AKT activation by N-cadherin regulates beta-catenin signaling and neuronal differentiation during cortical development

Background During cerebral cortical development, neural precursor-precursor interactions in the ventricular zone neurogenic niche coordinate signaling pathways that regulate proliferation and differentiation. Previous studies with shRNA knockdown approaches indicated that N-cadherin adhesion between cortical precursors regulates β-catenin signaling, but the underlying mechanisms remained poorly understood. Results Here, with conditional knockout approaches, we find further supporting evidence that N-cadherin maintains β-catenin signaling during cortical development. Using shRNA to N-cadherin and dominant negative N-cadherin overexpression in cell culture, we find that N-cadherin regulates Wnt-stimulated β-catenin signaling in a cell-autonomous fashion. Knockdown or inhibition of N-cadherin with function-blocking antibodies leads to reduced activation of the Wnt co-receptor LRP6. We also find that N-cadherin regulates β-catenin via AKT, as reduction of N-cadherin causes decreased AKT activation and reduced phosphorylation of AKT targets GSK3β and β-catenin. Inhibition of AKT signaling in neural precursors in vivo leads to reduced β-catenin-dependent transcriptional activation, increased migration from the ventricular zone, premature neuronal differentiation, and increased apoptotic cell death. Conclusions These results show that N-cadherin regulates β-catenin signaling through both Wnt and AKT, and suggest a previously unrecognized role for AKT in neuronal differentiation and cell survival during cortical development.


Background
In the developing cerebral cortex, neural precursors reside in the ventricular zone (VZ), a neurogenic zone of neuroepithelial cells that lines the lateral ventricles. The specialized microenvironment or niche within the VZ coordinates signaling pathways that regulate the self-renewal and differentiation of neural precursors. Recent studies suggest that the adherens junctions and apical domains of neural precursors contain signaling molecules that regulate interactions between neural precursors critical in maintaining precursor identity [1,2]. Knockdown of the adherens junction proteins N-cadherin [3] or αE-catenin [4] in cortical VZ cells caused reduced β-catenin signaling, increased exit from the cell cycle, premature neuronal differentiation, and increased migration out of the VZ. Although these findings suggest a novel role for N-cadherin-containing junctions in the positive regulation of β-catenin signaling during cortical development, little is known about the molecular mechanisms by which N-cadherin maintains β-catenin signaling.
Recent studies suggest that one way N-cadherin adhesion can influence β-catenin signaling is through regulating the AKT signaling pathway. AKT signaling impacts many fundamental processes regulating growth and metabolism; in organ development, AKT functions in the regulation of normal organ size by mediating signaling downstream of insulin and insulin-like growth factor I [5]. In mammals, the AKT family is comprised of three highly conserved members, AKT1, AKT2, and AKT3. While AKT1 −/− mice display a general reduction in sizes of all organs [6], AKT3 −/− mice have a selective decrease in brain size [7]. Unlike AKT2 mutant mice, neither AKT1 nor AKT3 mutant mice showed abnormalities in insulin signaling [7], suggesting that brain size regulation by AKT was mediated through alternative non-insulin signaling mechanisms. Recently, findings of active mutations of AKT3 in patients with hemimegalencephaly (HMG), an overgrowth of one hemisphere during brain development, provided further support for the role of AKT signaling in human brain size regulation [8][9][10].
N-cadherin adhesion can lead to phosphatidylinositol 3-kinase (PI3K)-mediated activation of AKT [11], and activated AKT signaling can stimulate β-catenin signaling. In intestinal stem cells (ISCs), loss of PTEN (phosphatase and tensin homolog), a negative regulator of PI3K/AKT, leads to β-catenin stabilization and upregulated β-catenin signaling [12]. In ISCs the mechanism by which AKT activates β-catenin signaling is via direct phosphorylation of βcatenin at residue Serine 552, which primes β-catenin for 14-3-3ζ binding and stabilization [13]. Moreover, AKT also phosphorylates and inactivates glycogen synthase kinase (GSK)3β at Serine 9; as GSK3β functions as a negative regulator of β-catenin stabilization, AKT phosphorylation of GSK3β can lead to β-catenin stabilization and nuclear βcatenin accumulation [14]. Our previous work suggested that N-cadherin engagement also activates β-catenin signaling through AKT in cortical precursors [3]. While AKT activity was recently described in dividing mouse developing neural precursors [8], how AKT signaling is regulated during neural development, and the underlying mechanisms through which AKT mediates brain growth, remains poorly understood.
The current study seeks to provide greater understanding of the mechanistic relationships between N-cadherin and β-catenin signaling in cortical development, using a combination of in vivo and cell culture approaches. Here we find further supporting in-vivo data that N-cadherin functions in the cortical VZ to maintain β-catenin signaling. We also find evidence using in vivo electroporation approaches and cell co-culture experiments for a cell-autonomous Ncadherin role in receiving Wnt signaling. In addition to its role in transducing Wnt signals through the Wnt coreceptor LRP6, we find that N-cadherin also regulates AKT phosphorylation and activation. Knockdown of N-cadherin leads to reduction of AKT phosphorylation as well as a reduction of Serine 552 phosphorylated β-catenin and Serine 9 phosphorylated GSK3β, both direct targets of active AKT. We show that both β-catenin Ser-552-P and GSK3β Ser-9-P are expressed in mitotic radial glial progenitor cells in the developing cortex, suggestive of activation of AKT signaling in these cells. Using in utero electroporation, we show that inhibition of AKT signaling using a dominant negative AKT (DN-AKT) leads to premature exit from the VZ, increased neuronal differentiation, and increased apoptotic cell death. Together, these studies suggest a pathway linking N-cadherin cell adhesion to the regulation of cell survival and differentiation via AKT activation.

Results
N-cadherin maintains β-catenin signaling in cortical precursors in vivo During cortical development, β-catenin signaling is active in neural precursors within the VZ [15] and regulates neural precursor proliferation [15,16] and differentiation [17]. Our previous studies suggested that N-cadherin, the primary cadherin of adherens junctions in cortical precursors [18], maintains β-catenin signaling and prevents neural precursors from premature differentiation [3]. Electroporation of N-cadherin-shRNA (Ncad-shRNA) in VZ precursors in vivo reduced the expression of an optimized β-catenin signaling reporter, TOPdGFP [3].
To examine the cell-autonomous role of N-cadherin in β-catenin signaling in VZ precursors, we co-electroporated expression plasmids for Cre recombinase and TOPdGFP [23] into the VZ of E13.5 Ncad Flox/Flox embryos. This approach enables conditional deletion of genes from cells receiving the Cre plasmid [4,15,24] and, as the reporter GFP is produced by the simultaneously introduced plasmids, the signaling readout is not affected by historical activation of the signaling pathway. Staining for GFP 24 hours after electroporation showed that, compared to cells electroporated with the pcDNA-lacZ control, Cre electroporation reduced β-catenin signaling ( Figure 1B). Together, these results support our previous finding that cell-autonomous N-cadherin is required for maintenance of β-catenin signaling in cortical neural progenitor cells.

N-cadherin functions in Wnt-induced β-catenin signaling in 293 T cells in a cell-autonomous manner
We observed previously that N-cadherin adhesion regulated endogenous β-catenin signaling activity in neural precursors [3]. The most well-characterized and common activator of β-catenin signaling is the family of secreted Wnt proteins. Wnt signaling results in inactivation of a phosphodestruction complex that normally serves to target free cytosolic β-catenin for ubiquitin-mediated degradation [25]. Our previous observations that increasing neural precursor cell density could increase β-catenin transcriptional activation [3] activity suggested that cell proximity/density could regulate Wnt/β-catenin signaling.
To examine further the relationship of cell proximity and Wnt/β-catenin signaling, we created signaler (Wnt "signaler") and reporter (Wnt "reporter") cells (that report activation of β-catenin-mediated transcription) by transfecting separate populations of 293 T cells with Wnt3a expression and pSuper8xTOPFlash reporter constructs and examined signaling in cells plated either on the opposite of a polycarbonate transwell filter ("short Figure 1 Conditional knockout of N-cadherin reduces β-catenin signaling in developing cortical precursors. (A) Immunostaining for d2EGFP (green) in E12.0 littermate control (Axin2-d2EGFP; Nes11Cre; Ncad Flox/+ ) and Ncad cKO brain (Axin2-d2EGFP; Nes11Cre; Ncad Flox/Flox ) embryonic cortex reveals that conditional tissue-wide knockout of N-cadherin leads to reduced EGFP expression in the developing ventricular zone (VZ) (DNA stained with DAPI, pseudocolored blue; bar = 100 μm). (B) Focal elimination of N-cadherin reduces β-catenin transcriptional activity. β-catenin mediated transcriptional activation was examined through expression of destabilized GFP controlled by the TOP promoter. E13.5 embryos were electroporated with pTOP-dGFP, pCAG-mCherry, and pCAGLacZ in the control treatment, and with pTOP-dGFP, pCAG-mCherry, and pCAG-Cre in the experimental condition, and analyzed at E14.5. The dot image below shows the positions of the individual electroporated cells (yellow dots represent double-labeled mCherry/dGFP + cells and red dots represent dGFP-cells (expressing mCherry only)). Only electroporated cells in the VZ were included in the analysis, and the proportion of cells expressing dGFP was compared to the total number of electroporated cells in the VZ. *P = 0.0242 by unpaired Student's t-test, n = 3 brains for each. Error bars represent 1 SEM. Bar = 50 μm. EP'd, electroporated.
distance") or on the bottom of the transwell chamber ("long-distance") ( Figure 2A). When plated on opposite sides of the filter, cells can make physical contacts with the cells on the opposite side through the 0.4 μm pores [26,27]. We found that Wnt3a-expressing signaler cells induced β-catenin signaling in reporter cells adherent on the opposite face of the transwell membrane, but not in reporter cells plated separately at the bottom of the chamber ( Figure 2A). Together with our previous findings that N-cadherin mediates density-dependent β-catenin signaling, this observation that Wnt induction of β-catenin signaling requires close cell-cell apposition suggested a role for N-cadherin cell adhesion in mediating Wnt signaling.
To examine specifically the role of N-cadherin in Wnt signaling, we conducted loss-of-function studies of Ncadherin in either signaler or reporter cells. Transfection of Ncad-shRNA or dominant negative N-cadherin with a C-terminal β-catenin binding domain truncation (N-cadΔC) in reporter cells reduced Wnt-induced β-catenin signaling in a cell-autonomous fashion ( Figure 2B). In contrast, N-cadherin knockdown or expression of N-cad ΔC in signaler cells resulted in a trend towards reduced Wnt-induced β-catenin signaling ( Figure 2C), suggesting only a modest role for non-cell-autonomous function for N-cadherin in Wnt signaling. Together, the in vivo electroporation and co-culture findings suggest that N-cadherin maintains Wnt-induced β-catenin signaling through a cellautonomous mechanism. These co-culture findings also suggest a role for non-autonomous N-cadherin in Wnt signaling, likely mediated in part by the requirement for close cell apposition.

Reduction of N-cadherin leads to reduction of AKT and LRP6 phosphorylation
Recent studies have suggested that cadherin adhesion can both positively [28] and negatively [25] regulate β-catenin signaling in a Wnt-independent manner. To explore further the downstream pathways linking cell-autonomous N-cadherin function to β-catenin signaling, we examined targets of Wnt-independent and Wnt-stimulated activation of β-catenin following knockdown or blockade of Ncadherin in cultured cells.
To examine the role of N-cadherin in regulation of βcatenin in the absence of Wnt stimulation, we transfected 293 T cells with N-cadherin-shRNA or EGFP-shRNA control. Western blot analysis after 24 hours incubation showed that, in the absence of exogenous Wnt, Ncad-shRNA reduced active AKT (phospho-AKT Ser 473). We also found reductions in two direct downstream targets of active AKT -GSK3β phosphorylated at Serine 9, and βcatenin phosphorylated at Serine 552 -following Ncadherin-shRNA transfection ( Figure 3A). Phosphorylation of β-catenin at Ser 552 increases its transactivation by increasing its stability [13], while phosphorylation of GSK3β leads to inactivation of the GSK3β and can result in accumulation of β-catenin [14]. We also observed a trend where Ncad-shRNA reduced total β-catenin levels, consistent with prior observations that phosphorylation at Ser 552 increasing β-catenin stability [13].
In 293T cell lines stably expressing Ncad-shRNA, we observed similar findings: reduction of N-cadherin expression led to reduced levels of active AKT (phosphor-Ser 473), GSK3β phospho-Ser 9, and β-catenin phosph-Ser 552 in the absence of Wnt ( Figure 3B). Together, these findings suggest that, in the absence of Wnt, Ncadherin can regulate AKT activity and β-catenin levels via phosphorylation of β-catenin at Ser 552.
In the absence of exogenous Wnt, we also observed in the stable N-cadherin knockdown lines that levels of phosphorylated LRP6, the Wnt co-receptor, were also reduced ( Figure 3B). As phosphorylation of LRP6 results from activation by Wnt [29], together with the observations that Wnt stimulation of signaling required N-cadherin ( Figure 2), this finding suggested that Wntstimulated β-catenin signaling via LRP6 is impacted by N-cadherin levels. When the stable N-cadherin knockdown cell lines were transfected with Wnt3a to activate the Wnt signaling pathway, we observed increased (See figure on previous page.) Figure 2 N-cadherin functions in Wnt-induced β-catenin signaling in a cell-autonomous manner. (A) Separate populations of 293 T cells were transfected with Wnt3a expression and pSuper8TOPFlash reporter constructs and plated either on the opposite of a polycarbonate transwell filter ("membrane bound") or on the bottom of the transwell assay ("separated"). In a TOPFlash reporter assay, Wnt3a-expressing signaler cells induced β-catenin signaling in reporter cells adherent on the opposite face of the transwell membrane, but not in reporter cells plated separately at the bottom of the chamber. P = 0.004 by repeated measures analysis of variance (ANOVA), ***P < 0.001, Neuman Keuls post-hoc test, n = 3. (B) Reduction of N-cadherin by shRNA (Ncad-shRNA) or by overexpression of C-terminal truncated N-cadherin (NcadΔC) in Wnt-responsive cell results in reduced Wnt-activated β-catenin transcriptional activation. Wnt3a transfected signaler cells were co-cultured with pTOPflash-transfected reporter cells co-transfected with either shRNA to N-cadherin (P = 0.0012 by repeated measures ANOVA, **P < 0.01, *P < 0.05 by Neuman Keuls post-hoc test; n = 4) or NcadΔC (P = 0.0006 by repeated measures ANOVA, ***P < 0.001, **P < 0.01 by Neuman Keuls post-hoc test; n = 3), and luciferase activity was measured 24 hours after co-culture. (C) Inhibition of N-cadherin in the Wnt-producing signaling cell does not affect Wnt-mediated β-catenin signaling. Wnt3a transfected signaler cells were co-transfected with either Ncad-shRNA (P = 0.0024 by repeated measures ANOVA, **P < 0.01 by Neuman Keuls post hoc test; n = 4) or NcadΔC (P = 0.0002 by repeated measures ANOVA, ***P < 0.001 by Neuman Keuls post-hoc test; n = 3), co-cultured with pTOPflash-transfected reporter cells, and luciferase activity was measured 24 hours after co-culture.
phosphorylation of LRP6, as expected. We then found that, in N-cadherin knockdown cells, Wnt3a-induced phosphorylation of LRP6 was also reduced. Together, these findings suggest that N-cadherin levels also regulate Wnt-induced phosphorylation of its co-receptor LRP6.
To examine the role of N-cadherin in Ser 552 phosphorylation of β-catenin in cortical neural stem cells (NSC), we utilized mouse cortical NSC derived from E14.5 mouse cortices [30,31]. Using two different shRNA constructs against N-cadherin (pSRNcad ("shRNA1") and pSM2cNcad ("shRNA2")), we found that reduction in N-cadherin caused a reduction in the amount of β-catenin phosphorylated at Ser 552 ( Figure 3C). To examine the role of N-cadherin in the phosphorylation and activation of the Wnt co-receptor LRP6 in NSC, we treated the NSC with either IgG control antibody or N-cadherin function-blocking antibodies (GC-4, Sigma; St. Louis, USA) and examined levels of phosphorylated LRP6 ( Figure 3D). We found that inhibition of Ncadherin dramatically reduced the levels of phospho-LRP6 in primary cortical NSC, providing further support for the notion that N-cadherin also functions in β-catenin signaling in neural stem/progenitors by Targets of AKT activity are expressed in dividing radial glia during cortical development These findings that N-cadherin functions in Ser 55 phosphorylation of β-catenin in cortical NSC suggest that AKT signaling may play an important role in cerebral cortical development. A recent study pointed to somatic activating mutations in AKT3 that caused HMG and highlighted the presence of phosphorylated AKT (P-Ser-473) in dividing cortical neural precursors suggestive of active AKT signaling [8]. To examine downstream targets of AKT signaling in the developing cortex, we examined the expression of GSK3β-Ser 9-P and β-catenin Ser 552-P in VZ precursors. We observed widespread localization of GSK3β Ser 9P and β-catenin Ser 552P in dividing apical neural progenitors in the VZ. P-GSK3β and P-β-catenin co-localized with the radial glial marker phospho-Vimentin 4A4 [8,32] (Figure 4). 74.3 ± 5.9% (1 SEM) of P-Vim expressing cells were also expressing P-β-catenin, and 82.5 ± 3.6% (1 SEM) of P-Vim-expressing cells were also P-GSK3β positive. While GSK3β Ser 9 and β-catenin Ser 552 have been shown to be direct targets of AKT signaling [12], our data do not rule out other kinases that might phosphorylate these residues (e.g. protein kinase C (PKC) and GSK3β Ser 9 [33], PKA and β-catenin Ser-552 [34]). However, together with the observations of active AKT in these same cells, our findings provide further support for active AKT signaling in radial glial precursors in the developing cortex.

AKT maintains β-catenin signaling and neural differentiation in vivo
To examine directly whether AKT regulates β-catenin signaling in cortical progenitors in vivo, we inhibited AKT signaling by in utero electroporation of a kinasedead dominant negative AKT (DN-AKT) and examined β-catenin signaling using a co-electroporated β-catenin signaling reporter, TOPdGFP. We found that DN-AKT reduced β-catenin signaling in cortical VZ cells, suggesting that AKT is required for maintaining β-catenin signaling during cortical development ( Figure 5). As βcatenin signaling is both necessary and sufficient to drive cortical precursor self-renewal, we next sought to examine whether modulating AKT signaling influenced precursor identity.
AKT signaling regulates normal organ size during development: AKT1 −/− mice display a general reduction in sizes of all organs [6]; AKT3 −/− mice present a selective decrease in brain size [7]; and three recent studies suggest that mutations and overactivation of AKT3 can cause enlargement of the brain in cases of human HMG [8][9][10]. Consistent with the idea that AKT activity regulates neural precursor proliferation, retroviral overexpression of AKT1 promoted retention of cortical progenitors in the VZ and subventricular zone and increased NSC self-renewal [35]. After in utero electroporation of DN-AKT, we observed increased cell exit from the VZ towards the pial surface and premature neuronal differentiation ( Figure 6). AKT signaling may also function in maintaining cell adhesion, as inhibition by DN-AKT decreases the fraction of cells in contact with the ventricle ( Figure 6G).
AKT signaling normally impacts a broad range of cellular processes including apoptotic cell death. Many members of the apoptotic signaling cascade are targets of AKT-mediated phosphorylation and inactivation [36], and GSK3β inhibition by AKT has been shown to protect both neurons [37] and neural precursor cells from apoptosis [38]. To examine whether inhibition of AKT signaling in the developing cortex alters cell apoptosis, we co-stained in utero electroporated cortices for the apoptotic marker cleaved caspase 3. We found that compared with control GFP-electroporated cells, where onlỹ 1% of electroporated cells were apoptotic, a much larger fraction of cells that received DN-AKT (~15%) exhibited signs of apoptosis (cleaved caspase 3 expression, condensed nucleus by DAPI) (Figure 7). These findings suggest that a critical role of AKT signaling in cortical development is in maintenance of cell survival during the neurogenic period. Further study is required to determine whether different threshold levels of AKT signaling may regulate the decisions to differentiate versus undergo programmed cell death in cortical development.

Discussion
A growing body of work points to a critical role for adherens junctions in maintaining neural progenitor identity [39], and a recent study provided evidence that regulation of N-cadherin expression by Foxp2 and Foxp4 initiates neuronal delamination from the VZ and differentiation [40]. Here we expand our understanding of the molecular mechanisms through which N-cadherin regulates the neural precursor niche in cerebral cortical development. Our observations suggest that cell-autonomous N-cadherin functions in both Wnt-mediated and AKTmediated β-catenin transcriptional activation. Furthermore, our findings suggest that N-cadherin mediated activation of AKT signaling functions in cortical precursors in vivo to promote β-catenin signaling, and inhibition of AKT leads to reduced β-catenin signaling and premature exit from the VZ, neuronal differentiation, and increased apoptosis ( Figure 8).
The Cre-mediated N-cadherin knockout in developing cortical precursors and the co-culture experiments presented provide two complementary lines of evidence suggesting that reduction or interference with N-cadherin reduces β-catenin signaling activity in a cell-autonomous manner. Our findings with co-cultured signaling/responder cells also suggest -but do not provide definitive evidence for -a role for N-cadherin-dependent cell contact in Wntsignaling. Our observations of a small trend of reduced Wnt-signaling when N-cadherin is reduced in Wntproducing cells are consistent with mechanisms where cell-cell adhesion enhances Wnt signaling, potentially by optimizing the physical interactions and/or proximity of ligand and receptor. In vivo, our studies are also consistent with a function for N-cadherin in physical retention of neural precursors in the VZ. However, our findings that cells still located in the VZ also show reduced β-catenin signaling ( Figures 1B and Figure 7 [4]) suggest that the model that N-cadherin maintenance of β-catenin occurs solely via physical retention of cells in the VZ is likely overly simplistic. We find that when N-cadherin or αEcatenin (another critical component of the VZ adherens junction) [4] are reduced, we observe reductions in β-  catenin signaling that likely precede exit from the VZ, as electroporated cells that have not yet left the VZ still exhibit reduced signaling.
Distinguishing between a purely adhesive versus signaling role of N-cadherin is challenging; loss of N-cadherin will cause breakdown of cell contacts and disruptions of neuroepithelial adhesion [41,42], and the possibility remains that adhesion and signaling function of N-cadherin cannot be separated. It is well known that regulators of cell polarity (such as scribble [43,44] and aPKC [45,46]) play a critical role in inhibiting neoplastic growth, and we believe that N-cadherin maintenance of cell adhesion in neural precursors may similarly link growth and polarity, but via control of β-catenin signaling and adhesion in a normal developmental context.
How cadherin adhesion can both negatively and positively regulate β-catenin signaling activity remains poorly understood. Observations that the forced overexpression of cadherins antagonizes β-catenin transcriptional activation suggested the simple model that cadherins set a threshold for β-catenin signaling. However, our findings suggest that N-cadherin functions both in Wnt-mediated activation of β-catenin via phosphorylation of the Wnt co-receptor LRP6 and in activation of β-catenin via AKT-mediated phosphorylation of Ser 552 of β-catenin. In contrast to the widely held viewpoint that cadherins are considered negative regulators of β-catenin signaling through sequestration [47], our findings lend additional support to a growing body of studies suggesting a positive role for a classical cadherin in β-catenin signaling [28,[48][49][50][51].
Our findings that N-cadherin functions to positively regulate Wnt-mediated β-catenin signaling through the Wnt co-receptor LRP6 support a recent study showing that Casein kinase 1 (CK1) association with LRP5/6 requires cadherin [51]. p120 catenin, via interactions with cadherin and CK1, appears to control assembly of the Wnt signaling complex [50]. As phosphorylation of LRP6 by CK1 isoforms plays an essential function in Wnt receptor complex activation [52], these findings together support a positive role for cadherin in Wnt/βcatenin signaling.
While Wnt-mediated β-catenin signaling in cortical development has been suggested by LRP6 mutant mice [53] and Wnt overexpression studies [54], the role of alternative regulation of β-catenin signaling in cortical development is not well investigated. Here, our evidence suggests that AKT regulates β-catenin signaling in cortical precursors in vivo, and inhibition of AKT activity leads to premature neuronal differentiation and exit from the VZ. Consistent with previously reported functions of AKT signaling in cell survival, we also observed that inhibition of AKT in vivo by electroporation leads to increased apoptotic cell death.
The alterations we observe in the proportions of cells that are progenitors or differentiated neurons after AKT inhibition may be a consequence of increased cell death in progenitor populations. However, it is difficult to extrapolate the effects of increased progenitor cell death on cells that differentiate from the progenitors, as their number would also be reduced as a consequence of progenitor death. Our findings of a relative increase in Tbr1 and Tbr2-expressing cells as well as increased migratory distance following DN-AKT supports a role in differentiation versus an exclusive role in cell survival.
Our results support recent findings that AKT and PI3K signaling functions in cortical growth, with gainof-function mutations implicated in human HMG [8][9][10], and suggest that abnormal activation of β- catenin signaling may underlie the molecular mechanisms driving cortical overgrowth in HMG.

Conclusions
Here, we provide evidence that cell-autonomous Ncadherin co-ordinates both Wnt-dependent and AKTmediated β-catenin activation. In the presence of Wnt, N-cadherin regulates LRP6 phosphorylation, while when Wnt is absent, N-cadherin also regulates β-catenin transcriptional activation via AKT-mediated phosphorylation of β-catenin Ser 552. Using in utero electroporation approaches, we found evidence that AKT regulates βcatenin signaling levels in VZ cortical precursors, and loss of function experiments showed that inhibition of AKT signaling causes increased apoptosis, premature neuronal differentiation, and exit from the VZ. Our findings that AKT functions in cortical precursors support the observations of recent studies suggesting a role for AKT in human megalenecephaly.

In utero electroporation
For in utero injection, timed-pregnant mice at E13.5 were anesthetized using inhalation of isoflurane mixed in a constant ratio with oxygen, after which abdominal fur was removed, and the uterine horns were exposed through a midline laparotomy incision. DNA solution (2.5 μL) in H 2 O containing 0.0125% fast green was injected through the uterine wall into the lateral ventricle of the embryos using a glass micropipette made from a microcapillary tube. After injection, Tweezer-trodes (BTX, 84 October Hill Rd. Holliston, MA 01746-1388) were applied across the outside of the uterus, oriented to flank the embryonic brain, and five 50 ms square pulses of 39 V with 950 ms intervals were delivered by an electroporator (BTX 830). Following injection and electroporation, the uterus is returned inside the abdomen and the abdominal muscle wall and skin sealed with sutures.
For the TOP-dGFP signaling studies, embryos were sacrificed 30 hours after electroporation. For the distribution and differentiation studies, embryos were sacrificed 24 hours after electroporation. Brains were fixed for 8 to 16 hours in 4% paraformaldehyde, cryoprotected in 30% sucrose dissolved in PBS overnight, and embedded in OCT (Optimal Cutting Temperature compound). A cryostat was used to make 12 μm coronal sections.
For studies of TOPdGFP expression in Ncad Flox/Flox mice, histograms in Figure 1B

Immunohistochemistry and tissue analysis
Brain sections were incubated with blocking solution (5% goat serum and 0.3% Triton X-100 in PBS) for 1 hour at room temperature, and then incubated with primary antibodies diluted in blocking solution for 2 hours at room temperature (for Pax6, Tbr2, Tbr1, mCherry and TOPdGFP) or overnight at 4°C (for Axin2-d2EGFP, phospho-AKT, phospho-GSK3β, phospho-βcatenin, phospho-Vimentin 4A4, and GFP in distribution assay). Primary antibodies used were anti- . Confocal images were captured using a Zeiss UV LSM510 confocal microscope Carl Zeiss Microscopy, LLC One Zeiss Drive Thornwood, NY 10594 USA). The analysis throughout the paper was carried out in a semi-blinded manner as previously described [3].
Primary neural stem cell culture, shRNA, and treatment with function-blocking antibodies For Western blots of N-cadherin, phospho-β-catenin S552, β-catenin, phospho-Akt S473 and Akt, E14.5 mouse primary cortical cultures were generated from E14.5 mouse embryos using minor modifications to the method described in [55]. After dissection, cells were disassociated with 0.25% Trypsin/EDTA (Sigma), washed by DMEM with 10% FBS, and resuspended in clone density media (CDM) containing DMEM (Cellgro Mediatech, Inc. 9345 Discovery Blvd. Manassas, VA 20109), 2 mM L-glutamine (Cellgro), 1 mM N-Acetyl-cysteine, 1 mM sodium pyruvate, and B27 and N2 supplement (final 1×) with 25 ng/mL FGF-2. Five million primary cortical precursors were transfected with 10 μg shRNA constructs against N-cadherin or EGFP control using Nucleofector following the manufacture's protocol (Lonza Lonza Inc. 90 Boroline Road Allendale, NJ 07401). Transfected cells were plated on Poly-D Lysine treated 6-well tissue culture plates and allowed to recover in the described media (as above) at 37°C for 24 hours before lysing for Western blots. Adherent cortical NSC cultures were generated from E14.5 mouse embryos as described in [30,31]. After dissection, cells were dissociated with Accutase (Millipore EMD Millipore Headquarters 290 Concord Road Billerica, MA 01821), washed three times with NS-A medium (Stem Cell Technologies #5750 STEMCELL Technologies Inc. 570 West Seventh Avenue, Suite 400, Vancouver, BC, V5Z 1B3, Canada), rinsed once more with NSC medium (NS-A supplemented with FGF-2, EGF, N-2A, B27, L-glutamine, Sodium pyruvate, N-acetyl cysteine, and Penicillin-streptomycin as in [31]) and plated on laminin-coated (treated with 2 μg/cm 2 laminin) dishes. To obtain material for protein analysis, 5 million mouse E13.5 or E14.5 primary cortical precursors was transfected by AMAXA Nucleofection (Amaxa Biosystems, Gaithersburg, MD, USA) following manufacturer protocols; cells were allowed to recover in the described media (as above) at 37°C for 24 hours before lysing for Western blots. Cells were plated at 1.8 × 10 5 /cm 2 for 24 hours, then treated with either function-blocking N-cadherin antibody (ACAM, GC-4, Sigma) or control IgG1κ isotype control (BD Pharmingen cat. 554721 BD Biosciences 2350 Qume Drive San Jose, CA 95131) (both at 20 μg/ml, final concentration) for 24 hours.  (2) on the bottom of a 12-well plate. After 24 hours, the 6.5 mm transwell membrane inserts with reporter cells attached were inverted and placed in a 12-well plate soaked in media. Signaler cells in the 24-well plate were lifted by 0.25% Trypsin/EDTA and one-quarter of the cells were plated on the inside of the transwell membrane inserts ("Membrane-bound" condition). The pore size of these inserts is 0.4 μm, through which cells plated on opposite sides of the insert can establish physical contacts [26,27]. For the "Separated" condition, the signaler cells were plated inside transwell membrane inserts and placed above the 12-well plate with the reporter cells plated on the bottom of the well [26]. After incubated for 24 hours at 37°C, the reporter cells were lysed for Dual luciferase assay (Dual-Luciferase W Reporter Assay System, Promega Promega Corporation 2800 Woods Hollow Road Madison, WI 53711 USA).
For 293 T co-culture with N-cadherin knockdown, the reporter and signaler cells were generated similarly to that described above (1 × 10 6 cells were transfected by 1.0 μg Super8xTOP FLASH Luciferase and 0.1 μg Super8x-FOP Renilla as reporter cells, and a further 1 × 10 6 cells were transfected by 1.0 μg Wnt3a or pcDNA3.0 control as signaler cells), with the difference that 0.25 μg N-cadherin/ EGFP-shRNA (or NCAD-ΔC/pIRES-GFP control) were transfected into either the reporter or signaler cells; pc-DNA3.0 empty vector was used to bring the total DNA mass to 1.6 μg in all the transfections. After 24 hours, the reporter and signaler cells were lifted by 0.25% Trypsin/ EDTA, and one-half of the signaler cells and one-sixth of the reporter cells were mixed and plated on a 24-well plate. The cells were lysed after another 48 hours for Dual luciferase assay.

T transfection
To collect the protein lysates for N-cadherin knockdown studies, 293 T cells were transiently transfected using Lipofectamine 2000 following the manufacturer's protocol (Life Technologies). 293 T cells (1 × 10 6 ) were plated on a 12-well tissue culture plate and transfected by 1.2 μg Wnt or pcDNA3.0 control together with 2.0 μg Ncadherin or EGFP shRNA construct. Cells were allowed to recover in cDMEM at 37°C for 24 hours before lysing for Western blots. For AKT studies, 1 × 10 6 293 T cells were plated on a 24-well tissue culture plate and immediately transfected by 1.0 μg HA-AKT-K179M or pcDNA3.0 control.