ER81 Expression in Breast Cancers and Hyperplasia

ER81 is a transcription factor that may contribute to breast cancer; however, little known about the role of ER81 in breast carcinogenesis. To investigate the role of ER81 in breast carcinogenesis, we examined ER81 expression in IDC, DCIS, ADH, HUT, and normal breast tissues by immunohistochemical staining. We found that ER81 overexpression was detected in 25.7% (9/35) of HUT, 41.2% (7/17) of ADH, 54.5% (12/22) of DCIS, and 63.0% (51/81) of IDC. In 20 of breast cancer tissues combined with DCIS, ADH, and HUT, ER81 expression was found in 14/20 (70%) IDC. In these 14 cases all cases were ER81 positive expression in DCIS, 13 of 14 cases were positively expressed of ER81 in ADH and 8 of 14 were positive for ER81 in HUT components. A statistical significance was found between NBT and HUT (P < .05) and HUT and ADH (P < .05). Clinical-pathological features analysis of breast cancer revealed that ER81 expression was significantly associated with Her2 amplification and was negatively associated with ER and PR expression. Our results demonstrated that ER81 overexpression was present in the early stage of breast development that suggested that ER81 overexpression may play an important role in breast carcinogenesis.


Introduction
Breast carcinogenesis is thought to undergo a transition from normal epithelium to invasive carcinoma (IDC) via hyperplasia of usual type (HUT), atypical ductal hyperplasia (ADH), and carcinoma in situ (DCIS) [1]. Over 14% of breast cancer diagnosed in the United States annually are DCIS, and approximately 50% of untreated DCIS will develop into an IDC within 24 years after the original biopsy [2]. However, it is unclear how IDC develop from these lesions.
ER81 (ETS-related 81), also called ETS variant 1 (ETV1), is a transcription factor that is a member of the ETS family of DNA-binding proteins [3][4][5]. Its association with cancer was first noted in Ewing tumors, in which the EWS gene can be translocated onto the ER81 gene and the resultant EWS-ER81 fusion protein exerts oncogenic properties [6]. From then on, many findings suggest that dysregulation of ER81 target genes in disparate tumors like Ewing sarcomas and prostate carcinomas are causally involved in tumorigenesis [7]. Of note, ER81 transcriptional activity is dramatically enhanced upon Her2/Neu (a receptor tyrosine kinase and proto-oncoprotein especially associated with breast cancer) overexpression [8,9]. Moreover, ER81 mRNA levels are increased in murine cell lines and tumors overexpressing Her2/Neu and also in many human breast cancer cell lines, which suggests that ER81 may contribute to breast tumorigenesis [10][11][12].
To gain more insight into ER81's role in breast tumorigenesis, we attempted to examine ER81 expression in invasive breast cancers, ductal carcinoma in situ, atypical ductal hyperplasia, and hyperplasia without atypia and normal breast tissues in this study.

Immunohistochemical Staining.
Immunohistochemistry was employed to detect the expression of ER81 for all breast tissues and the expression of ER, PR for all the 81 cases of invasive ductal carcinomas tissues. EnVision Systems was adopted for the staining. Briefly, 4 μm sections were taken from formalin-fixed, paraffin-embedded tissue blocks. The deparaffinized sections were pretreated with heat-induced epitope retrieval and then treated with 30 mL/L hydrogen peroxidase in methanol for 30 min to block endogenous peroxidase activity. The sections were further blocked with 10 mL/L normal goat serum for 30 min, followed by incubation with primary antibody (mouse antihuman monoclonal antibody ER81, SANTA CRUZ BIOTECHNOLOGY, INC; mouse antihuman monoclonal antibody ER and PR, Maixin-Bio, Fuzhou, China) at 4 • C overnight. The sections were then washed in 0.01 mol/L phosphate buffer solutions (PBS, pH 7.2) and sequentially incubated with Envision (Envision kit, DakoCytomation, Inc, Carpinteria, California, USA) for 30 min. The reaction product was visualized by diaminobenzidine tetrahydrochloride (DAB). All slides were counterstained with hematoxylin, dehydrated, and mounted. PBS substituting for the primary antibody was used as the negative control.

Assessment of Immunohistochemical Staining.
Specific staining was evaluated independently by two investigators. We used a semiquantitative manner to assess the ER81 staining, yielding an immunoreactive score (IRS) ranging from 0 to 9. IRS was calculated by multiplying the number of positive cytoplasmic staining of cells (0 = none, 1 =< 10%, 2 = 10-50%, 3 => 50% positive tumor cells) by the staining intensity (1 = weak, 2 = moderate, 3 = strong). Then we considered IRS 0 score as ER81 expression "−", IRS 1-2 score as ER81 expression "+", IRS 3-5 score as ER81 expression "++", IRS 6-9 score as ER81 expression "+++". Positive reaction in a normal epithelium yielded a maximum IRS of 2. Therefore breast hyperplastic cells and breast cancer cells in other groups were considered ER81 positive with IRS ≥ 3 as suggested by Going et al. [13]. The ER and PR positive staining should be localized to the nucleus. Specimens in which more than 10% of cells showed positive immunoreactivity were considered to be immunoreactive.

Fluorescence In Situ Hybridization (FISH).
A HER2/neu probe kit (China Medical Technologies, Inc, Beijing, China) was used for FISH analysis for all the 81 cases of invasive ductal carcinomas tissues. Tissue sections were baked overnight at 56 • C, dewaxed in xylene, dehydrated and air-dried. The

Results
Formalin-fixed tissue sections from a spectrum of mammary lesions were analyzed for ER81 expression. As shown in Table 1     development. In order to confirm the association between ER81 expression and breast tumorigenesis, we employed 20 IDC tissues combined with DCIS, ADH, HUT, and NBT to examine ER81 expression. We considered a threshold IRS set at ≥3 as ER81 positive expression. In this group, ER81 expression was found in 14/20 (70%) IDC. In these 14 cases, ER81 expression was found in 14/14 of adjacent DCIS,  Adjacent normal breast component demonstrated ER81 negative expression. By K-W analysis for the expression level of ER81, we found a statistical significance between NBT and HUT (P < .05) and HUT and ADH (P < .05), but no statistical significance was found between ADH and DCIS (P > .05) or DCIS and IDC (P > .05) ( Table 2 and Figures  4, 5). The relationship between ER81 expression and clinicalpathological features such as ER, PR, and Her2 in breast cancer is listed in Table 3. The results revealed that ER81 expression was significantly associated with Her2 amplification and was negatively associated with ER and PR expression. No correlation was found between ER81 expression and patient ages, menopause status, tumor sizes, nodal status, and histological stage (Table 3).

Discussions
Increasing lines of evidence suggest that breast cancer develop through a multistep model of carcinogenesis, that is, from normal breast epithelia to hyperplasia without atypia, hyperplasia with atypia, ductal carcinoma in situ, to invasive carcinoma [1,14,15]. In experiments carried out by DeOme et al. [16], when hyperplastic breast alveolar nodules (HAN), the breast epithelial cells infected by murine mammary tumor virus, were transplanted to cleared mammary fat pads, half of them developed into carcinomas by 13-21 weeks, which happened more frequently than normal breast tissues. In breast biopsies harbouring malignancy, infiltrating carcinoma is often found side-by-side with in situ carcinoma and/or benign proliferations. These lesions occasionally show morphological transition and continuity with the invasive carcinoma. Karpas et al. [17] evaluated 645 breast biopsies (226 malignant and 419 benign) and found atypical hyperplasia in 62% of malignant biopsies, but only in 4% of benign biopsies. Similarly, Kern and Brooks [18] found a greater incidence of atypical ductal hyperplasia (ADH) in cancer-bearing breasts. Foote and Stewart [19] found that papillary hyperplasia with atypia occurred five times more frequently in the cancerous breasts. In a similar study, Ryan and Coady [20] found that hyperplasia was four times more common in the cancerous breasts. These histopathological studies data provide convincing evidence that some forms of proliferative lesions are often found in association with invasive cancer and that ADH provides a significantly increased relative risk of subsequent invasive carcinoma. However, little is known about the molecular genetic mechanisms involved in the transformation from hyperplasia to cancer, which is important for the early diagnosis and molecularly targeted therapy of breast cancers [21]. ER81 is a downstream gene of Her2/Neu, a receptor tyrosine kinase and proto-oncoprotein. And Her2 is especially associated with breast cancer. Of note, ER81 transcriptional activity is dramatically enhanced upon Her2/Neu overexpression [8,9]. On the other hand, ER81 can target Her2 and upregulate Her2 expression in breast tumors, suggesting the existence of a feed-forward loop in the upregulation of HER2/Neu [22]. Moreover, ER81 mRNA levels are increased in murine cell lines and tumors overexpressing Her2/Neu and also in many human breast cancer cell lines, suggesting that ER81 may contribute to breast tumorigenesis [10][11][12]. Shin et al. found that ER81 downregulation suppresses proliferation of Her2-positive MDA-MB-231 breast cancer cells in vitro and tumor formation in vivo, proving for the first time the existence of a critical role of ER81 in breast cancer cell physiology [5]. Although transgenic mice overexpressing ER81 in the breast do not develop mammary tumors, ER81 overexpression may prime breast cells to become malignant, for instance upon additional overexpression of Her2/Neu [5].
In this study, we investigated the role of ER81 in breast carcinogenesis by two steps: (1) examining ER81 overexpression in IDC, DCIS, ADH, HUT, and normal breast tissues which represents different stage of breast cancer development. As a result, weak staining was observed in normal breast tissues yielding a maximum IRS of 2. If cells in other groups were considered ER81 positive with IRS ≥ 3, ER81 overexpression was detected in 25.7% (9/35) of HUT, 41.2% (7/17) of ADH, 54.5% (12/22) of pure DCIS, and 63.0% (51/81) of IDC. Although there was ER81 expression in HUT, all ER81 positive tissues were moderate staining. The expression level of ER81 was increased with the progression of the lesion. It is implied that ER81 overexpression are present in the early stage of breast development. (2) Examining ER81 overexpression in breast cancer and the adjacent hyperplasic components (each component represents one stage of breast cancer development) in a single tumor. In this group, ER81 expression was found in 70% (14/20) IDC. In these 14 cases all cases were ER81 positive expression in DCIS, 13 of 14, cases were positively expressed of ER81 in ADH, and 8 of 14 were positive for ER81 in HUT components. A statistical significance was found between NBT and HUT (P < .05) and HUT and ADH (P < .05), but no statistical significance was found between ADH and DCIS (P > .05) or DCIS and IDC (P > .05) confirming that ER81 may involve in breast carcinogenesis.
In addition, we analyzed the relationship between ER81 expression and clinical-pathological features of breast cancer including Her2 amplification and ER, PR expression. The results revealed that ER81 expression was significantly associated with Her2 amplification and was negatively associated with ER and PR expression. No correlation was found between ER81 expression and patient ages, menopause status, tumor sizes, nodal status, and histological stage. In Her2 positive amplification group, the number of ER81 positive expressed cases was more than that in Her2 negative amplification group. As we know, overexpression of ER81 in itself does not lead to breast tumor formation [23], possibly because ER81 requires stimulation in order to become transcriptionally competent and the activation of ER81 is inducible by the Her2 → Ras → Raf → MAP kinase signaling pathway [9,24,25]. These results suggest that Her2 and ER81 synergize breast carcinogenesis.
Conclusively, ER81 overexpression was present in the early stage of breast development. Together with previous study results, it is suggested that ER81 may play an important role in breast carcinogenesis.

Abbreviations
IDC: invasive ductal carcinoma DCIS: ductal carcinoma in situ ADH: atypical ductal hyperplasia HUT: hyperplasia of usual type NBT: normal breast tissue IRS: immunoreactive score ER: estrogen receptor PR: progesterone receptor.