Divergence Of Small Gtpase Pathways In Dendrites

Actin nucleation may also be achieved by rac, through homologs of the WASP family termed WAVE/Scar proteins. WAVE proteins do not possess a rac-binding motif, and therefore require intermediate factors. One such intermediate is the formation of a pentameric complex consisting of the rac-target Sra1/CYFIP1, together with Nap, Abi, HSPC300 and WAVE38. The relevance of this complex in dendrites is not known yet (however, see below and Figure 17.2; Colorplate 10). The other possibility for rac to activate WAVE is via the insulin receptor substrate of 53 kDa (IRSp53) which was shown by Miki et al.39 to bind and control of WAVE, thereby initiating actin nucleation in vitro. IRSp53 is prominently expressed in neurons where it is a major constituent of the postsynaptic density40, indicating that the IRSp53/WAVE pathway is likely to be involved in rac-dependent control of spine morphology. IRSp53 is targeted to synapses by interaction with PSD-9541-43. By siRNA and expression of dominant negative constructs, Choi et al. demonstrated that IRSp53 contributes to spine formation by increasing linear spine density on dendrites, and rac/IRSp53/WAVE2 signaling may provide the necessary amount of actin nucleation to achieve this. However, the example of IRSp53 also illustrates the complexity of molecular events, as its SH3 domain binds not only to WAVE2, but also to Mena (implicated in formation of filopodia44), Eps8 (a barbed end capping protein45 ), and shank (a postsynaptic scaffold promoting spine formation and recruitment of postsynaptic components; refs. 46,47; see Figure 17.2; Colorplate

10). Interaction with Mena and shank does not depend on activation by rac, but instead requires GTP-bound cdc42 which binds to a different site on IRSp53. This promiscuity in terms of upstream activators as well as downstream effectors makes it difficult to clearly identify a single signaling pathway. Instead, it is likely that IRSp53 fulfils multiple functions depending on the local and temporal availability of activators as well as effectors. Thus, Mena might be a signal for filopodia formation on dendrites in early phases of development, as it is expressed only transiently during neuronal development. The N-terminal actin bundling domain of IRSp53 would support this process48. In later phases, shank might replace Mena as ligand for the SH3 domain, leading to increased assembly of PSD proteins. Consistently, co-expression of shank1 with IRSp53 in heterologous cells interferes with the formation of filopodia induced by the expression of IRSp53 alone46.

Several studies which highlight the effect of rho family proteins on spine formation (presumably via regulation of the actin cytoskeleton) demonstrate that these GTPases should also be involved in postsynaptic complex formation, which is usually visualized as clustering of PSD-95. Thus kalirin expression is not only required for the maintenance of spines, but also for the maintenance of the pre- and postsynaptic protein complexes, visualized as PSD-95 and bassoon clusters49. So far it is unclear whether formation of PSD-95 clusters is secondary to achievement of a certain status of actin polymerization and branching in the spine, or whether the activity of small GTPases also has a direct effect on clustering or association of PSD scaffolds. The cdc42-dependent association of IRSp53 with shank indicates that small GTPases can have a direct effect on postsynaptic protein assembly. However, it remains to be shown whether other postsynaptic scaffolds are influenced by GTPases in a similar manner.

Figure 17.2. Possible Signaling Pathways Used by the Small GTPases rac and cdc42 in Neuronal Dendrites. Activation is denoted by a black arrow, inhibition by an interrupted black line. The final downstream effect of each pathway is indicated by a blue arrow. Note that for IRSp53, activation of Eps8 and WAVE has been described only in connection with rac signaling, whereas linkage to shank or Mena has been described after activation by cdc42. Several known effectors were omitted for clarity. See Colorplate 10.

Figure 17.2. Possible Signaling Pathways Used by the Small GTPases rac and cdc42 in Neuronal Dendrites. Activation is denoted by a black arrow, inhibition by an interrupted black line. The final downstream effect of each pathway is indicated by a blue arrow. Note that for IRSp53, activation of Eps8 and WAVE has been described only in connection with rac signaling, whereas linkage to shank or Mena has been described after activation by cdc42. Several known effectors were omitted for clarity. See Colorplate 10.


Shank proteins, like other multidomain postsynaptic scaffolds, link receptors and the cytoskeleton. The role of shank proteins for the generation of dendritic spines and as important scaffolds within the PSD is described in Chapter 18. Remarkably, overexpression of shank induces enhanced structural and functional maturation of the postsynaptic receptor complex, concomitant with an earlier morphological maturation of dendritic spines50,51. In aspiny neurons, overexpression of shank is sufficient to induce spine-like structures52, suggesting that shank is required in the transition from immature dendritic filopodia, which do not carry postsynaptic specializations, to proper, mushroom-shaped, postsynaptic spines. Thus the regulation of shank should be of crucial importance during synapse formation. Whereas so far no signaling mechanisms have been identified which act on the shank proteins themselves, an interesting feature of all three shank family members is the prominent dendritic localization of their coding mRNAs53. This became evident in in situ hybridization experiments by a strong labeling obtained in the so-called molecular layers of the hippocampus and the cerebellum, i.e., regions which contain mostly dendrites, axons, and synapses, but few cell bodies. Labeling was observed in dendritic fields originating from neurons in the CA1-CA3 regions, as well as the dentate gyrus in the hippocampus, and from Purkinje cells in the cerebellum. Dendritic targeting of mRNAs could be reproduced in an expression model in cultured neurons, leading to the conclusion that a 200 bp targeting sequence within the shank1 3' untranslated region is sufficient for targeting of the mRNAs. It remains to be shown if local translation of shank mRNAs contributes to maturation of the postsynapse; intriguingly, during postnatal development of the cerebellum high levels of shank2/ProSAP1 mRNA in the molecular layer were observed at a phase of intense synaptogenesis, i.e., during the second week post partum53.

The number of other mRNAs coding for PSD proteins which are localized in dendrites is rather limited; the most notable example is the a-subunit of the Ca2+/calmodulin-dependent protein kinase (aCaMKII); dendritic presence of this mRNA is required for synaptic plasticity, as detected in long-term potentiation paradigms54. In addition, mRNA coding for the arg3.1/arc gene product, a protein of unknown function, is rapidly induced and transported into distal dendrites upon neuronal activity55. Out of the multidomain structural proteins of the PSD, mRNAs coding for homer256 and the SAPAP/GKAP isoform 3 (SAPAP3 57,58 ) have been found in hippocampal dendrites besides the shank mRNAs mentioned above. Based on sophisticated PCR approaches and molecular imaging techniques, claims have been made for the dendritic localization of many more mRNAs. However, confirmation of their dendritic localization by in situ hybridization either on brain sections or cultivated neurons is missing, suggesting that only a rather small (but growing) subset of mRNAs may be translated in dendrites in significant quantities.

The concept of dendritically localized mRNAs, which may be translated "on demand" requires strong translational control mechanisms, so that translation should be suppressed during transport, and only be allowed upon a certain stimulus such as activation of cell surface receptors or Ca influx. Several imaging studies on dendritic RNA transport particles, as well as biochemical purification of these particles, now suggest that mRNAs are transported within huge RNP particles59. These include putative translational regulators such as the fragile X mental retardation protein (FMRP) and its homologs FXR1 and FXR2, transacting factors for RNA localization such as the RNA-binding proteins staufen, PUR, and elongation factor EF1a. One study implied that ribosomes are co-transported in these RNPs and that disassembly of the particle on a depolarizing stimulus would allow for translation of the transported mRNAs by these ribosomes60.


As protein synthesis is tightly regulated in dendrites, it probably does not come as a surprise that the same holds true for protein degradation. Several studies have now begun to illuminate how neurons creatively use the forces of protein destruction in order to restructure their synaptic connections. Depending on previous neuronal activity, individual components of the PSD are selectively targeted for destruction by ubiquitinylation, followed by proteolysis in the 26S proteasome. One example is Spar, a GAP for Rap GTPases (i.e., a RapGAP). Spar is an actin-binding protein which is present in spines and contributes significantly to spine morphology25. Strong neuronal activity induces the expression of the protein kinase SNK which phosphorylates Spar and thus marks it for proteasome-dependent degradation61. Ehlers4 systematically analyzed changes in the protein composition of the PSD from cultivated neurons which were induced by electrical activity. By pharmacological inhibition of proteasome activity and pulldown assays with polyubiquitin-binding proteins, he could identify a subgroup of PSD scaffold proteins which were targeted for degradation. Among these are the A-kinase anchoring protein AKAP, as well as shank and GKAP/SAPAP proteins. GKAP/SAPAP proteins are physically associated with shank, and both are believed to constitute some core fraction of PSD scaffold proteins. Selective removal of both proteins may allow for restructuring of the PSD. Interestingly, both shank and SAPAP3 mRNAs are localized to neuronal dendrites, suggesting that the expression or the turnover of these interacting proteins may be co-regulated. Further work is required to sort out which signaling pathways contribute to selection of PSD proteins for ubiquitinylation. In particular, it will be interesting to find out which types of E3 ubiquitin ligases confer substrate specificity to the process of polyubiquitinylation at postsynaptic sites.

Supplements For Diabetics

Supplements For Diabetics

All you need is a proper diet of fresh fruits and vegetables and get plenty of exercise and you'll be fine. Ever heard those words from your doctor? If that's all heshe recommends then you're missing out an important ingredient for health that he's not telling you. Fact is that you can adhere to the strictest diet, watch everything you eat and get the exercise of amarathon runner and still come down with diabetic complications. Diet, exercise and standard drug treatments simply aren't enough to help keep your diabetes under control.

Get My Free Ebook

Post a comment