEO1 cfs1Δ (YKT2171), and P-NEO1 cfs1Δ erd1Δ (YKT2172), all of which carry P-GFP-PIK1 and SEC7-mRFP1. Bar, 5 μm. DIC, differential interference contrast.
The cytoplasmic leaflet of the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant contains PS and decreased level of PI4P.
(A) and (B) PS is exposed to the cytoplasmic leaflet of the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant. (A) Localization of mRFP1-Lact-C2. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were wild type (WT) (YKT2139), P-NEO1 (YKT2158), and P-NEO1 cfs1Δ erd1Δ (YKT2142), all of which carry GFP-SNC1-pm and pRS416-mRFP1-Lact-C2 (pKT1755). (B) Localization of mCherry-evt-2 PH. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2139), P-NEO1 (YKT2158), and P-NEO1 cfs1Δ erd1Δ (YKT2142), all of which carry GFP-SNC1-pm and pRS316-mCherry-evt-2 PH (pKT2205). (C) and (D) PI4P was absent from the TGN in the Neo1p-depleted cfs1Δ erd1Δ mutant. (C) Localization of Osh2p-PH-GFP. Left panels: Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2159), P-NEO1 (YKT2160), P-NEO1 cfs1Δ (YKT2161), P-NEO1 cfs1Δ erd1Δ (YKT2162), cfs1Δ erd1Δ (YKT2163), and erd1Δ (YKT2164), all of which carry OSH2-PH-GFP. Right panel: The percentage of cells with Osh2p-PH-GFP only at the PM was determined (n = 200) and is expressed as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). (D) Localization of Osh2p-PH-GFP and Sec7p-mRFP1. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2165), P-NEO1 (YKT2166), P-NEO1 cfs1Δ (YKT2167), and P-NEO1 cfs1Δ erd1Δ (YKT2168), all of which carry OSH2-PH-GFP and SEC7-mRFP1. (E) Colocalization of GFP-Pik1p and Sec7p-mRFP1 in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were WT (YKT2169), P-NEO1 (YKT2170), P-NEO1 cfs1Δ (YKT2171), and P-NEO1 cfs1Δ erd1Δ (YKT2172), all of which carry P-GFP-PIK1 and SEC7-mRFP1. Bar, 5 μm. DIC, differential interference contrast.We next examined the localization of phosphatidylinositol-4-phosphate (PI4P) in the TGN. PI4P, which is produced by the PI4P kinase Pik1p, plays a pivotal role in the vesicle transport from the TGN to the PM and endosomes [47]. A tandem dimer of the PH domain of Osh2p was shown to detect the Golgi pool of PI4P in addition to the PM pool [48]. Osh2p-PH-GFP was mainly localized to Golgi-like structures in wild-type cells but was mainly localized to the PM in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants (Fig 3C). Cell counting indicated that more than 57% of the cells expressed the Osh2p-PH-GFP signal only at the PM in both the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants. The Neo1p-depleted cfs1Δ mutant also demonstrated mislocalization to a significant extent (~32%). We confirmed that Osh2p-PH-GFP was not localized to the TGN in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants of Sec7p-mRFP1-expressing cells (Fig 3D). These results suggest that PI4P is absent from the TGN, and this may cause membrane trafficking defects in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants, which is consistent with our previous finding that Pik1p depletion caused intracellular accumulation of GFP-Snc1p-pm [46]. We then examined the localization of GFP-Pik1p. Interestingly, GFP-Pik1p was normally colocalized with Sec7p-mRFP1 in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants (Fig 3E). In the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants, 89% (n = 224) and 95% (n = 213) of GFP-Pik1p structures were colocalized with Sec7p-mRFP1 structures, respectively. These results suggest that either Pik1p is inactivated or limited phosphatidylinositol (PI) is present in the TGN of these mutants. Related to this, the neo1Δ cfs1Δ mutant may have defects in the production of PI4P, as partial depletion of Pik1p inhibited the growth of this mutant (S2 Fig).
Pi homeostasis in the Golgi lumen is crucial for viability of the Neo1p-depleted cfs1Δ mutant
To gain insight to the synthetic lethality of the neo1Δ cfs1Δ erd1Δ mutant, we performed multicopy suppressor screening. A neo1Δ cfs1Δ erd1Δ strain that contains a NEO1/URA3 plasmid was transformed with a genomic library, and the transformants were then screened on a plate with 5-fluoroorotic acid (5-FOA), as described in the “Materials and Methods” section. As a result, PHO87, PHO90, ERS1, SAR1, and YIP1 were isolated (Fig 4A). Further, these multicopy suppressors were found to suppress the growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant (Fig 4B), but not those of the Neo1p-depleted mutant (Fig 4C), which suggests that the multicopy suppressors suppressed the erd1Δ mutation or the synthetic growth defects that are associated with the Neo1p-depleted cfs1Δ erd1Δ mutant. Pho87p and Pho90p are low-affinity Pi transporters at the PM [49], [50], [51]. ERS1 was isolated as a multicopy suppressor of the erd1Δ mutant [52] and Ers1p may function as a cystine transporter in vacuoles [53]. Sar1p regulates the formation of COPII vesicles from the ER [54], [55], and Yip1p is also implicated in COPII vesicle biogenesis [56]. We confirmed that all these multicopy suppressors suppressed the intracellular accumulation of GFP-Snc1p-pm and mislocalization of Osh2p-PH-GFP in the Neo1p-depleted cfs1Δ erd1Δ mutant (Fig 4D).
Fig 4
Isolation of multicopy suppressors in the neo1Δ cfs1Δ erd1Δ mutant.
(A) Suppression of the growth defects of the neo1Δ cfs1Δ erd1Δ mutant. A spot assay for growth was performed as in Fig 1B; 8-μl drops of 4-fold serial dilutions were spotted onto SD-L and SD-L+5-fluoroorotic acid (FOA) plates and then incubated at 30°C for 4 d. Only cells that lost the plasmid harboring the NEO1 gene grew on the 5-FOA medium. The strain used was the neo1Δ cfs1Δ erd1Δ mutant with YCplac33-NEO1 (pKT1470) (YKT2173). This strain was transformed with YEplac181(Vector), pKT2197(ERD1), pKT2198(PHO87), pKT2199(PHO90), pKT2200(ERS1), pKT2201(SAR1), and pKT2202(YIP1). (B) Suppression of the growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant. A spot assay was performed as in (A), except that 4-μl drops were spotted onto YPGA (galactose) and YPDA (glucose) plates and then incubated at 30°C for 1.5 d. The strain used was the P-NEO1 cfs1Δ erd1Δ mutant (YKT2174). (C) Suppression of the growth defects of the Neo1p-depleted mutant. A spot assay was performed as in (B). The strain used was the P-NEO1 mutant (YKT2134). (D) Suppression of the intracellular accumulation of GFP-Snc1p-pm and mislocalization of Osh2p-PH-GFP in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C for 20 h, followed by microscopic observation. The strain used was the P-NEO1 cfs1Δ erd1Δ mutant with either GFP-Snc1p-pm (YKT2174) or Osh2p-PH-GFP (YKT2162). Each strain was transformed with the plasmids in (A). Left panel: Representative cells are shown. Right panels: The percentage of cells with intracellular accumulation of GFP-Snc1p-pm or Osh2p-PH-GFP only in the PM was determined (n = 200), and the results are expressed as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). Bar, 5 μm. DIC, differential interference contrast.
Isolation of multicopy suppressors in the neo1Δ cfs1Δ erd1Δ mutant.
(A) Suppression of the growth defects of the neo1Δ cfs1Δ erd1Δ mutant. A spot assay for growth was performed as in Fig 1B; 8-μl drops of 4-fold serial dilutions were spotted onto SD-L and SD-L+5-fluoroorotic acid (FOA) plates and then incubated at 30°C for 4 d. Only cells that lost the plasmid harboring the NEO1 gene grew on the 5-FOA medium. The strain used was the neo1Δ cfs1Δ erd1Δ mutant with YCplac33-NEO1 (pKT1470) (YKT2173). This strain was transformed with YEplac181(Vector), pKT2197(ERD1), pKT2198(PHO87), pKT2199(PHO90), pKT2200(ERS1), pKT2201(SAR1), and pKT2202(YIP1). (B) Suppression of the growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant. A spot assay was performed as in (A), except that 4-μl drops were spotted onto YPGA (galactose) and YPDA (glucose) plates and then incubated at 30°C for 1.5 d. The strain used was the P-NEO1 cfs1Δ erd1Δ mutant (YKT2174). (C) Suppression of the growth defects of the Neo1p-depleted mutant. A spot assay was performed as in (B). The strain used was the P-NEO1 mutant (YKT2134). (D) Suppression of the intracellular accumulation of GFP-Snc1p-pm and mislocalization of Osh2p-PH-GFP in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C for 20 h, followed by microscopic observation. The strain used was the P-NEO1 cfs1Δ erd1Δ mutant with either GFP-Snc1p-pm (YKT2174) or Osh2p-PH-GFP (YKT2162). Each strain was transformed with the plasmids in (A). Left panel: Representative cells are shown. Right panels: The percentage of cells with intracellular accumulation of GFP-Snc1p-pm or Osh2p-PH-GFP only in the PM was determined (n = 200), and the results are expressed as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). Bar, 5 μm. DIC, differential interference contrast.Isolation of PHO87 and PHO90 is interesting, as Erd1p is proposed to transport Pi from the lumen of the Golgi to the cytoplasm [20]. Since Pho87p and Pho90p are low-affinity Pi transporters at the PM, our results suggest that the synthetic lethality of the neo1Δ cfs1Δ erd1Δ mutant may potentially be caused by decreased concentration of cytoplasmic Pi (Fig 5A). This possibility was examined in the following experiments. We first examined whether increased concentrations of extracellular Pi suppressed the growth defect of the Neo1p-depleted cfs1Δ erd1Δ mutant. The YPD medium contained approximately 5 mM Pi [57]. However, the Neo1p-depleted cfs1Δ erd1Δ mutant did not grow in the YPDA medium with either 20 or 50 mM Pi (Fig 5B). If the erd1Δ mutation decreased cytoplasmic Pi, and if this caused the synthetic lethality with the Neo1p-depleted cfs1Δ mutation, the Neo1p-depleted cfs1Δ mutant would express sensitivity to conditions or mutations that decrease cytoplasmic Pi. However, the Neo1p-depleted cfs1Δ mutant grew in a low-Pi medium (1 and 3 mM) (Fig 5C). We next introduced mutations into the Pi transporters of the Neo1p-depleted cfs1Δ mutant. In addition to Pho87p and Pho90p, two more Pi transporters reside at the PM, namely Pho84p and Pho89p, which are high-affinity transporters [58], [59]. Even the Neo1p-depleted cfs1Δ mutant with mutations in all four transporters grew normally (Fig 5D). These results suggest that increased Pi in the Golgi lumen, rather than decreased Pi in the cytoplasm, may be responsible for the synthetic lethality of the erd1Δ mutation with the Neo1p-depleted cfs1Δ mutation, and this possibility was examined in subsequent experiments.
Fig 5
The synthetic lethality of the erd1Δ mutation with the Neo1p-depleted cfs1Δ mutation is not caused by reduced levels of cytoplasmic Pi.
(A) The synthetic lethality of erd1Δ with Neo1p-depleted cfs1Δ may be caused by reduced levels of cytoplasmic Pi. In this model, the overexpression of PHO87/90 would suppress the growth defect by increasing cytoplasmic Pi. Membrane proteins to be transported to the PM are shown in a grey rod shape. (B) The growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant were not suppressed by increased concentration of extracellular Pi. Cells were grown and spotted, as in Fig 1B, in YPGA (galactose) or YPDA (glucose) medium with either 20 mM Pi or 50 mM Pi and then incubated at 30°C for 1.5 d. The strains used were wild type (WT) (YKT38), P-NEO1 (YKT2134), P-NEO1 cfs1Δ (YKT2135), and P-NEO1 cfs1Δ erd1Δ (YKT2136). (C) The Neo1p-depleted cfs1Δ mutant can grow in low-Pi conditions. Cells were grown and spotted as in (B), except that 8-μl drops of 4-fold serial dilutions were spotted onto SD plates with 0, 1, or 3 mM Pi, followed by incubation at 30°C for 2 d. Standard SD medium contains approximately 7.3 mM Pi according to the manufacturer. The strains used were the same strains as in (B). (D) The Neo1p-depleted cfs1Δ mutation is not synthetically lethal with mutations of the Pi transporters at the PM. Cells were grown and spotted as in (B) onto YPGA (galactose) and YPDA (glucose) plates, followed by incubation at 30°C for 1.5 d. The strains used were P-NEO1 cfs1Δ (YKT2135), P-NEO1 cfs1Δ erd1Δ (YKT2136), P-NEO1 cfs1Δ pho84Δ (YKT2176), P-NEO1 cfs1Δ pho84Δ pho87Δ pho90Δ (YKT2177), and P-NEO1 cfs1Δ pho84Δ pho87Δ pho89Δ pho90Δ (YKT2178). (E) Localization of GFP-Pho87p and GFP-Pho90p in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were wild type (WT) (YKT2179), P-NEO1 (YKT2180), P-NEO1 cfs1Δ (YKT2181), and P-NEO1 cfs1Δ erd1Δ (YKT2182), all of which carry mRFP1-SNC1-pm and pRS416 P-GFP-PHO87 (pKT2203) (upper panel) or mRFP1-SNC1-pm and pRS416 P-GFP-PHO90 (pKT2204) (lower panel). Bar, 5 μm. DIC, differential interference contrast. (F) The synthetic lethality of erd1Δ with Neo1p-depleted cfs1Δ may be caused by increased levels of luminal Pi in the Golgi. In this model, the overexpression of PHO87/90 would suppress the growth defect by decreasing luminal Pi. Our data seem to be consistent with this model.
The synthetic lethality of the erd1Δ mutation with the Neo1p-depleted cfs1Δ mutation is not caused by reduced levels of cytoplasmic Pi.
(A) The synthetic lethality of erd1Δ with Neo1p-depleted cfs1Δ may be caused by reduced levels of cytoplasmic Pi. In this model, the overexpression of PHO87/90 would suppress the growth defect by increasing cytoplasmic Pi. Membrane proteins to be transported to the PM are shown in a grey rod shape. (B) The growth defects of the Neo1p-depleted cfs1Δ erd1Δ mutant were not suppressed by increased concentration of extracellular Pi. Cells were grown and spotted, as in Fig 1B, in YPGA (galactose) or YPDA (glucose) medium with either 20 mM Pi or 50 mM Pi and then incubated at 30°C for 1.5 d. The strains used were wild type (WT) (YKT38), P-NEO1 (YKT2134), P-NEO1 cfs1Δ (YKT2135), and P-NEO1 cfs1Δ erd1Δ (YKT2136). (C) The Neo1p-depleted cfs1Δ mutant can grow in low-Pi conditions. Cells were grown and spotted as in (B), except that 8-μl drops of 4-fold serial dilutions were spotted onto SD plates with 0, 1, or 3 mM Pi, followed by incubation at 30°C for 2 d. Standard SD medium contains approximately 7.3 mM Pi according to the manufacturer. The strains used were the same strains as in (B). (D) The Neo1p-depleted cfs1Δ mutation is not synthetically lethal with mutations of the Pi transporters at the PM. Cells were grown and spotted as in (B) onto YPGA (galactose) and YPDA (glucose) plates, followed by incubation at 30°C for 1.5 d. The strains used were P-NEO1 cfs1Δ (YKT2135), P-NEO1 cfs1Δ erd1Δ (YKT2136), P-NEO1 cfs1Δ pho84Δ (YKT2176), P-NEO1 cfs1Δ pho84Δ pho87Δ pho90Δ (YKT2177), and P-NEO1 cfs1Δ pho84Δ pho87Δ pho89Δ pho90Δ (YKT2178). (E) Localization of GFP-Pho87p and GFP-Pho90p in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C as in Fig 1C. The strains used were wild type (WT) (YKT2179), P-NEO1 (YKT2180), P-NEO1 cfs1Δ (YKT2181), and P-NEO1 cfs1Δ erd1Δ (YKT2182), all of which carry mRFP1-SNC1-pm and pRS416 P-GFP-PHO87 (pKT2203) (upper panel) or mRFP1-SNC1-pm and pRS416 P-GFP-PHO90 (pKT2204) (lower panel). Bar, 5 μm. DIC, differential interference contrast. (F) The synthetic lethality of erd1Δ with Neo1p-depleted cfs1Δ may be caused by increased levels of luminal Pi in the Golgi. In this model, the overexpression of PHO87/90 would suppress the growth defect by decreasing luminal Pi. Our data seem to be consistent with this model.Considering that PM proteins accumulated in the TGN of the Neo1p-depleted cfs1Δ erd1Δ mutant (Figs 1 and 2), Pho87p and Pho90p may also accumulate in the TGN. We then examined the localization of GFP-Pho87p and GFP-Pho90p, which are expressed under the control of the TPI1 promoter. In wild-type cells, GFP-Pho87p and GFP-Pho90p were mainly localized to the plasma membrane, but dotty structures were observed. mRFP1-Snc1-pm was exclusively localized to the plasma membrane. In the P-NEO1 mutant, GFP-Pho87p (85%, n = 200) and GFP-Pho90p (82%, n = 200) were localized to regions of intracellular accumulation of mRFP1-Snc1-pm (Fig 5E). Similarly, in the Neo1p-depleted cfs1Δ erd1Δ mutant, GFP-Pho87p (79%, n = 200) and GFP-Pho90p (84%, n = 200) were also colocalized with mRFP1-Snc1-pm (Fig 5E). These results suggest that Pho87p and Pho90p are accumulated in and localized to the TGN of the Neo1p-depleted cfs1Δ erd1Δ mutant.Overall, our results suggest that the overexpression of Pho87p or Pho90p accelerates transport of the luminally accumulated Pi to the cytoplasm, which results in the suppression of the growth defects of the neo1Δ cfs1Δ erd1Δ mutant (Fig 5F). We propose that the elevated Pi levels of the Golgi lumen are responsible for the lethality of the Neo1p-depleted cfs1Δ erd1Δ mutant.
The Neo1p-depleted cfs1Δ erd1Δ mutant also showed defects in the ER structure
Isolation of SAR1 and YIP1 as multicopy suppressors led us to examine phenotypes of the ER. To examine the morphology of the ER, Hmg1p-GFP was expressed. Hmg1p is a hydroxymethylglutaryl-CoA (HMG-CoA) reductase, which catalyzes the conversion of HMG-CoA to mevalonate [60]. Hmg1p localizes to the perinuclear ER [61]. However, its localization was disorganized in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants (Fig 6A). The perinuclear localization of Hmg1p-GFP was absent, and some cells exhibited string-like structures in their cytoplasm. We also examined the nuclear membrane by observing Nup188p, which is a nucleoporin of the nuclear pore complex [62]. Similar to Hmg1p-GFP, nuclear localization of Nup188p-mRFP1 was absent in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants (Fig 6B). We next examined the localization of the cortical ER marker protein Rtn1p, which is a member of the reticulon family and is involved in shaping ER tubules [63]. Interestingly, Neo1p-depleted (78%, n = 200) and Neo1p-depleted cfs1Δ erd1Δ (29%, n = 200) mutants exhibited abnormal dotty aggregates of Rtn1p-GFP (Fig 6C), which was also observed in the cho2Δ mutant that is defective in PC biosynthesis [64]. These structures were not observed in either wild-type or Neo1p-depleted cfs1Δ cells. These results indicate that the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants have structural defects in the ER. However, vesicle transport from the ER was not severely affected, since mRFP1-Snc1-pm did not accumulate in the Hmg1p-GFP structure of these mutants (< 1%, 200 cells) (Fig 6D).
Fig 6
Disorganization of the ER structure in the Neo1p-depleted cfs1Δ erd1Δ mutant.
(A) Localization of Hmg1p-GFP. Upper panel: Strains expressing Hmg1p-GFP were cultured in YPDA as in Fig 1C. The strains used were wild type (WT) (YKT2183), P-NEO1 (YKT2184), P-NEO1 cfs1Δ (YKT2185), P-NEO1 cfs1Δ erd1Δ (YKT2186), cfs1Δ erd1Δ (YKT2187), and erd1Δ (YKT2188), all of which carry HMG1-GFP. Lower panel: The percentage of cells that lacked perinuclear localization of Hmg1p-GFP was determined (n = 200) and is expressed as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). (B) Localization of Nup188p-mRFP1. Cells were cultured in YPDA as in Fig 1C. The strains used were WT (YKT2189), P-NEO1 (YKT2190), P-NEO1 cfs1Δ (YKT2191), and P-NEO1 cfs1Δ erd1Δ (YKT2192), all of which carry HMG1-GFP and NUP188-mRFP1. (C) Localization of Rtn1p-GFP. Cells were cultured in YPDA as in Fig 1C. The strains used were WT (YKT2193), P-NEO1 (YKT2194), P-NEO1 cfs1Δ (YKT2195), and P-NEO1 cfs1Δ erd1Δ (YKT2196), all of which carry RTN1-GFP. (D) GFP-Snc1p-pm was not colocalized with Hmg1p-GFP in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA as in Fig 1C. The strains used were WT (YKT2197), P-NEO1 (YKT2198), and P-NEO1 cfs1Δ erd1Δ (YKT2199), all of which carry HMG1-GFP and mRFP1-SNC1-pm. Bar, 5 μm. DIC, differential interference contrast.
Disorganization of the ER structure in the Neo1p-depleted cfs1Δ erd1Δ mutant.
(A) Localization of Hmg1p-GFP. Upper panel: Strains expressing Hmg1p-GFP were cultured in YPDA as in Fig 1C. The strains used were wild type (WT) (YKT2183), P-NEO1 (YKT2184), P-NEO1 cfs1Δ (YKT2185), P-NEO1 cfs1Δ erd1Δ (YKT2186), cfs1Δ erd1Δ (YKT2187), and erd1Δ (YKT2188), all of which carry HMG1-GFP. Lower panel: The percentage of cells that lacked perinuclear localization of Hmg1p-GFP was determined (n = 200) and is expressed as the mean ± standard deviation of five independent experiments. Asterisks indicate a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). (B) Localization of Nup188p-mRFP1. Cells were cultured in YPDA as in Fig 1C. The strains used were WT (YKT2189), P-NEO1 (YKT2190), P-NEO1 cfs1Δ (YKT2191), and P-NEO1 cfs1Δ erd1Δ (YKT2192), all of which carry HMG1-GFP and NUP188-mRFP1. (C) Localization of Rtn1p-GFP. Cells were cultured in YPDA as in Fig 1C. The strains used were WT (YKT2193), P-NEO1 (YKT2194), P-NEO1 cfs1Δ (YKT2195), and P-NEO1 cfs1Δ erd1Δ (YKT2196), all of which carry RTN1-GFP. (D) GFP-Snc1p-pm was not colocalized with Hmg1p-GFP in the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA as in Fig 1C. The strains used were WT (YKT2197), P-NEO1 (YKT2198), and P-NEO1 cfs1Δ erd1Δ (YKT2199), all of which carry HMG1-GFP and mRFP1-SNC1-pm. Bar, 5 μm. DIC, differential interference contrast.Erd1p was originally isolated as a mutant that secretes a luminal ER protein, such as the ER chaperone Kar2p/BiP [36, 37]. Thus, the erd1Δ mutant is defective in the retrieval of Kar2p from the cis-Golgi region to the ER. The neo1 temperature-sensitive mutant also secretes Kar2p and mislocalizes Rer1p [14], which is a retrieval receptor for ER membrane proteins [65]. Failure to retain Kar2p in the ER accumulates unfolding proteins in the ER, which induces the unfolded protein response (UPR) [66], [67]. We examined UPR in the Neo1p-depleted cfs1Δ erd1Δ mutant with UPR elements (UPRE) that are fused to GFP [68]. Treatment with a reducing agent, such as dithiothreitol (DTT), accumulates unfolding proteins and induces UPRE-GFP expression. The Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants exhibited strong induction of UPR (Fig 7A). The erd1Δ mutation caused weak induction, as previously reported [69]. ERD2 encodes a receptor for retrieval of luminal ER proteins from the cis-Golgi region [70]. The Erd2p-depleted mutant exhibited a strong induction of UPR and the abnormal structure of the ER, which is reminiscent of the Neo1p-depleted cfs1Δ erd1Δ mutant, but it exhibited normal transport of Snc1p-pm to the PM (S3 Fig). These morphological and functional defects in the ER of the Neo1p-depleted cfs1Δ erd1Δ mutant may be caused by defective membrane trafficking from the cis-Golgi to the ER, similar to the case with the neo1 mutants [14].
Fig 7
UPR is induced in the Neo1p-depleted cfs1Δ erd1Δ mutant.
(A) Induction of UPR in the Neo1p-depleted cfs1Δ erd1Δ mutant. Yeast cells that express GFP under the control of 2× UPRE were cultured in YPDA as in Fig 1C. For treatment of wild-type cells with dithiothreitol (DTT), early log phase cells equivalent to 1.5 ml culture of OD600nm = 1.0 were collected and incubated in YPDA that contains 5 μM DTT for 4 h. The strains used were WT (YKT2200), P-NEO1 (YKT2201), P-NEO1 cfs1Δ (YKT2202), P-NEO1 cfs1Δ erd1Δ (YKT2203), cfs1Δ erd1Δ (YKT2204), and erd1Δ (YKT2205), all of which carry 2× UPRE-GFP. (B) Suppression of the abnormal localization of Hmg1p-GFP by the multicopy suppressors of the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C for 20 h. The strain used was the P-NEO1 cfs1Δ erd1Δ (YKT2186) mutant, which contains HMG1-GFP and multicopy suppressors, as in Fig 4B. Left panel: Representative cells are shown. Right panel: The percentage of cells with abnormal localization of Hmg1p-GFP was determined (n = 200) and is expressed as the mean ± standard deviation of five independent experiments. An asterisk indicates a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). Bar, 5 μm. DIC, differential interference contrast.
UPR is induced in the Neo1p-depleted cfs1Δ erd1Δ mutant.
(A) Induction of UPR in the Neo1p-depleted cfs1Δ erd1Δ mutant. Yeast cells that express GFP under the control of 2× UPRE were cultured in YPDA as in Fig 1C. For treatment of wild-type cells with dithiothreitol (DTT), early log phase cells equivalent to 1.5 ml culture of OD600nm = 1.0 were collected and incubated in YPDA that contains 5 μM DTT for 4 h. The strains used were WT (YKT2200), P-NEO1 (YKT2201), P-NEO1 cfs1Δ (YKT2202), P-NEO1 cfs1Δ erd1Δ (YKT2203), cfs1Δ erd1Δ (YKT2204), and erd1Δ (YKT2205), all of which carry 2× UPRE-GFP. (B) Suppression of the abnormal localization of Hmg1p-GFP by the multicopy suppressors of the Neo1p-depleted cfs1Δ erd1Δ mutant. Cells were cultured in YPDA at 30°C for 20 h. The strain used was the P-NEO1 cfs1Δ erd1Δ (YKT2186) mutant, which contains HMG1-GFP and multicopy suppressors, as in Fig 4B. Left panel: Representative cells are shown. Right panel: The percentage of cells with abnormal localization of Hmg1p-GFP was determined (n = 200) and is expressed as the mean ± standard deviation of five independent experiments. An asterisk indicates a significant difference, as determined by the Tukey–Kramer test (*: p < 0.01). Bar, 5 μm. DIC, differential interference contrast.We also examined whether the multicopy suppressors suppressed the morphological defects of the ER in the Neo1p-depleted cfs1Δ erd1Δ mutant. All the isolated suppressors suppressed the abnormal structure of the ER (Fig 7B). The mechanism of suppression by SAR1 and YIP1 is unknown, but continued transport of membrane proteins, including Ers1p, Pho87p, and Pho90p, from the ER to the Golgi may result in the suppression. Alternatively, enhanced vesicle transport from the ER increases the volume of the Golgi apparatus in the Neo1p-depleted cfs1Δ erd1Δ mutant, as membrane transport from the Golgi is inhibited in this mutant. This may dilute Pi in the Golgi lumen, which results in the suppression of membrane trafficking defects through the Golgi apparatus.
Discussion
In this study, we isolated erd1 as a mutation that is synthetically lethal with neo1Δ cfs1Δ mutations. The Neo1p-depleted cfs1Δ erd1Δ mutant exhibited severe defects in Golgi functions, including the anterograde vesicle transport from the TGN and the retrograde transport from the cis-Golgi region. Our results suggest that Erd1p is a new factor that is functionally relevant to phospholipid asymmetry. Although flippase activity has not been demonstrated for Neo1p, neo1 temperature-sensitive mutants exhibited loss of PE asymmetry in the plasma membrane [16]. Likewise, genetic studies demonstrated that Neo1p is functionally relevant to Drs2p [11], [17], [35], [71], which has been most extensively characterized as a flippase [72]. Thus, Cfs1p seems to function in an antagonistic manner to flippases; Cfs1p may be a scramblase or positive regulator of a floppase. Alternatively, Cfs1p may segregate the Neo1p and Drs2p functions, as Drs2p may replace Neo1p in the neo1Δ cfs1Δ mutant [35]. In any case, transbilayer phospholipid distribution in the Golgi membrane may not be normally regulated in the neo1Δ cfs1Δ mutant.The role of flippases in membrane trafficking has been extensively studied due to its possible involvement in vesicle formation [8], [9]. The DRS2/CDC50 complex genetically and physically interacts with the Arf1 small GTPase and its regulators [73], [74], [75]. The drs2Δ mutant is defective in the formation of clathrin-coated vesicles of the TGN [73], [76], [77]. Based on the lipid transport activity to the cytoplasmic leaflet, two models have been proposed for the function of flippase in the vesicle formation process: (1) local phospholipid flipping induces membrane curvature, which promotes vesicle formation, and (2) transported lipids (e.g., PS) recruit components, such as adaptors and coat proteins, to facilitate vesicle formation [9], [17], [78]. However, both models are yet to be proven at the molecular level.Erd1p is required for transport of Pi from the Golgi lumen to the cytoplasm [20]. Pi is a product of the glycosylation reaction, and Pi accumulation in the Golgi lumen seems to be responsible for impaired protein glycosylation and defective retrieval of Kar2p/Bip from the cis-Golgi compartment of the erd1Δ mutant. Our results suggest that Pi accumulation in the Golgi lumen is responsible for the synthetic lethality of erd1Δ with neo1Δ cfs1Δ mutations. The combination of abnormal regulation of phospholipid asymmetry and defects of Pi homeostasis leads to severe defects of the vesicle transport process from Golgi membranes. The transbilayer phospholipid distribution in the Golgi membrane is unknown, but asymmetric phospholipid distribution may regulate the function and activity of membrane proteins, including glycosyltransferases, ion transporters, and cargo receptors. Thus, flippase-regulated phospholipid redistribution and luminal Pi concentration seem to coordinately regulate Golgi membrane functions. This mechanism may differ substantially from the direct involvement of flippases in the vesicle formation process. The functional relevance between luminal Pi and phospholipid asymmetry is unknown, but increased Pi alters the ionic environment, which may affect protein-lipid interactions at the luminal surface of the Golgi membrane. Thus, simultaneous changes of the ionic and lipid environments appear to drastically affect membrane trafficking in the Golgi apparatus.PM proteins were not transported out of the TGN in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants, which may be due to the lack of PI4P at the TGN. Interestingly, the PI4 kinase Pik1p was normally colocalized with Sec7p, which is a binding partner of Pik1p [79]. Thus, Pik1p is either not activated or PI, which is its substrate, is not sufficient in the cytoplasmic leaflet of the Golgi membrane. PI is also used as a donor of inositol phosphate for the synthesis of complex sphingolipids, inositol phosphoceramide (IPC) and mannosyl-di-inositol phosphoceramide (MIP2C), which is a process that is thought to occur in the luminal leaflet of the Golgi membrane [80]. Therefore, regulation of the transbilayer distribution of PI is an important factor to facilitate Golgi functions. However, PI has not been examined in flippase assays, as a fluorescence-labeled PI is not yet available. PI may be mainly distributed in the luminal leaflet of the Golgi membrane in the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants. Regarding future research, an examination of the potential involvement of Neo1p in flipping PI is of particular interest. We previously showed that inositol depletion from the growth medium suppressed the growth defects of Cdc50p-depleted lem3Δ crf1Δ and Neo1p-depleted mutants, but not those of the Cdc50p- and Neo1p-depleted mutants [81]. Interestingly, this suppression pattern was the same for the cfs1Δ mutation [18]. The suppression mechanism by inositol depletion remains unknown, but it may be relevant to the metabolism or transbilayer distribution of PI, and Cfs1p may also be involved in this process.
Synthetic lethality with the Neo1p-depleted cfs1Δ mutation is specific to erd1Δ.
(A) Neo1p-depleted cfs1Δ is not synthetically lethal with other mutations involved in ER retention. Cells were spotted onto YPGA (galactose) and YPDA (glucose) plates and grown as described in Fig 1B. The strains used were P-NEO1 cfs1Δ (WT) (YKT2085) and P-NEO1 cfs1Δ carrying erd1Δ (YKT2136), bst1Δ (YKT2206), emp24Δ (YKT2207), eps1Δ (YKT2208), erp1Δ (YKT2209), erp2Δ (YKT2210), and rer1Δ (YKT2211). (B) Neo1p-depleted cfs1Δ is not synthetically lethal with mutations involved in Golgi glycosylation. Cells were spotted onto YPGA (galactose) and YPDA (glucose) plates and grown as described in (A). The strains used were P-NEO1 cfs1Δ (WT) (YKT2085) and P-NEO1 cfs1Δ carrying erd1Δ (YKT2136), van1Δ (YKT2212), mnn10Δ (YKT2213), mnn2Δ (YKT2214), mnn5Δ (YKT2215), mnn4Δ (YKT2216), mnn6Δ (YKT2217), and mnn1Δ (YKT2218).(TIF)Click here for additional data file.
Partial depletion of Pik1p impairs growth of the neo1Δ cfs1Δ mutant.
Cells were grown and spotted as in Fig 4A onto synthetic medium containing 2% glucose (Gal 0%) or 2% raffinose and 0.01% (Gal 0.01%) or 0.1% (Gal 0.1%) galactose, followed by incubation at 30°C for 2 d. The strains used were wild type (WT) (YKT38), P-PIK1 (YKT2219), P-PIK1 cfs1Δ (YKT2220), and P-PIK1neo1Δ cfs1Δ (YKT2221).(TIF)Click here for additional data file.
Induction of UPR and abnormal structure of the ER in the Erd2p-depleted mutant.
(A) The growth defect of the Erd2p-depleted mutant. Cells were grown and spotted as in Fig 1B onto YPGA (galactose) and YPDA (glucose) plates, followed by incubation at 30°C for 1.5 d. The strains used were wild type (WT) (YKT38) and P-ERD2 (YKT2222). (B) Induction of UPR in the Erd2p-depleted mutant. Yeast cells that express GFP under the control of 2× UPRE were cultured in YPDA as in Fig 1C. The strain used was the P-ERD2 (YKT2222), which carries 2× UPRE-GFP. (C) Localization of Hmg1p-GFP and Snc1-pm in the Erd2p-depleted mutant. Cells were cultured in YPDA as in Fig 1C. The strains used were the P-ERD2 with HMG1-GFP and mRFP1-SNC1-pm (YKT2223) or GFP-SNC1-pm and SEC7-mRFP1 (YKT2224). Bar, 5 μm. DIC, differential interference contrast.(TIF)Click here for additional data file.
Saccharomyces cerevisiae strains used in this study.
(PDF)Click here for additional data file.
Plasmids used in this study.
(PDF)Click here for additional data file.16 Apr 2020PONE-D-20-06448A complex genetic interaction implicates that phospholipid asymmetry and phosphate homeostasis regulate Golgi functionsPLOS ONEDear Dr. Tanaka,Thank you for submitting your manuscript to PLOS ONE. I am very sorry for the late response due to the difficulties in obtaining the comments from the reviewers amid the COVID-19 pandemic . After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.As you will see from their comments, although both reviewers are very impressed by the findings of an unexpected functional link between phosphate homeostasis and phospholipid transport relevant for a proper functioning of the Golgi, they are not fully convinced by your statements or interpretations of the data.We would appreciate receiving your revised manuscript by the end of June. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocolsPlease include the following items when submitting your revised manuscript:A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.We look forward to receiving your revised manuscript.Kind regards,Reiko Sugiura, M.D., PhD.Academic EditorPLOS ONEJournal requirements:When submitting your revision, we need you to address these additional requirements:1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf[Note: HTML markup is below. Please do not edit.]Reviewers' comments:Reviewer's Responses to QuestionsComments to the Author1. Is the manuscript technically sound, and do the data support the conclusions?The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.Reviewer #1: NoReviewer #2: Yes**********2. Has the statistical analysis been performed appropriately and rigorously?Reviewer #1: YesReviewer #2: Yes**********3. Have the authors made all data underlying the findings in their manuscript fully available?The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.Reviewer #1: YesReviewer #2: Yes**********4. Is the manuscript presented in an intelligible fashion and written in standard English?PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.Reviewer #1: NoReviewer #2: Yes**********5. Review Comments to the AuthorPlease use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)Reviewer #1: This study reports on the genetic interactions between a putative flippase, Neo1p, its antagonistic regulator Cfs1p, and a Golgi-resident candidate transporter of inorganic phosphate (Pi), Erd1p. The authors present evidence that cells lacking Neo1p and Cfs1p are particularly sensitive to imbalances in Golgi Pi homeostasis, causing defects in cargo trafficking from the Golgi, an aberrant subcellular distribution of PS and PI4P, and induction of an unfolded protein response. While the study does not provide a coherent mechanistic basis for these observations, the experimental data are of good quality and point at an unexpected link between phosphate homeostasis and phospholipid transport relevant for a proper functioning of the Golgi. However, the manuscript contains some verbose or fuzzy statements and not all conclusions are justified by the experimental data.COMMENTS1) I have trouble grasping the meaning of some key statements in the manuscript. For instance, in the Abstract and on p. 4 (bottom) and p. 26 (middle), the authors state “our results suggest that flippase-mediated phospholipid redistribution is functionally vital not only in the cytoplasmic leaflet but also in the luminal leaflet of the Golgi membrane”. I do not see the added value of such a cryptic statement. The study also does not provide any concrete data on the lipid composition of the luminal leaflet in the Golgi of wildtype and mutant cells. I urge the authors to formulate a less ambiguous take-home message, one that focuses more on their principal experimental findings.2) On p. 19 (bottom) and p. 20 (top), the authors state that their findings demonstrate that overexpression of Pho87p and Pho90p accelerates transport of luminally accumulated Pi to the cytoplasm, and conclude that elevated Pi levels in the Golgi lumen are responsible for the lethality of the Neo1p-depleted csf1/erd1 mutant cells. However, their experimental data provide only indirect evidence for this. Therefore, they should significantly tune down these claims, for instance by replacing the words “demonstrate” for “suggest” and “conclude” for “propose”.3) To dissect the complex genetic interactions that implicate phospholipid asymmetry and phosphate homeostasis in regulating Golgi function, the authors analyzed the consequences of Neo1p depletion on Golgi function in wildtype, csf1 and csf1/erd1 mutant cells. As Erd1p has been suggested to transport Pi from the lumen of the Golgi, analyzing also the consequences of Neo1p depletion in csf1 mutant cells on Golgi functions would have been particularly informative. However, such experiments are not part of the present study. Why not?4) p. 14 (top) “…the TGN is exposed to the accumulated PS” should read “…PS is accumulated in the cytosolic leaflet of the TGN”5) p. 14 (end 1st para) “…the cytoplasmic leaflet of the TGN…. is exposed to PS” should be reformulated as indicated above.6) Fig. 3 caption “The TGN… is exposed to PS and has no PI4P” should read “The cytoplasmic leaflet of the TGN … contains PS and has no PI4P”Reviewer #2: Neo1 is a P-type ATPase from the P4 subfamily (P4-ATPase), most members of this subfamily being lipid flippases, i.e. they catalyse lipid transport from the exoplasmic to cytoplasmic leaflet of eukaryotic cell membranes. This activity turns out to be critical for numerous cell functions, ranging from establishment of cell polarity to the regulation of membrane trafficking events. However, the function of Neo1 remains elusive probably partly due to the fact that Neo1 is the only essential P4-ATPase in yeast. The authors previously identified Cfs1, a member of the PQ-loop protein family, as a suppressor of the neo1Δ growth defect. In the present manuscript, to gain further insights into the relationships between Neo1 and Cfs1, Miyasaka and colleagues report on the identification of the erd1 mutation as synthetically lethal with neo1Δcfs1Δ. Erd1 is believed to transport phosphate from the lumen of the Golgi apparatus to the cytosol. Miyasaka and colleagues characterized the neo1Δcfs1Δerd1Δ mutant. Interestingly, the neo1Δcfs1Δerd1Δ and neo1Δ mutants exhibited PI4P defects at the TGN, possibly explaining the trafficking defects of these mutants. The authors further identified genes that suppress the neo1Δcfs1Δerd1Δ phenotype, but not the neo1Δ phenotype, suggesting they suppress the erd1 mutation. Given the identity of the identified suppressors, this work provides a possible mechanism for the synthetic lethality of the Neo1-depleted cfs1Δerd1Δ mutant, namely elevated Pi Golgi levels. Indeed, some of the suppressor genes identified belong to a conserved family of plasma membrane transporters involved in phosphate uptake. Additionally, some of the identified suppressors are proteins involved in the ER to Golgi transport, in line with the ER structural defects observed for the Neo1-depleted cfs1Δerd1Δ mutant and for the neo1 mutant.Overall, this work proposes a connection between phospholipid asymmetry and luminal Pi levels and significantly contributes to the field. The approach is smart and elegant, the work is very neat, and the manuscript well organized.My specific comments are as follows:- In the abstract (lines 32-34) and in the discussion (lines 488-489), the authors propose that phospholipid redistribution is not only important in the cytosolic leaflet but also in the luminal leaflet. Do the authors mean that phospholipid asymmetry has an important impact not only on the cytosolic face of the membrane but also on the luminal one? I find the word ‘leaflet’ misleading in this case, as it restricts the impact of phospholipid asymmetry to the membrane itself- Introduction, line 47: spelling mistake ‘thorough’. But in fact, what do the authors mean by trafficking ‘through’ the membranes?- Introduction, lines 65-69. It is mentioned that ‘in the neo1Δcfs1Δ mutant, phospholipid asymmetry may not be normally regulated’. As cfs1 is a suppressor of neo1, it does not sound so obvious. Why should we envision defects in the neo1Δcfs1Δ mutant? Where does this assumption originate from? I think it would help the general reader to explain a bit better the rationale followed here. This is further substantiated by the fact that at lines 157-158 in the results section, it is claimed that the cfs1Δ mutation completely suppresses PE and PS exposure.- Results, lines 182-184. The sentence is odd, as it starts by saying that Pdr5 accumulated in the Neo1-depleted cfs1Δerd1Δ mutant, and finishes with ‘although the accumulation of Pdr5-GFP was relatively low’. In addition, it’s not really clear form the figure that the accumulation is low.- Results, lines 184-185. What is the evidence at this stage that the defect in protein trafficking is in the secretory rather than in the endocytic pathway? Especially given that Drs2/Cdc50 mutations interfere with endocytic retrieval of Snc1?- This is probably a very naive question, which is however relevant to several experiments displayed in the manuscript: in the legend to figure 1B, it is mentioned that cells are grown in YPDA, and then plated either on YPDA or YPGA; however, as Neo1 is an essential gene, how can cells grow with Neo1 under the control of a GAL promoter in a medium containing glucose as a carbon source?- Legend to figure 1C, line 198. It is mentioned that ‘strains are cultured for 20h…’ In which medium are cells cultured?- Results, line 248. ‘…the TGN is exposed to the accumulated PS’. The wording sounds odd to me if the aim is to say that PS is exposed to the cytosolic leaflet of the TGN. This wording can be found elsewhere in the manuscript. Moreover, how does PS distribute, using Lact-C2 and evt-2 probes, in Neo1-depleted cfs1Δerd1Δ mutant?- Results, line 264. It is suggested that absence of PI4P in the TGN may be the reason for membrane trafficking defects in the Neo1-depleted cfs1Δerd1Δ mutant. Isn’t it also true for the neo1Δ mutant? I suggest it’s worth recalling it so that a connection is made between the Neo1 flippase and PI4P.- Title of Fig 3. It is claimed that the TGN ‘has no PI4P’. I guess it would be fairer to say that PI4P levels dropped dramatically in the TGN, in the absence of sensitive assays to measure PI4P levels.- In my opinion, the manuscript would gain in clarity if a simple schematic recalling the direction of phosphate transport envisioned for Erd1 and demonstrated for PHO transporters were added. It’s not immediately obvious that PHO transporters are involved in the uptake of phosphate and that would help understand the rationale of the experiments (lines 333-336).- Results, line 342. What is the phosphate concentration in a ‘normal’ SD medium (Fig 5B)?- Results, lines 360-361. It’s probably a bit bold to claim that the authors demonstrated that ‘…Pho87 and Pho90 accelerated transport of the luminally accumulated Pi to the cytoplasm…’ in the absence of transport assays.- Legend to Fig 6A, line 409. Why the authors do not provide statistic tests here as for the other figures (figure 4 for instance)? Is the observed difference significant?- Results, lines 427-428. I’m not sure about the wording of the sentence. Is it the induction of UPR by DTT in the WT which is referred to? Why not comparing induction of UPR in the Neo1-depleted cfs1Δerd1Δ mutant with that of the WT in the absence of DTT? And the induction of UPR in the Neo1-depleted cfs1Δerd1Δ mutant does not seen higher than that of WT +DTT?**********6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.If you choose “no”, your identity will remain anonymous but your review may still be made public.Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.Reviewer #1: NoReviewer #2: Yes: Guillaume Lenoir[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.9 Jun 2020We thank the reviewers for valuable comments on our manuscript. We have thoroughly revised our manuscript according to the reviewers’ comments as described below.Our responses to the comments of the Reviewer #11) I have trouble grasping the meaning of some key statements in the manuscript. For instance, in the Abstract and on p. 4 (bottom) and p. 26 (middle), the authors state “our results suggest that flippase-mediated phospholipid redistribution is functionally vital not only in the cytoplasmic leaflet but also in the luminal leaflet of the Golgi membrane”. I do not see the added value of such a cryptic statement. The study also does not provide any concrete data on the lipid composition of the luminal leaflet in the Golgi of wildtype and mutant cells. I urge the authors to formulate a less ambiguous take-home message, one that focuses more on their principal experimental findings.We agree to the reviewer’s comments that our data do not lead us to any conclusion concerning a functional difference between the cytoplasmic leaflet and the luminal leaflet of the Golgi membrane. According to the reviewer’s suggestion, we have changed the sentences to “Our results suggest that phospholipid asymmetry and luminal Pi concentration coordinately regulate Golgi functions” in the Abstract (lines 32-33), the Introduction (lines 71-72), and the Discussion (lines 494-495).2) On p. 19 (bottom) and p. 20 (top), the authors state that their findings demonstrate that overexpression of Pho87p and Pho90p accelerates transport of luminally accumulated Pi to the cytoplasm, and conclude that elevated Pi levels in the Golgi lumen are responsible for the lethality of the Neo1p-depleted csf1/erd1 mutant cells. However, their experimental data provide only indirect evidence for this. Therefore, they should significantly tune down these claims, for instance by replacing the words “demonstrate” for “suggest” and “conclude” for “propose”.We agree to the reviewer’s comment that we should tone down our statement on the significance of elevated Pi levels in the Golgi lumen for the lethality of the Neo1p-depleted cfs1 erd1 mutant. According to the reviewer’s suggestion, we have replaced the words “demonstrate” for “suggest” and “conclude” for “propose” (lines 362-365).3) To dissect the complex genetic interactions that implicate phospholipid asymmetry and phosphate homeostasis in regulating Golgi function, the authors analyzed the consequences of Neo1p depletion on Golgi function in wildtype, csf1 and csf1/erd1 mutant cells. As Erd1p has been suggested to transport Pi from the lumen of the Golgi, analyzing also the consequences of Neo1p depletion in csf1 mutant cells on Golgi functions would have been particularly informative. However, such experiments are not part of the present study. Why not?Although the reviewer suggests “analyzing the consequences of Neo1p depletion in csf1 mutant”, we already analyzed this mutant in the original version. Judging from the context, we think that the reviewer suggests “analyzing the consequences of Neo1p depletion in erd1 mutant”. However, because the Neo1p depletion in wild-type cells shows severe defects in Golgi functions, the Neo1p-depleted erd1Δ mutant would show similar severe defects unless the erd1Δ mutation suppresses the neo1Δ mutation. We actually constructed the Neo1p-depleted erd1Δ mutant expressing GFP-Snc1p-pm. This mutant exhibited massive accumulation of GFP-Snc1p-pm as the Neo1p-depleted and Neo1p-depleted cfs1Δ erd1Δ mutants. This result has been shown in Figure 1C in the revised version.4) p. 14 (top) “…the TGN is exposed to the accumulated PS” should read “…PS is accumulated in the cytosolic leaflet of the TGN”5) p. 14 (end 1st para) “…the cytoplasmic leaflet of the TGN…. is exposed to PS” should be reformulated as indicated above.6) Fig. 3 caption “The TGN… is exposed to PS and has no PI4P” should read “The cytoplasmic leaflet of the TGN … contains PS and has no PI4P”We thank the reviewer for pointing out our mistakes. We have corrected these points accordingly (lines 250, 254-255, 276, and 278).Our responses to the comments of the Reviewer #2- In the abstract (lines 32-34) and in the discussion (lines 488-489), the authors propose that phospholipid redistribution is not only important in the cytosolic leaflet but also in the luminal leaflet. Do the authors mean that phospholipid asymmetry has an important impact not only on the cytosolic face of the membrane but also on the luminal one? I find the word ‘leaflet’ misleading in this case, as it restricts the impact of phospholipid asymmetry to the membrane itselfThis comment has also been made by the Reviewer #1. We agree to the reviewer’s comments that our data do not lead us to any conclusion concerning a functional difference between the cytoplasmic leaflet and the luminal leaflet of the Golgi membrane. According to the reviewer’s suggestion, we have changed the sentences to “Our results suggest that phospholipid asymmetry and luminal Pi concentration coordinately regulate Golgi functions” in the Abstract (lines 32-33), the Introduction (lines 71-72), and the Discussion (lines 494-495).- Introduction, line 47: spelling mistake ‘thorough’. But in fact, what do the authors mean by trafficking ‘through’ the membranes?We thank the reviewer for pointing out our mistake. As questioned by the reviewer, we do not mean any specific membrane trafficking pathway by “through”. Therefore, this word has been deleted. (line 47)- Introduction, lines 65-69. It is mentioned that ‘in the neo1Δcfs1Δ mutant, phospholipid asymmetry may not be normally regulated’. As cfs1 is a suppressor of neo1, it does not sound so obvious. Why should we envision defects in the neo1Δcfs1Δ mutant? Where does this assumption originate from? I think it would help the general reader to explain a bit better the rationale followed here. This is further substantiated by the fact that at lines 157-158 in the results section, it is claimed that the cfs1Δ mutation completely suppresses PE and PS exposure.As pointed out by the reviewer, our statement that “in the neo1Δ cfs1Δ mutant, phospholipid asymmetry may not be normally regulated” may not sound obvious. To make the sentence more consistent to readers, we have changed this sentence to “However, the neo1Δ cfs1Δ mutant may not be equivalent to the wild type in its Golgi functions, otherwise a set of these two genes including the essential NEO1 gene would be dispensable.” (lines 65-67)- Results, lines 182-184. The sentence is odd, as it starts by saying that Pdr5 accumulated in the Neo1-depleted cfs1Δerd1Δ mutant, and finishes with ‘although the accumulation of Pdr5-GFP was relatively low’. In addition, it’s not really clear form the figure that the accumulation is low.We thank the reviewer for these comments. To make how the accumulation of Pdr5-GFP was relatively low clearer, we have modified the sentence as follows. “…, although the accumulation of Pdr5p-GFP was relatively low compared to GFP-Snc1p-pm and GFP-Sso1p” (line 184). As to Fig 1D, the number of cells that showed clear accumulation of Pdr5p-GFP was lower compared to GFP-Snc1p-pm and GFP-Sso1p, but the cells that showed accumulation accumulated Pdr5p-GFP to an extent similar to those of GFP-Snc1p-pm and GFP-Sso1p.- Results, lines 184-185. What is the evidence at this stage that the defect in protein trafficking is in the secretory rather than in the endocytic pathway? Especially given that Drs2/Cdc50 mutations interfere with endocytic retrieval of Snc1?Because we used GFP-Snc1p-pm, which is a mutant version that is not endocytosed, this protein does not enter the endocytic recycling pathway. However, it is known that Pdr5p is endocytosed and delivered to vacuoles. Thus, we have changed the sentence to “These results suggest that the Neo1p-depleted cfs1Δ erd1Δ mutant exhibits major defects in membrane trafficking pathways.” (lines 184-186)- This is probably a very naive question, which is however relevant to several experiments displayed in the manuscript: in the legend to figure 1B, it is mentioned that cells are grown in YPDA, and then plated either on YPDA or YPGA; however, as Neo1 is an essential gene, how can cells grow with Neo1 under the control of a GAL promoter in a medium containing glucose as a carbon source?Because the catabolite repression is slow in the GAL1 promoter, and because the GAL1 promoter is a strong promoter, we need preincubation of the cells in YPDA medium to deplete Neo1p for ~12 h. Otherwise, the PGAL1-NEO1 cells grown in YPGA continue to grow for ~12 h even in the YPDA medium. We have added “to deplete Neo1p” in the revised version (lines 193 and 199)- Legend to figure 1C, line 198. It is mentioned that ‘strains are cultured for 20h…’ In which medium are cells cultured?As written above, the cells were cultured in YPDA medium to deplete Neo1p. In the sentence pointed out by the reviewer, there was a redundant description that indicated that the cells were cultured more than 20 h in YPDA. This description, “and were cultured”, which is our mistake, has been deleted in the revised version (line 199). Because the repression of the GAL1 promoter is not 100% in YPDA medium, some mutant requires depletion time of more than 12 hours. As shown in Fig 1B, the PGAL1-NEO1 cfs1Δ erd1Δ mutant showed some residual growth compared to the PGAL1-NEO1 mutant, suggesting that the 12 h depletion is not enough to completely inhibit cell growth of the PGAL1-NEO1 cfs1Δ erd1Δ mutant. To observe the terminal phenotype of the PGAL1-NEO1 cfs1Δ erd1Δ mutant, we determined that twenty hours incubation in YPDA is appropriate for the PGAL1-NEO1 cfs1Δ erd1Δ mutant (line 199).- Results, line 248. ‘…the TGN is exposed to the accumulated PS’. The wording sounds odd to me if the aim is to say that PS is exposed to the cytosolic leaflet of the TGN. This wording can be found elsewhere in the manuscript. Moreover, how does PS distribute, using Lact-C2 and evt-2 probes, in Neo1-depleted cfs1Δerd1Δ mutant?These our mistakes were also pointed out by the reviewer #1, and all of them have been corrected appropriately to describe that PS is exposed to the cytoplasmic leaflet of the TGN (lines 250, 254-255, 276, and 278). As to the PS distribution using Lact-C2 and evt-2 probes, these probes were similarly colocalized with GFP-Snc1p-pm in the Neo1-depleted cfs1Δ erd1Δ mutant (Fig 3A and B). As shown in Fig 2C, GFP-Snc1p-pm was colocalized with Sec7p-mRFP1. These results suggest that PS is exposed to the cytoplasmic leaflet of the TGN in the Neo1-depleted cfs1Δ erd1Δ mutant (lines 254-255).- Results, line 264. It is suggested that absence of PI4P in the TGN may be the reason for membrane trafficking defects in the Neo1-depleted cfs1Δerd1Δ mutant. Isn’t it also true for the neo1Δ mutant? I suggest it’s worth recalling it so that a connection is made between the Neo1 flippase and PI4P.We thank the reviewer for pointing out that the Neo1p-depleted mutant should be included in the sentence. It is important to describe that the absence of PI4P may cause membrane trafficking defects in the Neo1p-depleted mutant. It has been added to the revised version (line 266).- Title of Fig 3. It is claimed that the TGN ‘has no PI4P’. I guess it would be fairer to say that PI4P levels dropped dramatically in the TGN, in the absence of sensitive assays to measure PI4P levels.We agree to the reviewer’s comment that we should not describe “has no PI4P”. According to the reviewer’s suggestion, we have changed it to “decreased level of PI4P” (line 277).- In my opinion, the manuscript would gain in clarity if a simple schematic recalling the direction of phosphate transport envisioned for Erd1 and demonstrated for PHO transporters were added. It’s not immediately obvious that PHO transporters are involved in the uptake of phosphate and that would help understand the rationale of the experiments (lines 333-336).We thank the reviewer for the comment. According to the reviewer’s suggestion, we have added the schematic model as Fig 5A. In addition, we have also added Fig 5F to explain the other model, which is consistent with our data.- Results, line 342. What is the phosphate concentration in a ‘normal’ SD medium (Fig 5B)?Phosphate concentration in SD medium is 7.3 mM according to the manufacturer. This has been added to Fig 5C and described in the legend in the revised version (lines 377 to 378).- Results, lines 360-361. It’s probably a bit bold to claim that the authors demonstrated that ‘…Pho87 and Pho90 accelerated transport of the luminally accumulated Pi to the cytoplasm…’ in the absence of transport assays.This comment has also been made by the reviewer #1. According to the reviewers’ suggestion, we have replaced the word “demonstrate” for “suggest” (line 362).- Legend to Fig 6A, line 409. Why the authors do not provide statistic tests here as for the other figures (figure 4 for instance)? Is the observed difference significant?According to the reviewer’s suggestion, we have performed statistic tests for Fig 6A (line 418) and Fig 1C (lines 204-205). Both tests gave a significant difference to the results as shown in the revised version.- Results, lines 427-428. I’m not sure about the wording of the sentence. Is it the induction of UPR by DTT in the WT which is referred to? Why not comparing induction of UPR in the Neo1-depleted cfs1Δerd1Δ mutant with that of the WT in the absence of DTT? And the induction of UPR in the Neo1-depleted cfs1Δerd1Δ mutant does not seen higher than that of WT +DTT?As pointed out by the reviewer, to compare the induction of UPR in the Neo1-depleted cfs1Δ erd1Δ mutant with that of the WT in the absence of DTT, we have deleted the words “compared to that induced by DTT” in the revised version (line 436).Submitted filename: Response to Reviewers.docxClick here for additional data file.30 Jun 2020PONE-D-20-06448R1A complex genetic interaction implicates that phospholipid asymmetry and phosphate homeostasis regulate Golgi functionsPLOS ONEDear Dr. Tanaka,Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.==============================Two reviewers' evaluations are now in as shown below. I am happy to inform you that two reviewers #1 and #2 suggest minor revisions. Please read carefully those arguments and revise the manuscript item-by-item.==============================Please submit your revised manuscript by the end of July. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.Please include the following items when submitting your revised manuscript:A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocolsWe look forward to receiving your revised manuscript.Kind regards,Reiko Sugiura, M.D., PhD.Academic EditorPLOS ONE[Note: HTML markup is below. Please do not edit.]Reviewers' comments:Reviewer's Responses to QuestionsComments to the Author1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.Reviewer #1: All comments have been addressedReviewer #2: All comments have been addressed**********2. Is the manuscript technically sound, and do the data support the conclusions?The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.Reviewer #1: YesReviewer #2: Yes**********3. Has the statistical analysis been performed appropriately and rigorously?Reviewer #1: YesReviewer #2: Yes**********4. Have the authors made all data underlying the findings in their manuscript fully available?The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.Reviewer #1: YesReviewer #2: Yes**********5. Is the manuscript presented in an intelligible fashion and written in standard English?PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.Reviewer #1: YesReviewer #2: Yes**********6. Review Comments to the AuthorPlease use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)Reviewer #1: I urge the authors to have the manuscript proof-read by a native English speaker.xxxxxxxxxxxxxxxxReviewer #2: The authors adequately addressed the issues I have raised.- However, I think it would help indicating the (perhaps putative) function of the genes which mutations are involved in ER retention (line 174), for instance in the legend to Figure S1A. The same holds for Fig S1B.- Line 205: '...fluorescent microscope' should read '...fluorescence microscope'?- Please indicate in the legend to Fig 5A what the rod-shaped grey drawing corresponds to.**********7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.If you choose “no”, your identity will remain anonymous but your review may still be made public.Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.Reviewer #1: NoReviewer #2: Yes: Guillaume Lenoir[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.4 Jul 2020Our responses to the comments of the Reviewer #1I urge the authors to have the manuscript proof-read by a native English speaker.The unrevised original version of our manuscript had already been checked by a leading company for English language editing. However, their work does not seem to be perfect, because the sentences pointed out by the reviewer #1 (e.g. comments 4, 5, and 6), which were also pointed out by the reviewer #2, were modified according to the editor’s suggestions. In the revised version, we finally made these sentences back almost to the original ones. We understand that this is because the editor would not be a specialist in our filed. Therefore, in the revised version, we checked the manuscript by ourselves for any kind of mistakes as much as possible.Our responses to the comments of the Reviewer #2- However, I think it would help indicating the (perhaps putative) function of the genes which mutations are involved in ER retention (line 174), for instance in the legend to Figure S1A. The same holds for Fig S1B.According to the reviewer’s suggestion, gene (protein) functions have been included in S1 Fig A and B.- Line 205: '...fluorescent microscope' should read '...fluorescence microscope'?We thank the reviewer for pointing out our mistake. It has been corrected in the revised version (line 200).- Please indicate in the legend to Fig 5A what the rod-shaped grey drawing corresponds to.According the reviewer’s suggestion, we have indicated in the legend that the rod-shaped grey drawing corresponds to membrane proteins to be transported to the plasma membrane (line 371).Submitted filename: Response to Reviewers.docxClick here for additional data file.9 Jul 2020A complex genetic interaction implicates that phospholipid asymmetry and phosphate homeostasis regulate Golgi functionsPONE-D-20-06448R2Dear Dr. Tanaka,We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.Kind regards,Reiko Sugiura, M.D., PhD.Academic EditorPLOS ONE15 Jul 2020PONE-D-20-06448R2A complex genetic interaction implicates that phospholipid asymmetry and phosphate homeostasis regulate Golgi functionsDear Dr. Tanaka:I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.If we can help with anything else, please email us at plosone@plos.org.Thank you for submitting your work to PLOS ONE and supporting open access.Kind regards,PLOS ONE Editorial Office Staffon behalf ofDr. Reiko SugiuraAcademic EditorPLOS ONE
Authors: Jolanda van Leeuwen; Carles Pons; Joseph C Mellor; Takafumi N Yamaguchi; Helena Friesen; John Koschwanez; Mojca Mattiazzi Ušaj; Maria Pechlaner; Mehmet Takar; Matej Ušaj; Benjamin VanderSluis; Kerry Andrusiak; Pritpal Bansal; Anastasia Baryshnikova; Claire E Boone; Jessica Cao; Atina Cote; Marinella Gebbia; Gene Horecka; Ira Horecka; Elena Kuzmin; Nicole Legro; Wendy Liang; Natascha van Lieshout; Margaret McNee; Bryan-Joseph San Luis; Fatemeh Shaeri; Ermira Shuteriqi; Song Sun; Lu Yang; Ji-Young Youn; Michael Yuen; Michael Costanzo; Anne-Claude Gingras; Patrick Aloy; Chris Oostenbrink; Andrew Murray; Todd R Graham; Chad L Myers; Brenda J Andrews; Frederick P Roth; Charles Boone Journal: Science Date: 2016-11-04 Impact factor: 47.728
Authors: Richa Sardana; Carolyn M Highland; Beth E Straight; Christopher F Chavez; J Christopher Fromme; Scott D Emr Journal: Elife Date: 2021-11-25 Impact factor: 8.140
Fig 1
Fig 2
Fig 3
Fig 4
Fig 5
Fig 6
Fig 7