-
PDF
- Split View
-
Views
-
Cite
Cite
Kazuo Kurokawa, Akihiko Nakano, The ER exit sites are specialized ER zones for the transport of cargo proteins from the ER to the Golgi apparatus, The Journal of Biochemistry, Volume 165, Issue 2, February 2019, Pages 109–114, https://doi.org/10.1093/jb/mvy080
Close - Share Icon Share
Abstract
The endoplasmic reticulum (ER) is a multifunctional organelle, including secretory protein biogenesis, lipid synthesis, drug metabolism, Ca2+ signalling and so on. Since the ER is a single continuous membrane structure, it includes distinct zones responsible for its different functions. The export of newly synthesized proteins from the ER is facilitated via coat protein complex II (COPII)-coated vesicles, which form in specialized zones within the ER, called the ER exit sites (ERES) or transitional ER. In this review, we highlight recent advances in our understanding of the structural organization of ERES, the correlation between the ERES and Golgi organization, and the faithful cargo transport mechanism from the ERES to the Golgi.
Eukaryotic cells are characterized with a variety of intracellular membrane structures called organelles, which are to be compartmentalized in various regions of the cytoplasm and are accountable for different functions within the cells. Many different proteins and lipids are dynamically transferred from one organelle to another for orchestrating and maintaining cell functions. The endoplasmic reticulum (ER) is a multifunctional organelle, including secretory protein biogenesis, lipid synthesis, drug metabolism, Ca2+ signalling and so on. These various roles require the ER to be a huge organelle which has a large surface and shows a throughout distribution in the cytoplasm. The ER generates a complex architecture composed of tubular and lamellar membrane structures, resulting in an enormous functional space. The organization of the ER varies not only in different cell types, but also in different regions within one cell. Although the ER is indeed a single continuous membrane compartment, it includes functionally and structurally distinct subdomains that we recognize as functional zones. Most obvious zones of the ER are smooth and rough ER. The latter is recognized by the interaction of ribosomes on it. Other specialized zones are the membrane contact sites between the ER and essentially every other organelles, and the ER exit sites (ERES) or transitional ER, where coat protein complex II (COPII)-coated vesicles form (1, 2). This review focuses on our present knowledge of the structural organization of ERES, the correlation between the ERES and Golgi organization, and a new model for ER-to-Golgi cargo transport at the ERES.
ER-to-Golgi Cargo Transport via COPII-Coated Vesicles Begins at the ERES
The ER is the first organelle in the secretory pathway. Approximately one-third of the eukaryotic proteins are synthesized, modified and folded into proper conformation in the ER, and then they are delivered to the Golgi apparatus as secretory cargoes. In the Golgi, cargo proteins are processed and glycosylated before being sorted to their final destinations, including the organelles of secretory and endocytic pathways, the plasma membrane and the extracellular space (3). The ER-to-Golgi transport of cargo is facilitated via COPII-coated vesicles, which typically have a diameter of 60–90 nm and concentrate at the ERES, specialized zones within the ER (1, 2) (Fig. 1A). Components responsible for COPII-coated vesicle formation are well conserved from yeast to higher plants and animals and include inner coat complex Sec23/24, outer coat complex Sec13/31, a small GTPase Sar1 and its specific guanine nucleotide exchange factor (GEF) Sec12 (Fig. 1A).
The ERES are specialized zones for cargo transport from the ER to the Golgi. (A) Activated Sar1-GTP binds to the ER membrane and recruits Sec23/24 to form the pre-budding complex, to which cargo is concentrated. The pre-budding complexes are clustered by the Sec13/31 to generate COPII-coated vesicles, which accumulate at the ERES. (B) Spatial correlation between the ERES and the Golgi in different species. The ERES locate adjacent to the first post-ER compartments, cis-Golgi or the ERGIC.
The formation of COPII-coated vesicle is triggered by Sar1, whose activation is initiated by its GDP-GTP exchange catalysed by the ER transmembrane GEF Sec12 (4, 5). GTP binding by Sar1 induces its conformational change, which exposes an N-terminal amphipathic α-helix and then inserts it into the ER membrane (6). Sar1-GTP on the ER membrane recruits Sec23/24 heterodimer, leading to the formation of the inner layer of the COPII coat through the interaction with Sec23 and the capture of cargo proteins through the contact with Sec24 (pre-budding complex) (7). Sec13/31 proteins assemble into the outer layer of the COPII coat on top of the pre-budding complex through their interactions with Sec23. Then the inner and outer layers polymerize to form a caged structure (8, 9). The inner and outer layers of the COPII coat have other functions; Sec23 is a GTPase-activating protein (GAP) for Sar1, and this GAP activity is enhanced by Sec31 (10). GTP hydrolysis by Sar1 in turn causes its dissociation from the membrane (10).
Another key component required for COPII-coated vesicle formation in vivo is a peripheral membrane protein, Sec16, which directly interacts with several COPII components (11–17). This suggests that Sec16 organizes and acts as a scaffold for assembly of COPII components. Notably, recent studies using yeast have proposed that Sec16 is a regulator for COPII-coated vesicle formation by controlling GAP activity towards Sar1 rather than a scaffold (18–20).
An in vitro reconstitution assay revealed that the GTP loading on Sar1 is required for budding of COPII-coated vesicles (2). Indeed, insertion of N terminus of Sar1-GTP into the synthetic liposome membranes deforms the liposomes, causing them to form various tubular structures by generating membrane curvature (6, 21–23). Introducing a GTP-locked form of Sar1 alone into permeabilized mammalian cells can induce expansion of tubules from the ER membrane (21, 24), thus activated Sar1 itself has an ability to deform the ER membrane. In addition, in vitro cell-free experiments with synthetic liposomes, proteoliposomes or in a planar lipid bilayer have shown that Sec23/24, Sec13/31 and Sar1-GTP are sufficient for formation of COPII-coated vesicles (25–27). Therefore, a role for Sar1 in COPII-coated vesicle formation is clear, but the mechanisms by which Sar1 contributes to cargo concentration, vesicle budding and scission and cargo transport have not been fully elucidated. Single-molecule imaging studies showed that multiple rounds of the Sar1 GDP–GTP cycle are required for efficient cargo concentration (7, 28) and for enhancing large cargo transport such as procollagen (29). It has been reported that the GTPase activity is necessary for COPII-coated vesicles to detach from the ER membrane in vitro and in semi-intact cells (10, 21, 22). There is a report, however, claiming that the GTPase activity is not necessarily required for this process (30). Importantly, a GTP-locked mutant of Sar1 [yeast Sar1 H77L and mammalian Sar1 H79G; note mammalian have Sar1A and Sar1B genes reviewed in ref. (31)] blocks cargo transport in vivo (32, 33). From this, it is suggested that the nucleotide-bound state of Sar1 must be spatially and temporally controlled during COPII-coated vesicle formation, cargo concentration, vesicle budding and scission and cargo transport in vivo. Indeed, we recently reported that Sar1 shows restricted accumulation at the rims of COPII-coated vesicle membranes, but is excluded from the rest of COPII-coated membranes in living yeast cells (34).This finding indicates that the nucleotide-bound state of Sar1 is spatially and temporally controlled in vivo.
Structural Organization of ERES Within the ER Membrane
The ERES are morphologically distinct from the surrounding ER (1). Although COPII-coated vesicle formation has been characterized in detail as described above, the structural organization of ERES remains elusive. Since the ER has an elaborate shape consisting of a blanching tubule network with flattened or fenestrated sheets, the structural organization of ERES might depend on the ER membrane morphology. The generation and maintenance of the morphological feature of the ER are facilitated by the curvature-stabilizing proteins such as reticulons and the Yop1/DP1-family proteins and by the atlastin family proteins which drive membrane fusion (35–39). The structural organization of ERES has been recently understood by live cell three-dimensional observation of the budding yeast Saccharomyces cerevisiae (40). The S.cerevisiae ERES appear as numerous scattered puncta throughout the ER network (40, 41). The ERES show significantly overlapped localization with reticulon Rtn1 which accumulates on the high-curvature regions of the ER membrane, both the ER tubules and the rims of the ER sheet, suggesting that the ERES are preferentially organized at the ER high-curvature domain (35, 37, 40). In the yeast mutant cells having fewer high-curvature ER domains, the ERES accumulate at the remaining high-curvature ER domains on the edge of expanded ER sheets (40). In addition, three-dimensional observation indicates that the ERES show the restricted distribution at the saddle-like ER membrane structures, harbouring both positive and negative curvatures, which probably consist of various, differently shaped lipids (40). Since vesicle budding requires not only positive curvature but also negative curvature to be constricted at the neck, the saddle-like shape of the ER membrane surface may facilitate COPII vesicle formation within the ERES. The preference of ERES to reside in the ER high-curvature domain might also be based on the structure of the Sar1–Sec23/24-cargo protein pre-budding complex, which is a concave surface associated with its membrane-orientated face (42). Insertion of Sar1-GTP amphipathic helix generates deformation of the ER membrane (6). However, the local membrane curvature has shown to increase Sar1 affinity to the membrane and enhance its GTPase activity (43), suggesting the existence of a positive feedback loop between membrane curvature and the Sar1 activity.
The Number and Spatial Organization of ERES Correlate with the Organization of the First Post-ER Compartment, cis-Golgi or the ERGIC
The number and spatial organization of the ERES are distinct in different species but they might be correlated with the organization of the Golgi apparatus (Fig. 1B). This correlation becomes clear by observation of the ERES arrangement in various species. The most of eukaryotes exhibit a tight association between adjacent compartments (cisternae) of the Golgi apparatus to form the stacks. Budding yeast, Pichia pastoris has a small number of ERES (two to six per cell), each juxtaposes to the stacked Golgi (44, 45). A similar organization of ERES and Golgi is found in the unicellular parasite Trypanosoma brucei and Toxoplasma gondii, which have one ERES adjacent to stacked Golgi cisternae (46, 47). In plant cells, a large number of stacked Golgi are scattered throughout the cytoplasm. Even a larger number of the ERES distribute along the ER network, and to many of them the cis-face of stacked Golgi is tightly associated, as if they behave together as a unit (48–50). However, vertebrate cells have a unique complex structure of the Golgi called the Golgi ribbon, which is composed of many stacked Golgi being tightly connected with each other. They usually locate at the perinuclear region in a microtubule-dependent manner. The ERES show hundreds of punctate structures on the surface of the ER membrane throughout the cell, most of which look quite distant from the Golgi ribbon. However, they are functionally interacting with the perinuclear Golgi by the ER-Golgi intermediate compartment (ERGIC) or vesicular tubular clusters (VTCs) (51). Most of ERGIC are found adjacent to the ERES. ERGIC is structurally distinct from either the ER or the Golgi and can be thought to play a role for the first post-ER compartment to receive cargo from the ERES (52, 53). Thus, there may well be spatial interconnection between the ERES and the first post-ER compartment, the ERGIC in vertebrate cells and cis-Golgi in other species. However, in the budding yeast S.cerevisiae, the Golgi is not arranged into stacks but consists of individual cisternae scattered in the cytoplasm (27, 54). The ERES of S.cerevisiae show numerous punctate structures on the surface of the ER membrane (40, 41). The dispersed Golgi cisternae of S.cerevisiae are moving around in the cytoplasm. We realized that cis-Golgi cisternae exhibit higher probability of staying in the close vicinity of ERES than trans cisternae (40). In addition, we have recently shown that cis-Golgi, but not trans-Golgi, frequently approaches the ERES, suggesting that of both the ERES and the first post-ER compartment would generate a functional unit for cargo transport from the ER to the Golgi.
ER-to-Golgi Cargo Transport via Physical Link Between COPII-Coated Vesicles and the First Post-ER Compartments
In the conventional model, ER-to-Golgi and ER-to-ERGIC cargo delivery has been believed to operate in a way that ER-derived COPII-coated vesicles are once released to the cytosol, move around in the cytosol, and then get tethered to and fuse with cis-Golgi or the ERGIC (55). However, COPII-coated vesicles free from the ER are very rare in electron microscopic observations of yeast cells, unlike COPI- and clathrin-coated vesicles which are easily found. Saccharomyces cerevisiae shows temporal formation of a functional unit between the ERES and cis-Golgi. Then, how can ER-to-Golgi cargo delivery be achieved faithfully? Our high-resolution 4D live imaging has indicated that mobile cis-Golgi approaches and contacts ERES and captures cargo directly from the ER (56) (Fig. 2). It must be much safer and much more efficient for the COPII-coated vesicles to be captured by cis-Golgi than being released into the cytosol. The important role of the physical link between the ERGIC and COPII-coated vesicles at the ERES has also been proposed in the procollagen export from the ER in mammalian cells (57–59). Importantly, the recruitment of ERGIC towards COPII-coated vesicles at the ERES during this process is mediated by TANGO1, which also controls the concentration of procollagen and formation of large COPII-coated vesicles and is localized to the ERES (58, 60). To summarize, the ERES work as functional zones, which link concentration of cargo, formation of carriers and transport of cargo.
The dynamic approaching motion of cis-Golgi to the ERES, called the ‘hug-and-kiss’ action, executes efficient cargo delivery from the ER to the Golgi apparatus in S.cerevisiae.
Conclusion and Perspective
In most of eukaryotic cells, there are multiple ERES, which are spatially coupled to cis-Golgi (or the ERGIC in vertebrate cells). A wide variety of cargo proteins are synthesized one after another in the ER and are then precisely transported from the ERES to the Golgi apparatus. Within a continuous membrane structure of the ER, the ERES operate as specialized zones, which play a pivotal role in faithful cargo transport from the ER to the Golgi. It is still an open question whether cargo proteins, depending on their properties, are selectively loaded into COPII-coated vesicles at a certain type of ERES, or are incorporated randomly. Pioneering work in yeast demonstrated that two distinct types of cargo proteins, GPI-anchored proteins and transmembrane proteins, both destined to the plasma membrane, are incorporated into distinct COPII-coated vesicles and concentrated into different ERES (61, 62). Live imaging at extremely high spatiotemporal resolution, which is now in our hands, will enable us to understand how various proteins are accurately and safely transported from the ER to the Golgi apparatus.
Funding
This work was supported by Grants-in-Aid for Scientific Research (JP25221103 and JP17H06420 to A.N.) and (JP16HD05419, JP17H06420 and JP25221103 to K.K.) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by the 4D Measurements for Multilayered Cellular Dynamics Project of RIKEN to A.N.
Conflict of Interest
None declared.

