DGK is distributed exclusively in the brain particularly
DGKβ  is distributed exclusively in the brain, particularly in the striatum, cerebral cortex, olfactory bulb, and hippocampus [9,10]. This isozyme accumulates at the perisynaptic sites of medium spiny neurons in the striatum . The GDC-0068 of DGKβ rapidly increases after 14 days of age, after which synaptic maturation begins to progress . Interestingly, Caricasole et al. reported that one of the splice variants of human DGKβ, which lacks the C-terminal region (CT), is associated with mood disorders . Moreover, DGKβ-knockout mice showed impairment in emotion responses (bipolar disorder remedy-sensitive mania-like behaviors) and long-term memory related to cognitive functions [13,14]. These results suggest that DGKβ is important for neuronal functioning and plays a role in several neuronal diseases. Furthermore, it has been reported that DGKβ induces neurite outgrowth/branching and spine formation [, , , ]. However, the precise molecular mechanisms underlying DGKβ-induced morphological changes remain elusive.
When DGKβ was overexpressed in COS-7 cells as a control experiment during studies of other DGK isozymes, we unexpectedly found that DGKβ strongly induced filopodium-like protrusion formation (see Fig. 1). Filopodium formation is known to play an important role in neurite outgrowth/branching [, , ] and spine formation . Therefore, the purpose of this study was to analyze the structure-function relationships of DGKβ in terms of filopodium formation. Moreover, we observed a correlation between the filopodium formation induced by DGKβ and its plasma membrane localization/F-actin colocalization. Furthermore, the interaction of DGKβ with chimaerins, which are Rac1-GTPases-activating proteins (GAPs) [22,23] that enhance filopodium formation, was determined.
Materials and methods
Discussion In the present study, we unexpectedly found that DGKβ strongly induced filopodium formation (Fig. 1). Moreover, the results indicate that CT, CD and C1D of DGKβ are essential for filopodium formation (Fig. 1 and Table 1). We also observed that the ability of DGKβ to induce the formation of filopodia (Fig. 1) is positively correlated with the extent of its plasma membrane localization and F-actin colocalization (Fig. 2, Fig. 3 and Table 1). The results suggest that DGKβ must be localized at the plasma membrane and colocalized with F-actin to induce filopodium formation. CT is essential for plasma membrane localization and F-actin colocalization of DGKβ. Moreover, CD and C1D play important roles for these events. Caricasole et al. reported that the 35-amino-acid, C-terminal region of DGKβ is necessary for its phorbol ester-induced translocation to the plasma membrane . Kano et al. reported that the C-terminal region of DGKβ is important for its plasma membrane localization in SHSY-5Y neuroblastoma cells . In addition to DGKβ, Merino et al. found that the C-terminal sequence of DGKα is essential for its binding to the plasma membranes of T-lymphocytes . Hence, it is possible that the C-termini of type I DGKs commonly function to localize these proteins to the plasma membrane. A kinase-dead mutant of DGKβ exhibited a market decrease in filopodium formation and plasma membrane localization compared to DGKβ-WT (Fig. 1, Fig. 2), indicating that the catalytic activity of DGKβ plays an important role in these activities. However, kinase inactivation did not completely inhibit the filopodium formation or plasma membrane localization. Therefore, DGKβ is able to contribute to these activities in a manner independent of its catalytic activity, at least in part. It is also possible that the kinase activity regulates only the plasma membrane localization and exerts its influence on filopodium formation at the plasma membrane in a kinase activity-independent manner. The importance of filopodium formation in spine formation  and neurite outgrowth/branching has been reported [, , ]. Interestingly, Kano et al. demonstrated that DGKβ induced neurite outgrowth and branching in SHSY-5Y neuroblastoma cells . Consistent with our results, a DGKβ mutant lacking the C-terminal tail lost its ability to induce neurite outgrowth and branching, and it was shown that the C1 and catalytic domains were important for these events. Therefore, these results support the conclusion that the ability of DGKβ to induce filopodium formation positively correlates with its ability to induce neurite formation and branching in neuronal cells.