Green Light for Traffic in the Early Secretory Pathway

The endomembrane system, which includes the endoplasmic reticulum (ER), the Golgi apparatus, vacuoles, and associated vesicles, functions in the biosynthesis and transport of lipids, proteins, and cell wall polysaccharides destined for various locations within the cell, the maintenance of cell integrity, and the detoxification of foreign molecules that gain entry into the cell. In fact, this pathway plays a major role in feeding the world through the synthesis of seed storage proteins (Vitale and Denecke, 1999). Anterograde traffic in the pathway begins with protein synthesis: nascent proteins destined for secretion are delivered to the ER and then transported to the Golgi apparatus, where they are further sorted for translocation to the cell exterior, or to cellular compartments such as protein storage vacuoles, lytic vacuoles, the plasma membrane, or the cell wall. Retrograde traffic, which runs in the opposite direction, allows for endocytosis of extracellular molecules and for recycling of membranes and proteins to maintain the integrity of the various cellular compartments (Vitale and Denecke, 1999).

The ER is the largest membrane system in eukaryotes, consisting of highly invaginated tubules and flattened cisternae connected to ribosomes (rough ER) and a network of numerous extensions and vesicles that connect to the nuclear envelope and the Golgi apparatus, and extend through plasmodesmata to neighboring cells. The Golgi apparatus consists of stacks of individual cisternae, flattened organelles with bulbous ends that are connected by a network of tubules. Both the number of cisternae within a stack and the number of stacks per cell can vary considerably between cell types and genera, and changes also may be triggered by developmental and environmental signals (Andreeva et al., 1998). Golgi stacks are polarized; the inter-cisternal distance increases, but the density of the cisternae increases, moving from the cis-face to the trans-face of an individual stack, and it was demonstrated recently that Golgi resident proteins can have either a predominantly cis or trans distribution (Wee et al., 1998; Nebenführ et al., 1999). Compounds delivered from the ER enter the Golgi apparatus at the cis-face, and move through the medial- and trans-Golgi to the trans-Golgi network.

Transport from the ER to the Golgi apparatus, between cisternae of the Golgi stack, and from the trans-Golgi network to final destinations within or outside the cell occurs via various protein-coated vesicles. The so-called COPI- and COPII-coated vesicles are involved in early traffic between the ER and Golgi stacks. In yeast and mammalian cells, there is good evidence that COPII-coated vesicles bud from the ER and transport material to the cis-Golgi, whereas COPI-coated vesicles are involved in traffic between cisternae of the Golgi apparatus and in retrograde transport from the Golgi back to the ER (Scales et al., 1997; Bannykh and Balch, 1998; Nickel and Wieland, 1998). Evidence is accumulating for COPI and COPII homologs playing similar roles in plant cells, but the situation in plants is neither well characterized nor well understood.

The study of the mechanisms of transport within the plant endomembrane system has been hampered by the difficulty of isolating intact organelles and active protein constituents involved in the trafficking of compounds through this complex pathway. In this issue of THE PLANT CELL, we highlight two articles representing significant advances in our ability to study the mechanisms of transport through the secretory pathway. On pages 2219–2235, Pimpl et al. show in situ localization of COPI-coated vesicles and demonstrate recruitment of coatomer, the protein complex that makes up the COPI vesicle coat, from a cytosolic fraction onto budding vesicles reconstituted in vitro. Batoko et al., on pages 2201–2217, present a green fluorescent protein (GFP)-based assay for visualizing membrane traffic in the secretory pathway in plants, and show that the function of a Rab1 GTPase is required for transport from the ER to the Golgi apparatus and for normal Golgi movement in plant cells.

The COPI vesicle protein coat, originally characterized in yeast, consists of a complex of seven polypeptides collectively termed “coatomer.” Pimpl et al. (2000) identified maize and Arabidopsis homologs of coatomer components, and used immunogold in situ labeling to localize plant coatomer predominantly to the cis- and medial-Golgi cisternae. COPI-coated vesicle formation in mammalian cells depends on coatomer and the activity of the GTPase Arf1p (Palmer et al., 1993). Using in vitro vesicle induction experiments, Pimpl et al. (2000) showed that mixed ER/Golgi membrane fractions, isolated from cauliflower, were capable of specific recruitment of coatomer components and Arf1p from the cytosolic fraction. In vitro COPI vesicle budding was dependent on the presence of ATP/GTP, and was inhibited by the drug brefeldin A (BFA). BFA also has been shown to prevent coatomer attachment and in vitro vesicle budding in mammalian cells (Dascher and Balch, 1994). Although plant coatomer homologs, COPII-coat protein homologues, and related proteins have been identified in plants, COPI- and COPII-coated vesicles have not been isolated or identified in situ. The work of Pimpl et al. (2000) thus represents an important step on the road to the isolation of these trafficking vesicles from plants, the identification of their components, and the elucidation of their mechanism of action.

In cultured mammalian cells, the COPII vesicles often arise from the ER membrane in peripheral regions of the cell, where they fuse with each other to form transient transport intermediates known as vesicular tubular clusters. These are transported along microtubules to the Golgi cluster, and as they go, COPI vesicles assemble on their membranes and selectively remove escaped ER resident proteins for recycling back to the ER (reviewed in Ellgaard et al., 1999). Intermediates similar to vesicular tubular clusters have not been identified in plants, and it is not clear where in the ER/Golgi system membrane exchange occurs. The data presented by Pimpl et al. (2000), localizing COPI vesicle components predominantly to the cis-Golgi, suggest that an analogous role is possible for COPI vesicles in plants.

The GFP assay of Batoko et al. (2000) is based on previous observations that GFP may be targeted to the ER or the Golgi apparatus with the use of signaling peptides (Boevink et al., 1998; Nebenführ et al., 1999) and that export of GFP from the ER could be inhibited by BFA (Boevink et al., 1999). Soluble proteins are targeted for delivery to the ER by the presence of a specific N-terminal signal peptide, which is removed as the nascent protein enters the ER lumen (Vitale and Denecke, 1999). ER-transported polypeptides subsequently are secreted to the apoplast by default, being retained within the ER only if they contain the C-terminal tetrapeptide HDEL, KDEL, or a variant of these sequences. Still other signal peptides designate proteins for transport to various other intracellular destinations, and N-glycosylation also plays a role. Batoko et al. (2000) prepared two GFP constructs: one that contained an N-terminal ER signal peptide and a C-terminal c-Myc-epitope tag (secGFP), and another that contained both the N-terminal ER signal peptide and a C-terminal HDEL peptide (GFP-HDEL). Using a transient expression assay in tobacco leaves, they first showed that secGFP was delivered to the ER but did not accumulate (being secreted by default) unless transport from the ER was inhibited by treatment with BFA, whereas GFP-HDEL was retained within the ER in the absence of BFA. Thus, secGFP could be used in an assay for correct functioning of the secretory pathway downstream of the ER; a malfunction (demonstrated by treatment with BFA) would cause the accumulation and visualization of secGFP within the ER. This assay was then used to test the function of a Rab1 GTPase from Arabidopsis, AtRab1b, in ER-Golgi trafficking by co-expression of secGFP with a dominant negative mutant of AtRab1b, AtRab1b(N121I).

Rab GTPases are a large class of eukaryotic proteins implicated in the regulation of vesicle trafficking in the secretory pathway (Bischoff et al., 1999). GTPases are molecular switches that cycle between an active (GTP binding) and an inactive (GDP binding) form. The substitution of isoleucine for asparagine at position 121 in AtRab1b(N121I) was predicted to alter the GTP binding domain and inhibit the activity of the endogenous protein. Transient co-expression of secGFP and AtRab1b(N121I) resulted in the accumulation of secGFP in the ER, suggesting that the mutant Rab1b protein inhibited the normal export of secGFP from the ER. Furthermore, secretion of secGFP could be restored in the presence of the mutant Rab1b protein by co-expression with the wild-type AtRab1b. In further experiments, Batoko et al. (2000) showed that an N-glycosylated membrane-bound GFP construct (N-ST-GFP), which localized to the Golgi when it was expressed alone or in conjunction with wild-type AtRab1b, also accumulated in the ER when co-expressed with the mutant protein AtRab1b(N121I).

One of the major differences between the secretory pathways of plant and animal cells is in the movement and localization of the Golgi stacks. Whereas in mature mammalian cells the Golgi stacks are organized on microtubules and remain clustered in a juxtanuclear position at the microtubule organizing center, in plant cells the Golgi apparatus is dispersed into tens or hundreds of individual stacks that are highly mobile (Andreeva et al., 1998). Golgi movement in plant cells occurs in bursts along an actin network that is co-extensive with the ER, and this movement may be integral to the mechanism of transport between the Golgi apparatus and the ER (Boevink et al., 1998; Nebenführ et al., 1999). Boevink et al. (1998) postulated that the Golgi stacks move along and pick up products from the ER like a mobile “vacuum cleaner.” Nebenführ et al. (1999) presented an alternate “stop-and-go” (or “filling station”) hypothesis that places emphasis on regulatory signals produced in the ER. In this model, Golgi stacks moving along actin tracks are induced to stop at active ER export sites by a signal produced in the ER; after the exchange of materials, the “stop signal” is turned off and the Golgi stack is allowed to resume movement. The most interesting observations of Batoko et al. (2000) arose when they recorded moving images of transformed cells. Surprisingly, these showed that the expression of AtRab1b(N121I) inhibited normal Golgi movement (see http://www.plants.ox.ac.uk/moore/index.html), suggesting a link between Golgi mobility, control of membrane traffic from the ER, and Rab1 GTPase activity.

In mammalian cells, Rab1 activity is required for docking of ER-derived COPII vesicles with cis-Golgi cisternae, via regulation of membrane-bound SNARE proteins (Schekman and Orci, 1996). The results of Batoko et al. (2000) are consistent with a similar role for AtRab1b in plant cells. Future experiments will be needed to establish whether the cessation of Golgi movement is specific to the inhibition of Rab1 GTPase function or whether it is a common occurrence whenever ER–Golgi membrane traffic is inhibited. It will be important to establish whether the inhibition of Golgi movement arises from the inhibition of membrane traffic (consistent with the filling-station model) or whether the inhibition of movement is responsible for the inhibition of membrane traffic (consistent with the vacuum cleaner model). In this regard, it will be interesting to investigate the effect on membrane traffic of drugs and mutants that inhibit Golgi movement and to determine the effect on Golgi movement of mutants in other regulatory factors such as Arf1 and Sar1 (which function in the formation of COPI- and COPII-coated vesicles, respectively). It also will be important to establish the site of COPII vesicle formation as Pimpl et al. (2000) have done for COPI. In addition to the first direct evidence for Rab1b GTPase function in ER–Golgi transport in plant cells, Batoko et al. (2000) present an excellent system for uncovering further mechanistic details of trafficking in the secretory pathway.