The exocyst is a complex of proteins originally identified in yeast that has been implicated in polarized exocytosis/secretion. Components of the exocyst have been implicated in neurite outgrowth, cell polarity, and cell viability. An exocyst component, sec15, has been isolated in a screen for genes required for synaptic specificity. Loss of sec15 causes a targeting defect of photoreceptors that coincides with mislocalization of specific cell adhesion and signaling molecules. Additionally, sec15 mutant neurons fail to localize other exocyst members like Sec5 and Sec8, but not Sec6, to neuronal terminals. However, loss of sec15 does not cause cell lethality in contrast to loss of sec5 or sec6. The data suggest a role for Sec15 in an exocyst-like subcomplex for the targeting and subcellular distribution of specific proteins. The data also show that functions of other exocyst components persist in the absence of sec15, suggesting that different exocyst components have separable functions (Mehta, 2005).

Exocyst components have been implicated in different developmental contexts where they appear to function in polarized exocytosis. Sec5 and Sec6 have been implicated in neurite growth, synaptic transmission, and polarization of the Drosophila oocyte (Beronja, 2005; Murthy, 2003; Murthy, 2004; Murthy, 2005). Furthermore, Sec6 was shown to regulate apical exocytosis in photoreceptor cells (Beronja, 2005). As with loss of sec15, loss of sec5, sec6, or sec15 function similarly disrupts the localization of Drosophila E-Cadherin, which accumulates in an enlarged recycling endosome (RE) compartment. Exocyst components play a role in regulating the localization of E-Cadherin in epithelial cells and the delivery of E-Cadherin from the REs. Sec5 regulates the recycling of basolateral E-Cadherin to the apical AJs. While a connection between exocyst and the Rab11 recycling endosomal compartment has been proposed both in Drosophila and mammalian cells, the characterization of sec15 phenotype demonstrates the functional significance of this interaction. Texocyst has been viewed as an important regulator of exocytosis, but the mechanisms by which the exocyst complex could regulate the exocytosis of specific cargoes to specific domains within the cell have remained poorly understood. A model is proposed whereby the interaction between Sec10 and Arm enables the exocyst to regulate E-Cadherin trafficking in Drosophila (Mehta, 2005 and references therein).

Sec15 plays multiple roles in recycling membrane components: (1) Studies with sec15 mutants point to a role of sec15 in recycling of Delta thus promoting Notch signaling during the asymmetric division of Drosophila sensory organ precursors (Jafar-Nejad, 2005); (2) Loss of function of sec15 in epithelial cells results in DE-Cad accumulation in an enlarged recycling endosomal compartment and inhibits DE-Cadherin delivery to the membrane (Langevin, 2005); (3) In fly photoreceptors Sec15 colocalizes with Rab11, acts as an effector of Rab11, and loss of Sec15 affects rhabdomere morphology (Wu, 2005); finally (4) sec15 is required for the delivery of specific cell adhesion and signaling molecules required for the establishment of synaptic specificity after the growth cones reach their target regions (Mehta, 2005)

More than twenty-five years ago, Novick (1980) identified 23 temperature-sensitive mutant complementation groups that caused a secretory defect in yeast. Many of the SEC mutants isolated in this work have now been studied in more detail (Schekman, 2004). Much is known about some of these genes, including SEC9, a SNAP-25 homolog, SEC17, an α-SNAP homolog, and SEC18, an NSF homolog, because they have been implicated in numerous secretory processes in yeast and many metazoans (Bennett, 1993). Another subset of SEC genes (SEC3, SEC5, SEC6, SEC8, SEC10, and SEC15) has been shown to encode members of a large protein complex, the exocyst or Sec6/8 complex, which also includes Exo70p and Exo84p. Although they have been studied quite extensively in yeast, their role in metazoans is ill-defined due to a dearth of functional studies (Mehta, 2005 and references therein).

In yeast, exocyst proteins are required for polarized exocytosis of secretory vesicles. Sec3p localizes to the membrane even when other exocyst members are lacking and is thought to represent a spatial landmark. Sec15p interacts with the secretory vesicle-associated rab GTPase, Sec4p, in its GTP bound form. A further interaction of Sec15p with Sec10p is thought to recruit other exocyst components and connect the complex to Sec3p (Guo, 1999). Hence, Sec15p is thought to target the vesicle to the correct exocytic site (Mehta, 2005).

In multicellular organisms, the exocyst proteins were found to be present in brain as well as most other tissues. In nonpolarized epithelial cells (Madin-Darby canine kidney [MDCK] cell culture), the complex is found in a soluble form in the cytoplasm. Upon cell-cell contact, it relocalizes to the interacting plasma membranes. After disruption of E-cadherin-mediated cell-cell contact, the complex dissociates again from the plasma membrane. In the same epithelial cell line, antibodies against Sec8 specifically inhibit vesicle delivery to the basolateral but not the apical membrane. These data suggest that the recruitment of the exocyst is a consequence of cell-cell adhesion and is essential for epithelial cell polarity (Mehta, 2005 and references therein).

In cultured hippocampal neurons, Sec6 immunoreactivity is present in the growth cone during neurite outgrowth as well as in periodic punctae on the axon prior to synaptogenesis. After formation of stable synapses, the Sec6 immunoreactivity disappears, suggesting that Sec6 protein and the corresponding complex are not required in mature synapses. In addition, overexpression of a dominant-negative Sec10 protein blocks neurite outgrowth in cultured PC12 cells. Taken together, the data from yeast and mammalian cells suggest that the complex may play a role in growth cone extension and possibly synaptogenesis (Mehta, 2005 are references therein).

To date, the only published mutants providing in vivo data in metazoans are knockouts of sec8 in mouse and sec5 in Drosophila. The sec8 knockout results in lethality at E7.5, precluding analyses of neuronal development. Loss of sec5 in Drosophila reveals no defect in neurotransmitter release, and larvae die soon after their maternal protein contribution is exhausted (Murthy, 2003). In cell culture, loss of sec5 blocks neurite outgrowth and incorporation of newly synthesized transmembrane proteins into the membrane, in agreement with a proposed role for the exocyst in neurite outgrowth (Hsu, 1999: Murthy, 2003). In the developing oocyte, sec5 is required for membrane trafficking and polarization (Murthy, 2004). Finally, clones of sec5 in the eye do not form eye tissue. Since Sec5 is a core component of the complex, Murthy (2003) proposes that the phenotype associated with the loss of sec5 represents the function of the entire exocyst complex. No functional data in yeast or metazoans so far indicate any role of individual exocyst components independent of the entire complex, suggesting that sec5 mutant phenotypes represent a generic consequence of mutations in exocyst members (Mehta, 2005).

This study describes the isolation of Drosophila sec15 mutants in a forward genetic screen designed to identify genes that affect synapse development. In contrast to the cell lethality associated with sec5 mutations, sec15 mutant photoreceptor neurons are viable and display surprisingly specific defects in a distinct neuronal targeting step. Loss of sec15 does not cause defects in neurite extension, but leads to the formation of synapses between inappropriate partners, causing a loss of synaptic specificity. The data indicate that sec15 is required for the delivery of specific cell adhesion and signaling molecules required for the establishment of synaptic specificity after the growth cones reach their target regions. They also suggest a model in which subcomplexes of the exocyst perform separable functions (Mehta, 2005).

It is surprising how many developmental processes are unaffected by the loss of sec15: neuronal differentiation, axonal outgrowth, and initial target recognition of the correct brain areas all appear normal. Subsequently, a specific neuronal sorting process that ensures synaptic specificity is disrupted. After the defective neuronal targeting step, the developmental program proceeds rather normally with synaptic partner selection and synapse formation. While it is known that these developmental processes are genetically separable, they have primarily been associated with cell adhesion molecules (CAMs). Based on the finding of targeting defects for specific CAMs and signaling molecules, it is proposed that a vesicular transport mechanism exists to spatiotemporally target certain CAMs as well as other proteins. Notably, soluble NSF-attachment receptors (SNAREs), which are required for most if not all vesicle docking and fusion processes, are unlikely to convey much spatiotemporal targeting information. This is primarily because target-SNAREs are distributed uniformly over membranes, even though vesicle fusion is restricted to limited subdomains, as is also the case for the target-SNARE syntaxin in photoreceptor terminals (Mehta, 2005).

Further evidence for targeting defects of a vesicular cargo transport mechanism comes from vertebrate cell culture experiments. Disrupting exocyst function caused defects in basolateral, but not apical, targeting of proteins in epithelial cells (Grindstaff, 1998). The exocyst has also been implicated in the trafficking of GLUT4 transporters in response to insulin stimulation (Inoue, 2003). Sec8 is involved in NMDA receptor insertion at dendrites through an interaction with SAP102 (Sans, 2003). In Drosophila photoreceptors, the basolateral compartment is the axon (Izaddoost, 2002), while in vertebrate neurons, the basolateral compartment is dendritic. It is therefore possible that the same subcomplex of exocyst (and other) components is required for correct targeting in the presynaptic compartment of Drosophila photoreceptors and vertebrate postsynaptic receptor targeting (Mehta, 2005).

The investigation of the establishment of specific synaptic contacts has largely been focused on CAMs and their regulating/modifying proteins, since these molecules are thought to convey information about spatiotemporal specificity. The finding of CAM misregulation in sec15 mutant neurons links the spatiotemporal regulation of CAMs to the cell biology of the neuron. Although vesicle trafficking of CAMs was not studied directly in this work, other studies in both yeast and vertebrate systems provides evidence that Sec15 and other exocyst members are associated with vesicles and play a role in vesicle trafficking (Ang, 2004 and Boyd, 2004). It is proposed that specificity may be established through unique vesicular trafficking mechanisms in addition to the transcriptional and posttranslational regulation of various isoforms of CAMs in certain cells at distinct times. Much is known about the specificity of the vesicular trafficking machinery at the ER and Golgi, where distinct cargoes need to be specifically targeted. In neurons, numerous synaptic proteins, including the CAM N-cadherin are specifically targeted to the active zones and other synaptic domains via a specialized vesicular transport mechanism. In Drosophila photoreceptors, the misregulation of CAMs prior to synapse formation has been observed in neurons that lack n-synaptobrevin, and specific vesicles for the targeted transport of synaptic components have been described in mammals. It is not known whether a distinct type of vesicle exists for the transport of CAMs and signaling molecules to the developing terminal, but it is proposed that specific subsets of CAMs are transported and integrated into the membrane by Sec15. This is not to say that synaptic specificity is the only role of Sec15 or the CAMs that were examined. The data shows mislocalization of CAMs in both the cell body as well as axons of sec15 mutant photoreceptors. The consequences of this cell body mislocalization were not investigated, except to make sure that such mislocalization does not impact cell viability, neuronal differentiation, axonal outgrowth, and initial target recognition of the correct brain areas by these photoreceptors. In this respect, it is interesting to note that some of the CAMs examined (N-cadherin, Flamingo) were correctly targeted in mutant photoreceptors while others (Dlar, Fas2, IrreC-rst) were not. Although the in vitro inhibition of microtubule polymerization by exocyst members (Wang, 2004) and defects in sec5 mutants (Murthy, 2003) with neurite extension may argue for a role in general protein or membrane trafficking for the exocyst, the curren data argue that this is not the case for Sec15. In addition to the correct localization of the CAMs mentioned above, sec15 mutant neurons are viable, extend axons, and assemble functional synapses. It seems unlikely that a loss of protein required for general protein trafficking would not affect these processes. In summary, the data describe a defect in synaptic specificity and concommitant mislocalization of proteins known to affect synaptic specificity in photoreceptors lacking Sec15. These defects can be explained by a Sec15-dependent intracellular vesicle trafficking mechanism for certain molecular components that are required for the establishment of synaptic specificity (Mehta, 2005).

Most studies consider individual exocyst components representative for the whole complex and assume that the entire eight member complex is responsible for the different roles proposed or demonstrated for the exocyst. Since mutations have been isolated in sec15, Drosophila is the first metazoan in which two independent gene disruptions of exocyst components allowed tests of this assumption. Furthermore, a gene disruption of Drosophila sec6 has recently been generated and, like sec5, observed to cause cell lethality in homozygous mutant eyes. The remarkable difference between sec5, sec6, and sec15 mutant phenotypes questions the idea of a mutation in an individual member having a 'generic exocyst' phenotype. One possibility that may explain this discrepancy would be an exocyst-independent function of sec15. The findings that Sec5 and Sec8 are mislocalized in sec15 mutant neuronal terminals argue against this possibility. Instead, it suggests that Sec15 participates with Sec5 and Sec8 in a specific developmental process. However, the sec15 phenotype is notably more specific than the sec5 or sec6 cell lethality. While Sec5 is suggested to have a role in general membrane trafficking (Murthy, 2003), the large number of normal processes observed in sec15 mutant neurons (cell viability, axon outgrowth, axon guidance, neurotransmitter release, etc.) argues against this being the case for Sec15. This implies that at least Sec5 and perhaps Sec8 sustain another essential function in the absence of Sec15. For the Sec15 interaction partner Sec10, a knockdown using RNAi as well as Sec10 overexpression were found to have no detectable phenotype in photoreceptors (Andrews, 2002). However, Sec10 overexpression in sec15 mutant photoreceptors causes cell lethality, indicating a genetic interaction. Moreover, the finding that Sec6 is normally localized in the absence of Sec15 is puzzling. Close investigation of the expression pattern in lamina cartridges revealed a clear postsynaptic localization of Sec6. This finding together with the differential expression patterns of Sec5, Sec6, Sec8, and Sec15 during development and adulthood argue against a single functional complex in Drosophila neurons. Similarly, immuno-EM in rat hippocampal neurons revealed that Sec6 localizes to secretory vesicles, while Sec8 has a diffuse, cytoplasmic distribution (Vik-Mo, 2003). Another precedent for subcomplexes of the exocyst having different functions can be found in the COG complex. The COG, exocyst, and GARP complexes share distant homology in the N termini of their subunits and are placed in a family of 'quatrefoil complexes' by Whyte because all three complexes have multiples of four subunits (Whyte, 2001). Distinct COG subcomplexes are proposed to act in intra-Golgi vesicle recycling and endosome to Golgi recycling (Whyte, 2002). A similar situation may exist for the exocyst, where subcomplexes are involved in trafficking different populations of vesicles. Although it is not known which assemblies of exocyst components exert which functions in vivo, the current results suggest that at least two different functional compositions of the known exocyst components exist. Given that Sec5 is considered the core member of the complex (Guo, 1999), loss of sec5 may represent either the loss of all exocyst functions as suggested by Murthy (2003), or of at least one essential function. Either way, the cell lethality associated with the loss of sec5 or sec6 masks possible later more specific functions of subcomplexes. Hence, the data indicate that a detailed analysis of the many proteins that are thought to be required for exocyst function will have to be initiated in order to understand their precise role in metazoans (Mehta, 2005).

GENE STRUCTURE

cDNA clone length - 2616 bp

Bases in 5' UTR - 71

Exons - 3

Bases in 3' UTR - 244

PROTEIN STRUCTURE

Amino Acids - 766

Structural Domains

Sec15 plays multiple key roles in a variety of processes in exocytosis by interacting with vesicle-associated small Rab GTPases, assisting in targeting vesicles to budding sites and recruiting, by way of Sec10, other components of the exocyst. in vitro and in vivo studies have shed light on some of these roles. The relatively large size (85 kDa) of Sec15 ensures that there are enough docking sites, each probably residing in either a single domain or a combination of domains, for other protein components necessary for its diverse functions. A segment named the C-terminal domain, containing nearly the entire C-terminal half of the protein binds to a set of Rabs in a GTP-dependent manner. Its atomic structure is composed of two distinct subdomains, only one of which (the C subdomain) harbors the Rab-binding site. The finding of a bipartite C-terminal domain was unexpected; there was no hint of this feature even from the BLAST Conserved Domain Database search for domains. The all-helical structure of the domain, with its two different subdomains, has a novel fold (Wu, 2005).

date revised: 12 March 2006

Home page: The Interactive Fly © 2006 Thomas Brody, Ph.D.

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