Steroid Receptor RNA Activator Protein (SRAP): a potential new prognostic marker for estrogen receptor-positive/node-negative/younger breast cancer patients.

INTRODUCTION: The steroid receptor RNA activator is a functional RNA suspected to participate in the mechanisms underlying breast tumor progression. This RNA is also able to encode for a protein, Steroid Receptor RNA Activator Protein (SRAP), whose exact function remains to be determined. Our aim was to assess, in a large breast cancer cohort, whether levels of this protein could be associated with outcome or established clinical parameters. METHODS: Following antibody validation, SRAP expression was assessed by tissue-microarray (TMA) analysis of 372 breast tumors. Clinical follow-up and parameters such as steroid receptor and node status were available for all the corresponding cases. Immunohistochemical scores were independently determined by three investigators and averaged. Statistical analyses were performed using standard univariate and multivariate tests. RESULTS: SRAP levels were significantly (Mann-Whitney rank sum test, P < 0.05) higher in estrogen receptor-alpha positive (ER+, n = 271), in progesterone receptor positive (PR+, n = 257) and in older patients (age > 64 years, n = 182). When considering ER+ tumors, PR+ tumors, or younger patients (< or = 64 years), cases with high SRAP expression had a significantly (Mantel-Cox test, P < 0.05) worse breast cancer specific survival (BCSS) than those with low SRAP levels. SRAP also appeared as a very powerful indicator of poor prognostic for BCSS in the subset of ER+, node negative and young breast cancer patients (Cox regression analysis, n = 60, BCSS Hazard Ratio = 8.61, P < 0.006). CONCLUSIONS: Our data suggest that SRAP levels might provide additional information on potential risk of recurrence and negative outcome in a specific set of patients with otherwise good prognosis when considering only estrogen receptor and nodal status.

Immunohistochemical validation of multiple phospho-specific epitopes for estrogen receptor alpha (ERalpha) in tissue microarrays of ERalpha positive human breast carcinomas.

Estrogen receptor alpha (ERalpha) activity is regulated by phosphorylation at several sites. Recently several antibodies specific for individual phosphorylated sites within ERalpha have became available. Such antibodies potentially provide invaluable tools to gain insight into the relevance in vivo of phosphorylated ERalpha in human breast tumors. However, validation of these antibodies for immunohistochemistry in particular is necessary in the first instance. In this study we have investigated the usefulness of several antibodies generated to specific phosphorylated sites within ERalpha for immunohistochemistry of formalin-fixed, paraffin-embedded human breast cancer biopsy samples. As well, these data demonstrate for the first time, the detection of multiple phosphorylated ERalpha forms in breast cancer (P-S104/106-ERalpha, P-S118-ERalpha, P-S167-ERalpha, P-S282-ERalpha, P-S294-ERalpha, P-T311-ERalpha, and P-S559-ERalpha) suggesting the possibility that profiling of phosphorylated ERalpha isoforms might be useful in selecting subgroups of breast cancer patients that would benefit from endocrine therapy.

Estrogen receptor alpha negative breast cancer patients: estrogen receptor beta as a therapeutic target.

Clinical management of breast cancer is increasingly guided by assessment of tumor phenotypic parameters. One of these is estrogen receptor (ER) status, currently defined by ERalpha expression. However with the discovery of a second ER, ERbeta and its variant isoforms, the definition of ER status is potentially more complex. In breast tumors there are two ERbeta expression cohorts. One where ERbeta is co-expressed with ERalpha and the other expressing ERbeta alone. In the latter subgroup of currently defined ER negative patients ERbeta has the potential to be a therapeutic target. Characterization of the nature and role of ERbeta in ERalpha negative tumors is essentially unexplored but available data suggest that the role of ERbeta may be different when co-expressed with ERalpha and when expressed alone. This review summarizes available data and explores the possibility that ERbeta signaling may be a therapeutic target in these tumors. Evidence so far supports the idea that the role of ERbeta in breast cancer is different in ERalpha negative compared to ERalpha positive tumors. However, cohort size and numbers of independent studies are small to date, and more studies are needed with better standardization of antibodies and protocols. Also, the ability to determine the role of ERbeta in ERalpha negative breast cancer and therefore assess ERbeta signaling pathways as therapeutic targets would be greatly facilitated by identification of specific downstream markers of ERbeta activity in breast cancer.

Estrogen receptor-alpha phosphorylated at Ser118 is present at the promoters of estrogen-regulated genes and is not altered due to HER-2 overexpression.

Detection of estrogen receptor (ER)-alpha phosphorylated at Ser(118) (P-Ser(118)-ER-alpha) may be an indicator of an intact ligand-dependent ER-alpha in breast tumors in vivo and may predict responsiveness to endocrine therapy. The current study addresses whether P-Ser(118)-ER-alpha is functionally involved in ER target gene transcription and if this is modulated by HER-2 overexpression. Using chromatin immunoprecipitation analysis, P-Ser(118)-ER-alpha was found associated with the promoters of several estrogen-regulated genes in MCF-7 breast cancer cells 30 minutes following estrogen treatment. Coactivators AIB1 and p300 were coimmunoprecipitated with P-Ser(118)-ER-alpha following estrogen treatment. The overexpression of HER-2 protein in MCF-7 cells did not affect estrogen induction of phosphorylation of Ser(118) or its presence at the promoters of several estrogen-regulated genes. U0126, an inhibitor of mitogen-activated protein kinase (MAPK) pathway, had no effect on P-Ser(118)-ER-alpha. The lack of effect of HER-2 overexpression on P-Ser(118)-ER-alpha expression in cell models is supported by similar levels of expression of P-Ser(118)-ER-alpha in ER(+)/HER-2-overexpressing and ER(+)/HER-2(-) breast tumors in vivo. Using inhibitors of cyclin-dependent kinase 7 (Cdk7), [(5,6-dichloro-1-beta-d-ribofuranosylbenzimidazole and 2-(R)-1-ethyl-2-hydroxyethylamino)-6-benzylamino-9-isopropylpurine], and IkappaB kinase-alpha (IKK-alpha; BAY-11-7082), we show that IKK-alpha, but not Cdk7, is at least in part involved in estrogen-mediated phosphorylation at Ser(118) in MCF-7 cells. Our data provide direct evidence for a functional role of P-Ser(118)-ER-alpha in estrogen-regulated signaling and do not support the hypothesis that resistance of breast tumors to tamoxifen therapy involves ligand independent activation of ER-alpha due to constitutive phosphorylation of Ser(118) by constitutive activation of MAPK pathway.

Steroid receptor RNA activator protein (SRAP) expression as a prognostic factor in ER+ human breast tumors.

BACKGROUND: The steroid receptor RNA activator protein (SRAP) is a newly described protein modulating the activity of multiple transcription factors including the estrogen receptor (ER). We have recently reported the immunodetection by Western blot of multiple SRAP peptides in breast tissue. High expression of these peptides, assessed by tissue micro-array (TMA) analysis, was associated with poor prognosis in patients whose primary tumors were ER positive (ER+). In such studies, it is recognized that intensity as well as specificity of the signal detected directly depends upon the antibody used as well as the position of the epitope recognized. To confirm the potential relevance of SRAP as a new prognostic factor, it is critical to establish whether similar results are obtained with independent antibodies. METHODS: Two commercial anti-SRAP antibodies (742A and 743A), respectively, recognizing the N- and C-terminal extremity of the protein, were first used to analyze by Western blot SRAP expression in protein extracts from frozen breast tumor tissue sections. These antibodies were further used to investigate by immunohistochemistry (IHC) SRAP location in paraffin-embedded breast tumors. Comparative TMA analysis of 170 ER+ tumors was eventually performed in order to establish the potential associations existing between SRAP expression and clinical outcome. RESULTS: Multiple SRAP peptides were differentially detected by Western blot. Both antibodies led to similar nuclear and cytoplasmic staining in breast tissue section. A solid correlation was found (Spearman r = 0.46, P < 0.001) between 742A and 743A IHC scores. Results from both antibodies independently showed that dividing expression levels into lower 25 percentile, 26-75 percentile, and highest 25 percentile demonstrated a hazard ratio (HR) of 1.82 (P = 0.0042) for 742A antibody and 1.35 (P = 0.14) for 743A antibody. When both scores are combined, double high expressor (by 742A and 743A) was associated with a poor prognosis of breast-cancer-specific survival (Mantel-Cox: P = 0.005, HR = 2.24). CONCLUSION: Overall, our data suggest the existence in breast tumor tissue of multiple SRAP-like peptides. Assessing their expression in primary breast tumors can predict clinical outcome in ER+ breast cancer patients.

Claudins 1, 3, and 4 protein expression in ER negative breast cancer correlates with markers of the basal phenotype.

In the present study we investigated the protein expression of claudins 1, 3, and 4 and their relationship to clinical variables and outcome in a cohort of ER-ve and ER+ve human invasive breast cancers. Immunohistochemical analysis was performed on tissue microarrays representing a total of 412 tumors and interpretable data was derived from 314, 299, and 306 tumors for claudins 1, 3, and 4, respectively. In the ER+ve subset, 5%, 89%, and 52%, and in the ER-ve subset, 39%, 79%, and 79% of tumors stained positively for claudins 1, 3, and 4, respectively (p < 0.0001, p = 0.026, p < 0.0001). Thus, in the two subsets, a significantly higher number of tumors were positive for claudins 3 and 4, compared to claudin 1. In addition, protein expressions of claudins 1 and 4 were significantly higher in those tumors that displayed characteristics of the basal-like subtype of breast cancers (ER-ve, Her-2-ve, EGFR+ve, CK5/6+ve). This study shows a unique pattern of expression for the different claudins in ER-ve and ER+ve tumors. Our data also suggests that increased expression of claudins 1 and 4 was associated with the basal-like subtype of breast cancers, a subtype generally linked to poor outcome.

Estrogen receptor-beta regulates psoriasin (S100A7) in human breast cancer.

We have previously observed a paradoxical relationship of the psoriasin/S100A7 gene with estrogen response in-vitro in ERalpha positive cells but its association with ERalpha negative status in-vivo raising the possibility that S100A7 might be regulated by ERbeta in breast cancer. Using doxycycline-inducible ERbeta and ERalpha expressing MCF-7 cells the hypothesis that psoriasin/S100A7 is ERbeta regulated was investigated To explore the relationship between psoriasin/S100A7 and ERbeta expression in-vivo, we also assessed a cohort of 233 ERalpha negative breast tumors using tissue microarrays and immunohistochemistry. Psoriasin/S100A7 was increased by 17beta-estradiol (E2) following ERbeta induction, in several clones of ERbeta over-expressing but not in the original MCF-7 cells, nor clones over-expressing ERalpha. The effect of E2 on psoriasin/S100A7 was inhibited by 4-hydroxytamoxifen and ICI 182780 but not with a selective ERalpha antagonist. An ERbeta selective-agonist but not an ERalpha selective-agonist, induced psoriasin/S100A7. This induction still occurred after stable down-regulation of ERalpha using siRNA in ERbeta inducible cells. E2 increased psoriasin/S100A7 mRNA but cycloheximide treatment inhibited this effect. A relationship between ERbeta and psoriasin/S100A7 was observed in the p53 immunohistochemically negative subset of invasive breast tumors in-vivo (r = 0.225, p = 0.046, n = 79). In conclusion we demonstrate that E2 induction of psoriasin/S100A7 can be specifically regulated through ERbeta in-vitro and associated with ERbeta in-vivo. These data support the hypothesis that psoriasin/S100A7 is specifically regulated by ERbeta activity and could be useful to guide future therapies targeting ERbeta in certain phenotypic subsets of breast cancers in-vivo.

Assessment of multiple different estrogen receptor-beta antibodies for their ability to immunoprecipitate under chromatin immunoprecipitation conditions.

Several different antibodies to total estrogen receptor (ER)beta, ERbeta1 and ERbeta2/cx have been tested and compared for their ability to immunoprecipitate ERbeta specific isoforms under chromatin immunoprecipitation conditions (ChIP). The rabbit polyclonal antibodies AP-ERbeta1 and AP-ERbeta2/cx, specific for ERbeta1 and ERbeta2/cx isoforms, respectively, were the most efficient for ChIP. The monoclonal antibody MCA1974/PPG5/10 was also able to ChIP ERbeta1, but less efficiently than AP-ERbeta1. All other antibodies tested were not suitable for ChIP analyses although most antibodies tested immunoprecipitated the appropriate ERbeta isoforms under standard conditions. To identify antibodies that can also be used to verify in-vivo expression profiles, a comparison of the antibodies to detect ERbeta isoforms by western blotting and immunohistochemistry was also undertaken. Under the tissue processing and autostaining conditions used at the Manitoba Breast Tumor Bank 385P/GC17, MCA1974/PPG5/10, Ab288/14C8 and MCA2279S/57/3 were found to be the best for IHC of ERbeta isoforms in human breast tissue biopsy sections, while Ab14021, AP-ERbeta1 and AP-ERbeta2/cx were best for western blot detection of ERbeta isoforms.

Differential expression and comparative subcellular localization of estrogen receptor beta isoforms in virally transformed and normal cultured human lens epithelial cells.

A number of variants of the wild-type (wt) estrogen receptor beta (ERbeta-1) coexist in a wide range of tissues. In the human these include, together with others, the expression of several isoforms (ERbeta-2-ERbeta-5) due to alternative splicing of exons encoding the carboxy terminus. In this study, we determined whether virally transformed cell cultures of human lens epithelial cells (HLE-B3) express both full length (or wt) and variant isoforms of ERbeta in comparison to normal secondary cultures of human lens epithelial cells (nHLE) and furthermore, identify the subcellular localization of the wtERbeta-1 and ERbeta isoform variants in HLE-B3 and nHLE cells, as well as from human breast adenocarcinoma cells (MCF-7) which provided a positive control. ERbeta isoform mRNA expression was evaluated by coupled RT-PCR. Subcellular localization of ERbeta isoforms was determined on formaldehyde-fixed, Saponin-permeabilized cells using conventional immunofluorescence techniques and affinity purified polyclonal antibodies specific for ERbeta-1 as well as to two of the truncated carboxy terminus isoforms (beta-2 and beta-5). Total RNA was extracted from HLE-B3 and nHLE cells and lens tissue, as well as from human breast adenocarcinoma cells (MCF-7) and subjected to RT-PCR using specific estrogen receptor primers intended to distinguish ERbeta-1-ERbeta-5 mRNA. The PCR products corresponded to wtERbeta-1 as well as to the isoform variants beta-2 and beta-5. The proportional distribution of wtERbeta-1, beta-2 and beta-5 PCR products differed between the normal lens epithelial cells and the SV-40 transformed lens epithelial cell line; the nHLE being similar to lens tissue with respect to relative expression of ERbeta isoform cDNAs. Confocal microscopy and immunofluorescence revealed ERbeta-2 was distributed throughout the cytosol and was associated with the nucleus of all cells examined, although sporadic immunostaining was observed with the nuclei of MCF-7. Prominent immunostaining of ERbeta-1 appeared in the mitochondria (along with weaker staining in the nucleus) of all cell types as authenticated by co-localization with Mitotrack-633. ERbeta-5 immunostaining was diffuse in the cytosol and also associated with the nuclei of all cell types. The differential subcellular partitioning of ERbeta-1 to the mitochondria and ERbeta-2 to the nucleus suggests a new aspect of regulation and function of the estrogen signalling system.

Reduced expression of oestrogen receptor beta in invasive breast cancer and its re-expression using DNA methyl transferase inhibitors in a cell line model.

To gain insights into the possible role of oestrogen receptor (ER) beta in breast carcinogenesis, immunohistochemical analysis of ER beta was performed on 512 breast specimens encompassing normal (n = 138), pure ductal carcinoma in situ (n = 16), invasive cancers (n = 319), lymph node metastases (n = 31), and recurrences (n = 8). Real-time polymerase chain reaction (PCR) was used to investigate the methylation status of the ER beta gene in the ER beta negative breast cancer cell lines SkBr3 and MDA-MB-435. A gradual reduction in, but not a complete loss of, ER beta expression was observed during the transition from normal and pre-invasive lesions to invasive cancers, where ER beta was lost in 21% of cases. This was more pronounced in invasive ductal than in lobular carcinomas, a significantly higher proportion of which were ER beta-positive (74% compared with 91%, respectively, p = 0.0004). Examination of paired primary cancers with their axillary lymph node metastases showed that if ER beta was present in the primary tumour, it persisted in the metastasis. Treatment of ER beta-negative cell lines with DNA methyl transferase inhibitors restored ER beta expression, providing experimental evidence that silencing of ER beta in breast carcinomas could be due to promoter hypermethylation. These results suggest that loss of ER beta expression is one of the hallmarks of breast carcinogenesis and that it may be a reversible process involving methylation.