Characterization of Stem Cell Genes in Solid Tumors

Abstract

Stemness genes regulate the maintenance of pluripotency and self-renewal in human embryonic stem cells (ESCs) and allow reprogramming of adult somatic cells into an ESC-like state. Accumulating evidence reveals that these genes are also abnormally expressed in cancers and play important roles in tumorigenicity, metastasis and chemoresistance. Understanding the role and expression pattern of stemness genes might be of utmost importance in different tumor cell types. SOX2 is one of the most prominent stemness factors in pluripotent stem cells, which subsequently was also reported as being involved in reprogramming and, as shown by us and other groups, found to express in different types of cancer (Bareiss, Paczulla et al. 2013, Schrock, Bode et al. 2014, Schaefer, Wang et al. 2015). In this work, we firstly describe a functional dependence of the Sex determining region Y-box 2 (SOX2) gene in regard to the expression and stability on the Ser/Thr-kinase AKT (Schaefer, Wang et al. 2015). In breast carcinoma, SOX2 expression has been linked to cancer stem cells (CSCs) and associated with poor clinical outcome. Here, we show that overexpression of AKT raises SOX2 levels, whereas knockdown or inhibition with specific inhibitors of AKT kinase depletes SOX2. Vice versa, SOX2 knockdown has no effect on AKT levels and phosphorylation status, indicating that AKT is upstream of SOX2. This functional dependence was confirmed in in vitro spheres and in vivo tumorgenicity assays, where either lentivirally mediated depletion of SOX2 or alternatively AKT-inhibition via treatment with various inhibitors could reduce sphere outgrowth and respectively tumor induction in a dose-dependent manner. Taken together, our results suggest that AKT-inhibitors efficiently target SOX2 and therefore may have the capability to eradicate tumor-initiating breast CSCs. Interestingly, we also found that the transcriptional activity of SOX2 is linked to CSC character in other cancer types. We constructed a lentiviral reporter vector, in which red fluorescent protein (RFP) is driven under the control of the SOX2 regulatory regions (either SRR1 or SRR2). These reporter-positive cells showed enhanced tumor spheres formation, whereas reporter negative cells generated fewer and smaller spheres in ovarian as well as breast carcinoma (Bareiss, Paczulla et al. 2013, Wang, Paczulla et al. 2015). Moreover, expression of stemness-related (NANOG, OCT4 and LIN28) and EMT genes (N-CADHERIN, TWIST, SNAIL) were enriched in reporter-positive versus -negative cells. Treatment with chemotherapeutic agents enhanced the percentage of reporter-positive cells. Interestingly, treatment with AKT inhibitors specifically decreased reporter positive cells(Schaefer, Wang et al. 2015). These facts indicate that SRR1/2 activity has the capability to identify CSCs in different tumor types. Next, we explored the importance of EVI1, a stem cell protein reported to regulate healthy and malignant blood stem cells, for its roles in solid tumor CSCs. We found that EVI1 is also aberrantly expressed in both breast and prostate carcinoma samples (Queisser, Hagedorn et al. 2017, Wang, Schaefer et al. 2017). Although EVI1 expresses in a majority of breast carcinoma samples independent of their ER status, it selectively impacts the biology of estrogen receptor (ER)-negative tumors, where it regulates cell growth via MAPK activity and metastatic behavior via KISS1 (Wang, Schaefer et al. 2017). Interestingly, EVI1 did not selectively mark breast CSCs since homogeneous expression was detected among all tumor cells. However, EVI1 appears to be co-expressed in and to co-regulate this compartment thus suggesting that targeted therapies against EVI1 will also eradicate this subpopulation. In healthy prostatic tissue, EVI1 expression was confined to reside within the prostate stem cell compartment located at the basal layer, as identified by stem cell marker CD44. In a prostate cancer progression cohort, EVI1 staining strongly increased with tumor progression. Functionally, EVI1 knockdown inhibited proliferation and cell cycle progression of the prostate cancer cell line PC3, as well as migratory capacity and anchorage independent growth of human prostate cancer cells. Moreover, EVI1 expression was induced in experimentally derived docetaxel-resistant prostate cancer cells (Queisser, Hagedorn et al. 2017). In summary, these data indicate EVI1 as a novel gene involved in breast and prostate cancer progression that may control carcinogenesis in part at the stem cell level.