Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Cholesterol is a major structural component of the plasma membrane (PM). The majority of PM cholesterol forms complexes with other PM lipids, making it inaccessible for intracellular transport. Transition of PM cholesterol between accessible and inaccessible pools maintains cellular homeostasis, but how cells monitor the accessibility of PM cholesterol remains unclear. We show that endoplasmic reticulum (ER)-anchored lipid transfer proteins, the GRAMD1s, sense and transport accessible PM cholesterol to the ER. GRAMD1s bind to one another and populate ER-PM contacts by sensing a transient expansion of the accessible pool of PM cholesterol via their GRAM domains. They then facilitate the transport of this cholesterol via their StART-like domains. Cells that lack all three GRAMD1s exhibit striking expansion of the accessible pool of PM cholesterol as a result of less efficient PM to ER transport of accessible cholesterol. Thus, GRAMD1s facilitate the movement of accessible PM cholesterol to the ER in order to counteract an acute increase of PM cholesterol, thereby activating non-vesicular cholesterol transport. eLife digest The human body contains trillions of cells. At the outer edge of each cell is the plasma membrane, which protects the cell from the external environment. This membrane is mostly made of fatty molecules known as lipids and about half of these lipids are specifically cholesterol. Human cells can either take up cholesterol that were obtained via the diet or produce it within a compartment of the cell called the endoplasmic reticulum. Cells need to monitor the cholesterol levels in both the endoplasmic reticulum and the plasma membrane in order to regulate the uptake or production of this lipid. For example, if there is too much of cholesterol in the plasma membrane, then the cell transports some to the endoplasmic reticulum to tell it to shut down cholesterol production. However, how these different areas of the cell communicate with each other, and transport cholesterol, has remained unclear. Naito et al. set out to look for key regulators of cholesterol transport and identified a group of endoplasmic reticulum proteins called GRAMD1 proteins. Cholesterol in the plasma membrane is either accessible or inaccessible, meaning it either can or cannot be moved back into the cell. The GRAMD1 proteins sense accessible cholesterol, and experiments with human cells grown in the laboratory showed that, specifically, the GRAMD1 proteins work together in a complex to sense accessible cholesterol at or near the plasma membrane. One particular part of the protein senses when the amount of accessible cholesterol reaches a certain level at the plasma membrane; when this threshold is reached, the complex flips a switch to start the transport of cholesterol to the endoplasmic reticulum and tell it to shut down cholesterol production. This coupling of sensing and transporting lipids by one protein complex also helps maintain the right ratio of accessible and inaccessible cholesterol in the plasma membrane to prevent cells from activating unwanted cell-signaling events. Getting rid of the GRAMD1 proteins in cells, or removing sensing part of these proteins, leads to inefficient transport of cholesterol. A better understanding of how GRAMD1 proteins sense the accessibility of cholesterol could potentially help identify new approaches to control cholesterol transport inside cells. This may in turn eventually lead to new treatments that counteract the defects in cholesterol metabolism seen in some forms of neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease. Introduction Sterol is one of the major membrane lipids in eukaryotes. In metazoans, cholesterol represents ~20% of total cellular lipids and is therefore essential for the structural integrity of cellular membranes and for cell physiology (van Meer et al., 2008; Vance, 2015). Sterol is distributed among cellular membranes primarily via non-vesicular transport, a process that is independent of membrane traffic (Baumann et al., 2005; Hao et al., 2002; Heino et al., 2000; Ikonen, 2008; Urbani and Simoni, 1990). Levels of sterol vary considerably between different cellular membranes. Between 60% and 80% of total cellular cholesterol is concentrated in the plasma membrane (PM), where it represents up to ~45% of total lipids in this bilayer (de Duve, 1971; Lange et al., 1989; Ray et al., 1969). Cellular cholesterol levels are maintained by regulated delivery and production, primarily through receptor-mediated endocytosis of low-density lipoproteins (LDLs) (Goldstein and Brown, 2015) and de novo synthesis in the endoplasmic reticulum (ER) that is controlled by the activation of SREBP transcription factors (Brown et al., 2018; Goldstein and Brown, 1990). Cholesterol is also supplied to cells via high-density lipoproteins (HDL) through the reverse cholesterol flux pathway (Acton et al., 1996; Phillips, 2014). Cholesterol within the bilayer membranes exists in two distinct chemical states: one being free and 'accessible' (also known as 'unsequestered' or 'chemically active'), and the other being 'inaccessible' (also known as 'sequestered' or 'chemically inactive') owing in part to the formation of complexes with other membrane lipids, including sphingomyelin and phospholipids (Chakrabarti et al., 2017; Das et al., 2014; Gay et al., 2015; Lange et al., 2013; Lange et al., 2004; McConnell and Radhakrishnan, 2003; Ohvo-Rekilä et al., 2002; Radhakrishnan and McConnell, 2000; Sokolov and Radhakrishnan, 2010). Most cholesterol in the PM is sequestered, but a small fraction of PM cholesterol (~15% of PM lipids) remains accessible for extraction and transport (Das et al., 2014). Although the majority of cellular cholesterol resides in the PM, the biosynthesis of cholesterol occurs exclusively in the ER. Thus, the ER must communicate with the PM to monitor levels of PM cholesterol and to adjust cholesterol biosynthesis in order to maintain lipid homeostasis. To achieve this, cells sense transient increases in the accessible pool of PM cholesterol and rapidly transport the newly expanded pool of accessible PM cholesterol to the ER. This suppresses cholesterol biosynthesis by inhibiting SREBP-2, a master regulator of de novo cholesterol synthesis, thereby avoiding cholesterol overaccumulation while maintaining PM cholesterol levels (Das et al., 2014; Infante and Radhakrishnan, 2017; Lange and Steck, 1997; Lange et al., 2014; Scheek et al., 1997; Slotte and Bierman, 1988). Artificially trapping the accessible pool of cholesterol in the PM results in dysregulated activation of SREBP-2 (Infante and Radhakrishnan, 2017; Johnson et al., 2019). Despite its critical importance, the intracellular transport machinery that senses the accessibility of PM cholesterol is unknown. This machinery is likely to respond to a sharp change in the accessibility of cholesterol on the cytoplasmic leaflet of the PM and to facilitate transport of accessible cholesterol from the PM to the ER, thereby helping the ER to communicate with the PM. Such a homeostatic system would also allow cells to monitor PM cholesterol accessibility in order to help to maintain cellular cholesterol homeostasis. The ER extends throughout the cytoplasm, forming physical contacts with virtually all other cellular organelles and the PM (Phillips and Voeltz, 2016; Wu et al., 2018). Growing evidence indicates that these membrane contact sites play critical roles in cellular physiology, including lipid exchange and delivery via non-vesicular lipid transport that is facilitated by lipid transfer proteins (LTPs) (Antonny et al., 2018; Drin, 2014; Elbaz and Schuldiner, 2011; Holthuis and Menon, 2014; Jeyasimman and Saheki, 2019; Kumar et al., 2018; Lahiri et al., 2015; Lev, 2012; Luo et al., 2019; Nishimura and Stefan, 2019; Petrungaro and Kornmann, 2019; Saheki et al., 2016; Saheki and De Camilli, 2017a; Saheki and De Camilli, 2017b; Wong et al., 2018). Thus, LTPs may participate in intracellular cholesterol transport and may help to maintain PM cholesterol homeostasis by regulating non-vesicular cholesterol transport between the PM and the ER at ER–PM contact sites. Decades of biochemical and genetic research into cholesterol metabolism has identified several key LTPs that bind to cholesterol and mediate its non-vesicular transport (Luo et al., 2019; Wong et al., 2018). These proteins include a family of 15 proteins that contain a StAR-related lipid transfer (StART) domain, which binds and transports a wide variety of lipids, including cholesterol, glycerolipids, and sphingolipids (Alpy and Tomasetto, 2014). Five members of this family, namely STARD1, STARD3, STARD4, STARD5, and STARD6, bind and transport cholesterol (Alpy et al., 2013; Iaea et al., 2017; Lin et al., 1995; Mesmin et al., 2011; Soccio et al., 2002; Stocco, 2001; Wilhelm et al., 2017), but they are not conserved in yeast. This lack of conservation suggests that there may be a more ancient family of sterol transfer proteins that control cholesterol homeostasis in all eukaryotes. A bioinformatics search for proteins that possess StART-like domains identified a novel family of evolutionarily conserved proteins that includes six Lam/Ltc proteins in budding yeast (Gatta et al., 2015; Murley et al., 2015), and five GRAM domain-containing proteins (GRAMDs) in metazoans. These GRAMDs include the StART-like domain-containing GRAMD1s, also known as Asters (GRAMD1a/Aster-A, GRAMD1b/Aster-B, and GRAMD1c/Aster-C), and two highly related proteins that lack a StART-like domain (GRAMD2 and GRAMD3). Lam/Ltc proteins and GRAMDs all possess an N-terminal GRAM domain, which has structural similarity to the PH domain and thus may sense or bind lipids (Begley et al., 2003; Tong et al., 2018), and a C-terminal transmembrane domain, which anchors the proteins to the ER. Structural and biochemical studies of yeast and mammalian StART-like domains have identified a hydrophobic cavity that can bind sterol (Gatta et al., 2018; Horenkamp et al., 2018; Jentsch et al., 2018; Sandhu et al., 2018; Tong et al., 2018). The StART-like domains of GRAMD1s bind and transport sterols in vitro (Horenkamp et al., 2018; Sandhu et al., 2018). Recent studies have demonstrated that some GRAMDs, including GRAMD1a, GRAMD1b, and GRAMD2, localize to ER–PM contact sites (Besprozvannaya et al., 2018; Sandhu et al., 2018). GRAMD2 facilitates STIM1 recruitment to ER–PM contacts and potentially regulates Ca2+ homeostasis (Besprozvannaya et al., 2018), whereas GRAMD1b facilitates the transport of HDL-derived cholesterol to the ER in the adrenal glands of mice (Sandhu et al., 2018). By contrast, yeast Lam/Ltc proteins sense cellular stress and potentially regulate cholesterol exchange between the ER and other membranes (Gatta et al., 2015; Murley et al., 2015; Murley et al., 2017). However, the role of these proteins in PM cholesterol or sterol homeostasis has been elusive. In this study, we provide evidence that GRAMD1s sense a transient expansion of the accessible pool of PM cholesterol and facilitate its transport to the ER at ER–PM contact sites, thereby contributing to PM cholesterol homeostasis. We found that GRAMDs form homo- and heteromeric complexes via their transmembrane domains and predicted the existence of luminal amphipathic helices that interact with each other within these complexes. We also found that GRAMD1s rapidly move to ER–PM contacts upon acute hydrolysis of sphingomyelin in the PM. We characterized the mechanisms of this acute recruitment and found that the GRAM domain acts as a coincidence detector of unsequestered/accessible cholesterol and anionic lipids in the PM, including phosphatidylserine, allowing the GRAMD1s to sense a transient expansion of the accessible pool of PM cholesterol once it increases above a certain threshold. We generated HeLa cells that lacked GRAMD1a/1b/1c (i.e., all of the GRAMDs that contain a StART-like domain) and determined the effect of knocking out these proteins on cholesterol metabolism, using a combination of cholesterol-sensing probes for live cell imaging and lipidomics of membrane extracts. Upon treatment with sphingomyelinase, which liberates the sphingomyelin-sequestered pool of PM cholesterol into the 'accessible' pool and thus stimulates its PM to ER transport, GRAMD1 triple knockout (TKO) cells exhibited exaggerated accumulation of the accessible pool of PM cholesterol and reduced suppression of SREBP-2 cleavage compared to wild-type control cells. This accumulation resulted from less efficient transport of accessible cholesterol from the PM to the ER. Using structure–function analysis, we demonstrated that GRAMD1s couple their PM-sensing property and cholesterol-transport function via their GRAM and StART-like domains, and that GRAMD1 complex formation ensures the progressive recruitment of GRAMD1 proteins to ER–PM contacts. Finally, we observed striking expansion of the accessible pool of PM cholesterol in GRAMD1 TKO cells at steady state. Drug-induced acute recruitment of GRAMD1b to ER–PM contacts was sufficient to facilitate removal of the expanded pool of accessible cholesterol from the PM in GRAMD1 TKO cells. Collectively, our findings provide evidence for novel cellular mechanisms by which GRAMD1s monitor and help to maintain PM cholesterol homeostasis in mammalian cells. As one of the key homeostatic regulators, GRAMD1s sense a transient expansion of the accessible pool of PM cholesterol and facilitate its transport to the ER, thereby contributing to PM cholesterol homeostasis at ER–PM contact sites. Results GRAMD proteins form homo- and heteromeric complexes Previous studies identified GRAMD1s as ER-resident proteins that are distributed throughout ER structures in a punctate pattern (Sandhu et al., 2018). GRAMDs (namely GRAMD1a, GRAMD1b, GRAMD1c, GRAMD2, and GRAMD3) all possess an N-terminal GRAM domain and a C-terminal transmembrane domain. In addition, the three GRAMD1 proteins (GRAMD1s) possess a StART-like domain (Figure 1A). Some LTPs are known to form homo- and heteromeric complexes. Thus, we reasoned that GRAMD1s may also interact with one another to form complexes. To further analyze the dynamics of these proteins on the ER at high spatial resolution, we tagged the GRAMD1s, as well as GRAMD3, with fluorescent proteins and analyzed their localization using spinning disc confocal microscopy coupled with structured illumination (SDC-SIM). Analysis of COS-7 cells expressing individual EGFP-tagged GRAMD1s or GRAMD3 (EGFP-GRAMD1a, EGFP-GRAMD1b, EGFP-GRAMD1c, or EGFP-GRAMD3) and a general ER marker (RFP-tagged Sec61β) revealed enrichment of GRAMD1s and GRAMD3 in similar discrete patches along ER tubules. By contrast, RFP-Sec61β localized to all domains of the ER, including the nuclear envelope and the peripheral tubular ER network (Hoyer et al., 2018) (Figure 1B and Figure 1—figure supplement 1A). When individual EGFP–GRAMD1s and either mRuby-tagged GRAMD1b (mRuby-GRAMD1b) (Figure 1C) or mCherry-tagged GRAMD3 (mCherry-GRAMD3) (Figure 1—figure supplement 1B) were co-expressed in COS-7 cells, the patches of EGFP and mRuby/mCherry significantly overlapped, indicating potential complex formation between these proteins on tubular ER. Figure 1 with 1 supplement see all Download asset Open asset GRAMD proteins form homo- and heteromeric complexes. (A) Domain structure of GRAMD proteins in comparison to yeast Lam6/Ltc1. (B) Confocal images of live COS-7 cells expressing the ER membrane marker RFP-Sec61β and EGFP–GRAMD protein constructs as indicated. Insets show at higher magnification the regions indicated by white dashed boxes. Note the presence of the patches of EGFP–GRAMDs throughout the tubular ER. Scale bars, 10 µm. (C) Confocal images of live COS-7 cells expressing mRuby-GRAMD1b and EGFP–GRAMD1s as indicated. Note the presence of mRuby–GRAMD1b patches that partially overlap with the patches of EGFP–GRAMD1s. Scale bars, 1 µm. (D, E) Extracts of HeLa cells transfected with the indicated constructs were subjected to anti-GFP immunoprecipitation (IP) and then processed for SDS-PAGE and immunoblotting (IB) with anti-GFP and anti-Myc antibodies. Inputs are 1% of the total cell lysates. Note the strong biochemical interaction between GRAMD1b and GRAMD1s (D) and between GRAMD3 and GRAMD1s (E). Immunoprecipitated EGFP-GRAMD1s, Myc-GRAMD1b and Myc-GRAMD3 are indicated by arrows. To test whether these proteins form complexes, we examined biochemical interactions between GRAMD1s and GRAMD3 using co-immunoprecipitation assays. HeLa cells co-transfected with individual EGFP–GRAMD1s together with either myc-tagged GRAMD1b (Myc–GRAMD1b) (Figure 1D and Figure 1—figure supplement 1C) or myc-tagged GRAMD3 (Myc–GRAMD3) (Figure 1E and Figure 1—figure supplement 1D) were lysed, and either anti-GFP (Figure 1D,E) or anti-Myc nanobodies (Figure 1—figure supplement 1C,D) were used to perform immunoprecipitation. Analysis of the resulting immunoprecipitates by western blotting (i.e. immunoblotting) revealed robust interaction between GRAMD1s and GRAMD1b (Figure 1D and Figure 1—figure supplement 1C), as well as between GRAMD1s and GRAMD3 (Figure 1E and Figure 1—figure supplement 1D). These results demonstrate that these proteins form both homo- and heteromeric complexes. Luminal helices and transmembrane domains of GRAMD proteins are important for their complex formation The formation of homo- and heteromeric complexes between GRAMD1s and GRAMD3 suggested the presence of amino-acid sequence within these proteins that facilitate their interaction. Secondary structure predictions indicated the presence of a conserved alpha helix within the luminal region of GRAMD1s (Figure 2A). Furthermore, helical wheel analysis of the luminal helix from GRAMD1b predicted that this protein contained an amphipathic helix, with charged and hydrophobic amino acids occupying opposite sides of the helix (Figure 2B and Figure 2—figure supplement 1A,B). It is known that some amphipathic helices mediate protein–protein interactions through their hydrophobic surfaces (Segrest et al., 1990). Therefore, we first asked whether the luminal helix was necessary for these proteins to form discrete patches on tubular ER. We focused on GRAMD1b as a model protein for analysis of the properties of the GRAMD1 luminal helices, generating a version of GRAMD1b that lacked the luminal helix (Δhelix), and a second version in which the five hydrophobic residues within the luminal helix were mutated to glutamic acid (5E), thereby disrupting the hydrophobic surface (Figure 2B and Figure 2—figure supplement 1C). Whereas GRAMD1b (wild-type control) formed patches on tubular ER, both GRAMD1b (Δhelix) and GRAMD1b (5E) exhibited diffuse localization patterns, with fewer discrete patches on tubular ER (Figure 2C). By contrast, a version of GRAMD1b in which the four hydrophobic residues preceding the luminal helix were mutated to glutamic acid (4E) formed patches that were similar to those formed by the control (Figure 2C), demonstrating that the 5E mutation specifically disrupted patch formation. Figure 2 with 1 supplement see all Download asset Open asset Luminal helix and transmembrane domain of GRAMD1b are important for homo- and heteromeric interaction. (A) Sequence alignment of the luminal region of GRAMD1s. This region is predicted by Phyre2 to contain an amphipathic helix (Kelley et al., 2015) as indicated. Blue and red asterisks mark hydrophobic amino acid residues that are partially conserved in GRAMD1s. The shared identities of the amino acid sequences of the amphipathic helices predicted by BLAST analysis were: 75% (GRAMD1a vs. GRAMD1b); 75% (GRAMD1a vs. GRAMD1c); and 80% (GRAMD1b vs. GRAMD1c). The effects of the mutations of these residues to glutamic acid (4E in the case of blue marks; 5E in the case of red marks) were tested in GRAMD1b. Black, red, blue, and pink/purple colors denote hydrophobic, acidic, basic, and hydrophilic amino acid residues, respectively. (B) Predicted luminal amphipathic helix region of wild-type GRAMD1b (left panel) and that with L693E, W696E, I699E, I700E and L707E (5E) mutations (right panel) are shown as helical wheel representations. Predictions were made with the Heliquest server (Gautier et al., 2008). (C) Confocal images of live COS-7 cells expressing RFP–Sec61β and EGFP fusions of various GRAMD1b constructs [control, wild-type GRAMD1b; Δhelix, GRAMD1b lacking the predicted luminal amphipathic helix; 4E, GRAMD1b with 4E mutations in the luminal region (W678E, L681E, L682E, Y688E); 5E, GRAMD1b with 5E mutations in the predicted luminal amphipathic helix]. Note the reduced formation of GRAMD1b patches in Δhelix and 5E mutants but not in the 4E mutant. Scale bars, 2 µm. (D) Overlay of the size exclusion chromatography (SEC) profiles of the recombinant EGFP-tagged luminal helix region of wild-type GRAMD1b (EGFP–helix) and EGFP–helix with the 5E mutations [EGFP–helix (5E)]. Note the difference in elution volumes, indicating the formation of complexes mediated by the wild-type luminal helix. (E) Blue native (BN)-PAGE analysis (left panel) and SDS-PAGE analysis (right panel) of SEC-purified EGFP–helix and EGFP–helix (5E). Black and red arrows indicate the major bands for EGFP–helix and EGFP–helix 5E, respectively. Note the difference in their migration pattern in BN-PAGE. CB, Colloidal blue staining. (F) Extracts of HeLa cells transfected with the constructs as indicated were subjected to anti-GFP immunoprecipitation (IP) and then processed for SDS-PAGE and immunoblotting (IB) with anti-GFP and anti-Myc antibodies. Inputs are 5% of the total cell lysates. Note that the interaction of GRAMD1b or GRAMD1a is much reduced in GRAMD1b Δhelix or 5E mutants and abolished in the GRAMD1b (TM swap) mutant (GRAMD1b with its transmembrane domain and luminal region replaced with those of Sec61β) when compared to the levels of interactions seen in cells with wild-type GRAMD1b. This reduction is smaller in the GRAMD1b 4E mutant. (G) Quantification of the co-immunoprecipitation experiments shown in (F). The ratio of the band intensity of the co-immunoprecipitated Myc–GRAMD1b (left) or Myc–GRAMD1a (right) over that of the indicated immunoprecipitated EGFP-tagged proteins were calculated. The values were then normalized by the ratio of the band intensity of Myc–GRAMD1b over that of EGFP–GRAMD1b (WT) (left) or by the ratio of the band intensity of Myc–GRAMD1a over that of EGFP–GRAMD1b (WT) (right) [mean ± SEM, n = 3 IPs for each sample]. (H) Confocal images of a live COS-7 cell expressing RFP-Sec61β and EGFP-tagged GRAMD1b (TM swap). Scale bars, 2 µm. (I) Confocal images of live COS-7 cells expressing mRuby–GRAMD1b and EGFP fusions of GRAMD1b constructs [Control, wild-type GRAMD1b; TM swap, GRAMD1b (TM swap)]. Note the abolished formation of GRAMD1b patches in TM swap mutants. Scale bars, 2 µm. (J) Model of the homo- and heteromeric interactions of GRAMD1a/b. Their complex formation is facilitated primarily by their luminal amphipathic helices and additionally mediated by their transmembrane domains. These regions are important for the ability of GRAMD1s to form complexes and patches on the tubular ER network. Figure 2—source data 1 Dataset for Figure 2. https://cdn.elifesciences.org/articles/51401/elife-51401-fig2-data1-v2.xlsx Download elife-51401-fig2-data1-v2.xlsx The potential ability of the luminal helices to interact directly with one another was examined using cell-free assays. Wild-type luminal helices (GRAMD1b674–718) and luminal helices with the 5E mutation (GRAMD1b674–718 5E) were purified individually as EGFP fusion proteins and analyzed by size exclusion chromatography (SEC). Whereas the predicted molecular weights of the fusion proteins were the same (~35 kDa), wild-type luminal helices (EGFP–helix: EGFP–GRAMD1b674–718) eluted at a much lower elution volume compared to 5E mutant luminal helices [EGFP–helix (5E): EGFP–GRAMD1b674–718 5E] (Figure 2D). Blue native PAGE analysis (BN-PAGE) of the purified proteins revealed that wild-type helices migrated slower than the 5E mutants, indicating that interaction between luminal helices depended on the hydrophobic surface of GRAMD1b (Figure 2E). By contrast, in the presence of SDS, the denatured forms of these proteins migrated similarly (SDS-PAGE). Slightly slower migration of 5E mutants on the gel was possibly due to the increased hydrophilicity of this fragment compared to wild-type (Guan et al., 2015) (Figure 2E). These results suggest that the luminal helix is probably amphipathic and is important for the formation of GRAMD1b complexes through its hydrophobic surface. Finally, the formation of GRAMD1 complexes was examined biochemically in cells using co-immunoprecipitation assays. Homomeric interactions between GRAMD1bs and heteromeric interactions between GRAMD1b and GRAMD1a were greatly reduced when the luminal helix of GRAMD1b was either removed (Δhelix) or mutated to the 5E version, supporting the important role of the luminal helix in homo- and heteromeric interactions of the GRAMD1s (Figure 2F,G). Residual interactions were mediated by the transmembrane domain of GRAMD1b, as replacing this domain and its luminal region with those from Sec61β (TM swap) (Figure 2J) completely abolished the ability of GRAMD1b to form homo- and heteromeric complexes (Figure 2F,G). Accordingly, GRAMD1b with the TM swap exhibited a diffuse localization pattern compared to that of wild-type GRAMD1b (Figure 2H), and failed to interact with wild-type GRAMD1b on tubular ER (Figure 2I). Thus, both transmembrane domains and luminal helices contributed to the formation of GRAMD1 complexes (Figure 2J). Taken together, these results revealed the biochemical mechanisms by which GRAMDs form homo- and heteromeric complexes. As key residues contributing to the hydrophobic surface of the luminal helix are conserved among GRAMD1s (Figure 2A and Figure 2—figure supplement 1A), they probably play a role in the heteromeric interactions of all of these proteins. The GRAM domain of GRAMD1s acts as a coincidence detector of unsequestered/accessible cholesterol and anionic lipids, and senses the accessibility of cholesterol Recent studies demonstrated that 'cholesterol loading' leads to the accumulation of GRAMD1s at ER–PM contact sites (Sandhu et al., 2018). Within 20 min of treating cells with a complex of cholesterol and methyl-β-cyclodextrin (cholesterol/MCD), GRAMD1b was indeed recruited to the PM (Figure 3A,B; Video 1). In addition, we found that GRAMD1a, GRAMD1c, and GRAMD3 were all recruited to ER–PM contacts upon cholesterol loading, with kinetics similar to GRAMD1b recruitment (Figure 3B). However, a version of GRAMD1b that lacked the GRAM domain (GRAMD1b ΔGRAM) failed to localize to the PM, even after 30 min, indicating the essential role of this domain in sensing PM cholesterol (Figure 3—figure supplement 1A; Video 2). Although these results suggest that PM cholesterol plays a critical role in recruiting GRAMDs to ER–PM contacts, all of the GRAMDs localize to tubular ER at rest, even though a significant amount of cholesterol is already present in the PM (Lange et al., 1989; Ray et al., 1969). Thus, their GRAM domains may possess unique abilities to sense the accessibility of PM cholesterol, rather than detecting the total levels of PM cholesterol. However, it is not known whether the GRAM domains are able to sense accessible cholesterol in the PM. Figure 3 with 2 supplements see all Download asset Open asset The GRAM domain of GRAMD1s acts as a coincidence detector of unsequestered/accessible cholesterol and anionic lipids, and senses a transient expansion of the accessible pool of cholesterol in the PM. (A) Confocal images of live HeLa cells expressing EGFP–GRAMD1b with or without cholesterol loading [the treatment with cholesterol/MCD complex (200 µM) for 30 min at 37°C]. Note the extensive recruitment of GRAMD1b to the PM upon cholesterol loading. Scale bars, 10 µm. (B) Time course of normalized EGFP signal, as assessed by total internal reflection fluorescence (TIRF) microscopy, from HeLa cells expressing EGFP–GRAMD protein constructs as indicated. Cholesterol loading [the treatment with cholesterol/MCD complex (200 µM)] is indicated. [mean ± SEM, n = 24 cells (EGFP–GRAMD1a), n = 29 cells (EGFP–GRAMD1b), n = 25 cells (EGFP–GRAMD1c), n = 28 cells (EGFP–GRAMD3); data are pooled from one experiment for GRAMD1a and two experiments for GRAMD1b, GRAMD1c and GRAMD3.] (C–F). Liposome sedimentation assays of the GRAM domain of GRAMD1b (GRAM1b) and GRAMD1a (GRAM1a). Liposomes containing the indicated mole% lipids were incubated with purified GRAM1b proteins (C, E) or purified GRAM1a proteins (D, F). Bound proteins [pellet, (P)] were separated from the unbound proteins [supernatant, (S)], run on SDS-PAGE and visualized by colloidal blue staining (mean ± SEM, n = 3 independent experiments for all the conditions). DOPC, phosphatidylcholine (1,2-dioleoyl-sn-glycero-3-phosphocholine); DOPS, phosphatidylserine (1,2-dioleoyl-sn-glycero-3-phospho-L-serine); Chol, cholesterol; SM, sphingomyelin (N-oleoyl-D-erythro-sphingosylphosphorylcholine). (G) Left: time course of normalized EGFP signal in response to sphingomyelinase (SMase), as assessed by TIRF microscopy of HeLa cells expressing EGFP–GRAMD1b or EGFP–GRAMD1b ΔGRAM. The treatment with SMase (100 mU/ml) is indicated. Right: values of ΔF/F0 corresponding to the end of the experiment as indicated by the arrow [mean ± SEM, n = 72 cells (EGFP–GRAMD1b), n = 64 cells (EGFP–GRAMD1b ΔGR
Abstract Branched ubiquitin (Ub) chains constitute a sizable fraction of Ub polymers in human cells. Despite their abundance, our understanding of branched Ub function in cell signaling has been stunted by the absence of accessible methods and tools. Here we identify cellular branched-chain-specific binding proteins and devise approaches to probe K48–K63-branched Ub function. We establish a method to monitor cleavage of linkages within complex Ub chains and unveil ATXN3 and MINDY as debranching enzymes. We engineer a K48–K63 branch-specific nanobody and reveal the molecular basis of its specificity in crystal structures of nanobody-branched Ub chain complexes. Using this nanobody, we detect increased K48–K63-Ub branching following valosin-containing protein (VCP)/p97 inhibition and after DNA damage. Together with our discovery that multiple VCP/p97-associated proteins bind to or debranch K48–K63-linked Ub, these results suggest a function for K48–K63-branched chains in VCP/p97-related processes.
Misfolded proteins in the endoplasmic reticulum (ER) are degraded by ER-associated degradation (ERAD). Although ERAD components involved in degradation of luminal substrates are well characterized, much less is known about quality control of membrane proteins. Here, we analyzed the degradation pathways of two short-lived ER membrane model proteins in mammalian cells. Using a CRISPR-Cas9 genome-wide library screen, we identified an ERAD branch required for quality control of a subset of membrane proteins. Using biochemical and mass spectrometry approaches, we showed that this ERAD branch is defined by an ER membrane complex consisting of the ubiquitin ligase RNF185, the ubiquitin-like domain containing proteins TMUB1/2 and TMEM259/Membralin, a poorly characterized protein. This complex cooperates with cytosolic ubiquitin ligase UBE3C and p97 ATPase in degrading their membrane substrates. Our data reveal that ERAD branches have remarkable specificity for their membrane substrates, suggesting that multiple, perhaps combinatorial, determinants are involved in substrate selection.
Endoplasmic-reticulum-associated protein degradation
Cholesterol is a major structural component of the plasma membrane (PM). The majority of PM cholesterol forms complexes with other PM lipids, making it inaccessible for intracellular transport. Transition of PM cholesterol between accessible and inaccessible pools maintains cellular homeostasis, but how cells monitor the accessibility of PM cholesterol remains unclear. We show that endoplasmic reticulum (ER)-anchored lipid transfer proteins, the GRAMD1s, sense and transport accessible PM cholesterol to the ER. GRAMD1s bind to one another and populate ER-PM contacts by sensing a transient expansion of the accessible pool of PM cholesterol via their GRAM domains. They then facilitate the transport of this cholesterol via their StART-like domains. Cells that lack all three GRAMD1s exhibit striking expansion of the accessible pool of PM cholesterol as a result of less efficient PM to ER transport of accessible cholesterol. Thus, GRAMD1s facilitate the movement of accessible PM cholesterol to the ER in order to counteract an acute increase of PM cholesterol, thereby activating non-vesicular cholesterol transport.
Abstract Nuclear architecture and functions depend on dynamic interactions between nuclear components (such as chromatin) and inner nuclear membrane (INM) proteins. Mutations in INM proteins interfering with these interactions result in disease. However, mechanisms controlling the levels and turnover of INM proteins remain unknown. Here, we describe a mechanism of regulated degradation of the INM SUN domain-containing protein 2 (SUN2). We show that Casein Kinase II and the C-terminal domain Nuclear Envelope Phosphatase 1 (CTDNEP1) have opposing effects on SUN2 levels by regulating SUN2 binding to the ubiquitin ligase Skp/Cullin1/F-Box βTrCP (SCF βTrCP ). Upon binding to phosphorylated SUN2, SCF βTrCP promotes its ubiquitination. Ubiquitinated SUN2 is membrane extracted by the AAA ATPase p97 and delivered to the proteasome for degradation. Importantly, accumulation of non-degradable SUN2 results in aberrant nuclear architecture, vulnerability to DNA damage and increased lagging chromosomes in mitosis. These findings uncover a central role of proteolysis in INM protein homeostasis.
(Molecular Cell 79, 768–781.e1–e7; September 3, 2020) The authors discovered two labeling errors in Figures 3C and 4A of this article. In Figure 3C, the symbols in the key were incorrect. In Figure 4A, the X and Y axes were swapped. Corrected figures appear below and with the article online. The authors apologize for the errors.Figure 3CCYP51A1TM Ubiquitination and Degradation Are Dependent on RNF185 and Membralin (original)View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4ARNF185 and Membralin Form a Novel ERAD Complex (corrected)View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4ARNF185 and Membralin Form a Novel ERAD Complex (original)View Large Image Figure ViewerDownload Hi-res image Download (PPT) Quality Control of ER Membrane Proteins by the RNF185/Membralin Ubiquitin Ligase Complexvan de Weijer et al.Molecular CellJuly 31, 2020In BriefMembrane proteins are diverse, and the components involved in their quality control in each case are unclear. By comparing the degradation of two short-lived ER proteins, van de Weijer et al. identified an ERAD complex composed of RNF185, TMUBs, and Membralin, a membrane protein essential to neuronal function. Full-Text PDF Open Access
Endoplasmic-reticulum-associated protein degradation
Misfolded, potentially toxic proteins in the lumen and membrane of the endoplasmic reticulum (ER) are eliminated by proteasomes in the cytosol through ER-associated degradation (ERAD). The ERAD process involves the recognition of substrates in the lumen and membrane of the ER, their translocation into the cytosol, ubiquitination, and delivery to the proteasome for degradation. These ERAD steps are performed by membrane-embedded ubiquitin-ligase complexes of different specificity that together cover a wide range of substrates. Besides misfolded proteins, ERAD further contributes to quality control by targeting unassembled and mislocalized proteins. ERAD also targets a restricted set of folded proteins to influence critical ER functions such as sterol biosynthesis, calcium homeostasis, or ER contacts with other organelles. This review describes the ubiquitin-ligase complexes and the principles guiding protein degradation by ERAD.
Endoplasmic-reticulum-associated protein degradation