Abstract
Background/Aim: The purpose of this study was to examine the expression of estrogen receptor α (ERα) and β (ERβ), androgen receptor (AR), SIRT1, SIRT2 and SIRT3 in prostate cancer (PCa). Materials and Methods: From October 2010 to January 2015, 70 patients who had undergone radical prostatectomy following a PCa diagnosis were enrolled in our study. Normal prostate tissue (NPT) and prostate cancer tissues (PCAT) were separated, and the expression of each receptor in each tissue was analyzed with immunochemical staining. Univariate and multivariate analyses were performed to identify factors affecting the development of PCa. Results: ERβ and AR were highly expressed in PCAT compared with NPT (p<0.05). SIRT2 was highly expressed in NPT and PCAT (p<0.05). Univariate and multivariate analyses showed that AR and SIRT2 affect PCa development. Conclusion: AR is a risk factor for PC, and SIRT2 is associated with a lower incidence of PCa.
Prostate cancer (PCa) is the most prevalent male malignancy in many regions of the world. Remarkable racial and ethnic differences in its incidence have been reported, ranging from 4.4 per 100,000 to 118.2 per 100,000 persons in India and the USA, respectively (1). According to data from the Korea National Statistical Office, in 2012, the prevalence of PCa in Korea over the previous 5 years was reported to be about 37.3 per 100,000 people, and in 2016, increased to about 46.2 per 100,000 individuals (2). Also, according to the The Global Cancer Observatory (GLOBOCAN) results, PCa was the second most frequently diagnosed cancer and the fifth leading cause of cancer mortality among men worldwide in 2012 (1). Therefore, various studies are being conducted to reveal the causes of PCa, and to identify diagnostic and prognostic factors for PCa.
Furthermore, there have been few studies on the sirtuin family (SIRT 1-7) to date. Many studies on androgen receptors, which are known to be associated with the onset and progression of PCa, have been published, however, there have not been many studies regarding the role of estrogen receptor α (ERα) and estrogen receptor β (ERβ) in PCa development.
Sirtuin is a highly conserved protein family consisting of seven members in mammals (SIRT 1-7) that function in cellular metabolism, chromatin stability, and DNA repair (3). The majority of sirtuins act as NAD+ dependent histone deacetylases, but some of them, can also act as nicotinamide adenine dinucleotide+ (NAD+) dependent Adenosine di-phosphate (ADP) ribosyl transferases (4). DNA methylation and histone tail modification are two key epigenetic processes that play vital roles in PCa progression (5).
It has been previously reported that SIRT1, a member of the class III histone deacetylases (HDACs), negatively regulates H2A.Z levels in cardiomyocytes through targeting this histone variant to degradation via the ubiquitin/proteasome pathway. In addition, it has been reported that SIRT1 is overexpressed in some cancers, but down-regulated in others. This supports its role as an oncogene or a tumor suppressor gene (TSG) (6).
SIRT2 overexpression prolongs longevity in mice hypomorphic for BubR1, and mice deficient in SIRT2 develop breast, liver, and other cancers, suggesting that SIRT2 functions in both aging and tumor suppression (7). Indeed, SIRT2 expression is decreased in several cancers, including human breast, liver, glioma, renal, prostate, lung, uterine, and basal cell carcinomas. Moreover, SIRT2 is mutated or deleted in ovarian, adenoid cystic, cervical, uterine, lung, pancreatic, gastric, esophageal, colorectal, liver, melanoma, testicular germ cell, thyroid, and breast cancers (7). Initially identified as a tumor suppressor in breast cancer, SIRT3 maintains the integrity of mitochondria during stress and Hif1α destabilization. However, little is known about SIRT3’s function in PCa (8).
The function of ERα in the prostate gland has been shown to be mediated via estrogen receptors (ERs) within differentiated basal, luminal, and stromal cell populations (9, 10). Human prostate stem cells express ERα and ERβ and exhibit proliferative responses to estrogens. Recently, our laboratory discovered that normal human prostate stem cells, although AR negative and resistant to androgen exposures, express ERα and ERβ, and transduce signals when exposed to 17b-estradiol (E2) or endocrine-disrupting agents that mimic estrogens (11).
The Androgen receptor (AR) ligand-induced transcription factor is expressed in primary prostate cancer and metastases. AR regulates multiple cellular events, proliferation, apoptosis, migration, invasion, and differentiation. Its expression in prostate cancer cells is regulated by steroid and peptide hormones (12). Therefore, in this study, the expression patterns of ERα, ERβ, AR, SIRT1, SIRT2, and SIRT3 in PCa and normal tissues were compared and analyzed. Based on the results of this study, we identified risk factors for PCa development.
Materials and Methods
From October 2010 to January 2015, 70 patients who had undergone radical prostatectomy following a PCa diagnosis were enrolled in our study. Normal prostate tissue (NPT) and prostate cancer tissues (PCAT) were separated, and the expression of each receptor in each tissue was analyzed with immunochemical staining. Univariate and multivariate analyses were performed to identify factors affecting the development of PCa.
All procedures performed in this study were in accordance with the Ethical standards of the Institutional Research Committee and the Declaration of Helsinki in 1964. Because this study was a retrospective study, informed consent was waived. This study was approved by the Gyeongsang National University Hospital Institutional Review Board (approval number: 2020-06-005).
Subjects. This study was conducted on 70 patients who had undergone radical prostatectomy (RP) following a PCa diagnosis at our hospital from October 2010 to January 2015. We collected demographic information such as age, hypertension, and clinical information such as tPSA levels, testosterone levels, Gleason score, as well as imaging information from transrectal sonography, prostate magnetic resonance imaging (MRI), and bone scan. In addition, normal and PCa tissues were separated using a prostate paraffin tissue block stored in our hospital. We analyzed the ERα, ERβ, AR, SIRT1, SIRT2, and SIRT3 expression patterns of each tissue with immunochemical staining.
Pathologic data analysis. Two pathologists analysed the staging of the prostate cancer in accordance with the American Joint Committee on Cancer (AJCC) TNM staging, and the Gleason score in accordance with the Gleason scoring system.
Tissue microarray. Resected tumor samples were fixed overnight in 20% buffered neutral formalin. The samples were grossly examined, dissected, and embedded in paraffin blocks. Representative portions of tumor and adjacent normal tissues were selected by microscopic examination of hematoxylin and eosin- stained sections from each specimen. Two 2.0-mm cores of each tumor per case and one 2.0-mm core of normal tissue were assayed.
Immunohistochemical analysis. Immunohistochemical (IHC) staining was performed on tumor tissues and adjacent normal tissues. Tissue microarray blocks were cut into 4-μm slices for IHC staining. After deparaffinization and rehydration, slides were incubated in 3% hydrogen peroxide for 10 min to block endogenous peroxidase activity, which results in nonspecific background staining. Sections were then heated for 20 min in a potential transformer (PT) module buffer No.4 (SIRT1, SIRT2) and a citrate buffer (AR, ERα, ERβ, SIRT3, pH=6.0) in a microwave oven (700 W). After blocking endogenous peroxidase activity with a peroxidase quenching solution, the sections were incubated overnight at 4°C with antibodies. The following monoclonal or polyclonal antibodies were used: SIRT1 (clone H-95, Santa Cruz Biotechnology, Seoul, Republic of Korea, dilution 1:750), SIRT2 (clone H-300, Santa Cruz Biotechnology, dilution 1:1,600) SIRT3 (clone C73E3, Cell Signaling Technology, Seoul, Republic of Korea, dilution 1:200), ERα (clone 33, Novus biologicals, Gyeonggi-do, Republic of Korea, dilution 1:200), ERβ (clone 14C8, Novus Biologicals, dilution 1:400), and AR (clone 156C135.2, Novus Biologicals, dilution 1:400).
Scoring of immunoreactivity. The expression levels of SIRT1-3 were semi-quantitatively scored by assessing the intensity of staining (0, no staining; 1, mild staining; 2, moderate staining; and 3, strong staining) and the percentage of positively stained cells (0, <30%; 1, 30-49%; 2, 50-69%; and 3 ≥70%). The sum index was obtained by combining the staining intensity and the percentage scores. The degree of expression was evaluated by calculating the mean value of measured scores, and a final mean score of ≥ 4 was considered to indicate positive expression in a specimen. Otherwise, the tumor was considered negative (13).
The expression levels of ERα, ERβ, and AR were scored by assessing nuclear staining using a scoring method incorporating both the proportion of positive cells and the intensity of nuclear positivity. We scored the proportion of tumor cells in each of the samples as 0 (negative), 1+ (weak positive), 2+ (moderately positive), or 3+ (strong positive). We defined positivity as an at least 1% nuclear staining of tumor cells. Two pathologists, who were blinded to the clinical information, performed the scoring.
Statistical analysis. Descriptive statistics frequency analysis was performed. Comparative analysis of the expression patterns in NT and PCAT for each factor such as ERα, ERβ, AR, SIRT1, SIRT2, and SIRT3, was conducted by using the Pearson’s Chi-square test and analyzed 95% confidence interval (CI) and p-value. Single and multivariate regression analysis was performed to identify risk factors for PCa expression among the six factors of ERα, ERβ, AR, SIRT1, SIRT2, and SIRT3. In addition, univariate and multivariate regression analysis was performed to identify factors that affect AR and SIRT2 expression. Statistical significance was defined as p<0.05. Statistical analysis was performed with the use of the SPSS statistics 21 software.
Results
The mean age of the patients in this study was 67.09 (±5.53) years, the mean body mass index (BMI) was 23.76 (±2.70), the mean PSA levels were 27.76 (±87.56), and the mean prostate volume was 37.76 (±15.76) (Table I). Compared to the number of patients in other stage groups, the number of stage II patients was reported to be relatively high, 44 patients (62.9%), and the number of Gleason group I patients was reported to be 32 (46.5%), which was higher than that of other groups (Table I). Resection margins were negative in 53 patients (75.7%) as compared to positive in 17 patients (24.3%), and systemic metastasis was confirmed in 6 patients (8.6%). The average follow-up period was about 50.01 (±29.33) months, and biochemical recurrence (BCR) was found in 20 (28.6%) patients. The average period until BCR was found to be about 5.01 (±12.53) months. Three patients (4.3%) progressed to castration-resistant PCa (CRPC), and the average period until CRPC was found to be 2.54 (±13.07) months (Table I).
Baseline characteristics.
The overall analysis of the results of the expression patterns in normal tissue (NT) and prostate cancer tissue (PCAT) in the 70 patients for the three types of receptors, ERα, ERβ, and AR, are presented in Table II. Thirteen patients (18.6%) had positive ERα expression in NT and 57 patients (7.1%) had negative expression. In addition, in the PCAT group, five patients (7.1%) had positive expression and 65 patients (92.9%) had negative expression. The differences in the expression in NT and PCAT were not statistically significant (p=0.075). In addition, nine patients (12.9%) had positive ERβ expression in NT, and 57 patients (87.7%) had negative expression. In PCAT, 50 patients (71.4%) had positive expression, and 20 patients (28.6%) had negative expression. The differences in the expression in NT and PCAT were statistically significant (p=0.000). In the AR group, 23 patients (32.9%) had positive expression in NT, and 47 patients (67.1%) had negative expression. In PCAT, 56 patients (80.0%) had positive expression and 14 patients (20.0%) had negative expression. The differences in the expression in NT and PCAT were statistically significant (p=0.000).
Androgen and estrogen receptor expression patterns in normal tissue and prostate cancer tissue.
Table III shows the expression patterns of SIRT1, SIRT2 and SIRT3 in NT and PCAT. Regarding SIRT1, 41 patients (56.0%) had positive expression in NT and 29 patients (41.0%) had negative expression. In addition, in the PCAT, 50 patients (71.4%) had positive expression and 20 patients (28.6%) had negative expression. The differences in the expression in NT and PCAT were not statistically significant (p=0.111). For SIRT2, 60 patients (85.7%) had positive expression in NT, and 10 patients (14.3%) had negative expression. In PCAT, 43 patients (61.4%) had a positive expression and 27 patients (38.6%) had a negative expression. The differences in the expression in NT and PCAT were statistically significant (p=0.001). Regarding SIRT3, 23 patients (32.9%) had positive expression in NT, and 47 patients (67.1%) had a negative expression. In PCAT, 29 patients (41.4%) had a positive expression and 47 patients (58.6%) had a negative expression. The differences in the expression in NT and PCAT were not statistically significant (p=0.294).
Comparison of SIRT expression in normal tissue and prostate cancer tissue.
In this study, univariate and multivariate analyses were performed to find out the factors affecting the expression of PCa among the ERα, ERβ, AR, SIRT1, SIRT2, and SIRT3 receptors (Table III). AR [OR=27.928 (95%CI=16.857-199.061)] (p=0.000) and SIRT2 [OR=0.047 (95%CI=0.013-0.166)] (p=0.000) were identified as factors influencing the expression of PCa (Table IV).
Univariate and multivariate analysis of estrogen receptor, androgen receptor and SIRT expression in normal tissue and prostate cancer.
Also, in this study, univariate and multivariate analyses were performed to find the baseline characteristics affecting AR and SIRT2 expression in NT and PCAT (Tables V and VI). We did not identify risk factors that influence AR and SIRT2 over and under expression in PCa.
Risk factor affecting androgen expression in normal tissue and prostate cancer.
Risk factors affecting SIRT 2 expression in normal tissue and prostate cancer.
When IHC was performed for each receptor, the expression patterns in NT and PCAT were described in each figure. In the case of ERα and ERβ, compared to the control group, when the cells were IHC-positive, the cells showed a similar staining pattern (Figures 1 and 2). The same pattern was found for AR (Figure 3). Also, the same pattern could be confirmed for SIRT1, SIRT2 and SIRT3 (Figures 4, 5 and 6).
Androgen receptor expression patterns. A) Negative expression on normal tissue, B) positive expression on normal tissue, C) negative expression on prostate cancer tissue, D) positive expression on prostate cancer tissue.
Estrogen receptor α expression patterns. A) Negative expression on normal tissue, B) positive expression on normal tissue, C) negative expression on prostate cancer tissue, D) positive expression on prostate cancer tissue.
Estrogen receptor β expression patterns. A) Negative expression on normal tissue, B) positive expression on normal tissue, C) negative expression on prostate cancer tissue, D) positive expression on prostate cancer tissue.
SIRT1 expression patterns. A) Negative expression on normal tissue, B) positive expression on normal tissue, C) negative expression on prostate cancer tissue, D) positive expression on prostate cancer tissue.
SIRT2 expression patterns. A) Negative expression on normal tissue, B) positive expression on normal tissue, C) negative expression on prostate cancer tissue, D) positive expression on prostate cancer tissue.
SIRT3 expression patterns. A) Negative expression on normal tissue, B) positive expression on normal tissue, C) negative expression on prostate cancer tissue, D) positive expression on prostate cancer tissue.
Discussion
In this study, the expression of ERα was not found to be statistically significantly different between NT and PCAT. Thus, this study confirms that there is no correlation between ERα and PCa expression. However, according to a review study by Bonkhoff et al., ERα expression was found in 43% and 62% of PCa patients with high-grade Gleason scores of 4 and 5, respectively. More than 50% of CRPC patients showed a greater than 25% ERα expression, and it was reported that ERα was involved in the progression of PCa (14). According to Kowalska et al. ERα affects zearalenone (ZEA), which promotes the invasiveness and migration of PCa, and has been reported to be involved in the progression of PCa (15). Other studies, including the one by Guang et al., have studied the association between PCa and ERα polymorphisms. They have reported that ERα (rs9340799 or rs2234693) increases the risk of PCa development (16). In this study, it was confirmed that ERα is not related to PCa development. However, considering the characteristics of the ERα expression pattern reported by other studies, this study showed that the proportion of patients with Gleason scores 1 and 2 was high and that the proportion of patients with a low pathological stage was high. In the future, additional studies on a large number of patients are needed to determine the relationship between ERα and PCa expression.
Univariate analysis showed that the expression of ERβ in PCAT was significantly higher compared to that in NT [OR=16.944 (95%CI=7.092-40.487)] (p=0.000). However, multivariate analysis did not show any statistically significant difference (p<0.05). Of interest, according to a study reported by Long et al., ERβ overexpression reduced inflammation induced by lipopolysaccharide (LPS) treatment via regulation of the levels of proinflammatory cytokines, including TNF-α, MCP-1, IL-1β and IL-6. Furthermore, it was demonstrated that the ERβ antagonist PHTPP increased the expression of proinflammatory cytokines. It was also observed that ERβ overexpression suppressed the viability and migration of PC-3 and DU145 prostate cancer cells and promoted apoptosis (17). Majumdar et al. have reported on the role of ERβ in prostate cancer cells. ERβ causes a selective effect on mitogen-activated protein kinase (MARK) signaling cascades, one of the chains of the nucleus expression pathway for prostate cancer stem cells. This study has also confirmed that ERβ is expressed in high levels in prostate stem cells (11). In this study, univariate analysis confirmed that the expression rate of ERβ in PCAT was high. However, multivariate analysis did not show that expression of ERβ is a significant factor. However, several studies have shown that ERβ affects PCa expression, and based on this, further studies are needed to verify this conclusion.
In this study, AR was identified as a factor influencing PCa development in both univariate and multivariate analyses. Culig et al. have reviewed the role of AR in PCa. The androgen receptor ligand-induced transcription factor is expressed in primary prostate cancer and in metastases. AR regulates multiple cellular processes including proliferation, apoptosis, migration, invasion, and differentiation. Its expression in prostate cancer cells is regulated by steroid and peptide hormones (12). Heinlein et al. have reported that AR is expressed in primary prostate cancer and can be detected during its progression in both hormone sensitive and hormone refractory cancers (18). In addition, PCa already uses the relationship between the expression of AR, and androgen deprivation therapy (ADT) is currently used as a standard method in prostate cancer treatment. Dal et al. have reported an alternative treatment for castration-resistant prostate cancer that does not respond to ADT (19). As such, studies on treatment methods considering the relationship between the AR and PCa are continuously published. Similar to the previous studies, AR was identified as a risk factor for prostate cancer in our study.
In this study, expression of SIRT1 was not statistically significantly different between NT and PCAT [OR=1.768 (95%CI=0.875-3.574)] (p=0.112). According to a study by Kumar et al. on SIRT1, the exact mechanism is unknown, but it is possible that SIRT1 is significantly overexpressed in PCa (20). Similarly, in a study conducted by You et al. on SIRT1, it was confirmed that SIRT1 was overexpressed in PCa (21). According to a study by Lovaas et al. on the expression of SIRT1 and matrix metalloproteinases (MMPs), SIRT1 was found to be an important modulator of MMP2 expression and to be related to PCa invasion (22). In addition, several studies have reported that SIRT1 expression is a risk factor for PCa (23, 24). On the other hand, a study conducted by Fu et al. reported that SIRT1 inhibits AR-induced PCa (25). The results on the role of SIRT1 expression in PCa are contradictory, but generally, the opinion that SIRT1 overexpression functions as a risk factor for PCa is dominant. In this study, the univariate analysis showed that SIRT1 was overexpressed in PCAT, but there was no statistically significant difference compared to NT. Considering that there are not many studies on the correlation between SIRT1 and PCa, future research on this subject is needed.
Regarding SIRT2 expression in NT and PCAT, univariate analysis [OR=0.265 (95%CI=0.116-0.605)] showed that it is associated with a statistically significantly lower risk for PCa (p=0.002). Furthermore, multivariate analysis [OR=0.047 (95%CI=0.013-0.166)] showed statistically significantly lower risk for PCa (p=0.000).
Damodaran et al. have reported that there is an association between the expression of SIRT2, total P300, and H3K18Ac with PCa, and as a result, they showed that overexpression of SIRT2 lowers the risk of primary and metastatic PCa (5). In this study, overexpression of SIRT2 was confirmed in PCAT. Based on the results of univariate and multivariate analyses, SIRT2 is considered to reduce the risk of PCa. Additional studies need to be performed to confirm these results.
We conducted univariate analysis for SIRT3 expression in NT and PCAT [OR=1.445 (95%CI =0.726-2.879)], and found no statistically significant differences (p=0.295). Although a few studies have examined the correlation of SIRT3 with PCa, in 2015, Yizhou et al. reported that SIRT3 overexpression lowers the risk for PCa (26). However, the relationship between SIRT3 and PCa has not been thoroughly examined and additional studies are needed.
The limitation of this study is that we only used paraffin blocks to study the expression of ERα, ERβ, AR, SIRT1, SIRT2, and SIRT3 expression by IHC, and further genetic analysis was not possible. Based on these results, further research should be performed.
Conclusion
AR is a risk factor for PC expression, and SIRT2 reduces PCa development.
Footnotes
Authors’ Contributions
JAE HWI CHOI: Conceived and designed the study, collected the data, contributed data or analysis tools, wrote the paper. SEE MIN CHOI: Collected the data. SIN WOO LEE: Collected the data. SEONG UK JEH: Performed the analysis. JAE SEOG HYUN: Supervision of writing. MIN HO LEE: Collected the data. CHUNWOO LEE: Collected the data. SUNG CHUL KAM: Interpretation of data. DONG CHUL KIM: Contributed to preparing specimens. JONG SIL LEE: IHC of specimens and analysis of results. JEONG SEOK HWA: Conceived and designed the study, supervision of writing.
Conflicts of Interest
The Authors have no conflicts of interest to declare regarding this study.
- Received October 26, 2020.
- Revision received January 1, 2021.
- Accepted January 29, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.