WNT Protein-independent Constitutive Nuclear Localization of β-Catenin Protein and Its Low Degradation Rate in Thalamic Neurons*

Nuclear localization of β-catenin is a hallmark of canonical Wnt signaling, a pathway that plays a crucial role in brain development and the neurogenesis of the adult brain. We recently showed that β-catenin accumulates specifically in mature thalamic neurons, where it regulates the expression of the Cav3.1 voltage-gated calcium channel gene. Here, we investigated the mechanisms underlying β-catenin accumulation in thalamic neurons. We report that a lack of soluble factors produced either by glia or cortical neurons does not impair nuclear β-catenin accumulation in thalamic neurons. We next found that the number of thalamic neurons with β-catenin nuclear localization did not change when the Wnt/Dishevelled signaling pathway was inhibited by Dickkopf1 or a dominant negative mutant of Dishevelled3. These results suggest a WNT-independent cell-autonomous mechanism. We found that the protein levels of APC, AXIN1, and GSK3β, components of the β-catenin degradation complex, were lower in the thalamus than in the cortex of the adult rat brain. Reduced levels of these proteins were also observed in cultured thalamic neurons compared with cortical cultures. Finally, pulse-chase experiments confirmed that cytoplasmic β-catenin turnover was slower in thalamic neurons than in cortical neurons. Altogether, our data indicate that the nuclear localization of β-catenin in thalamic neurons is their cell-intrinsic feature, which was WNT-independent but associated with low levels of proteins involved in β-catenin labeling for ubiquitination and subsequent degradation.

␤-Catenin protein performs two well established cellular functions. It stabilizes cell-cell contacts by interacting with cadherin cell adhesion molecules, and it activates gene transcription as a cofactor of LEF1/TCF transcription factors (1). ␤-Catenin stability and subcellular localization are determined predominantly by its degradation rate, regulated by the canonical Wnt signaling pathway (2)(3)(4). In the absence of secreted WNT molecules, membranous ␤-catenin maintains cell adhesion, whereas its cytoplasmic pool is rapidly marked for degra-dation by the APC-AXIN-GSK3 complex. ␤-Catenin phosphorylated by GSK3 at its N terminus is degraded via the ubiquitinproteasome pathway. Following the binding of WNT ligands to the cell surface receptor Frizzled, their LRP5/6 co-receptors activate the Dishevelled (DVL) 2 protein, which recruits AXIN to the membrane. The APC-AXIN-GSK3 complex is consequently disassembled, and ␤-catenin is stabilized. Subsequently, ␤-catenin enters the nucleus to activate Wnt target genes by interacting with LEF1/TCF transcription factors (2,5).
The Wnt/␤-catenin pathway is implicated in developmental patterning, stem cell proliferation and differentiation, and carcinogenesis (6,7). In the developing nervous system, Wnt signaling is involved in brain patterning, organogenesis, and neurogenesis (8 -15). The dorsal thalamus is a unique part of the brain with regard to the prevalence of Wnt pathway components during organogenesis and in adulthood. Thalamic neurons originate in WNT3A cell lineage, which extensively migrate from the dorsal midline of the neural tube and populate the dorsal half of the adult brain (16). The activity of the Wnt pathway is indispensable for dorsal thalamus primordia because the thalamus is totally disorganized in mice with a mutation in LRP6 (17). Continued Wnt signaling is required for the induction of the dorsal thalamus-specific marker Gbx2 (9,18,19). The other, highly expressed genes in the early and adult dorsal thalamus are Lef1 and Tcf7l2 (19 -23), which encode Wnt-effector transcription factors. Additionally, we recently showed that ␤-catenin accumulates specifically in the nuclei of thalamic neurons and regulates the expression of the gene encoding the Ca v 3.1 voltage-gated calcium channel (23). This channel is involved in T-type currents characteristic of the adult thalamus (24,25).
The early development of the dorsal thalamus involves activation of proximal steps of the Wnt signaling cascade (9,13). The constitutive accumulation of ␤-catenin in the nuclei of postnatal dorsal thalamic neurons (23,26) suggests the persistent activation of the Wnt pathway in this region of the adult brain. However, we report here that neither disruption of the thalamic environment nor inhibition of Wnt/DVL signal transduction affects cytosolic or nuclear levels of ␤-catenin in thalamic neurons, suggesting a mechanism downstream of the WNT receptor. We show that the ␤-catenin degradation rate is lowered in thalamic neurons, which appears to be a consequence of low levels of the APC-AXIN1-GSK3␤ complex in these cells. We conclude that the nuclear localization of ␤-catenin in thalamic neurons is a cell-autonomous and cellintrinsic feature and does not require WNT stimulation.

EXPERIMENTAL PROCEDURES
Primary Neuronal Cultures-Dissociated primary thalamic cultures were prepared from embryonic day 19 rat brains and cultured according to procedures described by Wisniewska et al. (23). To reduce glial cell growth in thalamic cultures, 2.5 M AraC was added the day after seeding and kept for 1 or 7 days in the medium. Cortical and hippocampal cultures were prepared as above and cultured at densities of 1850 cells/mm 2 and 740 cells/mm 2 on coverslips coated with laminin (2 g/ml; Roche Applied Science) and poly-D-lysine (30 g/ml) or poly-L-lysine (30 g/ml), respectively. Neurons were grown in Neurobasal medium supplemented with B27 (Invitrogen), 0.5 mM glutamine, 12.5 mM glutamate, and penicillin/streptomycin (Sigma). All cultures were maintained at 37°C in a humid atmosphere with 5% CO 2 .
RNA Isolation and Real-time PCR-RNA was isolated with the RNeasy kit from Qiagen (RNeasy Plus Mini kit). cDNA was then synthesized (SuperScript III RNase H Reverse Transcriptase; Invitrogen) and examined by real-time PCR in a 7900 HT Real-Time PCR System using SYBR Green dye (Applied Biosystems). The results were analyzed by absolute quantification with a relative standard curve. To detect the rat Cacna1g tran-script we used commercial primers (Qiagen). For rat Lef1, the primers were the following: forward CCCACACGGACAGC-GACCTA and reverse TAGGCTCCTGTTCCTTTCTCT; for Gapdh, forward TGACTCTACCCACGGCAAGTTCAA and reverse ACGACATACTCAGCACCAGCATCA; and for Axin2, forward AAACCCGCCACCAAGACCTACATA and reverse TTTCCTCCATCACCGCCTGAATCT.
Adenoviral Vectors and Cell Transduction-The AdEasy adenoviral expression system was obtained from Stratagene. pShuttle vector was modified by inserting truncated RSV promoter, woodchuck post-transcriptional regulatory element, and SV40 polyadenylation signal sequence as described previously (27). Next, Gfp and mouse Myc-Axin2 (Addgene) cDNA was cloned into modified pShuttle vector. Axin2 DNA sequence was PCR derived, using primers 5Ј-AGTAGC-GCCGTGTTAGTGACTCT and 3Ј-GGCAAAGTGGAGAG-GATCGACTGA containing XbaI and XhoI recognition sites. Viruses were generated from pShuttle vectors using standard procedure described elsewhere (28). Then they were purified using a Vivapure AdenoPACK 20 kit (Sartorius) and tittered with an AdEasy Viral Titer kit (Stratagene) yielding 10 10 infection units/ml. Viral stocks were maintained at Ϫ80°C before use. At 4 DIV of thalamic culture, half of the culture medium was removed, and purified virus was added to each well at 100 multiplicity of infection and incubated for 6 h at 37°C. Then the virus solution was removed and replaced by fresh neuronal culture medium. After 72 h mRNA was isolated.
Cell Treatment and Transfection-Thalamic neurons at 7 DIV and L Wnt-3A cells (purchased and cultured according to the American Type Culture Collection (ATCC)) were treated for 3 h with rat DKK1 (R&D Systems) at concentrations of 200 and 500 ng/ml, with 50 ng/ml WNT3A and with WNT3A and DKK1 simultaneously. As a control, 0.1% BSA in PBS was used.
In the experiments with mutated DVL3, L Wnt-3A cells were seeded at a density of 5 ϫ 10 4 cells in each well of a 24-well plate 1 day prior to transfection. Thalamic neurons at 7 DIV and L Wnt-3A cells were then transfected with 0.4 g of plasmid pCG plus 0.2 g of GFP, Myc-AXIN2, or dominant negative (dn) DVL3 (FLAG-(333-716)-DVL3), respectively, per well. Myc-AXIN2 and dnDVL3 were obtained from Addgene. Pure medium (50 l) and 1.5 l of Lipofectamine 2000 (Invitrogen) were added to the DNA, mixed gently, and incubated for 25 min at room temperature. Media from the cells were collected and replaced by media without antibiotics and additionally without glutamate in the case of neurons. The DNA mixtures were then added to the cells for 4 h. Afterward, the cells were washed and kept in the medium collected before transfection. After 24 h, the cells were fixed. The transfected neurons were also treated for 3 h with 50 ng/ml WNT3A and then fixed.
HEK Cell Transfection, Vectors, and Luciferase Assay-HEK 293 cells were purchased from ATCC and cultured according to the ATCC instructions. 3.5 ϫ 10 4 cells were seeded in each well of a 24-well plate 1 day prior to transfection. Transfection was performed according to Wisniewska et al. (23). Briefly, 0.1 g of control pRL-SV40 vector with the Renilla luciferase gene, 0.25 g of reporter plasmid TOP, FOP, or pTAluc with firefly luciferase gene, and 0.5 g of pCG plasmids were mixed with polyethyleneimine reagent and added to each well. Forty-eight hours after transfection, the medium was replaced by media collected from L cells ordered from ATCC, L Wnt-3A cells expressing WNT3A (ATCC), thalamic and cortical cultures, and freshly prepared neuronal medium. The cells were lysed either 6 or 14 h after treatment. The luciferase assay was performed using the specific luciferase substrates luciferin and coelenterazine (Luxbiotechnology) in a LUMIstar Galaxy luminometer.
Pulse-chase Assay-At 7 DIV, cortical and thalamic cultures (incubated overnight with AraC, 1 day after seeding) were washed twice with warm (37°C) PBS and incubated for 1 h in methionine-and cysteine-free DMEM (Sigma) supplemented with B27 (Invitrogen), 0.5 mM glutamine, and 30 g/ml L-cysteine. For protein labeling, 0.6 ml of Met/Cys-free DMEM with 100 Ci/ml [ 35 S]methionine (Hartmann Analytic) was added per each 35-mm-diameter cell dish, and the cells were incubated at 37°C in a humid atmosphere with 5% CO 2 for 1 h. After labeling, the cells were washed twice with warm PBS and chased in complete Neurobasal medium. In each experiment, four 35-mm-diameter dishes of cortical neurons and three 35-mmdiameter dishes with thalamic neurons were washed twice with PBS and lysed at the indicated time points. To release cytoplasmic proteins, cells were incubated for 15 min at 4°C with 0.1% saponin buffer containing 25 mM HEPES and 75 mM potassium acetate, pH 7.9 (29). Then the cells were lysed in standard radioimmunoprecipitation assay buffer. Both buffers contained protease inhibitor mixture (Roche Applied Science).
Immunoprecipitation and Autoradiography-Protein lysates were incubated with 0.8 g of anti-␤-catenin rabbit antibody (Santa Cruz Biotechnology) for 2 h at 4°C. Then, 20 l of protein G-agarose (Roche Applied Science) was added, and lysates were gently rotated overnight at 4°C. Beads were washed three times with cold buffer containing 50 mM Tris, 150 mM NaCl, and 0.75% CHAPSO, pH 7.4, and boiled in SDS sample buffer. The proteins were separated by standard SDS-PAGE. The gels were dried, exposed overnight to phosphorimaging screens, and scanned in Storm 820 (Amersham Biosciences). Quantification was performed with ImageQuant 5.0 software and normalized to the intensity at time point 0, which was set as 1.

Dissociated Thalamic Neurons Cultured in Vitro Retain High
Level of ␤-catenin-To search for the mechanisms leading to ␤-catenin accumulation in thalamic neurons, we first established dissociated thalamic neuronal cultures. The best survival rate (63 Ϯ 8% at 12 DIV) was obtained for neurons isolated from embryonic day 19 rat brain and cultured in the presence of cortical conditioned medium. This medium essentially had a trophic effect on thalamic neurons in vitro because under in vivo condition thalamic cell survival is regulated by cortex-derived factors (30,31). The localization of ␤-catenin in thalamic neurons was compared with cortical and hippocampal cells at 7 DIV. Neurons obtained in four independent experiments were analyzed by fluorescence microscopy (Fig. 1A). In each experiment, Ͼ100 cells were counted to estimate the percentage of ␤-catenin-positive nuclei. We observed nuclear ␤-catenin in 39 Ϯ 3.7% of thalamic neurons. In contrast, ␤-catenin was not detected in the nuclei of cortical or hippocampal neurons. Similarly, Western blot analysis showed ␤-catenin in nuclear fraction in thalamic neurons, but not in cortical or hippocampal cultures (Fig. 1B). Moreover, nuclear ␤-catenin was able to activate gene expression, because overexpression of AXIN2, which strongly induces ␤-catenin degradation, resulted in about 2-fold reduction of mRNA for Lef1 and Cacna1g (supplemental Fig. S1), which are known ␤-catenin targets (23).
Nuclear Localization of ␤-catenin Is a Cell-autonomous Feature of Thalamic Neurons-Because thalamic neurons were maintained in the presence of conditioned medium obtained from cortical cultures and co-cultured with glial cells, we tested the possibility that a paracrine mechanism was responsible for ␤-catenin accumulation. Thalamic cells were cultured for 7 DIV in the presence or absence of the conditioned cortical medium and stained with ␤-catenin-specific antibody (Fig. 2A). The number of nuclear ␤-catenin-positive cells was then estimated. The lack of conditioned cortical medium did not lead to any decrease in the number of nuclei containing ␤-catenin (Fig.  2, A and B). This implies that although the medium contains factors that promote thalamic neurons survival, they are not necessary for ␤-catenin accumulation.
We next tested whether the stabilization of ␤-catenin in thalamic neurons was a consequence of their stimulation by factors released by glia. AraC treatment, which resulted in a dramatic decrease in the number of glial cells in the cultures, did not change the number of nuclear ␤-catenin-positive neurons (Fig. 2, A and B), although such treatment impaired neuronal survival and morphology. Thus, neither glial nor cortical neuron-derived factors had an effect on the number of ␤-cateninpositive nuclei in thalamic neurons, indicating that the accumulation of this protein is cell-autonomous.

Nuclear Accumulation of ␤-catenin in Thalamic Neurons
Does Not Result from WNT Stimulation-To determine whether the accumulation of nuclear ␤-catenin in thalamic neurons depends on autocrine activation of the Frizzled receptor by WNT molecules, we performed two sets of experiments. In the first set, we tested the ability of the medium from thalamic cultures to activate WNT-responsive luciferase TOP reporter. HEK 293 cells were transfected with either the reporter TOP or its mutant version FOP (32). Then the transfected cells were incubated with a freshly prepared neuronal medium or with the medium collected either from 7 DIV thalamic cultures or from cortical cultures. TOP reporter activity was low and comparable under all conditions, indicating that the Wnt pathway was not activated (Fig. 3A). As a positive control, we used the medium from L Wnt-3A cells that constitutively produce WNT3A. This medium led to 300-fold activation of the TOP reporter. Thus, thalamic medium does not contain factors that activate the Wnt/␤-catenin pathway.
In the second set of experiments, we tested the effect of the canonical Wnt pathway antagonists Dickkopf1 (DKK1) and dominant negative mutant of DVL3 (dnDVL3) on the number of intracellular ␤-catenin-positive cells. DKK1 inhibits canonical Wnt signaling by binding to the LRP6 co-receptor, preventing the formation of the active WNT-Frizzled-LRP6 receptor complex (33,34). DnDVL3 protein, lacking the DIX and PDZ domains, inhibits ␤-catenin accumulation by diminishing the recruitment of the ␤-catenin degradation complex to the LRP6 co-receptor (35,36). DKK1 efficiently reduced ␤-catenin accumulation in L Wnt-3A cells (Fig. 3B, left) and in thalamic neurons stimulated with WNT3A (supplemental Fig. S2A), demonstrating its inhibitory action on the WNT receptor. However, DKK1 treatment had no inhibitory effect on spontaneously accumulated ␤-catenin in thalamic neurons (Fig. 3B, right). Similarly, dnDVL3 markedly reduced the number of L Wnt-3A cells with intracellular ␤-catenin (Fig. 3C, left) as well as of thalamic neurons upon WNT3A treatment (supplemental Fig.  S2B), whereas it did not inhibit spontaneously accumulated ␤-catenin in thalamic neurons (Fig. 3C, right). A positive control, AXIN2, which strongly induces ␤-catenin degradation, reduced the number of ␤-catenin-positive L Wnt-3A cells as well as of neurons with constitutive ␤-catenin. Thus, ␤-catenin accumulation in thalamic neurons must be regulated by mechanisms downstream of Frizzled receptor activation induced by WNT.

Proteins Labeling ␤-catenin for Degradation Are Present at Low Levels in the Thalamus and in Cultured Thalamic Neurons-
The level of ␤-catenin in the cell is strictly regulated by the APC-AXIN-GSK3 protein complex that labels cytoplasmic ␤-catenin for degradation. Using immunoblotting, we estimated the relative levels of APC, AXIN1, and GSK3␤ in the thalamus and cortex isolated from adult rats. In the total protein extracts, we detected ϳ2.5-fold lower levels of APC and GSK3␤ in the thalamus than in the cortex (Fig. 4, A and C). Lower levels of AXIN1 in the thalamus compared with the cortex were also observed. Using specific antibody we also examined the phosphorylation state of GSK3␤ on serine 9 (Ser(P)-9 GSK3␤) in the brain. The level of Ser(P)-9 GSK3␤, which is an inactive form of GSK3␤, was 2.7-fold higher in the thalamus than in the cortex.
To verify whether the reduced levels and activity of APC, AXIN1, and GSK3␤ observed in the adult thalamus occur in neurons, but not in glial cells, we compared the levels of these proteins in thalamic and cortical cultures in vitro. To inhibit glial cells growth, neurons were grown in the presence of AraC. The cellular extracts were immunoblotted, and infrared imaging was used to quantify intensity of the bands. A 3.4-fold lower level of APC and 2-fold lower levels of AXIN1 and GSK3␤ were observed in thalamic cultures compared with cortical cultures. We also noticed a 1.4-fold higher level of inactive GSK3␤ (Ser(P)-9 GSK3␤) in thalamic neurons, although this result did not reach statistic significance (Fig. 4, B and D).
In summary, protein components of the ␤-catenin degradation complex, such as APC, AXIN1, and GSK3␤, were present at low levels in the thalamus compared with the cortex as well as in thalamic cultures compared with cortical neurons. This suggests that the machinery responsible for labeling ␤-catenin for degradation is not effective in the thalamic neurons, allowing ␤-catenin accumulation in the nuclei of these cells.
␤-catenin Turnover Is Slower in Thalamic Neurons Than in Cortical Cells-To corroborate the hypothesis that ␤-catenin accumulates in thalamic neurons because of the low degradation rate, we chased the amount of newly synthesized cytoplasmic ␤-catenin in thalamic and cortical cultures. The comparison of ␤-catenin, which was pulse-labeled with radioactive methionine, revealed a significant difference in its lifetime between cultures (Fig. 5). In thalamic cultures, cytoplasmic ␤-catenin degraded significantly more slowly than in cortical neurons during a chase period of 90 min. After extraction of cytoplasm from the cells the remainder contained ␤-catenin mostly from the plasma membrane. Therefore, this noncytoplasmic ␤-catenin was defined as membrane pool of ␤-catenin as shown in Fig. 5. We noticed that the membrane pool of ␤-catenin was similarly stable in both cultures, indicating that cytoplasmic ␤-catenin in thalamic neurons was not supplied from the membrane. These data showed that indeed cytoplasmic ␤-catenin has a slower degradation rate in thalamic neurons.

DISCUSSION
We and others showed that ␤-catenin is abundant in the nuclei of postmitotic thalamic neurons (23,26). The nuclear localization of ␤-catenin in mature neurons is unusual. In the adult brain, nuclear ␤-catenin is generally restricted to neural progenitor cells as a mediator of the Wnt signaling pathway (37,38). Thus, the nuclear pattern of ␤-catenin in postmitotic neurons of the adult thalamus would suggest persistent Wnt pathway activity in this brain region. In the present study, we found that the nuclear localization of ␤-catenin in thalamic neurons is constitutive and WNT-independent. Nuclear ␤-catenin was observed both in neurons of the adult thalamus and in isolated and dissociated thalamic neurons cultured in vitro. Previously, ␤-catenin was shown to translocate to the nucleus of hippocampal neurons upon tetanic stimulation or glutamate treatment (39 -41). Here, we found that the presence of ␤-catenin in the nuclei of thalamic neurons did not require any cell stimulation and was stable upon withdrawal of glial-and cortical-derived factors. Thus, the nuclear localization of ␤-catenin in thalamic neurons is constitutive and independent of cortical and glial stimuli in the thalamic environment.
Apart from the adult thalamus, the constitutive nuclear pattern of ␤-catenin has not been identified in adult tissues other than stem cells and tumors (42)(43)(44). In adult stem cells, ␤-catenin localizes in cell nuclei following WNT stimulation. Inhibition of the WNT receptor by transgenic expression of DKK1 or dnWNT1 in adult mice induced a complete loss of progenitor stem cells in the colon and brain (45,46). In our studies, soluble DKK1 and dnDVL3, another WNT receptor inhibitor, did not affect the number of thalamic neurons with nuclear ␤-catenin in vitro, indicating a WNT-independent mechanism. Moreover, media collected from thalamic cultures were not able to activate luciferase expression under the Wnt/ ␤-catenin-responsive promoter TOP, suggesting a lack of involvement of WNT or other soluble factors in ␤-catenin nuclear localization in thalamic neurons. Our findings showed that nuclear ␤-catenin in thalamic neurons is present in a constitutive manner, independent of WNT stimulation. This indicates the existence of an autonomous mechanism operating downstream of the WNT receptor. does not affect ␤-catenin spontaneous accumulation in thalamic neurons. L Wnt-3A cells (positive control) and neurons at 6 or 7 DIV were treated with DKK1 (200 or 500 ng/ml) for 3 h. The cells were fixed and stained with anti-␤-catenin antibody (red) and anti-NeuN antibody (green) to visualize neurons and Hoechst to stain nuclei (blue). Arrowheads point to cells with accumulated ␤-catenin. Pictures represent confocal images. Intracellular ␤-catenin was analyzed in three or four independent experiments (at least 100 neurons and 400 L Wnt-3A cells were counted in each experiment). Scale bar, 10 m. Error bars indicate S.D. *, p Ͻ 0.05 (one-way ANOVA followed by Tukey post hoc test). C, dnDVL3 does not affect ␤-catenin spontaneous accumulation in the nuclei of thalamic neurons. L Wnt-3A cells (positive control) and neurons at 6 or 7 DIV were transfected with GFP (negative control), AXIN2 (positive control), and dnDVL3 for 24 h. Then the cells were fixed and stained with anti-␤-catenin antibody (red) and anti-Myc-tag or anti-FLAG antibody (green) to visualize cells with AXIN2 or dnDVL3. Cell nuclei were stained with Hoechst (blue). Arrows point to transfected cells. Scale bar, 10 m. Pictures represent confocal images. Intracellular ␤-catenin was analyzed in three to five independent experiments (at least 200 cells were counted for each transfection variant). Error bars indicate S.D. ***, p Ͻ 0.001 (one-way ANOVA followed by Tukey post hoc test).
To understand the potential cell-intrinsic mechanisms leading to constitutive localization of ␤-catenin in the nuclei of thalamic neurons, we investigated the level and activity of proteins involved in ␤-catenin degradation (2-4). We found that APC, AXIN1, and GSK3␤, which are involved in the labeling of ␤-catenin for degradation, were reduced in the adult thalamus and thalamic neurons cultured in vitro. Loss-of-function mutations in APC and AXIN1 were described as factors that promote constitutive ␤-catenin cytoplasmic/nuclear accumulation in colon cancer (6,7,47,48). Similarly, brain-specific APC FIGURE 4. Protein levels of APC, AXIN1, and GSK3␤ are low in the thalamus of the adult rat brain and in thalamic cultures. A, Western blot analysis of APC, AXIN1, GSK3␤ and inactive (S9-P) GSK3␤ in protein lysates obtained from the thalamus (Th) and cortex (Cx). All proteins were detected by specific antibodies and developed by chemiluminescence. B, Western blot analysis of APC, AXIN1, GSK3␤ and inactive (S9-P) GSK3␤ in protein lysates obtained from thalamic and cortical neuronal cultures. All proteins were detected by specific antibodies and developed by the Odyssey Infrared Imaging System. C and D, densitometric analysis of APC, AXIN1, GSK3␤ and Ser(P)-9 GSK3␤ performed for six to eight animals (C) and for four neuronal cultures (D). The band intensity was normalized to GAPDH. The mean for the cortex is set as 1. Error bars indicate S.D. *, p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001 (two-tailed t test). FIGURE 5. Cytoplasmic ␤-catenin is degraded more slowly in thalamic neurons than in cortical neurons. A, autoradiograms of ␤-catenin labeled with [ 35 S]methionine, chased at the indicated time points. Protein lysates were obtained from thalamic (Th) and cortical (Cx) neuronal cultures and then immunoprecipitated with specific antibody. B, immunoprecipitates from at least three thalamic and cortical cultures analyzed after SDS-PAGE by autoradiography. In both cultures, labeled ␤-catenin chased at 0 h was set as 1. Error bars indicate S.D. **, p Ͻ 0.01 (two-tailed t test). and GSK3␤ knock-out mice exhibited intracellular stability of ␤-catenin (49 -51). Thus, decreased levels of APC-AXIN1-GSK3␤ in thalamic neurons appear to be a physiological mechanism leading to nuclear localization of ␤-catenin because of its slower cytoplasmic turnover. An increased phosphorylation level of Ser-9 in GSK3␤, which represents an inactive form of GSK3␤, may further reduce the efficiency of ␤-catenin phosphorylation by this kinase in thalamic neurons.
We conclude that cell-autonomous, Wnt-independent, and constitutive nuclear ␤-catenin localization is a signature of thalamic neurons. The fact that ␤-catenin and LEF1/TCF factors regulate expression of Cacna1g encoding the thalamic Ca v 3.1 calcium channel (23) suggests an involvement of ␤-catenin in determining a unique profile of gene expression in these specific neurons. Further research is needed to elucidate the role of nuclear ␤-catenin in the thalamus.