Synapses are specialized intercellular junctions that are formed when a pre-synaptic terminal contacts a post-synaptic neuron. These pre- and post-synaptic connections are mediated by numerous cell adhesion molecules (CAMs). Several CAMs undergo proteolytic shedding of their extracellular NH2-terminal domains and a subsequent intramembranous cleavage event mediated by presenilin-dependent gamma-secretase (Kim et al. 2002; Marambaud et al. 2003; Maretzky et al. 2005a,b; Uemura et al. 2006). This process may be enhanced with neuronal activity (Tian et al. 2007; Conant et al. 2010; Kim et al. 2010). Ectodomain shedding and presenilin-dependent gamma-secretase cleavage of synaptic CAMs would comprise a rapid and elegant means by which neurons might remodel spine structure in response to synaptic transmission. This change ultimately leads to long-term changes in synaptic function, which are required for higher order processes in the brain such as learning and memory (Shiosaka 2004; Zhang et al. 2005; Lee et al. 2008).
Nectin is a Ca2+-independent, immunoglobulin-like adhesion molecule involved in various cell–cell adherens junctions (Takai and Nakanishi 2003; Takai et al. 2003). Nectins are composed of four members – nectin-1, 2, 3 and 4. In the CNS, nectin-1 and 3 localize at the pre- and post-synaptic sides of puncta adherentia junctions (PAJs) formed in the CA3 pyramidal region of adult mouse hippocampus (Mizoguchi et al. 2002). At the synapse, nectin co-localizes with afadin, an F-actin binding protein (Mizoguchi et al. 2002; Lim et al. 2008). The addition of nectin-1 or nectin-3 inhibitors to cultured rat hippocampal neurons alters the cellular distribution of synaptophysin and postsynaptic density 95 (PSD-95) and decreases the size, but increases the number, of synapses (Mizoguchi et al. 2002). Mutations in the nectin-1 gene cause cleft lip/palate-ectodermal dysplasia and, in some cases, mental retardation consistent with a role in the development of ectodermally derived tissues (Suzuki et al. 2000; Sozen et al. 2001). While nectin-1 −/− and nectin-3 −/− mice show no dramatic organismal phenotypes (Inagaki et al. 2005), both mutant mice exhibit an abnormal mossy fiber trajectory and a reduction in the number of synaptic PAJs between the mossy fiber terminals and the dendrites of the CA3 pyramidal cells (Honda et al. 2006). Hippocampal neurons derived from nectin-1-null mice exhibit a reduction in spine head width and an increase in spine length (Togashi et al. 2006). Nectin-1 is initially expressed at excitatory and inhibitory synapses but is progressively lost at inhibitory synapses during their maturation (Lim et al. 2008). These data suggest that nectin plays an important role in synaptogenesis.
Nectin-1 undergoes ectodomain shedding by alpha-secretase and subsequent proteolytic processing by gamma-secretase (Kim et al. 2002, 2010, 2011; Tanaka et al. 2002). Ectodomain shedding and intramembrane cleavage of nectin-1 occur in both pre-synaptic and post-synaptic compartments in constitutive and regulated manners (Kim et al. 2010, 2011). The activity-dependent cleavage of nectin-1 mediated by alpha-secretase requires an influx of Ca2+ through NMDA receptors and an activation of calmodulin in mature cortical neurons (Kim et al. 2010). ADAM10 is the major alpha-sheddase responsible for nectin-1 ectodomain cleavage in neurons and in the brain (Kim et al. 2010).
Currently, it is not well understood how ectodomain shedding of synaptic adhesion molecules modulates synapse formation and synaptic plasticity. In this study, we examined the roles of nectin-1 shedding in synaptogenesis in cultured hippocampal neurons. Through a series of truncation mutants and alanine scanning point mutants, we identified constructs that were refractory to ectodomain shedding. By expressing these point mutants, we dissected the role of nectin-1 shedding in synaptogenesis. The results of our mutational analyses indicate that shedding of nectin-1 plays a role in the regulation of the density of dendritic spines. These observations support the critical role of nectin-1 in the formation of synapses and suggest that it functions as a regulator of synaptic plasticity through its modulation of synaptic connections.
- Top of page
- Materials and methods
- Supporting Information
In this study, we showed that the processing of nectin-1 occurs by multiple endoproteolytic events both in vivo (Fig. 1b) and in vitro (Fig. 1d), and that the shedding of nectin-1 is developmentally regulated in mouse brains (Fig. 1b), implying the importance of nectin-1 processing during brain development. Nectin-1 localizes to both post-synaptic and pre-synaptic membranes (Figure S2), and ectodomain shedding and intramembrane cleavage of nectin-1 occurs in both compartments (Kim et al. 2011). Therefore, the proteolytic machinery formed by metalloproteases and gamma-secretase must be present and functioning in both sides of the synapse to cleave nectin-1. Recently, we reported that nectin-1 recruits membrane palmitoylated protein 3 (MPP3) to cell–cell contact sites, mediated by a postsynaptic density-95/discs large/zonula occludens-1 (PDZ)-binding motif at the carboxyl terminus of nectin-1 (Dudak et al. 2011). ADAM10 is one of the major sheddases responsible for ectodomain shedding of nectin-1 (Kim et al. 2010), resulting in the generation of a 30 kDa CTF. ADAM10 interacts with synapse-associated protein 97 (SAP97) (Marcello et al. 2007), which increases the cell surface expression of ADAM10. SAP97 interacts with MPP3 through the L27 domains of MPP3 (Karnak et al. 2002). It is possible that the nectin–MPP3 complex may play a role in the recruitment of the ADAM10-SAP97 complex to cell–cell contact sites, and that this recruitment of ADAM10, in turn, increases ectodomain shedding of nectin-1 and other cell adhesion molecules associated with nectin-based adhesion junctions. However, besides ADAM10, it is still unknown which secretases are responsible for the shedding of nectin-1 and whether these cleavage events are sequential or independent of one another in vivo.
Using molecular approaches, we identified the regions containing the two main cleavage sites in the extracellular domain of nectin-1 (Fig. 3b and c). By alanine scanning mutagenesis, we identified point mutations that disrupt nectin-1 cleavage, substantially reducing alpha-CTF and/or CTF-1 (Fig. 4b). Over-expression of wild type nectin-1 had no effect on the dendritic spine density, whereas the expression of cleavage resistant mutants altered the density of dendritic spines (Fig. 5a). These observations suggest that ectodomain shedding of nectin-1 is a regulatory step to turn off nectin function at synapses. At synapses, nectin-1 shedding mediated by ADAM10 is regulated by the activation of NMDA receptors (Kim et al. 2010) or by chemical long-term potentiation (Kim et al. 2011). These data suggest that secretase-mediated cleavage of synaptic adhesion molecules, such as nectin-1, may allow the activated synapse to undergo either an increase or decrease in spine size or density observed during induction of long-term potentiation and long-term depression.
Interestingly, three residues (T310, Y311, and S323) that most robustly affect nectin-1 processing contain hydroxyl groups. Simple removal of the hydroxyl group by alanine mutation suggests a critical role of the hydroxyl groups themselves, perhaps through participation in hydrogen bonds. These residues could play a critical role in intramolecular hydrogen bonding in positioning the substrate for proteolytic cleavage of nectin-1. However, it is not clear whether these mutations reduce interactions with secretases or simply reduce the ability of secretases to cut. Thus, further investigation is required. Nevertheless, our studies indicate the role of nectin-1 shedding in dendritic spine morphogenesis and perhaps in related synaptic functions.
It is known that soluble extracellular domains of CAMs play roles in various physiological functions (Herreman et al. 1999; Dihne et al. 2003; Kalus et al. 2003; Tian et al. 2007; Conant et al. 2010). Therefore, the soluble fragments of nectin-1 generated by multiple sheddases may also participate in various biological functions, since different soluble fragments interact with different ligands to generate distinct signals. Ectodomain shedding of nectin-1 generates at least two soluble ectodomains derived from either α- or CTF1-cleavage events. These two soluble forms may have different physiological functions based on the observation of APP processing. APP undergoes independent α- or β-secretase cleavage events, releasing two soluble forms of APP: APPsα or APPsβ respectively. APPsα exhibits neurotrophic and neuroprotective properties (Furukawa et al. 1996; Meziane et al. 1998; Stein et al. 2004), whereas APPsβ seems to have a proapoptotic function (Nikolaev et al. 2009). The nectin-soluble ectodomains may act as a signaling protein where they bind to their receptors and mediate signals in an autocrine or paracrine fashion. In addition to the signaling function, these nectin-soluble ectodomains could act as regulators in the postulated cell-cell interaction by binding to either full-length nectin-1 or -3 through homo- or hetero-trans-dimerization. It has been shown that a fusion protein composed of the ectodomain of nectin-1 and the Fc portion of IgG trans-interacts with cellular nectin-1 and nectin-3, and induces filopodia and lamellipodia by activating Rap1, Cdc42, and Rac small G-proteins through the activation of c-Src in Madin-Darby canine kidney (MDCK) and fibroblast cells (Kawakatsu et al. 2002, 2005; Honda et al. 2003a,b). Therefore, binding of nectin ectodomains to full-length nectins may generate similar signaling pathways, resulting in local junctional remodeling. Interestingly, nectin-1 has three isoforms: α, β, and γ. The β form is missing 59 amino acid residues including a conserved postsynaptic density-95/discs large/zonula occludens-1 (PDZ)-binding motif of four amino acid residues at the carboxyl terminus, whereas the γ form is the secreted protein, which lacks the transmembrane region and the entire C-terminus. However, the biological role of the γ form is completely unknown. The presence of nectin-1γ in nature highly suggests that receptor(s) may exist for the soluble nectins and that the association between receptor(s) and soluble nectins may play an important role in cellular functions. It will be interesting to determine what physiological roles the shed ectodomains of nectins and nectin-1γ play and whether they exhibit different roles in vivo. Future studies will be necessary to determine whether extracellular domains of nectin-1 play a role in synaptic plasticity.
In conclusion, nectin-1 ectodomain shedding is regulated by multiple sheddases in vitro and in vivo, and ectodomain shedding of nectin-1 is a regulatory step to turn off nectin function at synapses. In turn, this process modulates the maintenance of dendritic spine densities in rat hippocampal cultures.
- Top of page
- Materials and methods
- Supporting Information
Figure S1. Analysis of nectin-1 point mutations from amino acid residues 301 to 351. HEK 293 cells were transfected with each mutant. Twenty-four hrs after transfection, cells were lysed in reducing sample buffer and analyzed on 12% SDSPAGE. Samples were transferred to nitrocellulose, and the blots were probed with antinectin-1 cyto-specific antibody. All experiments were repeated five times.
Figure S2. Analysis of nectin-1 point mutants in neuro-2A cell lines. Neuro-2A cells were transfected with nectin-1 point mutations that alter nectin-1 processing in HEK 293 cells. The cells were collected in reducing sample buffer 24 hrs after transfection and analyzed by Western blotting. The blot was probed with antinectin-1 cyto-specific antibody.
Figure S3. Nectin-1 localizes to both pre- and postsynaptic sites. A. Neurons at 28 DIV were examined at the ultrastructural level by electron microscopy. Synapses were identified by the presence of thickened presynaptic and postsynaptic specializations with intervening dense material indicated by arrows. Nectin-1 immunostaining was observed both in pre- and postsynaptic sites using antibody against the intracellular domain of nectin-1. Over 30 synapses were examined and the immunoreactivity of nectin-1 was observed in all synapses. B. Nectin-1 immunostaining was abrogated by incubation of the antibody with its cognate antigenic peptide.
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.