Expression analysis of the estrogen receptor target genes in renal cell carcinoma

Cellular State Transformations Using Deep Learning for Precision Medicine Applications

We introduce the Transcriptome State Perturbation Generator (TSPG) as a novel deep-learning method to identify changes in genomic expression that occur between tissue states using generative adversarial networks. TSPG learns the transcriptome perturbations from RNA-sequencing data required to shift from a source to a target class. We apply TSPG as an effective method of detecting biologically relevant alternate expression patterns between normal and tumor human tissue samples. We demonstrate that the application of TSPG to expression data obtained from a biopsy sample of a patient's kidney cancer can identify patient-specific differentially expressed genes between their individual tumor sample and a target class of healthy kidney gene expression. By utilizing TSPG in a precision medicine application in which the patient sample is not replicated (i.e., n=1), we present a novel technique of determining significant transcriptional aberrations that can be used to help identify potential targeted therapies.

•

We present the Transcriptome State Perturbation Generator (TSPG) application

•

We apply TSPG to The Cancer Genome Atlas data to perturb gene expression states

•

TSPG was used to learn patient-specific (n = 1) gene expression tumor alterations

Deep learning has shown tremendous success in image and natural language processing; however, attempts to apply the tools of machine learning to better understanding biological systems are still in the stage of early adoption. We propose a novel deep-learning tool that can be used to process samples of RNA-sequencing data. By applying the Transcriptome State Perturbation Generator to human samples, we show that deep learning derives insight into the gene expression shifts required for transition between two biological conditions (e.g., normal versus tumor). RNA-sequencing data derived from a single patient's tumor were analyzed using this tool to determine gene expression aberrations specific to that patient's tumor. As medicine shifts from cohort-based population studies to individual-based precision treatments, our example demonstrates that deep learning is a powerful ally in the quest to understand how complex biological systems have shifted for a single patient.

Expression and localization of estrogen receptors in human renal cell carcinoma and their clinical significance

This study aims to (1) evaluate the immunohistochemical expression of ERα, ERα36 and ERβ in combination in human renal cell carcinoma (RCC) and nearby non-tumorous tissue (2) correlate their expression pattern with the clinicopathological parameters and prognosis of the patients; this may provide a new insight into prediction of the disease outcome and understanding its progression. The three markers showed positive cytoplasmic (± membranous) staining pattern in tumor cells. The tubules in the nearby non-tumorous tissue showed either nuclear (± cytoplasmic) staining pattern (ERα and ERβ) or only cytoplasmic staining pattern (ERα36). The mean of cytoplasmic expression of ERα, ERα36 and ERβ was significantly higher in association with poor prognostic factors: larger tumor size (P<0.0001) for each, late clinical stage (P<0.0001) for each, higher nuclear grade (P = 0.003, P = 0.002 and P = 0.022) respectively, and presence of lymphovascular invasion (P<0.0001, P = 0.006 and P<0.0001) respectively. We have demonstrated for the first time that patients whose tumors express high cytoplasmic levels of ERα, ERα36 or ERβ experience shorter overall survival and disease-free survival. The independent role of ER subunits as markers of poor prognosis is proven only for ERβ and ERα36 but not ERα. In conclusion, our results indicate that the main staining pattern of ERα, ERα36 and ERβ in RCC is cytoplasmic with relation of this pattern to bad prognosis. So we can suggest the assessment of these receptors as markers of poor prognosis in RCC patients.

Salinomycin overcomes acquired tamoxifen resistance through AIB1 and inhibits cancer cell invasion in endocrine resistant breast cancer

Salinomycin is a monocarboxylic polyether ionophore isolated from Streptomyces albus. It has been widely used as an antibiotic in veterinary medicine in poultry. A recent study demonstrated that salinomycin selectively inhibits human breast cancer stem cells; one possible mechanism of tamoxifen resistance. Our results show that salinomycin is effective in inhibiting MCF-7/LCC2 and MCF-7/LCC9 cell lines which are well-established endocrine resistant cells and has a synergistic effect in combination with tamoxifen using MTT proliferation assay. The inhibitory effect of salinomycin on the reduction of critical ER co-activator; amplified breast 1 (AIB1) mRNA and protein expression is overcoming tamoxifen resistance. Moreover, salinomycin significantly inhibits cell invasion in Matrigel invasion assay. The effect was mediated at least in part by the decrease of matrix metalopeptidase 9 (MMP-9) which is one critical enzyme facilitated in the cell invasion process. In conclusion, salinomycin should be developed as a novel agent used alone or in combination for endocrine-resistant breast cancer.

The serine protease prostasin (PRSS8) is a potential biomarker for early detection of ovarian cancer

Background

Ovarian cancer (OVC) is the deadliest of all gynecologic cancers, primarily as a consequence of asymptomatic progression. The complex nature of OVC creates challenges for early detection, and there is a lack of specific and sensitive biomarkers suitable for screening and detecting early stage OVC.

Methods

Potential OVC biomarkers were identified by bioinformatic analysis. Candidates were further screened for differential expression in a library of OVC cell lines. OVC-specific overexpression of a candidate gene, PRSS8, which encodes prostasin, was confirmed against 18 major human cancer types from 390 cancer samples by qRT-PCR. PRSS8 expression profiles stratified by OVC tumor stage-, grade- and subtype were generated using cDNA samples from 159 OVC samples. Cell-specific expression and localization of prostasin was determined by immunohistological tissue array analysis of more than 500 normal, benign, and cancerous ovarian tissues. The presence of prostasin in normal, benign, and OVC serum samples was also determined.

Results

Gene expression analysis indicated that PRSS8 was expressed in OVC at levels more than 100 fold greater than found in normal or benign ovarian lesions. This overexpression signature was found in early stages of OVC and was maintained in higher stages and grades of OVC. The PRSS8 overexpression signature was specific for OVC and urinary bladder cancer among 18 human cancer types. The majority of ovarian cell lines overexpressed PRSS8. In situ hybridization and histopathology studies of OVC tissues indicated that overexpression of prostasin was largely localized to tumor epithelium and was absent in neighboring stroma. Significantly higher levels of prostasin were found in early stage OVC serum samples compared to benign ovarian and normal donor samples.

Conclusions

The abundant amounts of secreted prostasin found in sera of early stage OVC can potentially be used as a minimally invasive screening biomarker for early stage OVC. Overexpression of PRSS8 mRNA and high levels of prostasin in multiple subtypes of early stage ovarian tumors may provide clinical biomarkers for early detection of OVC, which can potentially be used with CA125 and HE4.

Electronic supplementary material

The online version of this article (doi:10.1186/s13048-016-0228-9) contains supplementary material, which is available to authorized users.

High ERα36 Expression Level and Membrane Location Predict Poor Prognosis in Renal Cell Carcinoma

Estrogen receptor alpha 36 (ERα36), a truncated variant of ERα, is located in cytoplasm and membrane that is different from other nuclear receptors of ERα family. ERα36 is involved in progression and treatment resistance of a variety of carcinomas. However, the clinical and prognostic significance of ERα36 in renal tumors have not been fully elucidated.Here, renal tumor tissues from 125 patients were collected and immunohistochemical stained with ERα36 antibody. ERα36 expression level and location in these cases were analyzed for their correlations with clinical characteristics. The differential diagnosis value was also assessed for benign and malignant renal tumors, as well as its prognostic value.The results showed that membrane ERα36 expression was rarely detected in benign tumors but predominantly observed in malignant renal tumors. Kaplan-Meier analysis indicated that significant correlations of high ERα36 level and ERα36 membrane expression were correlated with both poor disease-free survival and overall survival. Univariate and multivariate analysis confirmed that both ERα36 high expression and membrane location can serve as unfavorable prognostic indicators for renal cell carcinoma.It is thus concluded that membrane ERα36 expression is valuable for differential diagnosis of malignant renal tumors from benign ones. Both ERα36 high expression and membrane location indicate poor prognosis in renal cell carcinoma.