The Role of ESRP1 in the Regulation of PHGDH in Estrogen Receptor-Positive Breast Cancer

Resistance to hormone therapy leads to a recurrence of estrogen receptor-positive breast cancer. We have demonstrated that the epithelial splicing regulatory protein 1 (ESRP1) significantly affects cell/tumor growth and metabolism and is associated with a poor prognosis in this breast cancer subtype. In this study, we aimed to investigate the ESRP1 protein-messenger RNA (mRNA) interaction in hormone therapy-resistant breast cancer. RNA-binding protein immunoprecipitation (RIP) followed by Clariom D (Applied Biosystems/Thermo Fisher Scientific) transcriptomics microarray (RIP-Chip) was performed to identify mRNA-binding partners of ESRP1. The integration of RIP-Chip and immunoprecipitation-mass spectrometry analyses identified phosphoglycerate dehydrogenase (PHGDH), a key metabolic enzyme, as a binding partner of ESRP1 in hormone-resistant breast cancer. Bioinformatic analysis showed ESRP1 binding to the 5' untranslated region of PHGDH. RNA electrophoresis mobility shift assay and RIP-quantitative reverse transcription-polymerase chain reaction further validated the ESRP1-PHGDH binding. In addition, knockdown of ESRP1 decreased PHGDH mRNA stability significantly, suggesting the posttranscriptional regulation of PHGDH by ESRP1. The presence or absence of ESRP1 levels significantly affected the stability in tamoxifen-resistant LCC2 and fulvestrant-resistant LCC9 cells. PHGDH knockdown in tamoxifen-resistant cells further reduced the oxygen consumption rate (ranging from P = .005 and P = .02), mimicking the effects of ESRP1 knockdown. Glycolytic parameters were also altered (ranging P = .001 and P = .005). ESRP1 levels did not affect the stability of PHGDH in T-47D cells, although knockdown of PHGDH affected the growth of these cells. In conclusion, to our knowledge, this study, for the first time, reports that ESRP1 binds to the 5' untranslated region of PHGDH, increasing its mRNA stability in hormone therapy-resistant estrogen receptor-positive breast cancer. These findings provide evidence for a novel mechanism of action of RNA-binding proteins such as ESRP1. These new insights could assist in developing novel strategies for the treatment of hormone therapy-resistant breast cancer.

Deconvolution of gene expression for microenvironmental cell types in thymomas

e20623

Background: Thymomas are rare neoplasms in the anterior mediastinum that account for less than 1% of tumors. The presence of immune cells within the tumor is one of the cardinal features of thymomas. However, the nature of the immune infiltration is not well characterized. Methods: Deconvolution of the gene expression of thymomas in the TCGA-THYM cohort was performed using the CIBERSORT and xCELL programs. Pairwise correlative analyses were performed to study the relationships of cell types and thymoma histological subtypes. Results: The differences in presence of immune cells was better highlighted using xCELL, which identifies 67 categories of cells rather than in the 23 categories identified by CIBERSORT. As expected, there were dramatic differences in the presence of T-cells (almost all subtypes) and B-cells between A/AB and type B (-1, -2, -3) thymomas. A significant difference (P < 0.0001) in different type B thymomas was observed for the presence of monocytes, macrophages, and M1, but not in M2 macrophages. Significant differences in dendritic cells were observed within subtypes of B thymoma. Similarly, statistically significant differences were observed between most subtypes in Immunoscore (except AB/B3) and MicroenvironmentScore (except A/B3; AB/B2 and AB/B3). However, StromalScore differences were observed only in comparisons of type A/AB with B type thymomas (except AB/B3). Interestingly, there also differences in postulated pre-adipocytes, hepatocytes, mesangial cells and osteoblasts in types A vs B, and melanocytes between subtypes of B thymomas. Conclusions: The differences in lymphoid content of the tumor microenvironment is associated with dramatic changes in cell type distribution in subtypes of thymoma. These differences in composition underpin their clinicopathological behavior including incidence of myasthenia and likelihood of recurrence.