* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Shaping the Endoplasmic Reticulum into a Social Network
Survey
Document related concepts
Cell encapsulation wikipedia , lookup
Lipid bilayer wikipedia , lookup
Protein phosphorylation wikipedia , lookup
G protein–coupled receptor wikipedia , lookup
Theories of general anaesthetic action wikipedia , lookup
Extracellular matrix wikipedia , lookup
Model lipid bilayer wikipedia , lookup
Cell nucleus wikipedia , lookup
Protein moonlighting wikipedia , lookup
Magnesium transporter wikipedia , lookup
Intrinsically disordered proteins wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
SNARE (protein) wikipedia , lookup
Cytokinesis wikipedia , lookup
Signal transduction wikipedia , lookup
Cell membrane wikipedia , lookup
Western blot wikipedia , lookup
Transcript
TICB 1254 No. of Pages 10 Review Shaping the Endoplasmic Reticulum into a Social Network Hong Zhang1,* and Junjie Hu1,* In eukaryotic cells, the endoplasmic reticulum (ER) is constructed as a network of tubules and sheets that exist in one continuous membrane system. Several classes of integral membrane protein have been shown to shape ER membranes. Functional studies using mutant proteins have begun to reveal the significance of ER morphology and membrane dynamics. In this review, we discuss the common protein modules and mechanisms that generate the characteristic shape of the ER. We also describe the cellular functions closely related to ER morphology, particularly contacts with other membrane systems, and their potential roles in the development of multicellular organisms. Unique Morphology and Diverse Roles of the ER: A Perfect Model for Connecting Membrane Dynamics to Organellar Functions The ER is a single membrane-bound organelle involved in many critical cellular processes, including protein synthesis, lipid synthesis, and calcium storage. Some organelles adopt globular shapes, but the ER consists of interconnected membrane tubules and sheets [1,2]. From the center of a cell, the ER starts as the outer nuclear membrane (ONM), which is the outer layer of the nuclear envelope (NE). Most of the ER sheets, which are cisternal structures with two closely apposed membranes, appear in the perinuclear region. ER tubules, which typically form a reticular network, exist in both the perinuclear and peripheral regions. The distribution of ER sheets and tubules is tightly regulated in the cell. ER morphology can vary substantially in different cell types or in response to different growth cues and conditions. Some cells exhibit special ER arrangements. In yeast and plant cells, a large portion of the ER gathers in the cortex; a region immediately beneath the plasma membrane [3]. The cortical ER usually contains a tubular network with interspersed patches of sheets. Muscle cells contain a specialized ER, termed the sarcoplasmic reticulum (SR), which runs through the myofibril matrix as tubules but merges into cisternal structures when engaging the plasma membrane [4]. In addition, the ER of neuronal axons or plant root hairs is predominantly tubular, but professional secretory cells, such as b-pancreatic cells or plasma cells, turn their ER into a massive stack of sheets [5]. Thus, ER morphology is clearly linked to function. How the characteristic shape of the ER is generated is a fundamental question in cell biology. Our understanding of ER shaping began with the identification of tubule-forming proteins, namely the reticulons (Rtns) and DP1/Yop1 [6] (Figure 1). The ER-bound dynamin-like GTPases atlastin (ATL) and Sey1p/RHD3 were [7_TD$IF]then found to fuse ER membranes [7,8]; a key process in tubular network formation. The tubule-forming proteins are also implicated in the generation of ER sheets, along with another set of ER-resident proteins that include Climp-63, kinectin, and p180 [9]. Additional candidates for regulating ER morphology are the Lunapark (Lnp) family [10–12], protrudin [13,14], Rab10 [15], and Rab18 [16]. Although recent research has provided some mechanistic insight Trends in Cell Biology, Month Year, Vol. xx, No. yy Trends ER sheets and tubules are generated by membrane curvature stabilization, and tubules are fused into a reticular network by membrane-bound GTPases. Mechanistic studies reveal that transmembrane hairpins, amphipathic helices and coiled coils are important tools for ER morphogenesis. The ER contacts other organelles at discrete sites, where lipid transfer and calcium exchange occur. Various aspects of organelle dynamics, including fission, maturation and positioning, are also regulated by formation of contacts with the ER. The morphology of the ER controls the shape and area of the contact with other organelles, and thus the function of contact sites. Formation and maintenance of ER morphology are of general importance. When ER morphology is perturbed, organelle contacts, membrane trafficking and other ER functions are affected, and neurodegenerative diseases may result. 1 National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China *Correspondence: [email protected] (H. Zhang) and [email protected] (J. Hu). http://dx.doi.org/10.1016/j.tcb.2016.06.002 © 2016 Elsevier Ltd. All rights reserved. 1 TICB 1254 No. of Pages 10 Rtns Protrudin ATL Sey1p/RHD3 Kinecn Glossary Key: RHD Rtns Reeps ER tubule Transmembrane hairpin (TMH) ER sheet Climp-63 RHD Amphipathic helix (APH) Coiled coil (CC) Reeps Lnp p180 Figure 1. Determinants of ER Morphology. Endoplasmic-reticulum-shaping proteins are shown in either tubules or sheets. Common protein modules, including transmembrane hairpins (TMH), amphipathic helices (APH), and coiled coils (CC), are highlighted. Variable regions within protein families are shown by dotted lines. RHD, reticulon-homology domain, formed by two TMHs with a connecting loop. into these proteins, how the cell achieves the complex and dynamic morphology of the ER remains a mystery. Another important question is how the ER takes advantage of its morphological features. The flattened surface of sheets is speculated to accommodate ribosome/polysome translation better than tubules; thus, sheets are mainly involved in protein synthesis [9]. [8_TD$IF]By contrast, the curved surface of tubules would be ideal for generating bent membranes, such as vesicles exiting the ER [17]. Genetic alteration of regulators of ER morphology allows us to better understand the correlation between ER shape and physiological function. Mechanisms of ER Shaping and Remodeling Formation of ER Tubules ER tubules are cylindrical structures with a diameter of 30 nm in yeast and 50 nm in mammals. Rtns were identified as ER tubule-forming proteins in an in vitro ER network formation assay using Xenopus membrane extracts [6]. Subsequent analysis uncovered a similar protein named REEP5/DP1 in mammals and Yop1p in yeast. Overexpression of these proteins results in more ER tubules, and deletion or depletion has the opposite effect. The role of Yop1p or Rtn1p in tubule formation was confirmed when purified reconstituted protein generated membrane tubules in vitro [18]. The Rtns (four genes in mammals) and REEPs (six genes in mammals) contain a reticulonhomology domain (RHD), which is composed of two tandem hydrophobic segments. The length of each hydrophobic segment (30–35 amino acids) allows [9_TD$IF]it to be embedded in a membrane [10_TD$IF]most likely as a wedge-shaped hairpin. This configuration occupies more space in the outer leaflet of a membrane bilayer than in the inner leaflet, curving the membrane. Searching mammalian proteins based on structural conservation of the RHD predicted that ADP-ribosylation factor-like 6 interacting protein 1 (Arl6IP1) and family with sequence similarity 134, member B (FAM134B) contain similar tandem transmembrane hairpins (TMHs, see Glossary) [19]. Like Yop1p, purified Arl6IP1 induces membrane tubule formation when reconstituted with lipids [19]. FAM134B has been proposed to regulate ER-phagy in addition to having membrane shaping activity [20]. Thus, ER tubules can be formed and stabilized by a large set of proteins that share a common domain. In addition to inducing curvature, TMHs target proteins to ER tubules. Some TMHs traverse the bilayer completely, but they are usually connected by just a few residues. These less asymmetric hairpins do not actively bend membranes, but rather sense and prefer the presence of tubulelike membrane environments. TMH-containing proteins that localize to ER tubules include the 2 Trends in Cell Biology, Month Year, Vol. xx, No. yy Amphipathic helix (APH): an /-helix with hydrophobic residues on one side and hydrophilic residues on the other side. These helices can insert shallowly into membranes, in some cases inducing curvature, and in other cases sensing it. Coiled coil (CC): a structural motif in proteins where /-helices are intertwined. Hydrophobic residues are usually packed in the core of the coil, and repeated sequences are common. CCs are often involved in oligomerization and tethering. ER-mitochondria encounter structure (ERMES): a complex involved in ER–mitochondria contact formation in yeast, comprising the integral ER membrane protein Mmm1, the cytosolic protein Mdm12 and two mitochondrial outer-membrane proteins (Mdm10 and Mdm34). Extended synaptotagmins (E-Syts): integral ER membrane proteins involved in ER–PM tethering. They contain a cytosolic synaptotagmin-like mitochondrial-lipid binding protein (SMP) domain followed by multiple C2 domains. The orthologous proteins in yeast are the Tricalbins. Oxysterol-binding protein (OSBP)related domain (ORD): a domain found in OSBP and the closely related OSBP-related proteins (ORPs) (Osh proteins in yeast) that sense sterols and/or transfer sterols at contact sites. ORD binds cholesterol and oxysterols. Transmembrane hairpin (TMH): connected transmembrane /-helices that do not traverse membranes completely, or have a very short linker. TMHs are often found in curved membranes, such as ER tubules. They help to induce membrane curvature or target integral membrane proteins to highly curved membranes. The reticulon-homology domain is formed by tandem TMHs. VAMP-associated proteins (VAPs): integral ER proteins that interact with tethering factors and/or effectors via the FFAT motif (two phenylalanines in an acidic trait) for formation of membrane contacts. VAPs contain a major sperm protein (MSP) domain that binds to PI(4)P and PS. Mammalian cells contain VAP-A and VAP-B. Their yeast orthologs are Scs2 and Scs22. VAMP, vesicle-associated membrane protein. TICB 1254 No. of Pages 10 ER fusogens ATL and Sey1p/RHD3, the microtubule-severing protein spastin, and the ER morphology regulators Lnp and protrudin[1_TD$IF] (Figure 1). Two mechanisms have been proposed for curvature generation in ER tubules by tubule-forming proteins: the insertion of RHD wedges and scaffold formation via protein oligomerization [2]. Although these proteins clearly form homo- and hetero-oligomers [21], the molecular architecture of individual RHDs and precise mechanisms underlying oligomer assembly remain to be investigated. Recent studies of Yop1p revealed that an amphipathic helix (APH) C-terminal to its RHD is essential for tubule formation [22]. The helix is protected by lipids[2_TD$IF] but not detergent[3_TD$IF] upon trypsin digestion[12_TD$IF] [18], and may insert into the membrane as an additional wedge. Sequence analysis showed that the APH is conserved among most ER tubule-forming proteins. In some family members[13_TD$IF], an APH is also the predicted N-terminal of the RHD. These RHDflanking elements provide additional stabilization. ER tubule formation does not solely rely on Rtns and Rtn-like proteins. In mammalian cells, tubules are constantly pulled out of the plane of ER membranes; an effect related to microtubule sliding or tip binding [23]. The small GTPase Rab10 marks the growing tip of ER tubules [15]. Overexpression of a dominant-negative form or depletion of Rab10 causes ER sheet expansion. When a microtubule-based force generates a new ER tubule, RHD-containing proteins are thought to move in to stabilize it. Precisely how Rab10 regulates this process remains to be determined. Formation of ER Sheets ER sheets are cisternal structures with a constant distance between two apposed membranes. Three mechanisms have been proposed for sheet formation [2] (Figure 1). First, membrane at the edge of ER sheets is highly bent, with a curvature similar to a tubule cross-section. Thus, the curvature of sheets and tubules is likely generated in the same way, possibly by the same proteins. Tubule-forming proteins outline patches of ER sheets in cells, confirming their dual roles in ER morphogenesis [9]. Second, the thickness of an ER cisterna is usually fixed, suggesting the existence of a luminal spacer. Integral membrane protein Climp-63, which localizes exclusively in sheets, is thought to use its luminal coiled coil (CC) domain to bridge the two apposed membranes [24]. Consistent with its role in sheet formation, overexpression of Climp-63 increases the number of ER sheets in cells [9]. Third, the surface of ER sheets is kept flattened, likely by the sheet-enriched integral membrane proteins kinectin and p180 [9]. These proteins possess a cytosolic CC domain, which may form rod-like scaffolds. The NE, another sheet-like structure in the ER, may use the same mechanisms to generate and maintain its shape. The ONM and inner nuclear membrane (INM) are connected at nuclear pores. As the thickness of the NE is 26 nm, which is nearly half that of ER sheets, the curvature at the nuclear pore is higher than at the sheet edge. The nuclear pore complex (NPC) likely serves as a scaffold to stabilize this curvature. In addition, a class of SUN proteins in the INM and KASH proteins in the ONM interact similar to Climp-63 [25]. Notably, some of the KASH proteins contain large CC repeats in the cytosolic region [26], analogous to those of kinectin and p180, suggesting that KASH proteins not only control luminal spacing, but may also smooth the ONM. Depletion of several ER-localized proteins, including Rab10, Rab18, protrudin, and Lnp, has been reported to expand ER sheets [10,13,15,16]. If these proteins participate directly in ER shaping, they either promote tubule formation or negatively regulate ER sheets. Rab10 appears to promote tubule formation, but how other proteins influence ER morphology is unknown. Experimentally, we need to determine: (i) whether the effect is direct, as processes such as the activation of ER stress also result in ER expansion; (ii) whether the change in morphology is due to the augmentation of new sheets or redistribution of existing sheets; and (iii) whether the change correlates with increased levels of sheet-forming proteins, such as Climp-63. Trends in Cell Biology, Month Year, Vol. xx, No. yy 3 TICB 1254 No. of Pages 10 Fusion of ER Membranes Homotypic fusion, the merging of identical membranes, occurs frequently in the ER, forming a tubular network and maintaining the continuity of ER membranes. ATLs act as ER fusogens [7,8] and yeast and plants use the functional orthologs of ATLs, synthetic enhancer of yop1p (Sey1p) and root hair defective 3 (RHD3), respectively. Purified Drosophila ATL, Sey1p, and RHD3 have membrane fusion activity in vitro, directly supporting their role in ER fusion [8,27,28]. Alteration of ATL or Sey1p/RHD3 profoundly impacts ER morphology. In yeast cells, deletion of Sey1p and Yop1p results in a conversion of the tubular network into sheets and large areas of the cortex devoid of ER [7,27]. Mutation or deletion of RHD3 leads to cable-like and less-mobile ER tubules [29]. Depletion of ATLs generates unbranched ER [7] or, in the case of Drosophila ATL, fragmented ER [8]. These morphological defects are attributed to a lack of connections between ER tubules. Structural studies of human ATL1 and Candida albicans Sey1p have provided significant insight into the mechanism by which these GTPases fuse membranes [30–33]. ATL and Sey1p/RHD3 contain an N-terminal GTPase and a helical bundle domain, followed by a TMH and C-terminal tail (CT). GTP binding-induced dimerization of the GTPase promotes membrane tethering, and GTP hydrolysis induces conformational changes in the helical bundle, forcing the membranes to merge. In addition, the TMH mediates oligomerization and has sequence-specific functions and the CT forms an APH, which destabilizes the lipid bilayer during fusion [34]. ER fusion is tightly regulated [35]. Not all tubule meetings proceed to fusion, but drastic loss of fusion activity causes severe morphological and functional defects. Although regulators of ATL and Sey1p/RHD3 are yet to be discovered, some insights have come from the intrinsic properties of these GTPases. Recent studies have revealed that tethered membranes continuously consume GTP but do not necessarily fuse, and GTPases form dimers and hydrolyze GTP in the same membranes [35]. These apparently futile efforts may be used to self-regulate fusion dynamics. ER morphology in Membrane Contact Sites To maintain cellular homeostasis, organelles are connected by vesicle-mediated trafficking and membrane contact sites (MCSs) at which two heterologous membranes are closely apposed (typically within 30 nm) but do not fuse. The ER forms MCSs with multiple membrane systems, including the plasma membrane (PM), mitochondria, Golgi, endosomes, and lipid droplets (LDs) (Figure 2). Association of the ER with other organelles generally involves protein–protein interactions and/or protein–phospholipid interactions. The areas and shapes of such contacts are dynamic and may correlate with the functional demands of the contact, which include lipid and calcium exchange, fission, and organelle movement. ER–PM Contact Sites The cortical ER, which is tightly associated with the PM, consists of tubules and highly fenestrated sheets. In yeast, the cortical ER normally covers 20–40% of the PM [3], and can be separated from the PM only when six tethering proteins, including Ist2, three tricalbins [Tcb1-3, orthologs of extended synaptotagmin (E-Syt)], Scs2, and Scs22 [orthologs of VAMP-associated proteins (VAPs)[4_TD$IF]], are simultaneously deleted. Several pairs of interactions have been identified for ER–PM contacts. The ER-resident calcium sensor stromal-interacting molecule 1 (STIM1) binds to the PM-localized Ca2+[6_TD$IF] channel Orai1 to replenish Ca2+ levels in the cytosol when Ca2+stores in the ER are depleted [36]. Binding of STIM1 with PI(4,5)P2 on the PM enhances contact formation [36]. ER–PM tethering can also be mediated by numerous lipid transfer proteins. ER-localized E-Syts engage PI(4,5)P2 on the PM 4 Trends in Cell Biology, Month Year, Vol. xx, No. yy TICB 1254 No. of Pages 10 ER-PM contacts PM STIM1 Orai1 VAP-A/VAP-B Scs2/Scs22 NIR2 E-syts/ Tcbs PI(4,5)P2 ORP5 ORP8/ Osh PI(4)P PA SR Golgi ER-Golgi contacts VAP LD OSBP PI(4)P Sheet ER ER-LD contacts Mitochondria Seipin/Fld1 Tu Endosome ER-endosome contacts ORP5 NPC1 VAP ORP1L Mmm1 STARD3 Rab7 le ER-mitochondrial contacts (ERMES) Mmm12 STARD3NL Protrudin bu IP3R Mdm10 Mdm34 MCU PI(3)P Figure 2. ER-Mediated Organelle Contacts. Schematic illustration of ER-mediated contact sites. The shapes and areas of ER sites that contact other organelles are dynamic. Both ER tubules and sheets may be involved in responses to different stimuli. Some of the tether factors and/or effectors that mediate contact formation are listed. In general, ER-resident proteins are listed on the left, and their corresponding factors are listed on the right. Gray symbols represent proteins that are not discussed in this review. Abbreviations: ER, endoplasmic reticulum; LD, lipid droplet; SR, sarcoplasmic reticulum. via their C2 domains [37], and their SMP domain is shown to be a lipid transfer module [38–40]. Such interactions are thought to be sensitive to the levels of cytosolic Ca2+ [41]. The tricalbins found in yeast are structurally and functionally homologous to E-Syts [42,43]. Following phospholipase C (PLC)-activating stimuli, Nir2 is recruited to ER–PM contact sites, where it transfers phosphatidic acid (PA) generated in the PM to the ER [44]. The relocation of Nir2 involves VAP in the ER and PA in the PM. The ER integral membrane proteins ORP5 and ORP8, which bind to PI (4)P in the PM via their pleckstrin homology (PH) domains, deliver phosphatidyl serine (PS) from the ER to the PM and counter-transport PI(4)P to the ER for degradation by the ER-resident PI(4) P phosphatase Sac1 [45]. ER tubules are critical for ER–PM contact sites. Deletion of the tubule-forming protein Rtn4a dramatically expands ER sheets and reduces store operated calcium entry, which relies on Orai– STIM interactions [46]. E-Syts are anchored in ER membrane through a typical TMH [37], which suggests a preference for ER tubules. However, in myocytes, where the SR and PM need to be continually linked, the contacts are mainly cisternal [47]. Thus, PM interactions are likely initiated by ER tubules but can evolve into sheets if necessary. Trends in Cell Biology, Month Year, Vol. xx, No. yy 5 TICB 1254 No. of Pages 10 ER–Mitochondria Contact Sites The ER attaches to mitochondria at a region known as the mitochondria-associated ER membrane (MAM). MAMs are important for calcium homeostasis and phospholipid biosynthesis. In yeast, ER–mitochondria contacts are tethered by the ERMES complex[5_TD$IF] [48]. The molecular basis of the contacts in mammalian cells is not clear. ER tubules, in particular, appear to be associated with mitochondrial dynamics and function. ER tubules wrap around mitochondria to create mitochondrial constriction sites, where fission factors such as the dynaminlike GTPase Drp1 are recruited and assembled [49]. In yeast cells, such a wrapping interaction involves the highly conserved Miro GTPase Gem1 in addition to ERMES, and regulates the distribution of mitochondria and mitochondrial DNA [50]. Coupling of calcium signaling between the two major intracellular Ca2+[5_TD$IF] stores is based on the crosstalk between the IP3 receptor on the ER side and the mitochondrial Ca2+ uniporter (MCU), and relies on the physical contact between the two organelles [51]. Interestingly, IP3 receptormediated ER-mitochondria contacts colocalize with ER tubules [52], supporting the tubulebased nature of the contacts. Mitochondria need phospholipid supplies from the ER, and lipid transfer may benefit from ER– mitochondria contact sites. Components of the ERMES complex contain SMP domains that bind to phospholipids [53], and a conserved ER membrane complex facilitates lipid transfer from the ER to mitochondria [54]. However, contact site-mediated transfer may not be essential in higher eukaryotes, because in Xenopus egg extracts PS can be transferred from the ER to mitochondria when the two membranes are no longer tethered [55]. ER–mitochondria contacts also act as sites for autophagosome formation. Recent studies revealed that autophagosomes form at MAMs in mammalian cells, and impaired ER–mitochondria contacts attenuate autophagosome formation [56]. In yeast, ERMES is dispensable for nonselective bulk autophagy. However, mitophagy, which involves selective engulfment of mitochondria by autophagosomes, requires ERMES-established mitochondria–ER contacts [57]. ERMES functions at the membrane expansion during mitophagy, suggesting that MAMs may transfer phospholipids to the forming autophagosome. ER–Golgi Contact Sites ER–Golgi contacts occur mostly between flattened ER sheets and the trans-most Golgi cisternae [58]. Different lipid transfer proteins are present at ER–Golgi contacts. Oxysterolbinding proteins (OSBPs) deliver sterol to the Golgi and transfer PI(4)P for degradation by Sac1 [59]. CERT transports ceramide from the ER to the Golgi for sphingomyelin synthesis [60]. The area of ER–Golgi apposition can be modulated by the levels of lipids and tethering proteins. Golgi structures are reported to be completely enwrapped by the ER upon 25-hydroxycholesterol treatment and expression of VAP-A and OSBP [59]. ER–endosome Contact Sites The ER network forms multiple contacts with endosomes to regulate endosome processes including fission, maturation, movement, cholesterol transfer, and receptor dephosphorylation [61]. These distinct functions appear to be conferred by different tethering complexes involving different endosome subpopulations. Tethering of ORP5 with the late endosomal cholesterol transporter NPC1 facilitates the transfer of cholesterol from late endosomes (LEs) to the ER [62]. Low cholesterol levels on LEs facilitate the interaction between ORP1L and VAP-A, which removes the dynein motor subunit p150Glued and associated motors from Rab7–RILP, resulting in plus-end-directed transport of LEs [63]. The integral ER membrane protein protrudin interacts with Rab7-GTP and PI(3)P on the endosome membrane to form ER–LE contact sites, at which the motor protein kinesin 1 is transferred to the motor adaptor 6 Trends in Cell Biology, Month Year, Vol. xx, No. yy TICB 1254 No. of Pages 10 FYCO1 on LEs, resulting in transport of LEs to the cell periphery [64]. The formation of ER–endosome contacts creates constrictions of endosomes, which facilitate endosome fission [65]. Contact sites may provide platforms for recruiting fission machinery or generating highly curved membrane regions favorable for fission via lipid and Ca2+ transfer. Contact sites also restrict cargo diffusion, contributing to partitioning endocytic cargo to distinct endosome domains [65]. Overexpression of Rtn4a causes ER tubules to become less dynamic and reduces endosome fission [65]. Deletion of RHD3 in plants causes similar defects [66]. Although most ER–endosome contacts occur with ER tubules, increased levels of the LE membrane-anchored proteins STARD3 and STARD3NL, which tether the ER by interacting with VAPs, result in a tight enclosure of endosomes by ER sheets [67]. This impairs the generation of LE tubules, resulting in less mobile LEs [67]. Therefore, ER dynamics and shape must be precisely controlled to facilitate endosome dynamics. ER–LD Contact Sites Nascent LDs pinch off from ER bulges where neutral lipids accumulate between the phospholipid bilayers. LDs likely form contacts with ER sheets, and these contacts probably link neural lipid synthesis with LD expansion. The ER-resident transmembrane proteins Fld1 and seipin are enriched at ER–LD contact sites in yeast and mammalian cells, respectively [68]. Interestingly, LDs also physically connect with the ER via a stalk, which may function as a conduit for the exchange of lipids and proteins. Factors tethering ER–LD contacts remain largely unknown. ER Morphology in Membrane Trafficking In addition to direct contacts, vesicle-based trafficking is a major pathway for exchanging materials between membrane systems. Newly synthesized proteins or lipids are packed into COPII-coated vesicles and leave the ER through ER exit sites (ERESs). Because the curvature of a vesicle is comparable to that of a tubule cross-section, COPII vesicles are speculated to be preferentially generated in ER tubules. ERESs are enriched in tubules [17]. Most ER tubules are peripherally localized, but a significant proportion is clustered near the nucleus due to dyneinbased retrograde movements [69]. COPII trafficking is expected to occur in the perinuclear region where ER sheets, on which proteins are synthesized, and the Golgi apparatus, where proteins are processed, both localize. Whether perinuclear tubules harbor the majority of ERES remains to be assessed. The connection between membrane trafficking and ER morphology is supported by an auxinsignaling defect occurring with the deletion of RHD3; the major ER fusogen in plants [66]. Auxin transport relies heavily on coordinated exocytosis and endocytosis. In rhd3 cells, ER mobility and endosome streaming are drastically affected [66], impairing endocytosis. ER export may also be inhibited by the lack of ER complexity in rhd3 cells. Similar defects were found in the endocytic pathway when neurolastin, a brain-specific homolog of ATL, was deleted in a mouse model, and the consequence was drastically reduced excitatory synapses and spine density [70]. In mammalian cells, Rtn1 is shown to indirectly influence the function of the Golgi by regulating Rab1 and Rab43 via TBC1D20 [71]. In addition to the general defects caused by determinants of ER morphology, individual family members have been implicated in modulating the trafficking of specific cargos. In Arabidopsis, RTNLB regulates ER export of the FLS2 immune receptor [72]; in mammalian cells, ATL1 mutations are linked to the defective transport of BMP receptor BMPRII [73] and REEPs affect trafficking of specific G-protein-coupled receptors [74]. Whether additional cargos will be affected in these cases is not clear, as the cargos available for testing are limited. Nevertheless, these results suggest that the ER tubular network actively participates in cargo sorting and provides a platform for vesicle budding. Trends in Cell Biology, Month Year, Vol. xx, No. yy 7 TICB 1254 No. of Pages 10 ER Morphology in Multicellular Development Outstanding Questions ER morphogenesis, especially tubular network formation, appears to be a highly conserved process in eukaryotes. ER shaping and remodeling are expected to play a fundamental role during development. Surprisingly, in the basic eukaryotic model, the growth of yeast cells is only slowed when Yop1p and Rtn1p are deleted, and sey1D cells seem to be normal [7]. The functions of these proteins could be fulfilled by yet unknown analogous proteins. ER membrane dynamics have also been suggested to be more critical when cells expand during development [75]. The cortical ER of cotyledon epidermal cells in rhd3 plants is indistinguishable from that of wild-type plants at early developmental stages, but as cells grow, the cable-like ER phenotype starts to appear. In Caenorhabditis elegans, which has much larger cells than yeast, depletion of both YOP-1 and RET-1 (the homolog of Rtn4a in C. elegans) dramatically reduces the embryo survival rate [76]. Similarly, deletion of Rtnl1 (the only widely expressed Rtn in Drosophila) causes ER sheet expansion and defects in distal motor axons [77]. Many proteins have been identified that are involved in ER morphogenesis, and their mechanisms of action have prompted much speculation. How do tubule-forming proteins assemble into oligomers? Is Climp-63 really a luminal spacer? If so, what is the molecular basis of its activity? How do Kinectin and p180 contribute to sheet formation? How do ER shaping proteins coordinate with each other? When genes that shape the ER have redundancy in the genome, the situation becomes more complicated. [14_TD$IF]It is thought that deletion or mutation of individual isoforms may not cause a prominent phenotype. In addition to RHD3, Arabidopsis has two tissue-specific RHD3-like proteins with ER fusogen activity [28]. However, mutation or deletion of RHD3 alone causes abnormal cortical ER, cell expansion defects, and short root hairs [28]. Although mammals have more than 10 RHD-containing proteins, knocking out just Rtn4 is sufficient to convert most ER tubules into sheets in MEF cells [46]. Several ER-shaping proteins have been implicated in hereditary spastic paraplegia (HSP); a neurodegenerative disease characterized by axon shortening in corticospinal motor neurons and progressive spasticity and weakness of the lower limbs [78]: ATL1 (SPG3A), Rtn2 (SPG12), and REEP1 (SPG31) [79]. A mouse model of REEP1 deletion and a Drosophila model of Rtn2 deletion exhibit characteristics of HSP [77,80]. In general, neurons contain long protrusions, and the integrity of the ER network is more vulnerable to functional impairment. Whether the defect lies in ER–PM contacts, membrane trafficking, or organelle mobility is currently unknown. Concluding Remarks The network of tubules and sheets in the ER are generated by shared mechanisms using common protein modules, such as TMHs, APHs, and CCs. Many functions of the ER, including organelle contacts, rely on its morphological features (see [15_TD$IF]Outstanding [16_TD$IF]Questions). Although ER sheets serve as a platform for protein synthesis, the tubular ER network provides advantages in membrane trafficking. Processes such as lipid synthesis, though not discussed here, also profoundly affect ER shaping [15,65], and are likely affected by ER shape. Recent advancements in super-resolution microscopy will help uncover new ER shaping and remodeling phenomena. The precise physiological relevance of ER morphology will also be gradually revealed by model multicellular organisms. Acknowledgments We are grateful to Drs. Isabel Hanson and Alicia Prater for editing the work and to Dr. Sha Sun for help with the figures. Dr. Hong Zhang was supported by grants from the National Natural Science Foundation of China (NSFC) (31421002, 31561143001, 31225018), the National Basic Research Program of China (2013CB910100), and an International Early Career Scientist grant from the Howard Hughes Medical Institute. J.H. is supported by the NSFC (31225006), the National Basic Research Program of China (2012CB910302), and an International Early Career Scientist grant from Howard Hughes Medical Institute. References 1. Baumann, O. and Walz, B. (2001) Endoplasmic reticulum of animal cells and its organization into structural and functional domains. Int. Rev. Cytol. 205, 149–214 8 Trends in Cell Biology, Month Year, Vol. xx, No. yy 2. Shibata, Y. et al. (2009) Mechanisms shaping the membranes of cellular organelles. Annu. Rev. Cell Dev. Biol. 25, 329–354 Formation of contacts between the [17_TD$IF]ER and another heterologous membrane is highly dynamic and requires different tether and effector proteins in response to distinct stimuli. What is the molecular machinery required for contact formation and stabilization? How is the contact maintained and disassociated? How are the shape and area of apposed membranes at the contact site regulated? How does contact formation communicate with vesiclemediated trafficking? What are the physiological functions of morphologically distinct ER domains? What is the proteome of ER tubules? What is the phenotype when Climp-63 or other sheet-enriched proteins are deleted at the organism level? Studies of ER morphology and its functions have been mainly focused on yeast and tissue culture cells. Very little emphasis has been placed on ER morphology in the context of multicellular organism development. How is ER morphology integrated with developmental cues and extracellular signaling? How is ER morphology coordinately regulated in different tissues? TICB 1254 No. of Pages 10 3. West, M. et al. (2011) A 3D analysis of yeast ER structure reveals how ER domains are organized by membrane curvature. J. Cell Biol. 193, 333–346 30. Bian, X. et al. (2011) Structures of the atlastin GTPase provide insight into homotypic fusion of endoplasmic reticulum membranes. Proc. Natl. Acad. Sci. U.S.A. 108, 3976–3981 4. Bennett, P.M. (2012) From myofibril to membrane; the transitional junction at the intercalated disc. Front. Biosci. 17, 1035–1050 31. Byrnes, L.J. and Sondermann, H. (2011) Structural basis for the nucleotide-dependent dimerization of the large G protein atlastin1/SPG3A. Proc. Natl. Acad. Sci. U.S.A. 108, 2216–2221 5. Shibata, Y. et al. (2006) Rough sheets and smooth tubules. Cell 126, 435–439 6. Voeltz, G.K. et al. (2006) A class of membrane proteins shaping the tubular endoplasmic reticulum. Cell 124, 573–586 7. Hu, J. et al. (2009) A class of dynamin-like GTPases involved in the generation of the tubular ER network. Cell 138, 549–561 8. Orso, G. et al. (2009) Homotypic fusion of ER membranes requires the dynamin-like GTPase atlastin. Nature 460, 978–983 9. Shibata, Y. et al. (2010) Mechanisms determining the morphology of the peripheral ER. Cell 143, 774–788 10. Chen, S. et al. (2012) ER network formation requires a balance of the dynamin-like GTPase Sey1p and the Lunapark family member Lnp1p. Nat. Cell Biol. 14, 707–716 11. Shemesh, T. et al. (2014) A model for the generation and interconversion of ER morphologies. Proc. Natl. Acad. Sci. U.S.A. 111, E5243–E5251 12. Chen, S. et al. (2015) Lunapark stabilizes nascent three-way junctions in the endoplasmic reticulum. Proc. Natl. Acad. Sci. U.S.A. 112, 418–423 13. Chang, J. et al. (2013) Protrudin binds atlastins and endoplasmic reticulum-shaping proteins and regulates network formation. Proc. Natl. Acad. Sci. U.S.A. 110, 14954–14959 14. Hashimoto, Y. et al. (2014) Protrudin regulates endoplasmic reticulum morphology and function associated with the pathogenesis of hereditary spastic paraplegia. J. Biol. Chem. 289, 12946–12961 15. English, A.R. and Voeltz, G.K. (2013) Rab10 GTPase regulates ER dynamics and morphology. Nat. Cell Biol. 15, 169–178 16. Gerondopoulos, A. et al. (2014) Rab18 and a Rab18 GEF complex are required for normal ER structure. J. Cell Biol. 205, 707–720 17. Okamoto, M. et al. (2012) High-curvature domains of the ER are important for the organization of ER exit sites in Saccharomyces cerevisiae. J. Cell Sci. 125, 3412–3420 18. Hu, J. et al. (2008) Membrane proteins of the endoplasmic reticulum induce high-curvature tubules. Science 319, 1247–1250 19. Yamamoto, Y. et al. (2014) Arl6IP1 has the ability to shape the mammalian ER membrane in a reticulon-like fashion. Biochem. J. 458, 69–79 32. Byrnes, L.J. et al. (2013) Structural basis for conformational switching and GTP loading of the large G protein atlastin. EMBO J. 32, 369–384 33. Yan, L. et al. (2015) Structures of the yeast dynamin-like GTPase Sey1p provide insight into homotypic ER fusion. J. Cell Biol. 210, 961–972 34. Liu, T.Y. et al. (2012) Lipid interaction of the C terminus and association of the transmembrane segments facilitate atlastinmediated homotypic endoplasmic reticulum fusion. Proc. Natl. Acad. Sci. U.S.A. 109, E2146–E2154 35. Liu, T.Y. et al. (2015) Cis and trans interactions between atlastin molecules during membrane fusion. Proc. Natl. Acad. Sci. U.S.A. 112, E1851–E1860 36. Hogan, P.G. et al. (2010) Molecular basis of calcium signaling in lymphocytes: STIM and ORAI. Annu. Rev. Immunol. 28, 491–533 37. Giordano, F. et al. (2013) PI(4,5)P(2)-dependent and Ca(2+)-regulated ER-PM interactions mediated by the extended synaptotagmins. Cell 153, 1494–1509 38. Schauder, C.M. et al. (2014) Structure of a lipid-bound extended synaptotagmin indicates a role in lipid transfer. Nature 510, 552–555 39. Yu, H. et al. (2016) Extended synaptotagmins are Ca2+-dependent lipid transfer proteins at membrane contact sites. Proc. Natl. Acad. Sci. U.S.A. 113, 4362–4367 40. Saheki, Y. et al. (2016) Control of plasma membrane lipid homeostasis by the extended synaptotagmins. Nat. Cell Biol. 18, 504–515 41. Idevall-Hagren, O. et al. (2015) Triggered Ca2+ influx is required for extended synaptotagmin 1-induced ER-plasma membrane tethering. EMBO J. 34, 2291–2305 42. Creutz, C.E. et al. (2004) Characterization of the yeast tricalbins: membrane-bound multi-C2-domain proteins that form complexes involved in membrane trafficking. Cell. Mol. Life Sci. 61, 1208–1220 43. Toulmay, A. and Prinz, W.A. (2012) A conserved membranebinding domain targets proteins to organelle contact sites. J. Cell Sci. 125, 49–58 20. Khaminets, A. et al. (2015) Regulation of endoplasmic reticulum turnover by selective autophagy. Nature 522, 354–358 44. Kim, Y.J. et al. (2015) Phosphatidylinositol-phosphatidic acid exchange by Nir2 at ER-PM contact sites maintains phosphoinositide signaling competence. Dev. Cell 33, 549–561 21. Shibata, Y. et al. (2008) The reticulon and DP1/Yop1p proteins form immobile oligomers in the tubular endoplasmic reticulum. J. Biol. Chem. 283, 18892–18904 45. Chung, J. et al. (2015) PI4P/phosphatidylserine countertransport at ORP5-and ORP8-mediated ER-plasma membrane contacts. Science 349, 428–432 22. Brady, J.P. et al. (2015) A conserved amphipathic helix is required for membrane tubule formation by Yop1p. Proc. Natl. Acad. Sci. U.S.A. 112, E639–E648 46. Jozsef, L. et al. (2014) Reticulon 4 is necessary for endoplasmic reticulum tubulation, STIM1-Orai1 coupling, and store-operated calcium entry. J. Biol. Chem. 289, 9380–9395 23. Westrate, L.M. et al. (2015) Form follows function: the importance of endoplasmic reticulum shape. Annu. Rev. Biochem. 84, 791–811 47. Flucher, B.E. (1992) Structural analysis of muscle development: transverse tubules, sarcoplasmic reticulum, and the triad. Dev. Biol. 154, 245–260 24. Klopfenstein, D.R. et al. (2001) Subdomain-specific localization of CLIMP-63 (p63) in the endoplasmic reticulum is mediated by its luminal alpha-helical segment. J. Cell Biol. 153, 1287–1300 48. Kornmann, B. et al. (2009) An ER-mitochondria tethering complex revealed by a synthetic biology screen. Science 325, 477–481 25. Starr, D.A. and Fridolfsson, H.N. (2010) Interactions between nuclei and the cytoskeleton are mediated by SUN-KASH nuclear-envelope bridges. Annu. Rev. Cell Dev. Biol. 26, 421–444 26. Chang, W. et al. (2015) Accessorizing and anchoring the LINC complex for multifunctionality. J. Cell Biol. 208, 11–22 49. Friedman, J.R. and Voeltz, G.K. (2011) The ER in 3D: a multifunctional dynamic membrane network. Trends Cell Biol. 21, 709–717 50. Murley, A. et al. (2013) ER-associated mitochondrial division links the distribution of mitochondria and mitochondrial DNA in yeast. Elife. 2, e00422 27. Anwar, K. et al. (2012) The dynamin-like GTPase Sey1p mediates homotypic ER fusion in S. cerevisiae. J. Cell Biol. 197, 209–217 51. Qi, H. et al. (2015) Optimal microdomain crosstalk between endoplasmic reticulum and mitochondria for Ca2+ oscillations. Sci. Rep. 5, 7984 28. Zhang, M. et al. (2013) ROOT HAIR DEFECTIVE3 family of dynamin-like GTPases mediates homotypic endoplasmic reticulum fusion and is essential for Arabidopsis development. Plant Physiol. 163, 713–720 52. Goetz, J.G. et al. (2007) Reversible interactions between smooth domains of the endoplasmic reticulum and mitochondria are regulated by physiological cytosolic Ca2+ levels. J. Cell Sci. 120, 3553–3564 29. Zheng, H. et al. (2004) A GFP-based assay reveals a role for RHD3 in transport between the endoplasmic reticulum and Golgi apparatus. Plant J. 37, 398–414 53. AhYoung, A.P. et al. (2015) Conserved SMP domains of the ERMES complex bind phospholipids and mediate tether assembly. Proc. Natl. Acad. Sci. U.S.A. 112, E3179–E3188 Trends in Cell Biology, Month Year, Vol. xx, No. yy 9 TICB 1254 No. of Pages 10 54. Lahiri, S. et al. (2014) A conserved endoplasmic reticulum membrane protein complex (EMC) facilitates phospholipid transfer from the ER to mitochondria. PLoS Biol. 12, e1001969 68. Wang, C.W. et al. (2014) Control of lipid droplet size in budding yeast requires the coll*aboration between Fld1 and Ldb16. J. Cell Sci. 127, 1214–1228 55. Junker, M. and Rapoport, T.A. (2015) Involvement of VAT-1 in phosphatidylserine transfer from the endoplasmic reticulum to mitochondria. Traffic. 16, 1306–1317 69. Wang, S. et al. (2013) Multiple mechanisms determine ER network morphology during the cell cycle in Xenopus egg extracts. J. Cell Biol. 203, 801–814 56. Hamasaki, M. et al. (2013) Autophagosomes form at ER-mitochondria contact sites. Nature 495, 389–393 70. Lomash, R.M. et al. (2015) Neurolastin, a dynamin family GTPase, regulates excitatory synapses and spine density. Cell Rep. 12, 743–751 57. Bockler, S. and Westermann, B. (2014) Mitochondrial ER contacts are crucial for mitophagy in yeast. Dev. Cell 28, 450–458 58. De Matteis, M.A. and Rega, L.R. (2015) Endoplasmic reticulumGolgi complex membrane contact sites. Curr. Opin. Cell Biol. 35, 43–50 59. Mesmin, B. et al. (2013) A four-step cycle driven by PI(4)P hydrolysis directs sterol/PI(4)P exchange by the ER-Golgi tether OSBP. Cell 155, 830–843 71. Haas, A.K. et al. (2007) Analysis of GTPase-activating proteins: Rab1 and Rab43 are key Rabs required to maintain a functional Golgi complex in human cells. J. Cell Sci. 120, 2997–3010 72. Lee, H.Y. et al. (2011) Arabidopsis RTNLB1 and RTNLB2 reticulon-like proteins regulate intracellular trafficking and activity of the FLS2 immune receptor. Plant Cell 23, 3374–3391 60. Hanada, K. et al. (2003) Molecular machinery for non-vesicular trafficking of ceramide. Nature 426, 803–809 73. Zhao, J. and Hedera, P. (2013) Hereditary spastic paraplegiacausing mutations in atlastin-1 interfere with BMPRII trafficking. Mol. Cell. Neurosci. 52, 87–96 61. Friedman, J.R. et al. (2013) Endoplasmic reticulum-endosome contact increases as endosomes traffic and mature. Mol. Biol. Cell. 24, 1030–1040 74. Bjork, S. et al. (2013) REEPs are membrane shaping adapter proteins that modulate specific g protein-coupled receptor trafficking by affecting ER cargo capacity. PloS One 8, e76366 62. Du, X.M. et al. (2011) A role for oxysterol-binding protein-related protein 5 in endosomal cholesterol trafficking. J. Cell Biol. 192, 121–135 75. Lai, Y.S. et al. (2014) ER stress signaling requires RHD3, a functionally conserved ER-shaping GTPase. J. Cell Sci. 127, 3227–3232 63. Rocha, N. et al. (2009) Cholesterol sensor ORP1L contacts the ER protein VAP to control Rab7-RILP-p150(Glued) and late endosome positioning. J. Cell Biol. 185, 1209–1225 76. Audhya, A. et al. (2007) A role for Rab5 in structuring the endoplasmic reticulum. J. Cell Biol. 178, 43–56 64. Raiborg, C. et al. (2015) Repeated ER-endosome contacts promote endosome translocation and neurite outgrowth. Nature 520, 234–238 65. Rowland, A.A. et al. (2014) ER contact sites define the position and timing of endosome fission. Cell 159, 1027–1041 66. Stefano, G. et al. (2015) ER network homeostasis is critical for plant endosome streaming and endocytosis. Cell Discov. 1, http://dx.doi.org/10.1038/celldisc.2015.33 67. Alpy, F. et al. (2013) STARD3 or STARD3NL and VAP form a novel molecular tether between late endosomes and the ER. J. Cell Sci. 126, 5500–5512 10 Trends in Cell Biology, Month Year, Vol. xx, No. yy 77. O'Sullivan, N.C. et al. (2012) Reticulon-like-1, the Drosophila orthologue of the hereditary spastic paraplegia gene reticulon 2, is required for organization of endoplasmic reticulum and of distal motor axons. Hum. Mol. Genet. 21, 3356–3365 78. Salinas, S. et al. (2008) Hereditary spastic paraplegia: clinical features and pathogenetic mechanisms. Lancet Neurol. 7, 1127–1138 79. Hubner, C.A. and Kurth, I. (2014) Membrane-shaping disorders: a common pathway in axon degeneration. Brain. 137, 3109– 3121 80. Beetz, C. et al. (2013) A spastic paraplegia mouse model reveals REEP1-dependent ER shaping. J. Clin. Invest. 123, 4273–4282