Update on the role and therapeutic potential of polycomb repressive complexes in (biliary tract) cancer

KEYWORDS:

• Biliary tract cancer
• BMI1
• polycomb repressive complex
• PTC-209

1. Introduction

In a recent review we discussed the role of the polycomb repressive complexes (PRC) 1 and 2 in biliary tract cancer (BTC) [1]. Both, the PRC1 and 2 are multi-protein complexes that epigenetically regulate gene expression by specific histone modifications associated with transcriptional silencing [2]. The PRC2 consists of the core components EED, EZH2 (enzymatic activity), and SUZ12 and performs tri-methylation of histone 3 at lysine 27 (H3K27me3). The PRC1 mono-ubiquitinylates histone 2 at lysine 119 (H2AK119ub) and comprises the core components CBX, RING1, PHC, BMI1, and RYBP/YAF2, of which RINGA1 has enzymatic activity and BMI1 serves as an indispensable factor [1,2].

BTC is still a highly lethal disease with fatal outcome arising from the epithelial cells of the bile ducts and can be classified based on the anatomical origin as intrahepatic, extrahepatic, and perihilar cholangiocarcinoma as well as gallbladder cancer (GBC) [3]. The standard therapy for advanced BTC is based on the two chemotherapeutics cisplatin and gemcitabine and achieves overall survival of (only) about 1 year [4]. For a better understanding of the disease and potential new therapeutic strategies, identification of underlying pathologic mechanisms and new therapeutic targets is central.

In our review we summarized the available literature regarding the role of PRCs in BTC and concluded that the PRC core components might represent valuable therapeutic targets. Consistently, overexpression of PRC 1 and 2 core components was associated with disadvantageous clinico-pathological features such as tumor size, metastasis status, and tumor-promoting molecular traits such as under-expression and loss of prominent tumor suppressors (p16INK4A, PTEN) in BTC patient samples. Of note, most studies identified EZH2 or BMI1 as the aberrantly expressed core components of the PRC2 and PRC1, respectively [1]. At the time of the publication of our review, Kreso et al. released their initial description of the first small-molecule BMI inhibitor (PTC-209) [5]. Due to the described overexpression of BMI1 in BTC samples and the cytotoxic effect of PTC-209 in colorectal cancer cells as described by Kreso et al., we speculated that PTC-209 might have an antitumor effect in BTC cells [1]. Since then, only few new studies reported on BMI1 in BTC [6–8] and only one study investigated the effect of PTC-209 in BTC [7]. Due to the lack of new studies regarding PRC2/EZH2 in BTC, we therefore focused this editorial on new findings regarding the role of the PRC1 core component BMI1 in BTC and additionally discuss the antitumor effects of the BMI1 inhibitor PTC-209 in cancer cells (summarized in Figure 1).

Figure 1. Effect of the BMI1 inhibitor PTC-209 on (biliary tract) cancer cells.

PTC-209 is a specific small-molecule inhibitor of the polycomb repressive complex 1 core factor BMI1 that displays several anti-tumor effects in biliary tract cancer cells (BTC, left) and cells of other cancer entities (right) – based on [5,7,9–11].

2. The PRC1 core component BMI1 and its role in cancer – a short update

Due to the promising results presented by Kreso et al. in their initial publication describing PTC-209 [5], we asked, whether PTC-209 has also a cytotoxic effect on BTC cells. Using an in vitro BTC cell model, we found BMI1 to be expressed in BTC cells and that treatment with PTC-209 reduced BMI1 and H2AK119ub protein levels [7]. Moreover, in seven of the tested cell lines, PTC-209 significantly reduced the proportion of viable BTC cells. Further investigation revealed that PTC-209 caused a cell cycle stop at the G1/S transition. A detailed gene expression analysis confirmed the effect of PTC-209 on the cell cycle in BTC cells by showing a significant downregulation of numerous cell cycle promoting G1-phase genes accompanied by upregulation of the two cell cycle inhibitors CDKN1A and CDKN2B after PTC-209 treatment. Interestingly, we also measured significantly diminished mRNA levels of genes responsible for DNA synthesis initiation and DNA repair. Combined with the observed synergistic cytotoxic effect with the standard chemotherapeutic cisplatin, these results suggest that PTC-209 might be an efficient adjuvant treatment option. The cancer stem cell (CSC) concept postulates that within a tumor, a rare subpopulation of cells with stem cell properties exists that is responsible for various problems in clinical cancer management such as therapeutic resistance, secondary tumor formation, and tumor recurrence [12]. In the initial study regarding PTC-209 by Kreso and coworkers, the authors described a profound effect of PTC-209 on the CSC fraction of colorectal cancer [5]. Since then, various studies investigated the role of BMI1 and/or the effect of BMI1 inhibition on the CSC and demonstrated a connection in glioma, hepatocellular carcinoma, and breast cancer [9,13,14]. Albeit rarely investigated, the current literature suggests a contribution of CSCs also for BTC [15]. We therefore examined the effect of PTC-209 on established functional CSC characteristics in BTC cells and found impaired sphere formation potential and a reduction of the aldehyde dehydrogenase-1 positive BTC cell subpopulation following PTC-209 treatment [7]. These results suggest that BMI1 is involved in the CSC phenotype in BTC cells – an observation that is in line with a study by Sasaki et al., in which the authors describe an essential role for BMI1 regarding anchorage-independent colony formation [16]. To our knowledge, our study is currently the only publication regarding PTC-209 in BTC. However, since our review, two other reports described a tumor-progressive role for BMI1 in BTC cells. Ma and coworkers were able to connect various regulatory noncoding RNAs with expression of BMI1 in GBC [6]. They showed that the long-noncoding RNA CCAT1 was upregulated in GBC samples and associated with aggressive tumor features and that silencing of CCAT1 using siRNA resulted in diminished BMI1 levels. In contrast, overexpression of CCAT enhanced BMI1 levels. They further elucidated this observation and identified micro-RNA (miR)-218-5p as an interaction partner of CCAT1. Of note, miR-218-5p was found to be downregulated in GBC samples. In their proposed model, CCAT1 can directly interact with miR-218-5p, thereby reducing the amount of miR-218-5p that is available to negatively regulate BMI1 expression. Following this concept, silencing of CCAT1 therefore results in an increased amount of miR-218-5p available to negatively regulate BMI1 expression. Similarly to the results of our study, siRNA-mediated silencing of BMI1 impaired cell cycle progression at the G1 checkpoint. Additionally, knockdown of BMI1 reduced the number of invasive GBC cells [6]. In another study in GBC cells, the authors found that trichostin-A (TSA) and suberoylanilide hydroxamic acid (SAHA) also caused cell cycle arrest at the G1/S-transition along with a decrease of BMI1 – among other targets [8]. Although this result was not further discussed, the reduced BMI1 levels following TSA or SAHA treatment are likely representing indirect effects, since both, TSA and SAHA, are established histone deacetylase inhibitors [8]. Nonetheless, the results of this study suggest a close relationship between different epigenetic histone modification complexes in BTC.

Besides the few studies regarding BMI1 and PTC-209 in BTC, additional studies investigated the effect of PTC-209 in other cancer entities (see also Figure 1 for overview). Bolomsky et al. found that BMI1 was overexpressed in multiple myeloma (MM) patients and associated with short overall survival [10]. PTC-209 reduced BMI1 levels and number of viable MM cells in vitro. Furthermore and in accordance with the already described studies, PTC-209 caused cell cycle arrest at the G1 checkpoint accompanied by downregulation of cell cycle promoting genes and upregulation of cell cycle inhibitors. Interestingly, PTC-209 diminished the colony formation potential of MM cells as already described also for BTC cells [7]. Of note, the anti-MM effect of PTC-209 was maintained even in the presence of the known MM-promoters insulin-like growth factor 1 and interleukin 6, demonstrating the effectiveness of BMI1 inhibition in MM cells. In accordance with our study, Bolomsky and coworkers also observed a synergistic cytotoxic effect of combined PTC-209 and a standard chemotherapeutic treatment. In breast cancer cells, it was demonstrated that the miR-200c/141 locus regulates and is regulated by BMI1 [11], which is especially interesting since both of these miRs were recently described to be also involved in BTC [17,18]. The authors showed that treatment with PTC-209 not only resulted in specific downregulation of BMI1 and H2AK119ub, but also in upregulation of miRs 200c and 141 [11]. Furthermore, PTC-209 induced expression of the tumor suppressors p16INK4a, p21 and p53. Srinivasan et al. also investigated the effect of PTC-209 in breast cancer using a CSC cell model [9]. Again, PTC-209 caused accumulation of cells in the G1 phase of the cell cycle as well as reduction of quantity and size of tumor spheres. Looking at the expression of CD49f as an established surface marker of breast cancer CSC, in vitro and in vivo experiments demonstrated that the CD49f-positive subpopulation of breast cancer cells was reduced after PTC-209 incubation. Additionally, PTC-209 markedly reduced size and weight of tumors in vivo [9].

3. Expert opinion

Although published data regarding BMI1 and PTC-209 in BTC as well as PTC-209 in other cancer entities are currently limited, the available results concordantly suggest a strong antitumor effect of BMI1 inhibition by PTC-209. PTC-209 caused cell cycle arrest at G1/S and showed anti-CSC effects in several cancer in vitro models. Major problems in the management of BTC include high therapeutic resistance, high rate of tumor recurrence as well as the formation of secondary tumors, the latter being – at least partly – caused by a complex process termed epithelial-to-mesenchymal-transition [3,18]. According to the CSC concept, the potential to cause such phenomena is attributed to a subpopulation of malignant cells with stem cell properties [12]. Given the fact that treatment with PTC-209 reduced CSC characteristics in BTC cells and the fact that under physiological circumstances, stable gene silencing caused by PRC1 is essential for the maintenance of embryonic stem cells as well as adult stem cells [2], future studies should intensively investigate the connection between BMI1 and CSC in BTC and other cancer entities. Moreover, the study conducted by Ma and colleagues suggests a direct contribution of BMI1 in the regulatory network of noncoding RNA species (long-noncoding RNAs, miRs). Deregulation of miRs (oncogenic and tumor suppressive) is well established in BTC [17]. In addition, there is growing evidence for a major contribution of long-noncoding RNAs in various types of cancer, including BTC [19–21]. Future studies should therefore try to unravel the complex network of noncoding RNA-based transcriptional control and the PRC1 to get a better understanding of the potentially massive role of the PRCs in cancer and BTC development and progression.

Declaration of Interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

The authors’ projects on PRC in biliary tract cancer were supported by funds of the Oesterreichische Nationalbank (Anniversary fund, project number: 14842) and the research fund of the Paracelsus Medical University Salzburg (Grant No A-12/02/006-KIE)