Expression of receptor-binding cancer antigen expressed on SiSo cells and estrogen receptor subtypes in the normal, hyperplastic, and carcinomatous endometrium

Authors


Yi-Jun Wu, MD, PhD, Department of General Surgery, The First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310003, People’s Republic of China. Email: wu1jun@sina.com

Abstract

The objectives were to study the expression of receptor-binding cancer antigen expressed on SiSo cells (RCAS1) and estrogen receptor (ER) subtypes in the normal, hyperplastic, and carcinomatous endometrium and to explore their possible role in carcinogenesis and progression of endometrial carcinoma. Immunohistochemistry and semiquantitative reverse transcriptase–polymerase chain reaction (RT-PCR) were applied to detect protein and messenger RNA expression of RCAS1, ER-α, and ER-β in normal, hyperplastic, and carcinomatous endometrium. Western blotting was also used to detect the RCAS1 protein expression. Immunohistochemistry showed that the high expressions of RCAS1 protein were 0% (0/20), 9.1% (2/22), 40% (8/20), and 68.0% (34/50) in normal, simple, and complex hyperplasia, atypical hyperplasia, and endometrial carcinoma, respectively. There was a significant difference between each group (P < 0.05). The high-level expression of RCAS1 was detected more frequently in endometrial cancer with deep myometrial invasion, vascular invasion, and positive ER-α (P < 0.05). Two staining patterns of RCAS1 were observed. All normal, simple, and complex hyperplastic endometrium showed P pattern, while all malignant endometrium were of the D pattern. In atypical endometrium, 25% (5/20) cases showed D pattern. The Western blotting and RT-PCR results correlated with the immunohistochemistry results. The expression and distribution of RCAS1 may be involved in the malignant transformation of endometrium, and RCAS1 coexpression with ER-α may be associated with development and metastasis of endometrial carcinoma.

Endometrial carcinoma is one of the most frequently diagnosed gynecological malignancies in the world. Dualistic model for endometrial carcinogenesis is widely accepted(1). Type I endometrioid endometrial carcinoma is hormone dependent and its hormone receptor is positive, with unopposed estrogen stimulation as the etiologic factor associated with the development of the carcinoma(2). On the contrary, type II nonendometrioid carcinomas follow an estrogen-unrelated pathway and arise from a background of atrophic endometrium. There are a series of constant pathology images from simple, complex, and atypical hyperplasia to type I endometrial carcinoma. The transition from endometrial hyperplasia to invasive adenocarcinoma is a multistep process, and many molecular events such as phosphatase and tensin homologue deleted on chromosome (PTEN) gene silencing, microsatellite instability associated with defects in DNA mismatch repair genes, or mutations in the K-ras gene that are reported may be involved in the process(3–5), but the detail mechanism is not clear.

A novel tumor-associated antigen, receptor-binding cancer antigen expressed on SiSo cells (RCAS1), acts as a ligand for putative receptor present on the cells of the immune system, such as natural killer cells and activated T cells(6). RCAS1 binding to its ligand can induce apoptosis of these receptor-expressing cells, so that RCAS1 probably plays an important role in the evasion of host immune surveillance by tumor cells. In many kinds of tumors, such as those of the lung, esophagus, gallbladder, and pancreas, RCAS1 expression was significantly related to overall survival of patients(7–10). It was also found that RCAS1 is identical to the estrogen receptor–binding fragment associated gene 9 (EBAG9)(11), which has been identified as an estrogen-responsive gene from complementary DNA (cDNA) library of MCF-7, human breast cancer cell line. Endocrine–immune interactions are recognized to play an important role in the development and progression of various hormone-dependent tumors, but the details of these interactions remain unknown. Estrogen receptor (ER) has been implicated in the etiology and progression of type I endometrial cancer and their status correlates well with response to hormonal manipulation and prognosis(12). The possible correlation between RCAS1 and ERs expression is still unclear. Therefore, investigations on the association between RCAS1 and ERs should be of interest. To clarify whether RCAS1 is involved in the process from precancerous state to invasive cancer in endometrium, we investigated expression of RCAS1 in the normal, hyperplastic, and malignant endometrium and evaluated the correlations of RCAS1 with ERs status and various clinicopathologic parameters.

Materials and methods

Tissue specimen

A total of 112 endometrium tissue specimens were obtained from patients who had undergone curettage or hysterectomy at the First Affiliated Hospital, School of Medicine of Zhejiang University between February 1999 and November 2005. These specimens included 20 normal endometrium (10 proliferative phase and 10 secretory phase), 42 endometrial hyperplasia (10 simple, 12 complex, and 20 atypical hyperplasia), and 50 endometrial adenocarcinoma (46 endometrioid adenocarcinoma, 2 serous adenocarcinoma, and 2 clear-cell adenocarcinoma). The median age of patients with normal endometrium, endometrial hyperplasia, and carcinoma was 51, 49, and 52 years, respectively. Regarding the clinical staging of 50 cases of endometrial carcinoma, 25, 16, and 9 cases were classified as stages I, II, and III, respectively, according to the FIGO (1998). All the patients did not accept any hormonal therapy.

Immunohistochemistry

Surgical specimens were fixed in 10% formalin solution and embedded by routine methods in paraffin for sectioning at a thickness of 4 μm. Immunohistochemical analysis was performed using the streptavidin–biotin amplification method with a Histofine Kit (Nichirei, Tokyo, Japan). Sections were deparaffinized and incubated for 30 min with 3% H2O2 in methanol to block endogenous peroxidase activity. After being rinsed in Tris-buffered saline, sections were irradiated in a microwave oven. After the sections were cooled and rinsed in Tris-buffered saline, they were incubated for 6 h at room temperature with monoclonal antibodies directed against RCAS1 (1:200 dilution) (Medical Biological Laboratories, Tokyo, Japan), ER-α (clone Mc-20 [1:100 dilution]; Santa Cruz Biotechnology, Santa Cruz, CA), ER-β (clone H-150 [1:100 dilution]; Santa Cruz Biotechnology). The antibody complex was visualized with 3,3′-diaminobenzidine tetrahydrochloride (DBA) solution. The ER-positive breast cancer tissue was used as positive controls. In a negative control, phosphate-buffered saline was substituted for primary antibody.

Two observers blindly and independently assessed the immunohistochemical expression of RCAS1, ER-α, and ER-β. To evaluate RCAS1, we defined a score that corresponded to the sum of: a) the percentage of positive cells (0 = 0–4% immunopositive cells; 1 = 5–24% positive cells; 3 = 25–49% positive cells; and 4 =≥50% positive cells) and b) the staining intensity (0 = negative; 1= weak; 2 = moderate; 3 = strong)(13). We classified tumors with scores of 0–3 as low level expression and tumors with scores of 4–7 as high-level expression. Immunoreactivity of ER-α and ER-β was classified into two categories: less than 10%, positive cells; and greater than 10%, reactive cells(14).

Reverse transcriptase–polymerase chain reaction

Frozen sections of 35 μm thickness were cut in a cryostat under strict RNAse-free conditions. One set of slides were stained with hematoxylin–eosin and read by the pathologist to define normal, hyperplastic, and tumor cells. Individual frozen sections were mounted on plain glass slides, and 30-gauge needle attached to l mL syringe was used to microdissect the defined cells in the sections. The procured cells were used for following reverse transcriptase–polymerase chain reaction and Western blot analysis. Total tissue RNA was isolated with Trizol Reagent (Gibco BRL, Carlsbad, CA) in accordance with the manufacturer’s instructions. The cDNA was prepared by reverse transcription of 2 μg of total RNA using oligo dT18 and 200 U of superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) at 42°C for 70 min according to the manufacturer’s suggestion. Polymerase chain reaction (PCR) was carried out in 50 μL final volume with 2 μL of denatured cDNA, 2.5 U of Taq DNA polymerase (Invitrogen), 1 μM of both primers, and Taq polymerase buffer containing 1.5 mM Mgcl2 with 200 μM of each deoxyribonucleotide triphosphate (dNTPs). The primers used were as follows: for RCAS1 (product size: 347 bp), sense: 5′-CTAGCAACAGTATTCTCATTCC-3′, antisense: 5′-CTACTAGAGAAACCTGTGCTCC-3′; for ER-α (product size: 401 bp), sense: 5′-ATGACCATGACCCTCCACAC-3′, antisense: 5′-GAACCGAGATGATGTAGCCAG-3′; for ER-β (product size: 228 bp), sense: 5′-GCATGGAACATCTGCTCAA-3′, antisense: 5′-ACGCTTCAGCTTGTGACCTC-3′; for β-actin, (product size: 619 bp), sense: 5′-CGCTGCGCTGGTCGTCGACA-3′, antisense: 5′-GTCACGCACGATTTCCCGCT-3′. PCR primers were from Sangong (Shanghai, China). After an initial 3-min denaturation step at 95°C, 28 cycles of PCR were carried out under the following condition: 30 sec denaturation at 94°C, 30 sec annealing at 58°C for RCAS1, 62°C for ER-α, 60°C for ER-β, 30 sec at 72 h, and followed by a 6-min extension at 72°C. The amplified PCR products were resolved by electrophoresis on 1.5% (wt/vol) agarose gels and identified with ethidium bromide staining. The gene expression of RCAS1, ER-α, ER-β, and β-actin was quantified by densitometric scanning, using the digital image analyzer EDAS290 (Kodak, Vancouver, Canada). The signal intensities of the specific messenger RNA (mRNA) were normalized by comparison with that of β-actin and calculated as relative amounts. Negative controls without RNA and reverse transcriptase were performed.

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis and Western blot analysis

Cell lysates were prepared for Western blot analysis of RCAS1 using whole tissue protein extraction kits (Active Motif, Carlsbad, CA). The concentration of protein in each cell lysate was determined using bicinchoninic acid (BCA)-protein assay kit (Pierce, Rockford, IL) with bovine serum albumin as the standard. An identical amount of protein (40 μg) from each sample was loaded on a 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (0.45 μm) (Schieicher and Schuell, Dassel, Germany). Nitrocellulose membranes were blocked with 5% bovine serum albumin (Sigma, St. Louis, MO) in Tris-buffered saline (TBS) (25 mM Tris–HCl, 150 mM sodium chloride, and pH 7.2) for 1 h at room temperature. Blots were incubated with anti-RCAS1 or anti–β-actin specific rabbit polyclonal immunoglobulin G primary antibody (Santa Cruz) at 1:500 dilution at 37°C for 2 h. Blots were washed three times and then incubated in horseradish peroxidase–conjugated goat anti-rabbit antibody (1:2000 dilution) for 2 h at room temperature. All blots were developed using the enhanced chemiluminescence reagents (SuperSignal Dura Kit; Pierce) following the manufacturer’s instructions.

Statistical analysis

Statistical difference between RCAS1 status and clinicopathologic parameters, ERs status was evaluated in a cross-table using the χ2-test or the Fisher exact test. P < 0.05 was considered statistically significant.

Results

RCAS1 protein expression in endometrium

Positive immunostaining for RCAS1 was detected both on the membranes of the cancer cells and in their cytoplasm. There were two patterns of RCAS1 immunostaining seen in the specimens of endometrium. Granular staining enriched in the glandular side of cytoplasm with polarity was defined as P pattern, and granular staining scattered diffusely in the cytoplasm and on the cell membranes was defined as D pattern. P pattern was predominantly observed in normal (Fig. 1A, B), simple (Fig. 1C) and complex hyperplastic endometrium, while, D pattern was seen in the malignant endometrium (Fig. 1E). Two patterns could be observed in atypical endometrium, and 25% (5/20) cases showed D pattern (Fig. 1D; Table 1). All control slides yielded negative staining.

Figure 1.

Immunohistochemical staining for RCAS1 expression in various endometria. A) In normal proliferative B) secretory endometrium C) simple hyperplasia, RCAS1 is lowly expressed and shows P pattern (granular staining enriched in glandular side of cytoplasm with polarity). D) In atypical hyperplastic endometrium, RCAS1 is strongly expressed and shows D pattern (granular staining scattered diffusely in the cytoplasm and on the cell membrane), while the surrounding complex hyperplasia appeared P pattern. E) RCAS1 is strongly expressed in carcinoma tissue and showed D pattern (original magnification ×200). F) RCAS1 is highly expressed in invasive lesion of carcinoma tissue (original magnification ×100).

Table 1.  RCAS1 protein expression in endometrium
Pathologic diagnosisNumber of samplesNumber of high expressionPattern
PDA
  1. P, P pattern, granular staining enriched in the glandular side of the cytoplasm with polarity; D, D pattern, granular staining diffused throughout the cytoplasm and on the cell membranes; and A, absence of staining.

Normal
 Proliferative100802
 Secretory100802
Hyperplasia
 Simple100901
 Complex1221200
 Atypical2081253
Carcinoma50340473

RCAS1 protein was less expressed in all normal, proliferative, and secretory endometrium. The rates of high-level expression of RCAS1 in simple and complex hyperplasia, atypical hyperplasia, and endometrial carcinoma were 9% (2/22), 40% (8/20), and 68% (34/50), respectively, and there was a significant difference between each group (P < 0.05).

The relation between high-level expression of RCAS1 and clinicopathologic variables

The relation between RCAS1 expression and several clinicopathologic variables was investigated (Table 2). The high-level expression of RCAS1 was found to be significantly related with the myometrial invasion and vascular invasion of endometrial adenocarcinoma (P < 0.05). The rate of RCAS1 high-level expression was significantly higher in cases of ER-α positive than that in ER-α negative (P < 0.05). There was no significant relationship between the high-level expression of RCAS1 and the patient’s age, menopausal status, clinicopathologic stage, histologic grade, lymph node metastasis, and ER-β status.

Table 2.  Association between high expression of RCAS1 and clinicopathologic parameters
ParametersCaseHigh expression rate of RCAS1 (%)P value
Menopausal status0.697
 Premenopausal2315 (65.2) 
 Postmenopausal2719 (70.4)
Clinical stage0.747
 I2516 (64.0) 
 II1611 (68.8)
 III97 (77.8)
Nodal status0.495
 Positive86 (75.0) 
 Negative4228 (66.7)
Grade0.921
 12517 (68.0) 
 21712 (70.6)
 385 (62.5)
Myometrial invasion0.017
 <1/22815 (53.6) 
 ≥1/22219 (86.4)
Vascular invasion0.026
 Positive1816 (88.9) 
 Negative3218 (56.3)
Endometrioid adenocarcinoma4634 (73.9)
Serous adenocarcinoma20 (0)
Clear-cell adenocarcinoma20 (0)
ER-α0.042
 Positive3628 (77.8) 
 Negative146 (42.9)
ER-β0.291
 Positive2919 (65.5) 
 Negative2115 (71.4)

Western blot analysis

Using the monoclonal anti-RCAS1 antibody, a 32-kd band, which corresponded to the molecular weight of RCAS1, was detected in normal, hyperplastic, and carcinomatous endometrium. In the carcinomatous endometrium, the bands were detected at a higher intensity compared with normal and hyperplastic endometrium (Fig. 2).

Figure 2.

Western blot analysis of RCAS1. A 32-kd band was detected. In the endometrial carcinoma, the bands were detected at a higher intensity compared with normal and hyperplastic endometrium. Lines 1 and 2, endometrial carcinoma; line 3, atypical hyperplasia; line 4, simple hyperplasia, and line 5, normal hyperplasia.

RCAS1, ER-α, and ER-β mRNA expression in normal, hyperplastic, and malignant endometrium

RCAS1, ER-α, ER-β, and β-actin mRNA expressions were detected as a specific-single band (347 bp, 401 bp, 228 bp, and 619 bp, respectively) (Fig. 3). Expressions of RCAS1 and ER-α mRNA in endometrial carcinoma were significantly higher than that of the normal and hyperplastic endometrium (P < 0.05), but expression of ER-β mRNA had no significant difference between each group (P > 0.05) (Fig. 4).

Figure 3.

RCAS1, ER-α, and ER-β expression in relation to β-actin mRNA expression in various endometrium. M, marker; lines 1 and 2, normal endometrium; lines 3 and 4, hyperplastic endometrium; and lines 5 and 6, endometrial carcinoma.

Figure 4.

RCAS1, ER-α, and ER-β expression in relation to β-actin mRNA expression in various endometria. Densitometric analysis was performed on ethidium bromide-stained agarose gels and was compared with that one of the housekeeping gene β-actin, and their ratios were reported. Bars show the mean ± SD.

Discussion

In this study, immunoreactivity showed that the high-level expression of RCAS1 was significantly more frequent in the adenocarcinoma specimens than in the normal and hyperplastic endometrium specimens, which was consistent with the previous report(15). Moreover, from normal endometrium to simple, complex, and atypical hyperplasia and carcinoma, we found that not only the level of the RCAS1 protein gradually increased but also its expression pattern changed. P pattern existed in all normal, simple, and complex hyperplastic endometrium, while D pattern appeared in malignant endometrium. In atypical endometrium, two patterns could be seen. These two staining patterns of RCAS1 were first described in gastric cancer of Nakamura et al.’s study(16). In his study, the staining patterns correlated with size of tumors, depth of tumor invasion, histologic type, and lymph node metastasis. He also inferred that the differences in the RCAS1 distribution might be related to differences of function, including a death ligand for immune cells, malignant transformation, or progression. Whereas in our study, only D pattern was seen in endometrial carcinoma—is it the expression difference between the gastric tumor and the endometrial tumor? The reason needs further investigation. Although the physiologic roles of RCAS1 in normal tissue are unknown, it is consistent that the D pattern seemed to be associated with carcinogenesis bad biological behavior. Nearly one-quarter of atypical hyperplasia progress to well-differentiated endometrial carcinoma(17). Our data demonstrated that in atypical hyperplasia, the expression level of RCAS1 was significantly higher than that of simple and complex hyperplasia; moreover, D pattern staining appeared. In addition, expression of RCAS1 mRNA in endometrial carcinoma was significantly higher than that of normal and hyperplastic endometrium, which correlated well with the immunohistochemical results. Thus, supporting the atypical hyperplasia is the key point of the process of endometrial malignant transformation. Expression of RCAS1 experienced an alteration from quantity to quality. These data suggest that RCAS1 expression might be involved in malignant transformation and may play a role in the early events of endometrial carcinoma, which agree with the conclusion of Sonoda et al.(15). The high-level expression of RCAS1 could be regarded as a high-risk factor of hyperplasia, especially of atypical hyperplasia, and it may be useful for diagnosing problematic cases and help to screen the high-risk people for further therapy.

Several reports showed that expression of RCAS1 was associated with poor prognosis and advanced disease. RCAS1 was strongly expressed in uterine cervical cancer with the invasive tendency(18). Studies on patients with gallbladder cancer have showed that RCAS1 expression is associated with the depth of tumor invasion and the presence of lymph node metastasis(9). Our present findings were in good agreement with these previous reports and showed that high expression of RCAS1 protein is significantly related with the myometrial invasion and vascular invasion. That is to say, the endometrial carcinoma that highly expresses RCAS1 may be invasive and progressive, and such characteristics may lead to poor prognosis. However, owing to the limited accumulation time of this study and the satisfactory prognosis of endometrial carcinoma (all of the patients studied here are still alive), we cannot determine the overall survival or make any conclusion regarding prognosis. Further prospective studies should be undertaken to analyze the relationship of the overall survival of patients with the expression level of RCAS1.

Estrogens are well-known to contribute immensely to the development of type I endometrial carcinoma. Biological effects of estrogens are mediated through an interaction with ERs (ER-α and ER-β). An imbalance in ER-α and ER-β expression is believed to be a possible critical step in estrogen-dependent tumorigenesis. ER-α mRNA and protein expression are reported to decrease stepwise from normal or grade 1 to grade 3 tumor lesions. But ER-β expression does not alter, suggesting a shift to a decreased ER-α/ER-β ratio(19). Contrasted with many of the other previous reports, our results demonstrated that ER-α mRNA increased stepwise from normal to tumor lesions, but expression of ER-β mRNA had no significant difference between each group. ERs activate transcription of various target genes (ie, estrogen-responsive genes) in a ligand-dependent manner by direct DNA interaction through the estrogen-responsive element(s) (ERE) or by tethering to other transcription factors. EBAG9/RCAS1 is assumed to be an estrogen-responsive gene. Transfection analyses have shown that the nucleotide sequences between −86 and −36 contain an ERE in the 5′ promoter region of the EBAG9 gene. EBAG9 was upregulated after estrogen treatment in MCF-7 cells, an effect that is mediated by the binding of ER-α to the ERE in the promoter region of the EBAG9 gene(20). In our study, the immunohistochemistry results showed that the rate of RCAS1 high-level expression was significantly higher in cases of ER-α positive than that of ER-α negative (P < 0.05), which demonstrated that the expression of RCAS1 was associated with ER-α. This result was identical to that reported in a previous study on adenocarcinoma of the ovary(21). However, because of the lack of mechanism examinations and relatively limited number of cases in this study, additional study is required to clarify the detail mechanism of RCAS1 action and modulation with ERs, ERE, and other ligands.

In summary, the present study demonstrated that RCAS1 expression played an important role in the process of endometrial malignant transformation and may be an early event of endometrial carcinoma. The high-level expression of RCAS1 was associated with myometrial invasion and vascular invasion in endometrial carcinoma. RCAS1 coexpression with ER-α might be related with tumor development and metastasis. To clarify the role of RCAS1 in endometrial carcinoma, further investigations are warranted.

Acknowledgments

We would like to thank Mr Yao Hang-ping and Mr Ding Wei for their excellent technical assistance. This work was supported by Natural Science Foundation of Zhejiang, China. (M303847).

Ancillary