We find that, similar to the results with cultured neurons, AAK1 siRNA increased proximal branching in vivo (Figures 6I and 6J). Next, we investigated Rabin8′s function on dendrite development and spine

maturation in hippocampal cultures. Immunostaining of endogeneous Rabin8 by anti-Rabin8 antibody showed that Rabin8 is enriched in the Golgi (colocalized with Golgi marker GM-130; Figure 7A), in agreement with the role of Rab8 in post-Golgi trafficking. We first examined its function by mutating the Rabin8 phosphorylation Imatinib site and expressing these mutants in dissociated hippocampal neurons. We made the Rabin8 phospho mutant, where S240 as well as T241, S242, and S243 were mutated to Alanine (Rabin8-AAAA), which cannot be phosphorylated (Figure 5F), or to Glutamate (Rabin8-EEEE) as a putative phosphomimetic mutant. We found that these Rabin8 mutants and Rabin8 siRNA (Figures S6A and S7A) did not affect dendrite branching (Figures

S6C–S6F), indicating that Rabin8 phosphorylation by NDR1 is likely not involved in limiting dendrite branching. The total dendrite length was reduced by Rabin8-AAAA but not Rabin8 siRNA (Figure S6F). Given that Rabin8 siRNA may not have sufficiently knocked down the Rabin8 level, these observations indicate that Rabin8 is involved in dendrite growth. Next, we found that the expression of Rabin8-AAAA but not Rabin8-EEEE resulted in increased Selleckchem Galunisertib filopodia and atypical spines, and Rabin8 siRNA increased filopodia density (Figures 7B and 7C). An increase in filopodia was accompanied by a reduction in mushroom spine density by Rabin-AAAA, a trend mafosfamide that was close to reaching significance (p = 0.07). These data indicate that Rabin8 phosphorylation by NDR1/2 contributes to spine development by reducing filopodia and increasing mushroom spines. Rabin8-AAAA and Rabin8 siRNA produce less pronounced defects on spines than does NDR1/2 loss of function, possibly

because other NDR1/2 substrates act in parallel to Rabin8 and contribute to spine morphogenesis. Alternatively, it is possible that these manipulations do not completely block Rabin8 function because of their incomplete knockdown or dominant negative effect. Given that Rabin-EEEE did not alter spine or dendrite development, this mutant construct may not be able to mimic phosphorylated Rabin8, a notion reinforced by our failed attempt to rescue NDR1siRNA + NDR2siRNA’s effect on spine development with Rabin8-EEEE (Figure S6B). Since Rabin8 is involved in spine maturation, we wanted to learn if it is present in spines with synapses. With immunostaining of postsynaptic marker PSD95 and endogenous Rabin8, we observe Rabin8 in the perinuclear region resembling Golgi and inside the proximal dendrites in neurons (Figure S6G). We cannot rule out the presence of Rabin8 in spines; however, the majority of Rabin8 is found in Golgi. (Figure S6G).