Molecular and Cellular Pathobiology SHON Is aNovel Estrogen-RegulatedOncogene inMammary Carcinoma That Predicts Patient Response to Endocrine Therapy

Endocrine therapies are the primary systemic intervention for patients with estrogen receptor–positive (ERþ) breast cancer. However, a significant proportion of initially responsive ERþ tumors develop resistance, with relapses occurring in up to 50% of patients. Lack of reliable predictive biomarkers remains an unfilled need for enhanced clinical management of this disease. In this study, we address this need in identifying a novel estrogen-regulated gene called SHON (secreted hominoid-specific oncogene). Enforced expression of SHON in breast cancer cells increased their proliferation, survival, migration, and invasion in vitro. Furthermore, SHON enhanced the oncogenicity of these cells in xenograft models of human breast cancer and was also sufficient to oncogenically transform MCF10A human mammary epithelial cells. Conversely, SHON attenuation mediated by RNA interferenceor antibody-based methods reduced the oncogenicity of breast cancer cells. Mechanistic investigations indicated that the oncogenic transforming properties of SHON were mediated by BCL-2 and NF-kB. In primary clinical specimens, SHON was immunohistochemically detected in 62% of breast cancers, in which its expression was positively correlated with ER expression. In this setting, SHON expression predicted a favorable response to endocrine therapy in high-risk patients with ERþ breast cancer. Taken together, our findings identify SHON as a novel human oncogene with predictive utility in ERþ breast cancer, perhaps offering a simple biomarker to predict the therapeutic efficacy of antiestrogen therapy in patients with breast cancer. Cancer Res; 73(23); 6951–62. 2013 AACR. Introduction Breast cancer is the most prevalent cancer in females worldwide. Estrogen signaling and estrogen receptors (ER) play a pivotal role in the initiation and progression of breast cancer, with themajority of human breast cancer being initially estrogen dependent. In accepted models of ER function, estrogen binding to either ERa or ERb results in dimerization and conformational change (1). The resulting estrogen–ER complex modulates gene transcription by binding to estrogen response elements in DNA, or by interaction with other transcription factors such as specificity protein 1 (SP1), activating protein 1 (AP-1), or NF-kB (2). ER signaling can also be modulated by other coregulatory proteins or through genomic and extranuclear actions (3, 4). Expression arrays have identified early and late estrogen responsive genes (5), including genes involved in cancer cell proliferation and survival, enhancing our understanding of themolecular events involved in estrogen action. Antiestrogens remain the most effective form of systemic intervention for patients with ER-positive (ERþ) tumors, which account for approximately 75% of breast cancers. Current approaches include the use of selective ER modulators (SERM; e.g., tamoxifen) that compete with estrogen for ER binding or aromatase inhibitors (e.g., letrozole, anastrozole, and exemestane), which inhibit the synthesis of estrogen. ERa, the currently used predictive biomarker for antiestrogen responsiveness, is not consistently accurate in predicting response due to de novo or acquired tumor resistance. Although a significant high proportion of ERþ breast tumors initially respond to antiestrogens, resistance Authors' Affiliations: Liggins Institute, University of Auckland, Auckland, New Zealand; Department of Clinical Oncology; Department of Histopathology, School of Molecular Medical Sciences, University of Nottingham and Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom; The Key Laboratory of Molecular Epigenetics of Ministry of Education, Northeast Normal University; The Institute of Genetics and Cytology, Northeast Normal University, Changchun; Institute of Genetics andDevelopmental Biology, Chinese Academy of Sciences, Beijing; Hefei National Laboratory for Physical Sciences at Microscale and School of Life Sciences, University of Science and Technology of China; Department of Pathology, Anhui Medical University, Hefei, China; Cancer Science Institute of Singapore, andDepartment of Pharmacology, National University of Singapore; and National Cancer Institute of Singapore, National University Health System, Singapore Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). Corresponding Authors: Dong-Xu Liu, Liggins Institute, University of Auckland, 85ParkRoad,Grafton,PrivateBag92019,Auckland1023,NewZealand. Phone: 64-9923-9603; Fax: 64-9373-7497; E-mail: dx.liu@auckland.ac.nz; andPeter E. Lobie, Cancer Science Institute of Singapore, National University of Singapore, Centre for Life Sciences, #03-06C, 28Medical Drive, Singapore 117456, Singapore. Phone: 65-6601-1046; E-mail: csipel@nus.edu.sg doi: 10.1158/0008-5472.CAN-13-0982 2013 American Association for Cancer Research. Cancer Research www.aacrjournals.org 6951 on July 19, 2017. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from eventually develops, and up to 40% to 50% of patients with ERþ tumors relapse (6–9), through molecular mechanisms that are yet to be fully understood. These tumors typically continue to express ERa (10) and demonstrate earlier metastatic recurrence (11). The expression of ERa–regulated progesterone receptor (PR) and BCL-2 genes is associated with improved survival of patients with breast cancer (12, 13) and serves as a favorable prognostic marker for endocrine therapy (12, 14). However, there is still no definitive methodology to distinguish ERþ tumors that will or will not respond to endocrine therapy. There appears to be as few as 168 human-specific genes among the approximately 23,000 or so genes identified from the completion of the human genome project (15). When expanded to include primate lineages, there may be only several hundred unique hominoid genes. The identification of such human/hominoid-specific genes and analysis of their molecular functions might increase the understanding of the contributions these genes likely make to human disease processes. Studies have demonstrated that some human/hominoid-specific genes and their regulated signaling pathways may serve as biomarkers in human disease (16) and a number of human/hominoid-specific genes, such as Tre2 (17) and TBC1D3 (18–20), have been linked to human cancer. However, only a few proteins translated from human/hominoid-specific genes have been studied and the functions of many others remain as yet uncharacterized. Here, we report the identification of a novel estrogen regulated oncogene, secreted hominoid-specific oncogene (SHON) in mammary carcinoma and determine the predictive significance of SHON expression for response to endocrine therapy in high-risk patients with ERþ breast cancer. Materials and Methods Cell lines All cell lines used in this study were obtained from the American Type Culture Collection and were cultured in conditions as recommended except the MCF-7 cell line that was cultured in RPMI 1640 supplemented with 10% heat-inactivated FBS, 100 U/mLpenicillin, 100mg/mL streptomycin, and 2 mmol/L L-glutamine in a 37 C humidified incubator with 5% CO2. Plasmid constructs PIKR2786 (Genentech UNQ ID: UNQ2786) represented by an expressed sequence tag (EST; GenBank AY358103) was originally identified, through bioinformatic analysis, as a potential secreted or transmembrane protein (21).We have demonstrated that UNQ2786 is a secreted protein and is a human mammary epithelial oncogene and have named it as SHON. The human SHON-expressing plasmids were constructed by using a protocol previously described (22). The Bcl-2 P1 promoter reporter plasmid (23) and the NF-kB reporter plasmid (24) were previously described. Tumor xenografts in nude mice The experimental protocols followed are described in the Supplementary Materials and Methods. Clinical samples, patient data analysis, and ethics statement This studywas approved by the NottinghamResearch Ethics Committee 2. Immunohistochemical (IHC) staining was carried out using a cohort of breast tissue specimens from Nottingham, United Kingdom (25, 26). Detailed patient demographics and clinicopathologic characteristics and methodology about these samples can be found in the Supplementary Materials and Methods. A full description of the materials andmethods used for this work is described in Supplementary Materials and Methods. Results Expression of SHON in human normal tissues and carcinoma cell lines Sequence homology searches using National Center for Biotechnology Information (NCBI) basic local alignment search tool (BLAST) revealed that SHON belongs to a hominoid-specific gene family with no known orthologs outside the primate lineage. SHON contains a predicted open reading frame of 282 bp encoding a 93-amino acid residue protein with a predicted molecular mass of 9.7 kDa (Fig. 1A). There is a possibility that an internal disulfide bond is formed between the two cysteines in the mature protein. SHON gene expression was detected in all 48 human tissues in a commercially available panel of cDNAs after 40 cycles of PCR amplification (Fig. 1B). SHONmRNAwas also expressed in all cancer cell lines tested (Fig. 1C, top). A rabbit polyclonal antibody against the C-terminal of the predicted mature SHON protein was produced and its specificity was validated (Supplementary Figs. S1 and S2). A specific band of the approximate expected size (12 kDa) was detected in Western blot analysis in all cancer cell lines tested (Fig. 1C, bottom). However, no SHON protein expression was detected in the normal human mammary epithelial cell line, MCF10A (Supplementary Fig. S2D). SHON is a secreted protein SHON is predicted to be a secreted protein with a leading signal peptide encompassing amino acid residues 1 to 21 by SignalP 3.0 (Fig. 1A; ref. 27). HIS-tagged SHON protein can be readily detected in both the whole cell lysate and