Skip to main content
NCBI home page
Journal of Cancer Research and Clinical Oncology logoLink to Journal of Cancer Research and Clinical Oncology
. 2019 Dec 21;146(2):315–327. doi: 10.1007/s00432-019-03114-8

Crosstalk of estrogen receptors and Wnt/β-catenin signaling in endometrial cancer

Mariz Kasoha 1,✉, Chrisoula Dernektsi 1, Anita Seibold 2, Rainer M Bohle 2, Zoltan Takacs 1, Iordache Ioan-Iulian 1, Erich-Franz Solomayer 1, Ingolf Juhasz-Böss 1
PMCID: PMC11804341  PMID: 31865530

Abstract

Purpose

To investigate the interaction between Wnt/β-catenin and estrogen signaling pathways in endometrial cancer (EC).

Methods

119 women were involved in this study, including 65 women with histologically confirmed EC and 54 healthy women as a control group. Serum protein levels of Dkk1 were measured using ELISA. Protein expression levels of Dkk1, β-catenin, ER-β isoforms (β1, β2, β5), and ER-α were tested in paraffin-embedded tissues using IHC. Gene expression levels of Dkk1, CTNNB, ESR1, and ESR2 were tested in fresh tumorous and normal endometrium tissues using RT-PCR.

Results

EC patients had significantly higher serum levels of Dkk1 protein compared with healthy women. Dkk1 and β-catenin showed different expression pattern in tumor cells compared to it in normal cells at the protein level but not at the gene level. Protein expression levels of ERβ2 and ERα were significantly lower in tumor cells compared with tumor-adjacent normal cells. Increased protein expression levels of ERα were associated with favorable clinicopathological features and better overall survival rate (OS). Protein expression levels of ERα were correlated with protein expression levels of Dkk1 and cytoplasmic β-catenin. The association between ERα expression levels and OS was no more significant when tested in regard to Dkk1- and cytoplasmic β-catenin expression levels.

Conclusions

Our data demonstrated that Wnt/β-catenin and estrogen signaling systems are dysregulated in EC showing; for the first time, a potential crosstalk between certain components of these two pathways, which in turn has affected the specificity of these molecules in disease characteristics. Understanding the signaling networks in EC is crucial in designing clinical trials to evaluate the efficacy of molecular-targeted agents and providing more successful therapies in the future.

Keywords: Endometrial cancer, Wnt/β-catenin, Estrogen isotypes, Molecular interactions

Introduction

Endometrial cancer (EC) is the most prevalent malignancy of the female genital tract in developed countries. In the USA, 61,380 new cases and 11,000 deaths were estimated in 2017 (American Cancer Society 2017). In Europe, nearly 100,000 newly diagnose cases were reported with an age-standardized incidence of 13.6 per 100,000 women (WHO, GLOBOCAN 2012).

The majority of EC patients (nearly 75%) is diagnosed at early stage and is potentially cured by surgery with or without adjuvant radiotherapy resulting in a 5-year survival rate of 95%. However, women diagnosed with advanced or recurrent disease have a poor prognosis with a 5-year survival rate of 17% (Siegel et al. 2016), which in turn might be due to the low responce of these women to the localized therapies. The efficacy of radiation therapy in advanced stage disease remains an ongoing debate and the available treatment options after initial therapy with no standard options at the time of disease progression or recurrence are still limited. Therefore, better understanding of the molecular features of EC is needed to improve patient’s; overall outcomes and to provide the best long-term survival using certain therapies based on molecular signatures.

There is a large body of literature available that addresses different molecular alterations in EC. Estrogens (ERs) are part of the nuclear receptor superfamily of ligand-inducible transcription factors and composed mainly of two receptors including ERα and ERβ, which are encoded by ESR1 and ESR2 gene, respectively (Zhao et al. 2010). The functions of ERs are strongly related to the development and functional maintenance of the human endometrium (Chen et al. 2008; Young 2013). Therefore, it has been suggested that in situ estrogen metabolism might play a crucial role in the development and progression of EC. Consequently, the expression profiles of ER isoforms and their correlation with the clinical outcomes were extensively tested in EC and controversial results were obtained (Hapangama et al. 2015). Wnt/β-catenin signaling pathway, also known as canonical pathway, includes a ligand-induced conformational change of the receptors, followed by phosphorylation of key target proteins. This signal transduction pathway is known to play key roles in many essential physiological and pathological processes such as organogenesis, cellular differentiation, maintaining epithelial integrity, malignant transformation, and angiogenesis (Logan and Nusse 2004). Several mutations and aberrant gene and protein expression of different Wnt components have been reported in EC indicating the contribution of this signal pathway in uterine carcinogenesis and tumor progression (Kiewisz et al. 2015).

Although an increasing number of reports have suggested potential convergence between Wnt and estrogen signaling, little evidence of their interactions have been documented. In the present study, we aimed to further elucidate the aberrations in gene and protein expression levels of estrogen receptors and in two essential components of the Wnt/β-catenin signaling pathway including β-catenin and dickkopf-1 (Dkk1). We also aimed to explore the hypothesized crosstalk between these two main signaling systems in EC.

Materials and methods

Ethics statement

Ethical statement of this study was provided by the local Ethic Committee of the Medical Association of Saarland (reference number: 207/10). All tissue and blood samples were obtained from individuals after signing a written consent that was approved by the ethic committee.

Patients and specimens

119 women were involved in this study, including 54 healthy women as a control group and 65 women with histologically confirmed endometrial cancer. All patients were postmenopausal and had no history of malignancies. Demographic, clinical, and pathological data of all patients were retrieved from their records at our department. Patient’s characteristics are listed in Table 1 and details on the case series are given in Fig. 1.

Table 1.

Characteristics of study subjects

Characteristics Number Percentage (%)
Histology
 Endometrioid carcinoma 58 89
 Serous carcinoma 7 11
T stage
 T1 49 75
 T2 6 10
 T3 10 15
N stage
 N0 41 63
 N1 13 20
 Nx 11 17
Grading
 G1 16 25
 G2 25 38
 G3 24 37
V status: (unknown: 14 cases)
 Negative 42 82
 Positive 9 18
L status: (unknown: 14 cases)
 Negative 38 75
 Positive 13 25
FIGO status
 = 5–10% 45 69
 = 15–25% 20 31
Myometrial invasion: (unknown: 3 cases)
 ≤ 50% 36 58
 > 50% 26 42
Open in a new tab

Fig. 1.

Fig. 1

Open in a new tab

Details on the study case series

Serum samples were available from 54 patients and 54 age-matched healthy women [68 (53–90) years vs 70 (52–91) years, respectively]. Control individuals were recruited from those who had undergone comprehensive medical screening at our department. They were postmenopausal women with no current or history of cancerous disease. All serum samples were frozen in aliquots at − 80 °C until being analyzed.

18 uterine tumorous tissues and 7 normal endometrium tissues were collected from 22 women during surgery. Tissues were directly flash frozen in liquid nitrogen and immediately transferred to − 80 °C until use.

Cryosections from all collected tissues were developed and consequently stained with hematoxylin–eosin (HE) using the HE protocol for frozen tissue sections, and then tested by an experienced pathologist at our university hospital to confirm the histological component of each.

Representative formalin-fixed paraffin-embedded blocks of 55 patients were retrieved from the archives of the institute of pathology of our university hospital. Consecutive sections of 4-µm thickness were prepared from the blocks for HE staining and for immunohistochemistry (IHC) analysis. All HE and IHC sections were tested by a pathologist for defining existing tumor and/or normal areas and for interpreting staining results.

Enzyme-linked immunosorbent assay (ELISA)

Serum levels of Dkk1 protein were measured in all collected serum samples by ELISA using commercially available kit, Human Dkk-1 Quantikine ELISA Kit (DKK100B) and Quantikine immunoassay control set 921 for human Dkk1 (QC241) produced at R&D Systems®, according to the manufacturer’s instructions. All measurements were performed in duplicate. The optical density was measured at 450 nm and referenced to 570 nm on a 96-well microplate reader (Sunrise-Tecan, Life Science). Dkk-1 levels were obtained with a four-parameter logistic curve fitted against a standard curve and multiplied by the dilution factor using Magellan 7.2 Ink Data Analysis Software (Life Science-Tecan).

Immunohistochemistry (IHC)

Protein expression levels of Dkk1, β-catenin, ER-β isoforms (β1, β2, β5), and ER-α were tested in paraffin-embedded tissues of 55 patients using IHC. IHC staining was manually performed using Dako REAL™ Detection System and Alkaline Phosphatase/RED, Rabbit/Mouse kit (Code K5005-DAKO). Briefly, following deparaffinization, antigen retrieval was done using Dako retrieval solution (S1699, pH 6). After that, sections were blocked for 30 min with 3% bovine serum albumin (BSA fraction V, Sigma-Aldrich) at room temperature. Incubation with the primary antibodies [anti Dkk1 antibody (H-120) and anti-β-catenin antibody (H-102) from Santa Cruz Biotechnology; mouse anti-human estrogen receptor alpha (MCA1799T), mouse anti-human estrogen receptor beta 1 (MCA1974GA), mouse anti-human estrogen receptor beta 2 (MCA2279GT), and mouse anti-human estrogen receptor beta 5 (MCA4676T) from AbD Serotec®] was done for 1 h at 37 °C. Secondary antibody detection and visualization were done according to standard protocols. Then, slides were washed, counterstained with hematoxylin, and coverslipped. Stained tissues were analyzed under Zeiss microscope (Axioskop 40, Carl Zeiss, Germany) and selected pictures were captured with attached digital camera (AxioCam MRC, Carl Zeiss, Germany) using Axiovision Documentation Rel.4.8 program (Fig. 2). Cytoplasmic staining of Dkk1 and cytoplasmic and nuclear staining of β-catenin, in addition to the nuclear staining of ER isoforms were quantified according to Remmele and Stegner (1987). Staining intensity was scored semi-quantitatively as negative (0), weak (1), moderate (2), or strong (3). The extent of staining was scored as percentage of cells stained (0, 0% of cells; 1, < 10% of cells; 2, 10–50% of cells; 3, 51–80% of cells; 4, > 80% of cells). The final immune reactive score (IRS) was determined by multiplying the scores of intensity and extent of staining. IRS values 0–2 were considered as negative staining, 3–4 as weak staining, 6–8 as moderate staining, and 9–12 as strong staining.

Fig. 2.

Fig. 2

Open in a new tab

IHC staining of Dkk1, β-catenin, ERα, ERβ1, ERβ2, and ERβ5 in EC tumor cells. Staining interpretation was done for the cytoplasmic expression of Dkk1 and β-catenin and for the nuclear expression of Erα and ER isoforms.*No case with negative cytoplasmic expression of β-catenin was reported within our tested cases

Real-time RT-PCR

The gene expression levels of Dkk1, CTNNB, ESR1, and ESR2 were tested in 18 uterine tumorous and in 7 normal endometrium fresh tissues. Cryosections from all collected tissues were developed and consequently stained with HE and tested by a pathologist at our university hospital to confirm the histological component of each.

Using TissueLyser LT Adapter and stainless steel beads (Qiagen), a maximum amount of 15–20 mg frozen tissues was immersed and homogenized in 300 µl of RNeasy lysis buffer (Qiagen, Valencia, CA, USA). RNA extraction was done using RNeasy MiniKit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. Thereafter, Ambion® TURBO DNA-free™ DNase kit (Life Technologies GmbH, Darmstadt, Germany) was used to remove contaminating DNA from RNA preparations. Extracted RNA concentrations and purity were determined using Thermo Scientific™ NanoDrop 2000. RNA integrity was tested using Agilent RNA 6000 Nano Reagents Part I (Agilent Technologies, Waldbronn, Germany) and Bioanalyser Agilent 2100 from Agilent Technologies. Synthesis of reverse-transcribed complementary DNA (cDNA) was performed using High Capacity cDNA Reverse Transcription Kit as described by the manufacturer (Applied Biosystem, Foster City, CA, USA). Then, PCR efficiency was tested and was in a range between 90 and 100% for all tested assays. Quantitative real-time PCR was performed in triplicate with dilutions of 10 ng cDNA using TaqMan® Gene Expression Assays by Life Technologies (Table 2) on an Applied Biosystems 7500 Fast Real-Time PCR System. All samples with a cycle threshold (Ct) coefficient of variation value > 5% were tested again. In addition, a no template control (NTC) was included in each run, and the resulting Ct values were normalized to actin beta (ACTB) mRNA expression. The RT-PCR data were analyzed using the 7500 Software v.2.0.5 (Life Technologies GmbH, Darmstadt, Germany). The relative RNA quantity in the samples of this study was calculated separately by the comparative ∆Ct method, where ∆Ct = ([Ct RNA of interest] − [Ct RNA of ACTB]).

Table 2.

TaqMan® gene expression assays used for quantitative RT-PCR

Gene name Gene symbols TaqMan assay ID
Dickkopf WNT signaling pathway inhibitor 1 DKK1 Hs00183740_m1
Catenin beta1 CTNNB1 Hs00355045_m1
Estrogen receptor alpha (ERα) ESR1 Hs01046816_m1
Estrogen receptor beta (ER-β) ESR2 Hs01100353_m1
Actin betaa ACTB Hs03929097_m1
Open in a new tab

aEndogenous gene

Statistical analysis

All statistical tests were performed at a two-sided comparison-wise significance level of 5% using SPSS version 21 (IBM, Armonk, NY, USA). Descriptive analysis was used for continuous variables using median and range and for categorical variables using absolute and relative frequencies. The null hypothesis was tested against its alternative using two-tailed Mann–Whitney U test/or Kruskal–Wallis test. Correlations between two continuous variables were evaluated using the non-parametric Spearman correlation coefficient. Times to event variables were described using Kaplan–Meier curves. Overall survival (OS) was the interval from the date of histological diagnosis to date of death or last follow-up. Progression-free survival (PFS) was the interval from histological diagnosis to the first date of disease recurrence/metastasis involvement or last follow-up or death. Patients were followed for a mean of 49 months (range 3–90 months).

Results

Serum levels of Dkk1 are increased in EC woman patients

Our results showed that EC patients had significantly higher levels of serum Dkk1 compared with healthy women [3085 pg/ml (749–6156) vs 1739 pg/ml (786–4109), respectively] (Fig. 3). However, we found no correlation between Dkk1 serum levels and different clinicopathologic characteristics including tumor grading, lymphatic metastasis, and myometrial invasion (data are not shown). In addition, Dkk1 serum levels were classified as high or low in relation to the median value and Kaplan–Meier method was applied for OS and PFS tests. No significant correlation between serum levels of Dkk1 and disease prognosis was observed (p = 0.507 and p = 0.370 for OS and PFS, respectively).

Fig. 3.

Fig. 3

Open in a new tab

Serum protein expression levels of Dkk1. Serum levels of Dkk1 are increased in patients with EC [68 (53–90) years] compared with healthy women [70 (52–91) years]: 3085 pg/ml (749–6156) vs 1739 pg/ml (786–4109), respectively

Protein- and gene expression levels of Dkk1, β-catenin, and ER isoforms in tumor tissues and in tumor-adjacent normal tissues

Protein expression levels of Dkk1, β-catenin, ERα, ERβ1, ERβ2, and ERβ5 were tested in paraffin-embedded tissues of 55 patients using IHC. Cytoplasmic staining of Dkk1 and cytoplasmic and nuclear staining of β-catenin, in addition to the nuclear staining of ER isoforms were interpreted. Our results revealed that cytoplasmic expression of Dkk1 and β-catenin, in addition to the nuclear expression of ER isoforms, in tumor cells, was positive in the majority of tested cases and ranged between weak and strong (IRS = 4–12). Nuclear expression of β-catenin was found to be positive in 9% of tested cases (5/55) including three cases with weak staining (IRS = 3–4) and two cases with moderate staining (IRS = 6–8). All five cases with positive nuclear β-catenin expression were from patients with endometrioid EC (EEC). On the other hand, tumor-adjacent normal tissues that were valid to be tested showed that cytoplasmic Dkk1, cytoplasmic β-catenin, and nuclear ER isoforms were overall positively stained. However, nuclear β-catenin showed negative staining in all tested cases. The expression levels of ERβ2 and ERα, but not of Dkk1, β-catenin, ERβ1, or ERβ5 were significantly lower in tumor cells compared with tumor-adjacent normal cells (Table 3).

Table 3.

Gene expression levels of Dkk1, CTNNB, ERS1, and ERS2 in all tested tissues

Parameter Normal endometrium tissues Endometrium tumor tissues p
∆CtDkk1 10.09 (8.98–11.49) (N = 7) 10.88 (2.57–13.08) (N = 18) NS
∆CtCTNNB 1.89 (1.11–2.09) (N = 7) 1.90 (− 0.59 to 3.22) (N = 18) NS
∆CtERS1 2.91 (1.67–3.87) (N = 6) 3.74 (0.29–8. 69) (N = 16) NS*
∆CtERS2 10.25 (9.78–11.77) (N = 6) 11.61 (6.11–14.21) (N = 17) NS
Open in a new tab

Results are showed as [mean, range]. p value is tested with Mann–Whitney U test. NS not significant (p > 0.05 two-tailed)

N number of tested cases

*p > 0.05 two-tailed/p = 0.048 one-tailed

Next, we analyzed the correlation between the expression level of each tested protein and clinicopathological characteristics including tumor grading, tumor size, lymphatic metastasis, and myometrial invasion. Our results revealed that decreased protein expression levels of cytoplasmic β-catenin were associated with advanced G status [tumors with G1 and G2 status (N = 36) vs tumors with G3 status (N = 19): IRS (mean-range) = 8 (3–12) vs 8 (4–12) respectively, p = 0.029] (Fig. 4a). In addition, increased protein expression levels of ERα were associated with better G and N status [tumors with G1 and G2 status (N = 36) vs tumors with G3 status (N = 19): IRS (mean-range) = 9 (0–12) vs 4 (0–12) respectively, p = 0.002; tumors with N0 status (N = 35) vs tumors with N1 status (N = 11): IRS (mean-range) = 8 (0–12) vs 4 (0–12), respectively, p = 0.016] (Fig. 4b, c). In addition, we found that increased IRS values of ERα, ERβ2, and cytoplasmic β-catenin were associated with better OS [p values: 0.012, 0.042, and 0.041, respectively] (Fig. 5a–c). Moreover, patients with lower protein expression levels of ERα had recurrence or metastasized earlier than those with higher protein expression levels (p = 0.008) (Fig. 5d) and patients with lower protein expression levels of ERβ2 showed marginal tendency to have recurrence or metastasized earlier than those with higher protein expression levels (p = 0.052) (Fig. 5e).

Fig. 4.

Fig. 4

Open in a new tab

The correlation of protein expression level of cytoplasmic β-catenin and nuclear ERα in tumor cells with G and N status. a Correlation between cytoplasmic protein expression levels of β-catenin and G status [tumors with G1 and G2 status vs tumors with G3 status: IRS (mean-range) = 8 (3–12) vs 8 (4–12), respectively, p = 0.029]. b, c Correlation between protein expression levels of ERα and G and N status, respectively [Tumors with G1 and G2 status vs tumors with G3 status: IRS (mean-range) = 9 (0–12) vs 4 (0–12), respectively, p = 0.002; tumors with N0 status vs tumors with N1 status: IRS (mean range) = 8 (0–12) vs 4 (0–12) respectively, p = 0.016]

Fig. 5.

Fig. 5

Open in a new tab

The correlation of protein expression level of cytoplasmic β-catenin and nuclear-ERα and ERβ2 in tumor cells with OS and PFS rates. a–c OS curves indicated the prolonged OS of patients with increased protein expression levels of ERβ2, ERα, and cytoplasmic β-catenin, respectively, compared with patients with lower protein expression levels. All p < 0.05. d PFS curve indicated patients with lower protein expression levels of ERα had recurrence or metastasized earlier than those with higher protein expression levels (p = 0.008). e PFS curve indicated patients with lower protein expression levels of ERβ2 tend to have recurrence or metastasized earlier than those with higher protein expression levels (p = 0.052).*Marginal trend toward significance

Then, we calculated the ratio between staining scores of ERα and staining scores of ERβ1, ERβ2, and ERβ5 separately. The obtained values were dichotomized into two groups including cases with ratio less than 1 and cases with ratio equal or more than 1. The correlation between different groups and OS or PFS was tested and data revealed that patients with ERα/Eβ1 or ERα/Eβ5 more than 1 had less rate of recurrent disease or later metastasis (p = 0.007, p = 0.002, p = 0.140, respectively) (Fig. 6a–c). However, we found no significant correlation with OS rates (data are not shown).

Fig. 6.

Fig. 6

Open in a new tab

The correlation between ERα/ERβ1, ERα/ERβ2, and ERα/Erβ5 and PFS rate. a PFS curve indicates that patients with ERα/Eβ1 more than 1 had less rate of recurrent disease or later metastasis (p = 0.007). b PFS curve indicates that rate of recurrent disease or later metastasis in patients with ERα/Eβ2 less than 1 did not differ from that in those with ERα/Eβ2 more than 1 (p = 0.140). c PFS curve indicates that patients with ERα/Eβ5 more than 1 had less rate of recurrent disease or later metastasis (p = 0.002)

After that, the correlation between the expression levels of different tested proteins in 55 tumor tissues were tested and results revealed that the protein expression levels of ERα were correlated with the protein expression levels of Dkk1 and cytoplasmic β-catenin (p = 0.011, correlation coefficient = 0.341 and p = 0.029, correlation coefficient = 0.295, respectively) (Fig. 7a, b).

Fig. 7.

Fig. 7

Open in a new tab

The correlation between protein expression level of ERα with it of Dkk1 and cytoplasmic β-catenin in 55 matched cases. a, b Protein expression levels of ERα were significantly correlated with protein expression levels of Dkk1 and β-catenin respectively

In the next step, our RT-PCR results showed that the gene expression levels of Dkk1, CTNNB, ERS1, and ERS2 in the endometrium tumor tissues did not significantly differ from their expression levels in normal endometrium tissues (Table 4). Nevertheless, we found that the expression levels of CTNNB were significantly correlated with expression levels of ERS1 in 16 tumor tissues (p = 0.030) (Fig. 8).

Table 4.

Protein expression levels of Dkk1, β-catenin, ERα, and ERβ variants (ERβ1, ERβ2, ERβ5) in paraffin-embedded tumor tissues and in tumor-adjacent normal tissues

Tested protein Tumor cells Tumor-adjacent normal cells p
Number of tested cases IRS Number of tested cases (%) IRS IRS Mean (range)
(0–2) Negative staining (%) (3–4) Weak staining (%) (6–8) Moderate staining (%) (9–12) Strong staining (%) IRS Mean (range) (%) (0–2) Negative staining (%) (3–4) Weak staining (%) (6–8) Moderate staining (%) (9–12) Strong staining (%)
Cytoplasmic Dkk1 55 7 40 40 13 6 (0–9) 14 0 36 64 0 7 (4–8) NS
Cytoplasmic β-catenin 55 0 9 62 29 8 (3–12) 13 0 23 54 23 8 (4–12) NS
Nuclear β-catenin 55 91 5 4 0 0 (0–6) 13 100 0 0 0 2 (0–2) NS*
Nuclear ERβ1 55 11 23 55 11 6 (0–12) 14 0 8 25 67 4 (3–12) NS
Nuclear ERβ2 55 3 9 46 42 8 (0–12) 12 0 55 45 0 12 (4–12) 0.017
Nuclear ERβ5 55 11 31 51 7 6 (0–12) 11 0 15 8 77 4 (3–8) NS
Nuclear ERα 55 13 13 32 42 8 (0–12) 13 0 57 36 7 12 (4–12) 0.004
Open in a new tab

IRS results are presented as mean (range). p value is tested with Mann–Whitney U test

*Nuclear β-catenin expression was compared in 13 matched tumor and tumor-adjacent normal tissues

Fig. 8.

Fig. 8

Open in a new tab

The correlation between ERS1 and CTNNB gene expression levels in 16 EC tissues

Finally, we tested if the observed correlation between protein expression level of ERα with that of Dkk1 and β-catenin might affect the association of ERα protein expression levels with OS. For this purpose, we have generated four groups of patients with regard to ERα- and Dkk1 protein expression [group 1: ERα-IRS = 6–12/Dkk1-IRS = 6–12 (+, +), group 2: ERα-IRS = 6–12/Dkk1-IRS = 0–4 (+, −), group 3: ERα-IRS = 0–4/Dkk1-IRS = 6–12 (−, +), group 4: ERα-IRS = 0–4/Dkk1-IRS = 0–4 (−, −)], and another four groups of patients with regard to ERα- and β-catenin protein expression [group 1: ERα-IRS = 6–12/β-catenin-IRS = 6–12 (+, +), group 2: ERα-IRS = 6–12/β-catenin-IRS = 0–4 (+, −), group 3: ERα-IRS = 0–4/β-catenin-IRS = 6–12 (−, +), group 4: ERα-IRS = 0–4/β-catenin-IRS = 0–4 (−, −)]. Results revealed that correlation between protein expression level of ERα and OS rate was no more significant (Fig. 9a, b).

Fig. 9.

Fig. 9

Open in a new tab

a The correlation between protein expression levels of ERα in regard to protein expression levels of Dkk1 with OS. b The correlation between protein expression levels of ERα in regard to protein expression levels of β-catenin with OS

Correlation between protein expression levels and gene expression levels of different tested markers

We interestingly found that the expression pattern of Dkk1 and β-catenin in tumor cells was different from that in normal endometrium cells when compared at protein level, but not at gene level (Fig. 10). Serum protein levels of Dkk1 were neither correlated with the expression levels in paraffin tumor tissues (p = 0.921) nor with its gene expression in fresh tumor tissues (p = 0.381) in 44 and 9 matched cases, respectively. In addition, no significant correlation between protein expression levels and gene expression levels of Dkk1, β-catenin, ERα, or ERβ was reported in 17, 17, 15, and 16 matched cases, respectively (all p > 0.05).

Fig. 10.

Fig. 10

Open in a new tab

Protein and gene expression levels of Dkk1 and β-catenin in tumor and normal cells. a Protein expression levels of Dkk1 and β-catenin in 13 matched tumor and normal tissues. Dkk1 staining scores in comparison with β-catenin staining scores were significantly different between normal and tumor cells. b Gene expression levels of Dkk1 and β-catenin in 16 tumor tissues and in 6 normal tissues. Gene expression of Dkk1 and β-catenin in tumor tissues presented the same pattern as that in normal tissues

Discussion

Available data have described a large number of aberrant signaling and genomic changes involved in EC pathogenesis. However, the occurrence frequency of these molecular disorders and their plausible crosstalk could be crucial to the development of targeted and more successful therapies in the future.

Wnt/β-catenin signaling is an evolutionarily conserved signal transduction pathway that involves many developmental and physiological processes. β-Catenin is the hallmark component of this signaling pathway. In the absence of Wnt signal, cytoplasmic β-catenin is phosphorylated promoting its rapid degradation by the proteasomes. On the other hand, activation of Wnt signaling leads to the accumulation of unphosphorylated β-catenin and its translocation into the nucleus where it activates the transcription of Wnt target genes that influence cell cycle regulation as well as invasion and metastasis of cancer cells (Teo and Kahn 2010). Wnt/β-catenin cascade is negatively regulated by a wide range of effectors including Dkk1, the founding member of dickkopf secreted proteins family (Niehrs 2006). Aberrant Wnt/β-catenin signaling at gene and/or protein level has been showed to be involved in EC carcinogenesis (Liu et al. 2014; Chen et al. 2013). In this study, we found that serum levels of Dkk1 were significantly increased in EC patients compared with healthy controls. However, no significant correlation between serum levels of Dkk1 and any of the clinicopathologic features, OS, and PFS was reported. Elevated serum levels of Dkk1 have been demonstrated by different studies in a range of cancer entities and this was frequently associated with worse prognosis (Kagey and He 2017). So far, our study is the second after Jiang et al.’s study which also found that Dkk1 serum level was significantly higher in EC patients than in healthy females and was associated with advanced staging according to the FIGO stage (Jiang et al. 2009). These observations suggest that serum Dkk1 might be a promising diagnostic marker for EC. However, the small tested numbers of patients involved either in our study or in Jiang et al.’s study (54 and 28 patients, respectively) could have some limits. Therefore, further large-scale, multicenter studies are warranted to support the present findings, suggesting the use of serum Dkk1 as a diagnostic marker of EC in the clinic.

Further, we have investigated if the aberrant protein levels of Dkk1 in the serum might be due to dysregulated Dkk1 gene or protein expression in tumor tissues. Our data revealed that cytoplasmic protein expression of Dkk1 was positive in all tested normal tissues as well as in the vast majority of tested tumor tissues showing no significant differences. In addition, no significant correlation between Dkk1-IRS and clinicopathological characteristics was reported. Although a large body of literature available has addressed the alterations in the Wnt/β-catenin signaling in EC, data on the biology of Dkk1 function are very limited. Yi et al. showed, using IHC, that Dkk1 protein expression level in EC was significantly lower than it in benign endometrium and was significantly correlated with disease histology and clinical stage but not with pathology categories or lymph nodes metastasis (Yi et al. 2009). A later study of the same researchers demonstrated that the transient transfection of Dkk1 siRNA in Ishikawa endometrial carcinoma cells resulted in a markedly increased cell invasion and migration when compared with untreated cells (Yi et al. 2013). Discrepancies between our observations and those of Yi et al. could be due to the differences in number and in clinicopathological characteristics of the included cases such as age, menopausal status, and histological classification.

β-Catenin has been suggested to be essential for establishing endometrial homeostasis and normal functioning of the uterus (Jeong et al. 2009). Most studies on β-catenin alterations in EC have focused mainly on the CTNNB1 mutations (Dellinger et al. 2012). Nuclear localization of β-catenin has been observed during the mid- and late proliferative phase of menstrual cycle in normal endometrium indicating its relation to the activated cell proliferation signals (Nei et al. 1999). Nuclear β-catenin expression in EC was identified in 25% and 34.2% of 40 and 128 EC cases, respectively, showing rational positive staining with or without mutations for this gene (Moreno-Bueno et al. 2002; Palacios et al. 2001). In a recent study of Sarkar et al., β-catenin membranous and cytoplasmic localization in proliferative endometrium and endometrial hyperplasia cases versus membranous, cytoplasmic, and nuclear localization in endometrial carcinoma cases was reported. In addition, nuclear positivity of β-catenin was shown to be associated with increased disease severity (Sarkar et al. 2018). In our study, we have found that cytoplasmic β-catenin was expressed in all tested normal and tumor tissues and nuclear β-catenin was absent in all tested normal tissues and in the vast majority of tumor tissues presenting no significant differences. However, increased expression of cytoplasmic β-catenin was associated with reduced G status and better OS. Dysregulated β-catenin has been implied in EC pathogenicity in accordance with its dual function in transcriptional activation and in cell adhesion linking cadherins to the actin cytoskeleton (Jamora and Fuchs 2002). β-Catenin immunostaining was identified in the cytoplasm and membrane of tumor cells in 85% (34/40) of EEC cases and increased mean staining scores were significantly associated with increased differentiation degree and decreased depth of invasion stage (Florescu et al. 2016). Moreover, data from Nesina et al. study showed that expression of cytoplasmic–nuclear β-catenin in 55 EC patients was significantly reduced in stage III compared with those at stages I and II of the disease (Nesina et al. 2018). Data from these studies as well as from our study indicate clearly the aberrant Wnt/β-catenin signaling in EC. Supporting this notion, we found that tumor cells presented differential staining profile of Dkk1- and β-catenin proteins compared with that in normal cells. In contrast, at gene expression level, we have found that Dkk1 and CTNNB1 expression levels had similar expression pattern in tumor cells and in normal cells. These observations suggest that alterations in Dkk1- and/or β-catenin expression levels associated with EC might have occurred at the protein level not at the gene level.

EC is hormone dependent. Nuclear ER (ERα and ERβ) are expressed distinctly in the endometrium leading to cellular proliferation and differentiation (Yang et al. 2002). Evidence for the involvement of ERα and ERβ in EC has been reported by different studies (Backes et al. 2016). However, some questions concerning utilization of ER testing on tumor tissues, as indicator for surgery extent and treatment scenarios remain and yet need to be addressed. Our results showed that ERα, ERβ1, ERβ2, and ERβ5 proteins were positively stained in all tested normal tissues and in the vast majority of tumor tissues. ERα and ERβ2 showed significantly lower staining positivity in tumor tissues compared with normal tissues. Moreover, decreased staining scores of ERα in tumor tissues were associated with worse disease status including tumor grading and lymph node metastasis. In addition, patients with higher tumor staining scores of ERα and ERβ2 had better OS and PFS compared with those who had lower scores. At gene expression levels, we found that ERα gene tends to be lower expressed in tumor tissues compared with normal tissues. Our IHC observations are, in part, in line with those from other studies. Collins and colleagues showed that the expression of ERβ1, ERβ2, and ERβ5 was readily detected in well, moderately, and poorly differentiated EC tissues and ERα was low/absent in poor-grade cancers. Moreover, mRNA amount of ERα was significantly lower in samples with poor grades compared with those with well or moderate grades (Collins et al. 2009). Moreover, while immunostaining of ERα was decreased significantly or even lost, in EC samples compared with normal endometrium samples, ERβ expression showed no significant differences (Hu et al. 2005). Skrzypczak et al. examined the expression of ERα and different ERβ splice variants in 21 human endometrium samples and 19 cases of EC using quantitative RT-PCR and found that ERα- and ERβ gene expression in normal and cancerous endometrium was not significantly different, presenting similar results to ours (Skrzypczak et al. 2004). On the contrary, data from other studies reported significant changes in mRNA amount of ERα and/or ERβ in EC samples compared with normal endometrium (Smuc and Rinzer 2009; Häring et al. 2012).

ERβ could have a different action and activity compared to ERα, since it generally counteracts the ERα-promoted cell hyperproliferation in tissues such as breast and uterus (Heldring et al. 2007). Therefore, the ERα/ERβ ratio has been used to predict disease outcomes in different malignancies including breast cancer and EC. In a large cohort of patients with EEC including 315 patients, a ratio of ERα/ERβ < 1 was related to a shorter disease-free survival (Jongen et al. 2009). Moreover, in a group of 121 EEC cases, an ERα/ERβ1 or ERα/ERβ2 ratio of 1 or less has been proved to be associated with a higher risk of death (Zannoni et al. 2013). In line with these results, our data showed that an ERα/ERβ1 ratio and an ERα/ERβ5 ratio of 1 or more were associated with favorable PFS and OS. The significance of the relative expression of both ER subtypes in EC is still controversially discussed. Therefore, further systematic and well-designed research is warranted.

β-Catenin has been found to be a potent activator of different nuclear receptors including ER (Mulholland et al. 2005). Potential crosstalk between Wnt/β-catenin and estrogen signaling in vivo has been reported in uterus using several genetically engineered mouse models. Hou et al. showed enhanced nuclear localization of β-catenin in epithelial cells of the endometrium after exogenous estrogen treatment in mice and these effects were inhibited by delivery of Sfrp2, a known Wnt antagonist, in the uterus (Hou et al. 2004). In a recent study by Goad et al., sustained activation of Wnt/β-catenin signaling in doxycycline-treated mutant uteri resulted in endometrial hyperplasia. However, the constitutive activation of the Wnt/β-catenin pathway in the presence of unopposed estrogen resulted in endometrial cancer indicating a synergistic action of additional downstream effectors of estrogen signaling for malignant transformation (Goad et al. 2018). In our study, we showed for the first time that the protein expression levels of ERα correlated positively with protein expression levels of Dkk1 and β-catenin in tumor cells, but not in normal cells. Moreover, the significant association between increased staining scores of ERα and favorable OS no more existed when combined with either staining scores of Dkk1 or staining scores of β-catenin. Together, all these observations underlie the potential interaction between Wnt/β-catenin and estrogen signaling in EC providing a platform for further research on understanding the sophisticated connections of signaling networks in EC.

Acknowledgements

We thank Barbara Linxweiler from the Department of Obstetrics and Gynecology, University Medical School of Saarland, for her excellent technical assistance during the course of this research.

Author contributions

MK: study conception and design, supervision, patient’s data collection and processing, sample collection, participation in performing the experiments, interpretation and analysis of results, and taking responsibility in the construction of the whole body of the manuscript; CD: acquisition of data, sample collection, doing the main part of the experiments; AS: IHC result interpretation and analysis; RMB: providing paraffin-embedded tissue samples, IHC method validation, and doing critical review; ZT: patient’s data collecting and processing; II: clarification of patients; ES: supervision and doing critical review; IJ: study conception and design, supervision, and doing critical review.

Compliance with ethical standards

Conflict of interest

All authors declare that there is no conflict of interest and they have seen and approved the manuscript submitted.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. American Cancer Society (2017) Cancer facts and figures, 2017. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statstics/annual-cancer-facts-and-figures/2017/cancer-facts-and-figures-2017.pdf. Accessed 11 July 2017
  2. Backes FJ, Walker CJ, Goodfellow PJ et al (2016) Estrogen receptor-alpha as a predictive biomarker in endometrioid endometrial cancer. Gynecol Oncol 141:312–317 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Chen GG, Zeng Q, Tse GM (2008) Estrogen and its receptors in cancer. Med Res Rev 28:954–974 [DOI] [PubMed] [Google Scholar]
  4. Chen H, Wang Y, Xue F (2013) Expression and the clinical significance of Wnt10a and Wnt10b in endometrial cancer are associated with the Wnt/β-catenin pathway. Oncol Rep 29:507–514 [DOI] [PubMed] [Google Scholar]
  5. Collins F, MacPherson S, Brown P et al (2009) Expression of oestrogen receptors, ERα, ERβ, and ERβ variants, in endometrial cancers and evidence that prostaglandin F may play a role in regulating expression of Erα. BMC Cancer. 10.1186/1471-2407-9-330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dellinger TH, Planutis K, Tewari KS et al (2012) Role of canonical Wnt signaling in endometrial carcinogenesis. Expert Rev Anticancer Ther 12:51–62 [DOI] [PubMed] [Google Scholar]
  7. Florescu MM, Pirici D, Simionescu CE et al (2016) E-cadherin and β-catenin immunoexpression in endometrioid endometrial carcinoma. Rom J Morphol Embryol 57:1235–1240 [PubMed] [Google Scholar]
  8. Goad J, Ko YA, Kumar M et al (2018) Oestrogen fuels the growth of endometrial hyperplastic lesions initiated by overactive Wnt/β-catenin signalling. Carcinogenesis 39:1105–1116 [DOI] [PubMed] [Google Scholar]
  9. Hapangama DK, Kamal AM, Bulmer JN (2015) Estrogen receptor β: the guardian of the endometrium. Hum Reprod Update 21:174–193 [DOI] [PubMed] [Google Scholar]
  10. Häring J, Skrzypczak M, Stegerer A et al (2012) Estrogen receptor β transcript variants associate with oncogene expression in endometrial cancer. Int J Mol Med 29:1127–1136 [DOI] [PubMed] [Google Scholar]
  11. Heldring N, Pike A, Andersson S et al (2007) Estrogen receptors: how do they signal and what are their targets. Physiol Rev 87:905–931 [DOI] [PubMed] [Google Scholar]
  12. Hou X, Tan Y, Li M et al (2004) Canonical Wnt signaling is critical to estrogen-mediated uterine growth. Mol Endocrinol 18:3035–3049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hu K, Zhong G, He F (2005) Expression of estrogen receptors ERalpha and ERbeta in endometrial hyperplasia and adenocarcinoma. Int J Gynecol Cancer 15:537–541 [DOI] [PubMed] [Google Scholar]
  14. Jamora C, Fuchs E (2002) Intercellular adhesion, signalling and the cytoskeleton. Nat Cell Biol 4:E101–E108 [DOI] [PubMed] [Google Scholar]
  15. Jeong JW, Lee HS, Franco HL et al (2009) β-Catenin mediates glandular formation and dysregulation of beta-catenin induces hyperplasia formation in the murine uterus. Oncogene 28:31–40 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jiang T, Wang S, Huang L et al (2009) Clinical significance of serum DKK-1 in patients with gynecological cancer. Int J Gynecol Cancer 19:1177–1181 [DOI] [PubMed] [Google Scholar]
  17. Jongen V, Briët J, de Jong R et al (2009) Expression of estrogen receptor-alpha and -beta and progesterone receptor-A and -B in a large cohort of patients with endometrioid endometrial cancer. Gynecol Oncol 112:537–542 [DOI] [PubMed] [Google Scholar]
  18. Kagey MH, He X (2017) Rationale for targeting the Wnt signalling modulator Dickkopf-1 for oncology. Br J Pharmacol. 10.1111/bph.13894 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kiewisz J, Wasniewski T, Kmiec Z (2015) Participation of WNT and β-catenin in physiological and pathological endometrial changes: association with angiogenesis. Biomed Res Int. 10.1155/2015/854056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Liu Y, Patel L, Mills GB et al (2014) Clinical significance of CTNNB1 mutation and Wnt pathway activation in endometrioid endometrial carcinoma. J Natl Cancer Inst. 10.1093/jnci/dju245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Logan CY, Nusse R (2004) The Wnt signaling pathway in development and disease. Annu Rev Cell Dev Biol 20:781–810 [DOI] [PubMed] [Google Scholar]
  22. Moreno-Bueno G, Hardisson D, Sánchez C et al (2002) Abnormalities of the APC/beta-catenin pathway in endometrial cancer. Oncogene 21:7981–7990 [DOI] [PubMed] [Google Scholar]
  23. Mulholland DJ, Dedhar S, Coetzee GA et al (2005) Interaction of nuclear receptors with the Wnt/beta-catenin/Tcf signaling axis: Wnt you like to know? Endocr Rev 26:898–915 [DOI] [PubMed] [Google Scholar]
  24. Nei H, Saito T, Yamasaki H et al (1999) Nuclear localization of beta-catenin in normal and carcinogenic endometrium. Mol Carcinog 25:207–218 [PubMed] [Google Scholar]
  25. Nesina IP, Iurchenko NP, Buchynska LG (2018) Markers of the epithelial–mesenchymal transition in cells of endometrial carcinoma. Exp Oncol 40:218–222 [PubMed] [Google Scholar]
  26. Niehrs C (2006) Function and biological roles of the Dickkopf family of Wnt modulators. Oncogene 25:7469–7481 [DOI] [PubMed] [Google Scholar]
  27. Palacios J, Catasús L, Moreno-Bueno G et al (2001) Beta- and gamma-catenin expression in endometrial carcinoma. Relationship with clinicopathological features and microsatellite instability. Virchows Arch 438:464–469 [DOI] [PubMed] [Google Scholar]
  28. Remmele W, Stegner HE (1987) Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue. Pathologe 8:138–140 [PubMed] [Google Scholar]
  29. Sarkar S, Sarkar R, Khandakar B et al (2018) Study of beta-catenin expression: In Endometrial hyperplasia and carcinoma. APALM. 10.21276/apalm.1899 [Google Scholar]
  30. Siegel RL, Miller KD, Jemal A (2016) Cancer statistics. CA Cancer J Clin 66:7–30 [DOI] [PubMed] [Google Scholar]
  31. Skrzypczak M, Bieche I, Szymczak S et al (2004) Evaluation of mRNA expression of estrogen receptor beta and its isoforms in human normal and neoplastic endometrium. Int J Cancer 110:783–787 [DOI] [PubMed] [Google Scholar]
  32. Smuc T, Rizner TL (2009) Aberrant pre-receptor regulation of estrogen and progesterone action in endometrial cancer. Mol Cell Endocrinol 301:74–82 [DOI] [PubMed] [Google Scholar]
  33. Teo JL, Kahn M (2010) The Wnt signaling pathway in cellular proliferation and differentiation: a tale of two coactivators. Adv Drug Deliv 62:1149–1155 [DOI] [PubMed] [Google Scholar]
  34. WHO, GLOBOCAN (2012) Estimated cancer incidence, mortality and prevalence worldwide in 2012. https://globocan.iarc.fr/Pages/fact_sheets_population.aspx. Accessed 03 Apr 2015
  35. Yang P, Kriatchko A, Roy SK (2002) Expression of ER-alpha and ER-beta in the hamster ovary: differential regulation by gonadotropins and ovarian steroid hormones. Endocrinology 143:2385–2398 [DOI] [PubMed] [Google Scholar]
  36. Yi N, Liao QP, Li T et al (2009) Novel expression profiles and invasiveness-related biology function of DKK1 in endometrial carcinoma. Oncol Rep 21:1421–1427 [DOI] [PubMed] [Google Scholar]
  37. Yi N, Liao QP, Li ZH et al (2013) RNA interference-mediated targeting of DKK1 gene expression in Ishikawa endometrial carcinoma cells causes increased tumor cell invasion and migration. Oncol Lett 6:756–762 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Young SL (2013) Oestrogen and progesterone action on endometrium: a translational approach to understanding endometrial receptivity. Reprod Biomed Online 27:497–505 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zannoni GF, Monterossi G, De Stefano I et al (2013) The expression ratios of estrogen receptor α (ERα) to estrogen receptor β1 (ERβ1) and ERα to ERβ2 identify poor clinical outcome in endometrioid endometrial cancer. Hum Pathol 44:1047–1054 [DOI] [PubMed] [Google Scholar]
  40. Zhao C, Dahlman-Wright K, Gustafsoon JÅ (2010) Estrogen signaling via estrogen receptor β. J Biol Chem 285(51):39575–39579 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Cancer Research and Clinical Oncology are provided here courtesy of Springer

RESOURCES