The exocyst (Sec6/8) complex is necessary for secretion in yeast and has been postulated to establish polarity by directing vesicle fusion to specific sites along the plasma membrane. The complex may also function in the nervous system, but its precise role is unknown. Exocyst function was investigated in Drosophila with mutations in one member of the complex, sec5. Null alleles die as growth-arrested larvae, whose neuromuscular junctions fail to expand. In culture, neurite outgrowth fails in sec5 mutants once maternal Sec5 is exhausted. Using a trafficking assay, impairments were found in the membrane addition of newly synthesized proteins. In contrast, synaptic vesicle fusion was not impaired. Thus, Sec5, although not involved in vesicle transport, nevertheless differentiates between two forms of vesicle trafficking at the membrane: trafficking for cell growth and membrane protein insertion depend on sec5, whereas transmitter secretion does not. In this regard, Sec5 differs from the homologs of other yeast exocytosis genes that are required for both neuronal trafficking pathways (Murthy, 2003). Sec5 is also required for membrane traffic and polarity in the Drosophila ovary. During oogenesis, Sec5 localization undergoes dynamic changes, correlating with the sites at which it is required for the traffic of membrane proteins. Germline clones of sec5 possess defects in membrane addition and the posterior positioning of the oocyte. Additionally, the impaired membrane trafficking of Gurken, the secreted ligand for the EGF receptor, and Yolkless, the vitellogenin receptor, results in defects in dorsal patterning and egg size. However, the cytoskeleton is correctly oriented. It is concluded that Sec5 is required for directed membrane traffic, and consequently for the establishment of polarity within the developing oocyte (Murthy, 2004).

Directed membrane traffic is essential for many developmental processes, including cell growth, cytokinesis and signaling between cells. Such processes require membrane traffic to particular domains of the cell surface, in order to insert proteins at restricted regions of the membrane, to enlarge particular regions of the cell membrane, or to signal asymmetrically to neighboring cells. The identification, therefore, of the molecules required for directed membrane traffic will be important for understanding organismal development and cell-cell signaling (Murthy, 2004).

The exocyst complex, a set of eight proteins first identified from secretory mutants in yeast, is an attractive candidate for mediating directed traffic. Yeast cells use an anisotropic secretory apparatus for polarized growth at a selected bud site. While the bud is growing, there is almost no increase in the surface area of the mother cell, indicating that all membrane addition occurs at the bud tip. Later, secretion is redirected to the neck between mother and bud. The exocyst complex marks these areas of membrane addition, localizing to the bud tip of a growing daughter cell and the bud neck at the time of cytokinesis. Mutations in each member of the exocyst complex block the polarized trafficking that allows the bud to grow, but do not disrupt bud site selection. Thus, the exocyst complex in yeast may provide a model for the directed membrane traffic of developing cells in higher organisms (Murthy, 2004 and references therein).

The components of the exocyst complex, Sec3, Sec5, Sec6, Sec8, Sec10, Sec15, Exo70 and Exo84, are conserved from yeast to mammals. In multicellular organisms, though less extensively studied, these proteins are implicated in establishing cell polarity. In epithelial cells, E-cadherin mediated adhesion is sufficient to initiate the segregation of apical from basolateral membrane proteins. The exocyst localizes to these sites of adhesion and is required for the polarized transport of proteins to the basolateral domain (Grindstaff, 1998). Similarly, overexpression of Sec10 in MDCK cells increased delivery of a membrane protein to the basal-lateral, but not apical, domain (Lipschutz, 2000). Thus, as in yeast, the mammalian complex may mediate fusion at a spatially restricted domain (Murthy, 2004 and references therein).

Neurons generate their polarity by directing membrane traffic to growing neurites and growth cones, and by sorting proteins differentially between the axon and dendrites. The exocyst has a punctate distribution in processes of cultured primary hippocampal neurons and localizes near, but not precisely with, synaptic vesicles (Hsu, 1996; Kee, 1997). A role for the complex in mammalian synapse formation and neurite outgrowth has been suggested by the early presence of Sec6/Sec8 immunoreactivity at sites of synaptogenesis in culture (Hazuka, 1999) and the impairment of neurite outgrowth with dominant-negative forms of sec10 and sec8 in PC12 cells (Vega, 2001). Though abundant in adult brain, the function of the complex in neurons after development remains uncertain. The exocyst, like SNARE complexes, may be needed for all fusions at the plasma membrane both in developing and mature neurons, or it may be required only for particular forms of traffic (Murthy, 2003 and references therein).

Thus far, resolving the neuronal function of the exocyst complex has been hampered by the absence of a functional genetic analysis. In the mouse, a sec8 mutant has been identified, but these mice die shortly after gastrulation of the embryo, precluding a more detailed analysis of the role of the complex (Friedrich, 1997). sec5 mutations have been identified and characterized in Drosophila in order to delineate the role of the protein in neurons and particularly at synapses. This protein is required for many aspects of membrane traffic within neurons, including the elaboration of neurites, but the release of neurotransmitter at the synapse is independent of this exocyst component (Murthy, 2003).

Within neurons, multiple pathways are known to transport proteins and transmitters to the cell surface. In some cases, the exocytotic vesicles derive from the Golgi apparatus and consist of newly synthesized materials, while in others the vesicles derive from the plasma membrane and cycle locally. Another distinction that has been drawn contrasts the constitutive and the regulated pathways to distinguish the ongoing transport of protein and lipid to the cell surface from the ability to secrete hormones and transmitters in response to specific stimuli. Subsequently, further distinctions in trafficking pathways have been invoked to explain different forms of regulation and fusion at different domains of the cell. Two important pathways were examined in the current study: (1) a basic pathway of exocytosis that brings new proteins to the cell surface and permits the cell to grow, and (2) synaptic transmission, a specialized form of exocytosis, regulated by Ca²⁺ entry, in which vesicles already present at synapses fuse with the membrane and recycle locally (Murthy, 2003).

The mechanistic basis for these distinct pathways remains obscure. Presently, all membrane trafficking steps are thought to possess a similar underlying mechanism for membrane fusion, and yet clear distinctions in these trafficking steps must also occur. In many cases, synaptic transmission is thought to consist of the basic exocytotic pathway with additional regulatory processes superimposed. This study has found a distinction between these two trafficking pathways in which the basic pathway depends on a protein, Sec5, that does not appear to be required for synaptic transmission (Murthy, 2003).

Mutations were isolated in sec5, a component of the Drosophila exocyst complex; the basic exocytotic pathway is impaired in these mutants, and yet vesicle fusion at the synapse persists. Despite a small amount of maternal protein remaining in the mutant at the end of its life span, dramatic phenotypes were observed with respect to membrane addition. Muscles lacking Sec5 do not increase their surface area and motor neurons do not increase their arborizations on the muscle. When cultured, neurons lacking Sec5 survived but proved incapable of extending neurites. This defect is not due to a general malaise of the mutant neurons: neurons at this time point were capable of synthesizing new proteins, conducting action potentials, and releasing neurotransmitter, and they are equally viable in culture. Instead, the likely mechanism for this deficit in cell growth is due to a specific defect in membrane traffic, as revealed in the trafficking assay with which the cell surface insertion of a newly made transmembrane protein was followed in first instar larvae. Endogenous proteins insert into the cell surface during the time the maternal contribution persists, however, when CD8-GFP is induced only after the maternal contribution is no longer functionally adequate, CD8-GFP fails to accumulate in the plasma membrane. These dramatic phenotypes in membrane addition are consistent with data in yeast (Novick, 1980, 1981), where exocyst mutants fail to add new membrane during cell division and accumulate vesicles in the presumptive daughter bud (Murthy, 2003).

What step in membrane trafficking requires Sec5? Members of the exocyst complex have been shown to be recruited to budding vesicles in the trans-Golgi network (Liljedahl, 2001), and antibodies to Sec6 and Sec8 are reported to interfere with the ability of these proteins to exit the Golgi (Yeaman, 2001). In the sec5 null mutants, however, normal levels of synthesis of the CD8 and Synaptotagmin-GFP constructs were observed. They were both distributed throughout the cytoplasm and transported down the axon with equal efficiency to wild-type. The requirement for Sec5 would therefore appear to lie at a later step in membrane traffic, most likely at docking to and fusion with the plasma membrane. This finding is consistent with the phenotype of exocyst mutants in yeast (Novick, 1980, 1981), where vesicles are appropriately transported into the bud but fail to fuse with the membrane (Murthy, 2003).

In marked contrast to the arrest of growth and the impairment of the pathway for membrane protein addition, synaptic transmission continued to be robust. Each action potential that arrived at the neuromuscular junction of a 96 hr AEL sec5^E10 homozygous larva caused an average of 33 ± 8 vesicles to fuse. Though Sec5 protein declined between 24 and 96 hr in these null mutants, the quantal content increased and, if anything, attained a level slightly higher than would be expected for a junction with the observed number of boutons. This persistence of vesicle fusion cannot be attributed to a residuum of maternally derived Sec5, since there were severe defects in the other assays conducted at these times. At the same time point that the larvae are capable of repeatedly secreting numerous synaptic vesicles, the lack of Sec5 so effectively prevented net membrane addition that almost no neurites could be extended and that little CD8 was successfully expressed on the cell surface. Thus, the synaptic response in larva lacking Sec5 would appear to derive from the combination of a defect in neuronal growth and an unaltered capacity for synaptic vesicle fusion. The growth defect prevented the creation of the normal number of boutons and release sites for a 96 hr animal; however, those boutons that had formed remained fully functional (Murthy, 2003).

What is reported for mutations in sec5 is likely to represent the function of the entire exocyst complex. From biochemical studies in yeast and mammalian cells, the component proteins of the exocyst are found primarily in a complex (Guo, 1999b; Hsu, 1996; TerBush, 1995). This model is consistent with the similarities of the mutant phenotypes for each component of the complex in yeast. Moreover, Sec5p binds in the center of the complex (Guo, 1999a), and its absence causes Sec10p and Exo84p to be dissociated from Sec8p (Guo, 1999b). It is hypothesized that loss of this individual component will be equivalent to loss of most or all exocyst function (Murthy, 2003).

Heretofore, synaptic transmission could be viewed as an elaboration of the basic exocytosis pathway, in which regulatory mechanisms are superimposed on a scaffold equivalent to the exocytotic machinery of yeast. Thus far, all of the genes studied at the nerve terminal that are orthologs of genes isolated in yeast for defects in the final steps of exocytosis have essential roles in synaptic vesicle fusion (Lin, 2000; Rothman, 1994). These genes include the proteins of the SNARE complex, syntaxin (yeast Sso1p), SNAP-25 (yeast Sec9p), and VAMP (yeast Snc1p), as well as nsec1 (yeast Sec1p), rab (yeast Sec4p), NSF (yeast Sec17p), and α-SNAP(yeast Sec18p). The same screen in yeast identified six of the eight members of the exocyst complex, but, as described above, the exocyst appears to be an exception to this established pattern: it is essential for exocytosis in yeast but dispensable for synaptic vesicle fusion (Murthy, 2003).

Why is vesicle fusion at the synapse distinct from the vesicle fusion required for neurite outgrowth or the addition of CD8 in the membrane? The data indicate that the synapse has evolved away from a need for the exocyst in fusion. In yeast, the exocyst complex is known to be required before the assembly of the SNARE complex and the fusion of secretory vesicles with the plasma membrane (Grote, 2000), which would place the exocyst complex upstream of the final stage of vesicle fusion. At the synapse, the complicated cytoskeletal network at the active zone may take the place of the exocyst complex in tethering vesicles at the release site and priming them for fusion. Such active zone proteins as Munc-13 and RIM, likely regulators of SNARE complex formation, which do not have homologs in yeast, may fulfill at the active zone the role assigned to the exocyst elsewhere in the cell. A second possibility is that the fundamental distinction between exocyst-dependent and-independent pathways in neurons depends on the source of the vesicle; the local cycling of vesicles not derived primarily from the TGN, but directly from the plasma membrane or through an endosome, may be independent of the exocyst complex. In this case, exocyst-independent fusions of endosome-derived vesicles may occur at many places in the cell and not just at synapses. However, TGN-derived vesicles, exemplified in this study by the CD8-GFP containing vesicles, would require the exocyst. It remains to be determined whether peptide-containing dense core vesicles, which fuse in a regulated manner like synaptic vesicles, but which, unlike synaptic vesicles, bud directly off of the TGN, can fuse with the membrane in the absence of Sec5. A recent study (Moskalenko, 2002) may point in this direction; inhibition of Ral function, which binds to the N terminus of Sec5, blocks the stimulated release of human growth hormone, contained in dense core vesicles, from PC12 cells (Murthy, 2003).

After the run down of maternal contribution in sec5 mutants, newly synthesized vesicles labeled with Synaptotagmin-GFP do not concentrate at the synapse. Though the boutons contain a large population of preexisting synaptic vesicles, as evidenced by immunostaining with the vesicle marker CSP and by the electrophysiology of the terminals, the boutons do not accumulate Synaptotagmin-GFP. This is attributed to the nature of the biosynthetic pathway for synaptic vesicles. The current model is that transport vesicles containing synaptic vesicle proteins fuse with the plasma membrane via the constitutive pathway of exocytosis from the TGN. Thereafter, they are internalized and sorted in endosomes to become mature synaptic vesicles. This model is supported by studies following the movement of radioactively labeled synaptophysin in neuroendocrine cells. It seems likely that in sec5 mutants, transport vesicles have not fused with the membrane and are not matured into synaptic vesicles. Therefore, the Synaptotagmin-GFP, though transported down and visible in the axon, does not enter the pool of mature synaptic vesicles that are clustered in the bouton. The transport vesicles that fail to fuse with the plasma membrane likely remain in the axon or are sent retrogradely back to the soma (Murthy, 2003).

Neurite outgrowth fails in the absence of Sec5. This role for the exocyst complex was predicted from its localization to the tips of neurites in cultured hippocampal neurons (Hazuka, 1999) and from studies with a dominant-negative sec10 construct in cultured PC12 cells (Vega, 2001). Neurite outgrowth requires a substantial increase in the surface area of the neuron and, not surprisingly, requires membrane addition. This membrane addition has been shown primarily to occur at the growth cone rather than in the soma or along the length of the neurite. Thus, as in yeast, the localization of the exocyst complex, both in Drosophila and hippocampal cultures (Hazuka, 1999), is consistent with a role in targeting vesicles to the site of membrane addition (Murthy, 2003).

The surface expression of CD8 in neuronal cell bodies and along the axon is impaired by loss of Sec5. These defects in traffic to the cell surface have parallels in studies of the MDCK epithelial cell line. In MDCK cells, the exocyst complex localizes to tight junctions between cells, which are thought to be the sites of membrane addition. Antibodies to Sec8 block the delivery of proteins to the basolateral membrane, with little effect on apical membrane addition. With regard to the trafficking and sorting of proteins, parallels have been drawn between the neuronal axon and the apical domain of epithelial cells. In the case of Sec5, however, that equation would not appear to apply, in that removing Sec5 affects surface expression of CD8 along the axon and in nerve terminals as well as in the soma. Alternatively, there may have been an exocyst dependence of apical transport in MDCK cells in the Grindstaff (1998) study that was resistant to the blockade by the antibody protocol. The examination of polarized transport in the epithelia of the sec5 mutants may clarify the question of the exocyst complex and apical transport (Murthy, 2003).

Intriguingly, the motor neurons of sec5 mutants failed to form new synaptic boutons after the maternal contribution of Sec5 was gone. This observation, and the apparent requirement for the exocyst complex in membrane addition but not synaptic transmission, suggests that sec5 and the exocyst complex may play a very particular role in synaptic plasticity. Anatomical plasticity, the formation of a distinct synaptic contact between a neuron and its targets, may require the exocyst protein to be targeted to the site from which the new branch must sprout. By recruiting new membrane and synaptic components to this area, the exocyst may permit the growth of a new synapse that will subsequently function independently of the exocyst. The identification of mutations in sec5 and other components of the complex should permit tests of this hypothesis (Murthy, 2003).

GENE STRUCTURE

cDNA clone length- 2878 bases

Bases in 5' UTR - 64

Exons- 4

Bases in 3' UTR- 129

PROTEIN STRUCTURE

Amino Acids- 894

Structural Domains

Information about Sec5 structure can be found at the SCOP database.

date revised: 2 April 2004

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.