A Complex Interplay between Notch Effectors and b -Catenin Signaling in Morular Differentiation of Endometrial Carcinoma Cells

,

b-catenin is phosphorylated by a destruction complex composed of glycogen synthase kinase-3b (GSK-3b), adenomatous polyposis coli, and axin. 18Wnt signaling disrupts the destruction complexes, leading to accumulation of nuclear and cytoplasmic b-catenin; in the nucleus, bcatenin interacts with the transcriptional coactivator, T-cell factor/lymphoid enhancer factor. 19b-Catenin signaling, via crosstalk with several other signaling pathways, is a critical factor for trans-differentiation toward the morular phenotype of endometrial carcinoma (Em Ca) cells.20e22 Moreover, morular differentiation is closely associated with induction of the epithelial-mesenchymal transition and cancer stem cell-like properties in Em Ca cells. 22Medroxyprogesterone acetate therapy suppresses Em Ca cell proliferation and concomitantly induces squamous differentiation areas within tumors. 23Given this observation, it is suggested that elucidating the molecular mechanism underlying morular differentiation might lead to novel targeted therapies for Em Ca with morular phenotypes; this, in turn, would be associated with an improved prognosis.
Given that GSK-3b contributes to modulation of both Notch and b-catenin signaling, 2,24,25 it was hypothesized that these two signaling pathways would co-operate to induce morular differentiation in Em Ca cells.To test this, the present study investigated the expression of b-catenin, NICD1, Hes1, and MAML2, with reference to their interaction, transcription, and function using Em Ca cell lines and clinical samples.

Plasmids, Cell Lines, and Reagents
Full-length cDNA of human NICD1 was amplified by PCR using Notch 1 cDNA (RIKEN Human cDNA Clone, W01F001C23; Dnaform Co, Kanagawa, Japan) and the Dual promoter TA cloning kit (Invitrogen, Carlsbad, CA), and was cloned into the p3xFLAG-CMV-14 vector (Sigma-Aldrich Chemicals, St. Louis, MO).Human Hes1 cDNA was generated by PCR using the Dual promoter TA cloning kit (Invitrogen) and cDNA obtained from Em Ca cell line, Ishikawa cells, and was cloned into the p3xFLAG-CMV-14 vector (Sigma-Aldrich Chemicals).Full-length cDNAs of human MAML1 (pF1KSDA0200), MAML2 (pF1KSDA1819), and MAML3 (pF1KA1816) were obtained from Kazusa DNA Research Institute (Chiba, Japan) and were cloned into the pF5A CMV-neo Flexi vector (Promega, Madison, WT).MAML2 cDNA was also cloned into the p3xFLAG-CMV-14 vector (Sigma-Aldrich Chemicals) by PCR-based methods.The human MAML2 promoter (encompassing À974 to 47 bp, where 1 represents the transcription start site) and human Hes1 promoter (À1642 to 71) were cloned into the pGL-3B vector (Promega) by PCRbased methods using genomic DNA from Em Ca cell lines.The identity of all constructs was confirmed by sequencing before use.The sequences of PCR primers used in this study are listed in Table 1 ½T1 . The pcDNA Q6 eb-catenin (deleted S45), pCI-p300, pcDNA3.1eHAeGSK-3b,26e28 Three Em Ca cell lines (Ishikawa, Hec6, and Hec251) were acquired from the National Institute of Biomedical Innovation (Osaka, Japan).Cell proliferation was examined by counting after seeding cells at low density.DAPT Q7 (g-Secretase inhibitor IX) and lithium chloride were purchased from Sigma-Aldrich Chemicals.

Clinical Cases
A total of 102 cases of endometrioid-type Em Cas, including 38 of grade (G) 1, 33 of G2, and 31 of G3, were reviewed from the case records of Kitasato University Hospital (Japan

Q8
) during the period from 2007 to 2021, according to the criteria of the 2014 World Health Organization classification. 29Of these, 38 cases of G1 or G2 Em Ca with morular lesions were selected.None of the G3 Em Ca cases had morular lesions within tumor tissues.All tissues were routinely fixed in 10% formalin and processed for embedding in paraffin.Approval for this study was given by the Ethics Committee of Kitasato University School of Medicine (Sagamihara, Japan; B20-81).

Antibodies
Antibodies used in this study are listed in Table 2 ½T2 .

Immunohistochemistry
Immunohistochemistry (IHC) was performed using a combination of the microwave-oven heating and polymer immunocomplex (Envision; Dako Q9 ) methods, as described previously, 22,28 using normal mouse and rabbit sera as negative control.
For evaluation of IHC findings, scoring of cytoplasmic and/or nuclear immunoreactivities in morular and the surrounding carcinoma (Sur Ca) components was derived by multiplying the percentage of immunopositive cells and the immunointensity values, as described previously. 22,28The evaluation of IHC findings was conducted by three observers (A.Y., A.M., and M.H.).

Q10
Life Sciences, Corning, NY).The lower chamber was filled with medium containing 10% serum.Cells were suspended in serum-free upper medium and placed in the upper chamber.After 24 hours, the number of cells stained by hematoxylin-eosin on the bottom surface of the polycarbonate membranes was determined using a light microscope, as described previously. 22,28ow Cytometry Cells were fixed using 70% alcohol and stained with propidium iodide (Sigma, St. Louis, MO) for cell cycle analysis.Cells were then analyzed by flow cytometry using a BD FACS Calibur (BD Biosciences

Q11
) and CellQuest Pro software version 3.3 (BD Biosciences), as described previously. 22,28Ascope Assay for MAML2 and Hes1 mRNA in Situ Hybridization Expression of MAML2 and Hes1 mRNA was analyzed using an RNAscope assay (Advanced Cell Diagnostics, Hayward, CA), according to manufacturer's instructions.The hybridization was performed with targeted probes: Hs-MAML2 (number 434971), Hs-Hes1 (number 31191), positive control probe (number 2010684), and negative control probe (number 310043) for 2 hours at 40 C. The numbers of intranuclear and intracytoplasmic in situ hybridization signals were counted in at least 50 cells and were then expressed as an average number of signals per cell, as described previously.22,28

Proximity Ligation Assay
The slides were heated in 10 mmol/L Tris-EDTA buffer (pH 9.0) for 3 Â 5 minute cycles using a microwave oven and then incubated overnight with primary antibodies.Paired antibody staining with either Hes1 (rabbit)/b-catenin (mouse) or MAML2 (rabbit)/b-catenin (mouse) was performed, with single antibody used as negative control.After washing, the slides were treated according to the manufacturer's protocol using the Duolink Detection kit with proximity ligation assay (PLA) PLUS and MINUS probes for mouse and rabbit (Olink Bioscience, Uppsala, Sweden), as described previously. 22,28The numbers of intranuclear and intracytoplasmic PLA signals were counted in at least 50 cells and were then expressed as an average number of signals per cell.

Statistical Analysis
Comparative data were analyzed using the U-test and Spearman correlation coefficient.The cutoff for statistical significance was set as P < 0.05.

MAML2 Acts as Transcriptional Coactivator of b-Catenin Signaling
All three MAML members (MAML1 to MAML3) are core transcriptional coactivators in the Notch signaling pathway.30e32 To examine whether they are also coactivators of b-catenin signaling, the authors first performed one-hybrid assays.In Ishikawa and Hec251 cells, pG5 luc activity was significantly increased by cotransfection of a DNA-BDefused b-catenin fragment (pMeb-cat) with MAML2 (Figure 1 ½F1 A).MAML2 also significantly enhanced b-cateninedependent transcription, as demonstrated using TOP reporter constructs (Figure 1B).In line with the results in the previous sentences, endogenous MAML2 expression was observed in all three Em Ca cell lines, in contrast to the more restricted expression of MAML1 and MAML3 (Figure 1C).
Because MAML1 can recruit other coregulators, such as the histone acetyltransferase p300, 9,33,34 the authors examined whether MAML2 and p300 co-operated to enhance b-cateninedependent transcription in Em Ca cells.Cotransfection of MAML2 and p300 resulted in the formation of several nuclear aggregates (Figure 1D).However, cotransfection of p300 significantly inhibited the activity of transcriptional complexes, including b-catenin and MAML2 in Ishikawa cells, in contrast to the significantly enhanced activity in Hec251 cells (Figure 1E).
These findings suggest that MAML2 is directly capable of enhancing b-cateninedependent transcription in Em Ca cells, whereas the ability of p300 and MAML2 to co-operate and increase b-cateninedependent transcription is cell type dependent.

Transcriptional Up-Regulation of Hes1 and MAML2 in Morular Lesions of Em Ca
To examine whether there was crosstalk between the Notch and b-catenin signaling pathways in morular lesions, the authors conducted IHC analysis for b-catenin and Notch effectors (including MAML2 and Hes1).There was frequent overlap of nuclear immunostaining for b-catenin, MAML2, and Hes1 in morular lesions of Em Ca tissues (Figure 2 ½F2 A).MAML2 and Hes1 expression at both protein and mRNA levels, as well as nuclear b-catenin, were significantly higher in morular lesions compared with those of Sur Ca areas (Figure 2, A and B).The IHC scores of the three markers were significantly and positively correlated (Table 3   ½T3 ).In Em Ca cell lines, Hes1 promoter activity was increased by transfection of NICD1, and the effects were further enhanced by cotransfection of MAML2 (Figure 2C).Similar

Notch/b-Catenin Signaling in Em Ca
The American Journal of Pathology -ajp.amjpathol.findings were also observed with MAML2 promoter activity (Figure 2D).In contrast, b-catenin had a relatively weak effect on the activity of these promoters, even when MAML2 was cotransfected (Figure 2, C and D).These findings suggest that Hes1 and MAML2 are transcriptionally up-regulated in response to NICD1 in morular lesions of Em Ca, and that the positive feedback loop between Hes1 and MAML2 is b-catenin independent.

Functional Interactions between b-Catenin, MAML2, and Hes1 in Morular Lesions of Em Ca
To examine whether there were any functional associations between b-catenin, MAML2, and Hes1 in Em Ca morular lesions, the authors performed a PLA assay, which can be used to quantify protein-protein interactions in cells or tissue sections. 35Several foci, indicating interactions between bcatenin and either MAML2 or Hes1, were observed in nuclear and cytoplasmic compartments in both morular and Sur Ca lesions (Figure 3 ½F3 A).The PLA scores for b-catenin/ MAML2 and b-catenin/Hes1 were significantly higher in nuclear compared with cytoplasmic compartments in morules; the opposite was observed in Sur Ca lesions.In the nuclear compartment, the PLA scores were significantly higher in morules compared with Sur Cas (Figure 3B).
In Em Ca cell lines, several enlarged dots in the nuclei were observed following cotransfection of b-catenin, MAML2, and Hes1 (Figure 3C); this is consistent with the associations between b-catenin, MAML2, and Hes1 in Ish-FLAG-MAML2#29 cells and H6-FLAG-Ges1#34 cells (Figure 3D).In addition, the one hybrid assay revealed that pG5 luc activity was increased by cotransfection of b-catenin with either Hes1 or MAML2, but not NICD1 (Figure 3E).
These findings suggest that in Em Ca, functional interactions between nuclear b-catenin, MAML2, and Hes1 are predominant in morular lesions, and less prevalent in Sur Ca.

Hes1 Overexpression Is Associated with Decreased Proliferation and Inhibition of Cell Mobility
Treatment of Ishikawa and Hec251 cells with DAPT (a gsecretase inhibitor) decreased endogenous expression of NICD1 and Hes1 (Figure 4 ½F4 A).This is consistent with the evidence of Notch1 activation following g-secretaseemediated cleavage of the Notch S3 site, 36 and the authors' results from the Hes1 promoter reporter assays.To examine the functional role of Hes1 in Em Ca cells, the authors established Hes1-overexpressing Hec6 cells; Hec6 cells were chosen because they have low endogenous Hes1 expression when compared with that of Ishikawa and Hec251 cells (Supplemental Figure S1).Although Hes1 did not induce significant morphologic changes compared with mock-transfected cells (Figure 4B), it did reduce the proliferative rate in the exponential growth phase (Figure 4C).This was accompanied by decreased expression of nonphosphorylated Rb and p27 kip1 and an increased expression of p21 waf1 (Figure 4D), a decreased proportion of cells in the G 1 and G 2 /M phases, and an increased proportion of cells in the sub-G 1 fraction (Figure 4E).In addition, the cells overexpressing Hes1 refilled wounded empty spaces more slowly (Figure 4F), consistent with the significantly   decreased migration rates compared with mock-transfected and parental cells (Figure 4G).These findings suggest that NICD1-mediated Hes1 overexpression contributes to reduced proliferation and inhibition of migration in Em Ca cells.

Inhibition of GSK-3b Activity Alters Proliferation and Migration, along with Changes in Expression of Notch Effectors and b-Catenin
Because GSK-3b is a major regulator of both Notch and Wnt/b-catenin signaling, 2,24,25 the authors examined the expression and functional role of GSK-3b in Em Ca cells.In Hec6 cells, treatment with lithium chloride (a GSK-3b inhibitor) engendered a significant switch toward a fibroblastic morphology (Figure 5 ½F5 A) compared with the control.This was accompanied by decreased proliferation, particularly in the exponential growth phase (Figure 5B), and a significantly attenuated migration rate (Figure 5C); the latter was consistent with the slower rate at which wounds were refilled (Figure 5D).The treatment also significantly increased total and phosphorylated nuclear and cytoplasmic GSK-3b (Figure 5, E and F), and increased the expression of NICD1, Hes1, and active b-catenin (Figure 5G).
Representative IHC findings for GSK-3b in Em Ca with morules are illustrated in Figure 5H.Strong nuclear/cytoplasmic GSK-3b immunoreactivity was frequently observed in all morular components of 38 cases.The nuclear/cytoplasmic GSK-3b score was significantly higher in morular lesions compared with that of the Sur Ca, and was    3).These findings suggest that GSK-3b is inactive in morular lesions of Em Ca.In turn, this may explain the reduced proliferation, attenuatted migration, and activation of Notch and b-catenin signaling in Em Ca.

Discussion
The present study clearly shows that high expression of MAML2 (rather than MAML1 and MAML3) is a feature of three Em Ca cell lines.Moreover, MAML2 formed highly transcriptionally active complexes with b-catenin in nuclei.In general, MAML1, MAML2, and MAML3 exhibit distinct expression patterns during embryonic development, indicating that they play specific roles in different tissues. 32,37,38Given that MAML2 is the most effective enhancer of Notch signaling, 29,31 it was suggested that MAML2 is the major driver of the cellular response to bcatenin signaling in Em Ca cells.Moreover, the functional consequence of the MAML2/p300 interaction with regard to b-cateninemediated transcription is cell type dependent.Interestingly, the MAML1/p300 interaction leads to mutual Left panels: Staining by HE (left side) and immunohistochemistry (IHC) for GSK-3b (right side) in Em Ca with morules.Note the distinct nuclear and cytoplasmic immunoreactivities of GSK-3b in morular lesions (black arrows), in contrast to the weak immunoreactivities in the surrounding carcinoma.Insets: Closed boxed areas in the morular lesions are magnified.Right panel: IHC scores for nuclear (Nu) and cytoplasmic (Cyt) GSK-3b between morular (Mo) and surrounding carcinoma (Sur Ca) categories.Statistical analyses were performed using the U-test.Data are presented as means AE SDs (B and CeE and H, right panels).*P < 0.05, **P < 0.01, and ****P < 0.0001.Scale bars: 10 mm (A, E, left panels, and H, left panels, insets); 25 mm (C and D, left panels, and H, left panels, main images).Original magnifications: Â400 (E, left panels, and H, left panels, insets); Â100 (H, left panels, main images).Hes, hairy and enhancer of split; NICD, Notch intracellular domain; pGSK-3b, phosphorylated GSK-3b.

Notch/b-Catenin Signaling in Em Ca
The American Journal of Pathology -ajp.amjpathol.modifications: MAML1 causes p300 phosphorylation, whereas p300 mediates MAML1 acetylation. 9,39n addition, NICD1-dependent activation of the Hes1 promoter was further enhanced by MAML2.MAML2 promoter activity was also increased by a combination of Hes1 and MAML2, as well as NICD1 and MAML2.Because colocalized expression of MAML2 and Hes1 mRNA and protein was observed in morular lesions of Em Ca tissues, it was suggested that an NICD1/Hes1/MAML2 signal loop may drive Notch signaling in these lesions.Although bcatenin can activate NICD-dependent gene expression via direct interaction, 40 the present results indicate that b-catenin induced only minor changes in Hes1 and MAML2 promoter activity in Em Ca cells.
There are many reports on the crosstalk between Notch and b-catenin signaling, 40e42 but the outcome of this coordination is context dependent.In our results, the functional interactions of nuclear b-catenin with Notch effectors (Hes1 and MAML2) were significantly higher in Em Ca morular lesions compared with Sur Ca areas.Moreover, bcatenin formed the active transcriptional complexes with Hes1 and MAML2, but not NICD1.Specific binding partners determine whether Hes1 acts as a repressor or activator, and this causes its cell typeespecific effects. 43Given these observations, it was suggested that Notch effectors (including Hes1) may modulate b-cateninemediated morular differentiation of Em Ca cells through formation of active transcriptional complexes with nuclear b-catenin.
Several results from the present study support the conclusion that GSK-3b is a key regulator of morular differentiation in Em Ca.First, treatment of Em Ca cells with GSK-3b inhibitor increased the levels of both total and phosphorylated GSK-3b in nuclear and cytoplasmic compartments.This is consistent with evidence that GSK-3b is in constant transit between the mitochondria, nucleus, and cytoplasm. 44We suggest that GSK-3b is inactive in morular lesions, because they have a higher nuclear/cytoplasmic ratio of GSK-3b abundance when compared with Sur Ca regions.Second, GSK-3b inhibition increased expression of NICD1, Hes1, and active b-catenin: this is consistent with the positive correlation between nuclear/cytoplasmic GSK-3b score and Hes1, MAML2, and nuclear b-catenin scores in morular Em Ca.In contrast, GSK-3b is a positive regulator of Notch signaling because it protects the intracellular domain from proteasome-mediated degradation. 25Third, GSK-3b inhibition also engendered a fibroblastic appearance and decreased cell proliferation and migration; these features are consistent with the phenotypic characteristics of morular lesions, which contain spindle-shaped cells that proliferate slowly.20e22 In addition, GSK3 activity is required for keratinocytes to form lamellipodia and migrate directionally in response to wound signaling. 45Finally, cells stably overexpressing Hes1 have decreased proliferation and migration.We, therefore, infer that the GSK-3b/NICD1/ Hes1 axis is closely associated with the phenotypic characteristics of morular differentiation.Further studies to characterize the role of GSK-3b in morular Em Ca are clearly warranted.
Together, the observations suggest a model for the complex interplay between Notch and b-catenin signaling during morular differentiation in Em Ca cells (Figure 6 ½F6 ). Inhibition of GSK-3b leads to activation of NICD1-mediated Hes1/ MAML2 signaling loops and stabilization of nuclear b-catenin.In turn, this results in the formation of active transcriptional complexes, including b-catenin, Hes1, and MAML2, which engender b-cateninemediated transdifferentiation toward the morular phenotype in Em Ca cells.

Disclosure Statement
None declared.

Figure 1
Figure 1 Interaction between b-catenin (b-cat) and mastermind-like 2 (MAML2) in endometrial carcinoma cells.A: Ishikawa and Hec251 cells were transfected with pG5 luciferase (luc) and pM

Figure 2
Figure 2 Up-regulation of mastermind-like 2 (MAML2) and hairy and enhancer of split 1 (Hes1) in morular lesions of endometrial carcinoma (Em Ca).A: Top panels: Staining with hematoxylin-eosin (HE) and immunohistochemistry (IHC) for the indicated proteins in a case of Em Ca with morules.Note the frequent overlaps of b-catenin, MAML2, and Hes1 in morular lesions but not the surrounding carcinoma (Sur Ca) lesions.Top panels: The closed boxed areas, a (morule) and b (Sur Ca), are magnified in the middle panels (a) and bottom panels (b).Bottom panel: IHC scores for the indicated proteins between morule and Sur Ca categories.Statistical analyses were performed using the U-test.B: Top panels: Staining by HE and RNAscope for MAML2 and Hes1 mRNA, as well as negative control, in a case of Em Ca with morule.Note the multiple fine dot signals for MAML2 and Hes1 mRNA in morule (a) but not Sur Ca lesions (b).Top panels: Closed boxed areas are magnified in the bottom panels.Bottom panel: Number of in situ hybridization signals for MAML2 and Hes1 per cell in morule and Sur Ca components of Em Ca.Statistical analyses were performed using the U-test.C: Ishikawa and Hec251 cells were transfected with Hec1 luciferase (luc), together with b-catenin delS45, Notch intracellular domain 1 (NICD1), or MAML2, using LipofectAMINE PLUS.The experiments were performed in triplicate.Statistical analyses were performed using the U-test.D: Ishikawa and Hec251 cells were transfected with MAML2 luc, together with b-catenin delS45, NICD1, Hes1, or MAML2, using LipofectAMINE PLUS.The experiments were performed in triplicate.Statistical analyses were performed using the Utest.C and D: See Materials and Methods for details of luciferase assays.Data are shown as means AE SDs (A and B, bottom panels, C, and D).*P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.Scale bars: 25 mm (A and B, top panels); 10 mm (A, middle and bottom panels, and B, bottom panels).Original magnification, Â100 (A and B, top panels); Â400 (A, middle and bottom panels, and B, bottom panels).Nu, nuclear.

Figure 3
Figure 3 Functional interactions between b-catenin (b-cat or Cat), mastermind-like 2 (MAML2; M2), and hairy and enhancer of split 1 (Hes1) in morular lesions of endometrial carcinoma (Em Ca).A: Proximity ligation assay (PLA) for functional interactions of the b-catenin/Hes1 and b-catenin/MAML, in Em Cas, using antieb-catenin (mouse), anti-MAML2 (rabbit), and anti-Hes1 antibodies (rabbit).Note the multiple fine small, aggregated dots in nuclear and cytoplasmic compartments of morular lesions but not surrounding carcinoma (Sur Ca).Top panels: The closed boxed areas, a (morule) and b (Sur Ca), are magnified in the bottom panels.B: Number of PLA scores for the indicated combinations between morule and Sur Ca categories.Statistical analyses were performed using the U-test.C: Immunofluorescence for a combination of b-catenin, MAML2, or Hes1 in Ishikawa cells.Note the small, aggregated dots in nuclei.Immunoreactivities are detected using antieb-catenin (rabbit) and anti-FLAG Q17 (FL; for FLAG-Hes1 and FLAG-MAML2) antibodies (mouse).D: Western blot (WB) analysis with antieb-catenin (top and bottom panels) and anti-FLAG antibodies (middle panel) after immunoprecipitation (IP) with the indicated antibodies using Ish-FLAG-MAML2#29 (top and middle panels) and H6-FLAG-Hes1#34 cell lysates (bottom panel).Input represents 5% of the total cell extract.Normal rabbit IgG was used as a negative control.The experiments were performed in triplicate.E: Ishikawa and Hec251 cells were transfected with pG5 luciferase (luc) and pM Q18 eb-cat, together with Notch intracellular domain 1 (NICD1), Hes1, or MAML2, using LipofectAMINE PLUS.The experiments were performed in triplicate.Statistical analyses were performed using the U-test.E: See Materials and Methods for details of luciferase assays.Data are shown as means AE SDs (B and E).*P < 0.05, **P < 0.01.Scale bars: 25 mm (A, top panels); 10 mm (A, bottom panels, and C).Original magnification: Â100 (A, top panels); Â400 (A, bottom panels, and C).Cyt, cytoplasm; HE, hematoxylin-eosin; Nu, nucleus.

Figure 4
Figure 4 Hairy and enhancer of split 1 Hes1) overexpression is associated with low cell proliferation and inhibition of migration.A: Western blot analysis for the indicated proteins in total lysates from Ishikawa (left side) and Hec251 cells (right side) treated with DAPT

Figure 5
Figure 5 Functional role of glycogen synthase kinase-3b (GSK-3b) in morular lesions of endometrial carcinoma (Em Ca).A: Phase-contrast images of H6 cells left untreated or treated with 40 mmol/L lithium chloride (LiCl).Note the change in cell morphology toward a fibroblastic appearance (arrows) following treatment.B: Hec6 cells were treated with LiCl for the indicated doses and times and seeded at low density.The cell numbers are counted at the indicated times [day (D) 0, D3, D6, and D9 after seeding] and are presented.The experiments were performed in triplicate.Statistical analyses were performed using the U-test.C: Migration rate was measured using a Transwell assay.Left panels: Hec6 cells treated with LiCl at the indicated doses were seeded in 24-well Transwell plates and incubated for 24 hours in medium without serum.Cells (arrows) were stained with hematoxylin-eosin (HE) and counted using a light microscope.Right panel: The numbers of migrated cells are presented.The experiments were performed in triplicate.Statistical analyses were performed using the U-test.D: Left panels: Wound-healing assay using Hec6 cells with LiCl treatment for the indicated dosages.A scratch wound was introduced to the middle of wells containing cells grown to confluency, and phase-contrast images were taken after 10 and 24 hours.The red dotted lines indicate the borders between confluent cell layers and wound areas.Right panel: The values of wound areas in 0 hours were set as 1.The fold wound areas determined by ImageJ software Q23 version 1.41 are presented.The experiments were performed in triplicate.Statistical analyses were performed using the U-test.E: Left panels: Immunofluorescence for transfected HA Q24 eGSK-3b in Ishikawa cells in the presence and absence of 40 mmol/L LiCl.GSK-3b immunoreactivity is detected using an anti-HA antibody.Note the distinct nuclear immunoreactivity of HAeGSK-3b (arrows) in Ishikawa cells treated with 40 mmol/L LiCl.Right panel: The numbers of nuclear HAeGSK-3bepositive cells per 10 transfected cells are presented.The experiments were performed in triplicate.Statistical analyses were performed using the U-test.F: Western blot analysis for the indicated proteins in cytoplasmic (C), membranous (M), and nuclear fractions (N) from Hec6 cells treated with 40 mmol/L LiCl for the indicated times.G: Western blot analysis for the indicated proteins in total lysates from Ishikawa cells treated with 40 mmol/L LiCl.H: Left panels: Staining by HE (left side) and immunohistochemistry (IHC) for GSK-3b (right side) in Em Ca with morules.Note the distinct nuclear and cytoplasmic immunoreactivities of GSK-3b in morular lesions (black arrows), in contrast to the weak immunoreactivities in the surrounding carcinoma.Insets: Closed boxed areas in the morular lesions are magnified.Right panel: IHC scores for nuclear (Nu) and cytoplasmic (Cyt) GSK-3b between morular (Mo) and surrounding carcinoma (Sur Ca) categories.Statistical analyses were performed using the U-test.Data are presented as means AE SDs (B and CeE and H, right panels).*P < 0.05, **P < 0.01, and ****P < 0.0001.Scale bars: 10 mm (A, E, left panels, and H, left panels, insets); 25 mm (C and D, left panels, and H, left panels, main images).Original magnifications: Â400 (E, left panels, and H, left panels, insets); Â100 (H, left panels, main images).Hes, hairy and enhancer of split; NICD, Notch intracellular domain; pGSK-3b, phosphorylated GSK-3b.

Figure 6
Figure 6 Schematic representation of the interplay between Notch and b-catenin signaling in endometrial carcinoma with morules.Inhibition of glycogen synthase kinase-3b (GSK-3b) leads to activation of b-catenin signaling and Notch intracellular domain 1 (NICD1)/mastermind-like 2 (MAML2)/hairy and enhancer of split 1 (Hes1) signaling loops.Increased expression of MAML2 and Hes1 enhances b-cateninedependent transcription, resulting in acceleration of b-catenin emediated morular differentiation.Hes1 overexpression also contributes to the establishment and maintenance of a morular phenotype that is associated with low proliferation and inhibition of migration capability.

Table 2
Summary of Antibodies Used in this Study Q26.

Table 1
Primer Sequences Used in this Study