University of Southern California Dissertations and Theses

Polycomb repressive complex 2 subunit stabilizes NANOG to maintain self-renewal in hepatocellular carcinoma tumor-initiating stem-like cells

(USC Thesis Other)

Polycomb repressive complex 2 subunit stabilizes NANOG to maintain self-renewal in hepatocellular carcinoma tumor-initiating stem-like cells

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Polycomb Repressive Complex 2 Subunit Stabilizes NANOG to Maintain Self-renewal in Hepatocellular Carcinoma Tumor-initiating Stem-like Cells by Cheng (James) Liu A Thesis Presented to the FACULTY OF THE USC KECK SCHOOL OF MEDICINE UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (MOLECULAR MICROBIOLOGY AND IMMUNOLOGY) May 2023 ii Acknowledgements I would like to thank my PI and advisor, Dr. Keigo Machida, for giving me the opportunity to conduct research for this thesis project. I am grateful for his continual guidance, support, and mentorship throughout my research process. I would like to thank Dr. Dawei Yeh for starting this project and setting the foundation where I was able to start my own research. I would also like to thank my amazing fellow lab members for their help and support. Lastly, I would like to thank my committee members Dr. Keigo Machida, Dr. Weiming Yuan, and Dr. Oliver Bell for their valuable insight and feedback. iii Table of Contents Acknowledgements .......................................................................................................... ii List of Tables ................................................................................................................... iv List of Figures ................................................................................................................... v List of Abbreviations ....................................................................................................... vi Abstract ........................................................................................................................... vii Chapter One: Introduction ................................................................................................ 1 Materials and Methods ..................................................................................................... 6 Results ............................................................................................................................ 15 Discussion ....................................................................................................................... 36 Conclusion and Future Directions .................................................................................. 40 References ...................................................................................................................... 43 iv List of Tables Table 1. Antibodies used for immunoprecipitation and western blot analysis. ........................ 14 Table 2. Summary of Top Drug Candidates. ............................................................................ 29 v List of Figures Figure 1. Proposed mechanism of TIC genesis and PRC2 stabilization of NANOG. ............... 5 Figure 2. NANOG and PRC2 binding domain mapping by co-immunoprecipitation western blot analysis. ................................................................................................................. 21 Figure 3. EED is essential for NANOG protein level and enhances TIC self-renewal. ........... 22 Figure 4. EED stabilizes NANOG protein post-transcriptionally. ........................................... 23 Figure 5. Co-IP western blot analysis of NANOG and its PEST domain-associated proteins. 25 Figure 6. Site-directed mutagenesis of critical NANOG serine residues. ................................ 26 Figure 7. Drug inhibitor screening by XTT viability assay. ..................................................... 27 Figure 8. National Cancer Institute’s 60 Cancer Cell Line GI50 of Top Drug Candidates. .... 35 vi List of Abbreviations CD133 Cluster of differentiation 133 DNA Deoxyribonucleic acid EED Embryonic ectoderm development protein EZH2 Enhancer of zeste homolog 2 FBXW F-box/WD repeat-containing protein HCV Hepatitis c virus IB Immunoblot IP Immunoprecipitation NSC Cancer chemotherapy national service center PCR Polymerase chain reaction PRC1 Polycomb repressive complex 1 PRC2 Polycomb repressive complex 2 RNA Ribonucleic acid shRNA Short hairpin RNA siRNA Silencing RNA SUZ12 Suppressor of zest 12 TIC(s) Tumor-initiating stem like cells XTT 2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino) carbonyl]-2H- tetrazolium hydroxide vii Abstract Tumor-initiating stem like cells (TICs) contribute to chemoresistance and tumor relapses in hepatocellular carcinoma (HCC). Our group has previously reported the toll-like receptor 4 induced upregulation of stemness factor NANOG which generated CD133+ TICs in alcohol and high fat diet mice. Additionally, polycomb repressive complex 2 (PRC2) which has long been reported to be involved in cancer stem-like cell formation via histone 3 lysine 27 trimethylation was found to be associated with many NANOG target gene loci. This led to our question of whether PRC2 and NANOG cooperate to maintain TIC self-renewal and other phenotypes such as chemoresistance. In this study, we discovered a unique binding of PRC2 core subunit, embryonic ectoderm development (EED), to NANOG N-terminus PEST domain which is essential to NANOG degradation through ubiquitin proteasome pathway. We further investigated the effects of EED knockdown to NANOG protein stability as well as TIC maintenance. Mechanistically, we examined known factors to NANOG degradation pathway and how EED may contribute to overabundance of NANOG in TICs. Lastly, we utilized cell viability assay screening to identify potential inhibitors of EED-NANOG interaction by examining the drug’s ability to selectively target TICs. Understanding the mechanism of TIC maintenance may present a novel therapeutic target and strategy to treat metastatic HCC. Altogether, this study showed a novel interaction between two known oncogenic proteins which contribute to cancer stem-like cells and identified new drugs that are selectively cytotoxic to TICs. 1 Introduction Hepatocellular carcinoma (HCC) is usually diagnosed at advanced stages when tumors are unresectable and greatly limits available treatment options in patients. Five-year survival rate of advanced, metastatic HCC can be as low as 2.5% and viable therapies only provide modest benefits in only 30% of patients (Chidambaranathan-Reghupaty et al., 2021). Globally, death due to late-stage metastatic HCC steadily increases with reported number of deaths close to 700,000 and 800,000 new cases every year. The prevention of hepatocellular carcinoma and treatment of advanced HCC remain a global healthcare challenge. Alcohol consumption increases risk of HCC development 3- to 10-folds and has shown synergistic effect with hepatitis c virus (HCV) infection to induce liver tumorigenesis (Matsushita, 2019). Alcoholism especially with high-fat diet leads to increased intestinal permeability and higher translocation of endotoxins into the bloodstream. This exposure activates Toll-like receptors (TLR) in the liver and induces alcohol-related liver disease (ALD) and liver cirrhosis (Nishimura et al., 2021). Clinical data of adults diagnosed with HCC clearly indicate alcoholism, high-fat diet, HCV infection synergistically contribute to chronic liver damage and development of HCC (Moussa et al., 2021). In advanced-stage metastatic HCC, multi-kinase inhibitors such as Sorafenib are the primary treatment option; however, patients often develop chemoresistance due to heterogeneous nature of HCC (Cabral et al., 2020). It is crucial, therefore, to understand the mechanisms of drug-resistance and explore novel therapeutic strategies targeting these subpopulations. Tumor-initiating stem-like cells (TICs) or cancer stem cells, a rare subpopulation within tumors, are drug-resistant, asymmetrically dividing, and tumor-initiating which contribute 2 chemo-resistant and metastatic disease (Bezuidenhout, 2019). Previous studies by our group have shown stemness marker CD133 + TICs were generated from HCC developed in mice infected with HCV and fed alcohol and high-fat diet. These factors induced “leaky gut” where endotoxins, primarily lipopolysaccharides (LPS), enter the bloodstream and activate Toll-like receptor 4 (TLR-4) in hepatocytes. TLR4 induces immune activation which further damages the liver resulting in the generation of TICs (Figure 1A). These HCC samples showed upregulation of stemness factor NANOG (Chen, C.L. et al., 2013). In a follow up study, NANOG ChIP-seq analyses revealed that NANOG regulates the expression of genes involved in mitochondrial metabolic pathways required to support TIC self-renewal and drug resistance (Chen, C.L. et al., 2016). These results illustrate the importance of understanding how NANOG is upregulated in TICs and the mechanisms for TIC self-renewal. The polycomb repressive complex 2 (PRC2) is widely studied in embryonic stem cells for its silencing of developmental transcription factors by tri-methylation of histone 3 lysine 27 (H3K27). Human PCR2 consists of four core subunits: Enhancer of zeste homolog 2 (EZH2), embryonic ectoderm development (EED), suppressor of zeste 12 (SUZ12), and retinoblastoma suppressor associated protein 46/48 (RbAp46/48) and several auxiliary subunits (Shi et al., 2017). In recent years, dysregulation of PRC2 and its epigenetic modification activity has been linked to tumorigenesis of many types of cancers including diffuse intrinsic pontine glioma (DIPG), prostate cancer, and HCC (Ammerpohl et al., 2012). Due to its complex structure and dependence on multiple subunits and cofactors, PRC2 requires recruitment to target chromatin site via PRC1 (polycomb repressive complex 1) and stability by complex formation of its core subunits in order to carry out its catalytic activity (Yu et al., 2019). Because of its unique role in cancers as targetability, multiple inhibitor candidates of EZH2 and EED have reached clinical 3 stage targeting wide range of cancers (Chu et al., 2022). In addition to the efficacy of these inhibitors by prohibiting PRC2-mediated H3K27me3, selective EZH2 inhibitor has shown to enhance anti-PDL1 immunotherapy treatment by upregulating tumor-infiltrating lymphocytes (TIL) and tumor specific neoantigens (Hong et al., 2019). These advances demonstrate the expansive therapeutic potential of PRC2 inhibitors and the need to delineate the multiple aspects of PRC2’s involvement in metastatic cancers. Polycomb group (PcG) proteins have long been associated with the maintenance of pluripotency and cell fate in embryonic stem cells. More specifically, PcG proteins including PRC1 and PRC2 have been found on NANOG lotus as well as NANOG target genes (Chen, L. et al., 2021). In glioblastoma, EZH2 inhibitor has shown to disrupt cell plasticity and self-renewal of glioma stem-like cells without apoptotic effects (Natsume et al., 2013). Another study demonstrated that inhibiting PRC2 core subunits sensitize sorafenib-resistant HCC cells to traditional chemotherapy (Kusakabe et al., 2021). Although these studies point to a link between PRC2 and NANOG in tumor-initiating cancer cells, the mechanisms of how these inhibitors induce therapeutic benefits remain to be elucidated. Many groups have demonstrated NANOG N-terminus PEST (proline (P), glutamic acid (E), serine (S), and threonine (T)) domain is essential to NANOG stability. PEST domain is the target site for ubiquitin proteosome degradation and upstream kinase phosphorylation (Ramakrishna et al., 2011). In embryonic stem cells, phosphorylation of NANOG PEST domain serine residues has shown to increase stability by interacting with prolyl isomerase, Pin1 (Moretto-Zita et al., 2010). Another study reported ERK1 phosphorylates NANOG to promote E3 ligase, F-box and WD40 repeat domain-containing 8 (FBXW8), binding and subsequent degradation of NANOG (Kim et al., 2014). In cancer stem-like cells, Pin1 and NANOG 4 upregulation have been correlated with advanced human glioma (Yang et al., 2013). Additionally, FBXW7 has also been reported to bind to NANOG in both embryonic stem cells as well as cancer stem-like cells (Takeishi, Nakayama 2014). In this study, we aim to understand how PRC2 subunits may interact with NANOG to promote self-renewal and explore the possibility of exploiting this relationship to specifically target the TIC subpopulation in metastatic HCC. We hypothesize PRC2 interacts and stabilizes NANOG through avoiding NANOG’s normal proteasome degradation pathway (Figure 1B). First, we plan to map PRC2 binding to NANOG as well as this interaction’s effect on NANOG stability. Second, we would screen drug libraries to identify potential inhibitor to specifically block this interaction and assess the therapeutic potential against TICs. Altogether, this study would further our understanding of TIC self-renewal and novel therapeutic targets to eradicate TICs in metastatic HCC. 5 Figure 1. Proposed mechanism of TIC genesis and PRC2 stabilization of NANOG. A.) the mechanism of TIC generation from alcohol and obesity induced liver cancer. LPS from the intestine enters the bloodstream and activates TLR4 in hepatocytes. The resulting tumor- initiating cells have CD133 surface marker and NANOG upregulation inducing self-renewal. These HCC TICs exhibit stem-like characteristics and are chemo-resistant posing a hurdle in HCC treatment. B.) hypothetical model of PRC2 stabilization of NANOG in TIC by avoiding ubiquitination and proteosome degradation. A drug candidate, inhibitor X, would theoretically inhibitor PRC2 binding to NANOG and its subsequent stabilization. This would sensitize TIC to traditional chemotherapy or be directly toxic to TICs specifically. 6 Materials and Methods Cell Lines and Cell Culture. Human hepatocellular carcinoma cell lines HepG2 and Huh7 were originally purchased from ATCC (Manassas, VA) and consistently used in our lab. The HCC cell lines were cultured in Dulbecco’s Modified Eagle’s Medium (Genclone) supplemented with 10% heat-inactivated fetal bovine serum (Genclone), 1% Glutamax (Gibco), and 1% penicillin-streptomycin (Gibco). Cells were maintained at 37°C and 5% CO2 in a humidified incubator. Sorafenib-resistant HepG2 and Huh7 cell lines were established here and used in various assays. HepG2 and Huh7 cells were seeded with 10 µM of sorafenib (BAY 43-9006) (Stelleckchem). The cells which successfully attached and grew were passaged 5 fives and cultured for at least 14 days consistently in 10 µM of sorafenib. The resistant HepG2 and Huh7 cells were cryopreserved and used in downstream assays. The cells were kept in 10 µM of sorafenib before each assay to ensure population identity. Plasmid Construction. Human NANOG protein coding sequence was amplified from HepG2 cells total RNA by RT- PCR using SuperScript III (Thermal Fisher Scientific) reverse transcriptase and Q5 DNA polymerase (New England Biolabs), respectively. PCR amplicon of NANOG was subsequently cloned into Flag-tagged eukaryotic expression vectors pRK5. This full length NANOG cds plasmid was used as a template to generate various deletion constructs through PCR cloning of serine-rich N terminal (aa 1-94, N), NK-2 type homeobox (aa 95-154, H), and C-terminal tryptophan-rich domains (aa 196-240, W) and domains either upstream (aa 155-195, C1) or downstream (aa 241-305, C2) of W domain. 7 The expression constructs for human EZH2 were generated by subcloning of corresponding protein coding sequence from pCMVHA hEZH2 (Addgene plasmid #24230) to Myc-tagged pRK5. The expression constructs for human SUZ12 and EED were generated through Gibson assembly of restriction digestion-linearized Myc-tagged pRK5 and PCR amplicon for corresponding protein coding sequence by using NEBuilder HiFi DNA Assembly Cloning Kit (New England Biolabs) following the manufacturer’s instruction. Various deletion mutant constructs were generated by PCR cloning using the full-length expression constructs of EZH2 or SUZ12 as a template. Myc-tagged EZH2 deletion mutants consist of the N-terminal homology domain I (aa 1-250, HI), homology domains II (aa 251-481, HII), CXC domain (aa 560-616, CXC), or C-terminal SET domain (aa 617-751, SET). Myc- tagged SUZ12 deletion mutants comprise N-terminal domain (aa 1-550, N) or C-terminal VEFS domain (aa 551-739, VEFS). Plasmid Transfection. Huh7 cells were grown to 70-90% confluency and transiently transfected using a lipid-based transfection reagent Lipofectamine 2000 (Invitrogen). Following manufacture protocol, mixtures of transfection complexes were prepared using a:1:1:50 ratio of Lipofectamine (µl), DNA (µg), and Opti-MEM (Gibco), respectively. After 20-minute incubation, the entire mixture was directly added to cells dropwise and plates were returned to CO2 incubators. 24 hours after transfection, growth medium was replaced with fresh media described above. Some sample groups were treated with 10 µM proteosome inhibitor MG132 (Selleckchem) after media change. 48 hours post-transfection, all sample cells were lysed using RIPA buffer (Thermo Scientific) with protease and phosphatase inhibitor cocktail (Thermo Scientific) and used for downstream assays or kept frozen at -80°C. 8 Lentiviral Transduction by shRNA. To prepare stable knock down of PRC2 subunits in Huh7 cells, lentiviral transduction of shRNA was used. The shRNA sequences were obtained from The RNAi Consortium shRNA Library (Broad Institute) and the oligonucleotides were synthesized (GenScript). The shRNA was cloned into pGreenPuro (CMV) lentivector (System Biosciences). Transfection mixture was prepared using packaging plasmid psPAX2 (Addgene #12260), envelope plasmid pMD2.G (Addgene #12259), shRNA vector plasmid, Lipofectamine 2000 (Invetrogen) and Opti-MEM (Gibco) following manufacturers. The mixture was incubated for 5 minutes at room temperature then added to HEK293T cells grown to 70% confluency. After 24 hours post-transfection, the media was replaced with fresh growth media. Growth media was harvested and replaced after 48 hours and 72 hours post-transfection. The collected media containing lentivirus was filtered using 0.4 µm syringe driven filter and stored at -80°C. Huh7 cells were infected with the lentivirus for 48 hours and selected by puromycin (Gibco) for 3 continuous passages. Knockdown cells were screened for GFP after antibiotics selection and cryopreserved for later use. RT-qPCR. Cells grown using DMEM media for 2 days then were lysed by TRIzol reagent (Invitrogen). Following manufacturer’s protocol, total RNA was collected and quantified by Nanodrop (Thermofisher) and to ensure purity of RNA samples. Next, Universal One-Step RT-qPCR Kit (NEB) was used per manufacturer’s protocol with the collected RNA and specific oligonucleotides (GenScript) in 96-well PCR plate (Thermo Scientific) to reverse transcribe the RNA into cDNA and quantify by SYBR reagent included in the kit. The plates were read by StepOnePlus™ Real-Time PCR System (Applied Biosystems) using recommended temperature 9 and duration setting including melt curve. The relative fold change of the samples was calculated from ∆∆Ct. Magnetic Sorting of TIC Population To isolate the tumor-initiating cells population, CD133 + microbeads kit (Miltenyi Biotec) was used. Cells were grown in DMEM media and pooled into 1×10 8 total cells. Following manufacturer’s protocol, the cells were incubated for 30 minutes in 4°C with 100 µl kit buffer and CD133 microbeads each. The cells were pelleted by centrifuge and resuspended in 500 µl selection buffer (0.5% BSA and 2mM EDTA in PBS, pH 7.2). Using magnetic separator with LS column (Miltenyi Biotec) attached, CD133 + cells were attached to the CD133 magnetic microbeads and retained in the column by the magnet. CD133 - population was collected as flow- through. After flushing out CD133 - population 3 times with 3 mL selection buffer, CD133 + population was collected off the magnet with 5 mL of selection buffer. The separated populations were put back in culture to recover and expand for two days. Cells were then used in downstream analysis or cryopreserved for further use. CD133 + cells were tracked from day of selection (Day 0) to maximum 10 Days post-selection. XTT Viability Assay. CD133 + and CD133 – Huh7 cells were seeded at concentration of 5 × 10 3 cells/ well in 100 µl of culture medium into 96-well plates and incubated for 16 hours. The drug candidates obtained from NCI drug library plates were added at 10 µM per well and 1% DMSO was included as negative control. After 48 hours of incubation, XTT cell viability kit (Cell Signaling) was used to indicate viability after treatment. Per manufacturer’s protocol, the electron coupling reagent and XTT reagents were mixed (1:50 volume ratio) and 50 µl of the mixture was added per well. 10 After 3 hours incubation, the absorbance was read by FLUOstar Omega microplate reader (BMG Labtech) at 450 nm. Immunoprecipitation Dynabeads Protein A/G magnetic beads (Invitrogen) were used to immunoprecipitate NANOG and its associated proteins. 200 µl whole cell lysate collected from each sample group was added to 20 µl of pre-washed magnetic beads. The mixture was incubated for 20 minutes at room temperature and beads were separated using magnetic rack (Invitrogen). 2 µl of primary antibody was added to the pre-cleared lysate (200 µl). 2 µl of corresponding isotype antibody was added as isotype control. The samples were incubated with rotation overnight at 4°C. The immunocomplex was then rotated at room temperature with 20 µl of pre-cleared beads. After 4 times of washing with wash buffer or 1X cell lysis buffer, the pellet was resuspended with 20 µl of 3X SDS sample buffer and proceed for western blot analysis. Western Blot. Cells were lysed with RIPA buffer (Thermo Scientific) and 1X protease and phosphatase inhibitor cocktail (Thermo Scientific). The protein concentration was quantified by Bradford assay (Biorad) and results were collected by FLUOstar Omega reader. 20µg of the protein sample was mixed with 6X Laemmli SDS sample buffer and water to make a 20µl sample. Alternatively, immunoprecipitated samples were mixed with 3X Laemmli SDS sample buffer. The mixture was boiled for 5 minutes at 95°C then loaded to SDS-PAGE gels for electrophoresis at 70V (15 minutes) and 140V (45 minutes). After electrophoresis, the gels were stacked and transfer to PVDF membrane (Thermo Scientific). The blots were blocked using 5% skim milk in Tris-buffered saline with 0.1% Tween® 20 detergent (TBST) for 1 hour in room temperature. Following five times of five-minute washing with TBST, the blots were incubated with primary 11 antibody per manufacturer’s recommended dilution. After second washing, corresponding secondary HRP antibody (Santa Cruz Antibodies) was added to the membrane and incubated for 1 hour in room temperature. ECL HRP substrate (Advansta) was used to soak the blots for 2 minutes, and the blots were visualized by ChemidocMP imager (Biorad). The blot images were then analyzed and prepared with ImageJ software. Antibodies used summarized in Table 1. Antibodies used for immunoprecipitation and western blot analysis. Site-directed Mutagenesis To delineate the effects of upstream kinases, several known phosphorylation site of NANOG was mutated from serine to alanine to prevent phosphorylation. QuickChange II site-directed mutagenesis kit (Agilent) was used per manufacturer’s protocol with primer sets (GenScript) corresponding to each selected phosphorylation site. pRK5 FLAG-NANOG plasmid (Addgene) constructed previously in the lab was used as the backbone of each mutagenesis. Ultracompetent E. coli cells were transformed with the resulting plasmid and ampicillin-resistant colonies were picked after overnight incubation. QIAprep Spin Miniprep Kit (Qiagen) was used to extract plasmid DNA and the mutant plasmids were sent for sequencing (Genewiz) to confirm only the target site was mutated. The plasmids containing the correct single mutation were then used in transfection and downstream analysis. Spheroid Formation Assay. Hepatoblasts transfected with EED overexpression vector or NANOG knockdown were cultured for 12 days in DMEM/F-12 media (Sigma). Cells were washed with 1× PBS and centrifuged at 150 g for 5 min to pellet the cells. The supernatant was aspirated, and cells were counted using Countess Cell Counter (Invitrogen). 100 cells were serially diluted to three separate wells in a 6 well Ultra-low attachment cell culture plate (Corning). DMEM/F-12 media (Sigma) was used 12 containing 10% FBS, Insulin (1 mg/ml), Dexamethasone (1000 ng/ml), Nicotinamide (10 mM), HEPES (5 mM), 1% penicillin/streptomycin, and Epidermal Growth Factor (20 ng/ml). Spheroids were allowed to grow for ten days in 37°C 5% CO2 incubator and then were serially diluted to 100 cells per well twice more. Each time the spheroids were counted after 12 days. To test single cell’s ability to form spheroids in different sample groups (Huh7), cells were grown out in standard complete DMEM media for 2 days at 37°C in 5% CO2 incubator. The cells were dissociated and passed through 40 µm filter to obtain single cell suspension. Cells were counted using 0.4% trypan blue (Invitrogen) and pelleted at 300×g for 5 minutes. The supernatant was completely aspirated, and the pellet resuspended to 10 viable cells/ml in DMEM/F12 media (Sigma) containing 10% FBS (Genclone), 1% penicillin/streptomycin (Invitrogen), 1% Glutamax (Invitrogen), and 20 ng/ml of epidermal growth factor (Invitrogen). In 96-well ultra-low adhesion plates (Corning), each well was seeded 100 µl of the cell suspension. Multiple plates were seeded per sample group and incubated at 37°C in 5% CO2 incubator (Day 0). On day 6, 100 µl of fresh DMEM/F-12 media was added directly to each well. On day 12, the number of spheroids were counted for each sample groups. The spheroid counts were divided by number of wells seeded and multiplied by 100 to obtain spheroid formation efficiency. NANOG Protein Stability by Cycloheximide Chase Analysis. Huh7 cells were either transduced with scrambled shRNA or EED knockdown via lentivirus. Seventy-two hours after transduction, 100 µM of cycloheximide (CHX) was added and cells were lysed at 0, 0.5, 1, 1.5, 2, and 2.5-hours post-treatment. After cell lysate collection, NANOG protein levels were detected by western immunoblotting corresponding to each time point. 13 Statistical Analysis. Statistical significance was determined using ANOVA single factor. Asterisks (*) indicate a p- value < 0.05 and (**) indicate a p-value <0.01. Student’s t-test was used to analyze data with statistical software such as GraphPad Prism. Error bars were added on graphs from calculated standard deviation (SD). 14 Table 1. Antibodies used for immunoprecipitation and western blot analysis. 15 Results Domain mapping of NANOG and PRC2 binding sites. We sought to investigate the binding domain of NANOG and PRC2 core subunits: EZH2, EED, and SUZ12. To accomplish this, first plasmids were constructed carrying expression vectors of various human NANOG domain deletion mutants with N-terminal Flag epitope tag as well as plasmids with full length hEZH2, hEED, and hSUZ12 containing N-terminal Myc- epitope tags (Figure 2A). HEK293T were co-transfected with plasmids containing NANOG deletion mutants and Myc-tagged EZH2, EED, or SUZ12. Cell lysates were immunoprecipitated (IP) with anti-Myc antibody followed by SDS-PAGE and immunoblotting with anti-Flag antibody. For EED, cell lysates were immunoprecipitated with anti-Flag antibody and immunoblotted with anti-Myc antibody. The results of Co-IP western showed W domain and Carboxyl terminal domains (C1 and C2) of NANOG are essential to EZH2 (Figure 2B) and SUZ12 (Figure 2C) binding. Surprisingly, only EED showed specific binding to NANOG N- terminal domain and other subunits do not need N-terminal domain to bind to NANOG (Figure 2D). Because PRC2 subunits interact with one another, some NANOG deletion mutants still displayed EED binding without N-terminus presence. EED binds to NANOG N-terminus PEST domain. Through the Co-IP western blot analysis, we revealed that EED interacted with NANOG at both N-terminus domains and the C1-W domains while EZH2 and SUZ12 bind with C1-W domains ( Figure 2A). This specific binding to NANOG N-terminal domain may suggest EED independently act on NANOG. The unique interaction between EED and NANOG drove us to further investigate whether EED and NANOG interaction is essential for NANOG stability and TIC maintenance in HCC. We summarized NANOG associated proteins as well as essential 16 phosphorylation sites reported in literature in combination with PRC2 subunit bindings reported here (Figure 3A). To investigate EED and NANOG interaction, we generated EED-knockdown Huh7 cells via shRNA lentiviral transduction. The sh-EED Huh7 cells showed reduced protein level of both NANOG and EED compared to sh-Scrambled control cell line (Figure 3B). This finding indicated that EED protein level is associated with NANOG protein level and reduction of EED resulted in similar effect in NANOG. To test if EED is an essential factor for TIC generation and maintenance, we seeded Huh7 cells treated with either lentiviral knockdown of NANOG or EED overexpression vector. These cells were serially diluted then seeded on ultra-low adhesion plates for 12 days. Spheroid colonies were counted and statistically analyzed for significance. Cells transfected with EED overexpression vector formed significantly more colonies and this effect was reversed when NANOG is knocked down (Figure 3C). This study demonstrated EED overexpression significantly increased TIC generation and self-renewal ability. However, this effect is linked with NANOG since the colony formation ability was greatly attenuated with NANOG knockdown. EED stabilizes NANOG protein level by blocking its proteosome degradation pathway. Next, we wanted to see if EED stabilizes NANOG through its binding of NANOG PEST domain and interfere with NANOG’s degradation pathway. First, cycloheximide (CHX) chase analysis was conducted using sh-Scrambled and sh-EED Huh7 cells generated via lentiviral transduction. 72-hours post transduction, cells were treated with cycloheximide to stop protein synthesis. From 0 to 2.5-hours post-treatment in 30-minute intervals, cells were lysed and collected for western blot analysis. EED knockdown slightly decreased NANOG protein stability 1 hour post treatment with cycloheximide (Figure 4A). 17 To further decipher whether EED affects NANOG at the mRNA or post-transcription level, we examined EED knockdown’s effect on NANOG and its downstream target’s mRNA level by RT-qPCR. Sh-EED and sh-scrambled Huh7 cells were cultured for 2 days and treated with 10 µM of MG132 (proteosome inhibitor) for 24 hours. The cells were lysed, and total RNA collected for cDNA synthesis and subsequent quantitative PCR analysis. Relative fold change was calculated and sh-EED was normalized with sh-scrambled. EED knockdown did not affect NANOG mRNA level; however, expression of known NANOG target gene, cytochrome c oxidase subunit 6A 2(COX6A2), significantly increased with EED knockdown. Moreover, this effect was rescued upon treatment with MG132 (Figure 4B). Because NANOG normally downregulates expression of COX6A2, our finding indicated that NANOG was degraded post- transcriptionally in EED knockdown group through proteosome degradation pathway. To confirm this at the protein level, we transfected EED knockdown Huh7 cells and sh- Scrambled Huh7 cells with Flag-tagged NANOG expression vector. 24 hours post transfection, some cells were treated with 10 µM of MG132. Whole cell lysates were collected 48 hours post- transfection and analyzed by western blot. NANOG protein level decreased upon EED knockdown compared to sh-Scrambled and this effect was again rescued by MG132 treatment (Figure 4C). Interestingly, EED and EZH2 protein levels were both significantly increased by MG132 treatment in both sh-EED and sh-Scrambled groups (Figure 4D). These results suggest EED knockdown caused NANOG to be degraded at greater level through proteasome degradation pathway and both NANOG and PRC2 subunits are tightly controlled by the same pathway. 18 FBXW8 binds to NANOG independently of EED. To elucidate which of the many reported proteins which bind to NANOG are affected by EED, we employed immunoprecipitation (IP) western blot analysis. Both sh-EED and sh-scrambled Huh7 cells were transfected with Flag-NANOG vector with/without MG132 treatment. The cell lysates were immunoprecipitated with anti-flag antibody to pull down NANOG and followed by western blot analysis of various targets. Using Protein G beads for IP, sh-scrambled Huh7 cells with MG132 treatment showed greater levels of both Pin1 and FBXW8 by immunoblotting while other groups showed similar levels to isotype control (Figure 5A). To reduce the isotype control background, Protein A beads were used instead and showed FBXW8 protein in all the groups at similar levels (Figure 5B). However, ubiquitin and Pin1 were not detected by the same method. To see if FBXW7 binding to NANOG is altered by EED knockdown, we co-transfected Myc-tagged FBXW7 vector with Flag-tagged NANOG in the same two Huh7 cell lines. NANOG was immunoprecipitated by anti-flag antibody and Protein A beads followed by immunoblotting with anti-Myc antibody. This method showed similar band intensity across the sample groups compared to isotype control (Figure 5C). Site-directed mutagenesis of critical NANOG phosphorylation serine residues. Next, we investigated whether EED’s stabilization of NANOG is through interfering with NANOG phosphorylation. To study this, we used site-directed mutagenesis to change four reported serine residues that affects NANOG stability (S52, S65, S71, and S79) to alanine which blocks phosphorylation of these sites. Flag-tagged NANOG mutants (S52A, S65A, S71A, and S79A) were constructed by molecular cloning and transformed in E. coli. The ampicillin- resistant transformants were picked and the plasmid DNA collected by miniprep. Two mutant plasmids, S65A and S79A, yielded low DNA concentration (~35 ng/µl) and returned poor 19 sequencing results. S52A and S71A were successfully sequenced (Genewiz) to confirm the mutant plasmids only contain the intended mutagenesis. However, both S52A and S71A did not receive the intended single site mutagenesis and contained off-target mutations shown by NCBI BLAST tool (Figure 6A) (Figure 6B). Alternative primer designs may be necessary, and the PCR cloning conditions may need to be optimized. Inhibitor drug candidate screening by XTT viability assay. To screen for a drug candidate selectively targeting TICs, we utilized National Cancer Institute (NCI) Developmental Therapeutics Program (DTP) available drug plates. We first screen through the drug plates to identify top selective compounds targeting CD133 + Huh7 cells then would move to IC50 to determine the effective dosage and finally to in vivo cancer models (Figure 7A). We seeded freshly sorted CD133 + and CD133 - Huh7 cells for 16 hours and treated the cells with 10 µM of drug candidate for 48 hours. XTT (sodium 3´- [1- (phenylaminocarbonyl)- 3,4- tetrazolium]-bis (4-methoxy6-nitro) benzene sulfonic acid hydrate) assay was used to determine cell viability compared to DMSO treated controls. From the first drug screening, we identified 5 compounds (NSC8090, NSC14540, NSC25154, NSC123127, and NSC169780) which exhibited the best selective cytotoxic effect against CD133 + Huh7 cells (Figure 7B). We then tested the same top compounds in HepG2 HCC cell and saw no difference in cell viability between treated and DMSO groups (Figure 7C). Notably, HepG2 cells contain wild type TP53 which may cause the differential response to the drug candidates. This could also be another sign that the hit compounds are targeting mechanisms specific to CD133 + TICs. 2 of the top compounds (NSC8090, NSC14540) have no previously reported anticancer effects while others (NSC25154, NSC123127, and NSC169780) have been approved as chemotherapy or used in combination with other chemotherapy agents ( Table 2). NSC25152 and NSC 123127 20 mainly interfere with DNA replication processes; NSC169780 is a known iron chelator used to prevent cardiotoxicity ( Table 2). These drug candidates exhibited novel inhibitory target and need to be further examined for their activities against TICs. 21 Figure 2. NANOG and PRC2 binding domain mapping by co-immunoprecipitation western blot analysis. A.) Graphical summary of Flag-tagged NANOG deletion mutant and PRC2 subunit binding to each mutant. B.) Co-IP western blot analysis of full length Myc-tagged EZH2 binding with various NANOG deletion mutants. HEK 293T cells were co-transfected with Myc-tagged full length human EZH2 construct along with various Flag-tagged NANOG deletion mutants. The cell lysates collected were immunoprecipitated with either anti-Myc or anti-Flag antibodies followed by immunoblotting with anti-Flag antibody. Whole cell lysates were directly immunoblotted with anti-Myc and anti-Flag antibodies as positive controls. C.) Co-IP western analysis of SUZ12 and NANOG was done as described above. Myc-tagged SUZ12 and various NANOG deletion mutants were co-transfected in HEK 293T cells. The cell lysate collected were immunoprecipitated and analyzed by western blot with anti-Myc or anti-Flag antibodies. D.) HEK 293 T cells were co-transfected with full length Myc-tagged EED and Flag-tagged NANOG deletion mutants. 22 Figure 3. EED is essential for NANOG protein level and enhances TIC self-renewal. A.) Graphical summary of binding between NANOG and PRC2 subunits and other elements from literature. Peptidyl-prolyl cis-trans isomerase NIMA-interacting 1 (PIN1) and F-box/WD repeat-containing protein (FBXW) are known to bind to NANOG PEST domain. Critical serine residues have been reported to be phosphorylated and affect NANOG stability. Here we reported EED has binding activity to NANOG PEST domain and EZH2/SUZ12 binds to the tryptophan- rich domain. B.) sh-Scrambled and sh-EED Huh7 cells were generated via shRNA lentiviral transduction. EED knockdown not only affected EED protein level but also significantly decreased NANOG protein level shown by western blot. C.) Spheroid formation assay of hepatoblasts transfected with EED overexpression vector or NANOG knockdown. EED overexpression significantly increased number of spheroids formed. However, this effect is attenuated when NANOG is knockdown. Empty vector and sh-scrambled were included as controls. Spheroids were counted and the numbers were statistically analyzed for significance. (* p< 0.05 and ** p< 0.01) 23 Figure 4. EED stabilizes NANOG protein post-transcriptionally. A.) Huh7 cells were transduced with scrambled shRNA and EED knockdown shRNA via lentivirus. 72 hours post transduction, 100 µM cycloheximide (CHX) was added to the cells to stop protein synthesis for 24 hours and whole cell lysates were collected from 0 to 2.5-hours post-treatment in 30-minute intervals. NANOG level at each time point was visualized by western blot analysis. B.) EED knockdown and sh-scrambled Huh7 cells were cultured, and 24 some were treated with 10 µM MG132 for 24 hours. Cells were lysed and total RNA collected by TRIzol reagent. The RNA was reverse transcribed to cDNA and subsequently used in quantitative-PCR with SYBR green qPCR master mix. Oligonucleotide primers designed for NANOG and COX6A2 were used to amplify each gene. Relative fold change was calculated by calculating ddCT of each expression value. (* P<0.05) C.D.) Sh-EED and sh-Scrambled Huh7 cells were transfected with flag-tagged NANOG vector for 48 hours. At 24 hours post- transfection, some groups were treated with 10 µM MG132. Whole cell lysates were collected and used in western blot analysis to quantify Flag-tagged NANOG, EZH2, and EED protein levels. ImageJ was used to analyzed and prepare the blot images. Relative fold change was calculated using relative intensity of target protein divided by relative intensity of untreated sh- scrambled control. E.F.G.H.) The band and corresponding background intensity was measured by ImageJ. Inverted value was calculated with the formula: Inverted intensity = 255- Intensity of Protein. Relative intensity was calculated by dividing inverted intensity of protein of interest by inverted intensity of Beta-Actin. 25 Figure 5. Co-IP western blot analysis of NANOG and its PEST domain-associated proteins. A.) Sh-EED and sh-scrambled Huh7 cells transfected with Flag-NANOG vector were treated or untreated with 10 µM MG132. The cell lysates were pre-cleared with magnetic Protein G beads then incubated with anti-flag antibody overnight. The immunocomplex was pulled down with Protein G beads followed by western blot analysis of FBXW8 and PIN1 proteins. B.) The same procedure was done as described above with Protein A beads. Western blot analysis of FBXW8, ubiquitin, and PIN1 were done to determine their association with NANOG under these conditions. C.) Sh-EED and sh-scrambled Huh7 cells co-transfected with Flag-NANOG and MYC-FBXW7 vector were treated or untreated with 10 µM MG132. Following the same immunoprecipitation procedure, the lysate was immunoblotted with anti-Myc antibody. 26 Figure 6. Site-directed mutagenesis of critical NANOG serine residues. A.) Site-directed mutagenesis of pRK5FLAG-NANOG plasmid at PEST domain Serine52, 65, 71, and 79. Specific primers were designed to mutate single site serine to alanine substitution and Flag-NANOG vector was used as template. E. Coli was transformed using the plasmids containing each serine to alanine substitute mutants. The plasmid DNA was collected using miniprep method and the resulting DNA was sent for Sanger sequencing. The sequencing results were compared to template plasmid used to determine if only the desired mutation occurred. Using NCBI BLAST Alignment tool, the query sequence was the S52A mutant plasmid DNA and subject was template Flag-NANOG plasmid. B.) Using the same method the query sequence was mutant S71A plasmid and subject was template Flag-NANOG plasmid. 27 Figure 7. Drug inhibitor screening by XTT viability assay. A.) Graphical summary of inhibitor drug screening strategy. Drug candidates from NCI drug library plates were screened for selective killing effect of CD133 + TICs. Top candidates will be screened further with cytotoxic concentration 50 (CC50) assay to determine the effective dosage. 28 Lastly, the drug candidates would be used in PDX mice models to determine efficacy in vivo. B.) Huh7 cells were sorted by CD133 magnetic beads and seeded into 96-well plates. Drug candidates were added to each well at 10 µM and incubated for 48 hours. XTT reagents were added to each well and incubated for 3 hours. Absorbance at 475 nm was recorded by plate reader and compared with DMSO treated control. C.) HepG2 cells were seeded at 5000 cells/well in 96-well plate and incubated for 18 hours. Drug candidates were added to each well at 10 µM and incubated for 48 hours. XTT reagents were added to each well and incubated for 3 hours. Absorbance at 475 nm was recorded by plate reader and compared with DMSO treated control. (* P< 0.05 and ** P< 0.01) 29 Table 2. Summary of Top Drug Candidates. Each top drug candidate is listed by their NSC reference number. The most common usage of each drug is listed along with their mechanism of action. NCBI Developmental Therapeutics Program database which includes NCI 60 cell line cytotoxicity data and in vivo mice study data. 30 31 32 33 34 35 Figure 8. National Cancer Institute’s 60 Cancer Cell Line GI50 of Top Drug Candidates. NCI conducts GI50 (50% growth inhibition concentration) for all compounds available for the Developmental Therapeutics Program. Each compound was assessed initially at a single high dose (10-5 M) in the full NCI 60 cell panel. Only compounds which satisfy pre-determined threshold inhibition criteria in a minimum number of cell lines will progress to the full 5-dose assay. Compounds which exhibit significant growth inhibition in the One-Dose Screen are evaluated against the 60-cell panel at five concentration levels. The human tumor cell lines of the cancer screening panel are grown in RPMI 1640 medium containing 5% fetal bovine serum and 2 mM L-glutamine. For a typical screening experiment, cells are inoculated into 96 well microtiter plates in 100 μL at plating densities ranging from 5,000 to 40,000 cells/well depending on the doubling time of individual cell lines. After cell inoculation, the microtiter plates are incubated at 37° C, 5 % CO2, 95 % air and 100 % relative humidity for 24 h prior to addition of experimental drugs. Following drug addition, the plates are incubated for an additional 48 h at 37°C, 5 % CO2, 95 % air, and 100 % relative humidity. Using the seven absorbance measurements [time zero, (Tz), control growth, (C), and test growth in the presence of drug at the five concentration levels (Ti)], the percentage growth is calculated at each of the drug concentrations levels. Percentage growth is calculated as: [(Ti-Tz)/(C-Tz)] x 100 for concentrations for which Ti>/=Tz [(Ti-Tz)/Tz] x 100 for concentrations for which Ti<Tz. The red box indicates cell lines with more sensitivity to the drug candidate above two-fold change. 36 Discussion In this study, we showed PRC2 core subunit EED specifically binds to NANOG PEST domain independently of other PRC2 components. Additionally, EED binding may contribute to NANOG stability and self-renewal ability in TICs. EED knockdown has shown to decrease NANOG protein level post-transcriptionally and key NANOG targeted gene silencing effects was also reversed. These effects were rescued when proteosome inhibitor MG132 was treated to EED Huh7 knockdown cells, indicating EED may stabilize NANOG through inhibiting the ubiquitin proteosome degradation pathway. In addition to increasing NANOG levels, MG132 treatment also increased protein levels of PRC2 subunits and EED knockdown was reverted shown by western blot analysis. This may be due to the inhibition degradation of remaining EED protein as well as EED stabilization from forming complexes with other subunits. Since PRC2 and its subunits have multiple functions in the cell, these compounding effects may be difficult to distinguish. One possible remedy to this MG132 rescue effect of EED is to utilize gene knockout to eliminate EED protein presence thus ensuring EED is not a factor in NANOG stability. Although our western blot analysis did not show EED knockdown affected other subunits, other groups have reported EED may be integral to PRC2 assembly, PRC1 recruitment, and its subsequent H3K27 trimethylation (Cao et al., 2014). Since PRC2 represses major cell fate and cell cycle gene expression in TICs, it would be essential to delineate changes to EED’s conventional function from knockdown or knockout. Together, many complementary experiments may be needed to delineate the role of EED in TIC self-renewal despite the well-established critical roles EED plays in both TIC and PRC2 maintenance. 37 Because of the strict regulation of NANOG protein, many factors either stabilize or contribute to degradation of NANOG. Moreover, NANOG is not stable in non-TICs which makes timing of cell lysate collection and addition of ubiquitination degradation inhibitor critical. While immunoprecipitation with protein G gives off strong background signals, protein A with additional wash steps could not capture any signals of NANOG binding partners except for FBXW8 even though some binding is expected (Figure 5). One solution to the specificity issue would be to construct overexpression Myc-tagged constructs of protein of interests and immunoprecipitate Myc-tagged protein first then immunoblot for Flag-NANOG. It is important in future analysis to include positive controls which immunoblots the immunoprecipitated Flag- tag as well as co-transfected Myc-tag construct to ensure transfection efficiency. We may also be able to increase the presence of NANOG and its binding partners by isolating CD133 + cells first before the transfection and IP-western analysis. We attempted to inhibit phosphorylation at critical serine residues via alanine substitution to examine whether upstream kinase phosphorylation of NANOG is affected by EED binding. However, the sequencing results of mutants showed no change at the target serine sites. More mutagenesis primers need to be designed to obtain a mutant with only the desired serine to alanine substitution and no off-target changes. Another challenge to this study is the many opposite and overlapping effects of NANOG phosphorylation. Many groups have reported various proteins including FBXW groups, Pin1, and other factors such as ubiquitin-specific peptidase 21 (USP21) which deubiquitinates NANOG induced by upstream phosphorylation (Liu et al., 2016). It is then essential to observe changes to each of the reported factor when serine residues are mutated. Even more, multiple serine sites may need to be substituted to induce changes to NANOG binding partners and its stability. Therefore, single and different 38 combinations of serine to alanine substitution mutants need to be constructed and IP-western of each suspected binding partners need to be assessed. Our drug screening identified 5 top candidates from NCI drug plates which displayed novel and selective cytotoxic effects against TICs. The candidates showing the most significant differential toxicity towards TICs compared to CD133 - population mainly inhibit DNA synthesis and replication. Two compounds identified from the diversity set have no previous reports on their anticancer activity which may involve unknown pathways against TICs. Lastly, NSC169780 is a known chemo-protectant used in combination with other chemotherapy treatments and has no previous reported anticancer activity. A separate viability assay using HepG2 cells of the same candidates showed no significant cytotoxic effect across the five compared to DMSO treated control (Figure 7C). HepG2 also carries wild-type tumor antigen p53 (TP53) which may contribute to its resistance to the drug candidates especially ones inhibiting DNA synthesis. Considering all these factors, it is plausible the candidates demonstrated selective cytotoxicity against CD133 + population through TIC specific mechanisms that not currently known. We included NCI 60-cell line panel growth inhibition (GI50) data of each of our top candidates (Figure 8). Notably, NSC14540 did not show any significant growth inhibition at high dosage screening of the 60 cell-line panel. This may indicate NSC14540 targets some essential and unique pathways in HCC TICs. NSC123127 is broadly cytotoxic to almost all the cancer cell lines with average GI50 of 9.5×10 -8 M but may not be TIC specific. Interestingly, all five candidates show higher cytotoxicity to cell lines linked with metastatic cancer stem-like cells. NCI-H460, COLO205, and MCF7 (NSC169780, NSC123127, NSC25154) are both enriched in CD133 + stem-like cells with upregulation of NANOG (Shi et al., 2012) (Vincent et al., 2014) 39 (Ling et al., 2012). Additionally, MOLT4 and HL60 (NSC169780, NSC123127, NSC25154, NSC8090) are leukemia cell lines with upregulated NOTCH1 and HL60 responds to all-trans retinoic acid (ATRA) which has been shown effective against HCC TICs (Wu et al., 2018) (Hou et al., 2020). Combining our viability screening with NCI GI 50 data, we have shown our drug candidates selectively target TICs via previously unreported mechanism. 40 Conclusion and Future Directions To study whether our drug candidates inhibit PRC2 and NANOG interaction we proposed to use fluorescence polarization assay screening. Fluorescence polarization (FP) is a well-established high-throughput assay which can screen for drug inhibitors of specific protein- protein interaction (Lyamu et al., 2020). We have purified NANOG protein and constructed green fluorescence (FITC) tagged EED molecule as fluorescent probe so we can quantify binding efficiency by microplate reader (Excitation 492/Emission 520). By deciding a competitive assay with drug candidates added to our target protein and EED-fluorescent probe, we can decipher whether EED and NANOG interaction is blocked by the inhibitor from changes to the fluorescence intensity emitted. In a 396-well format, we can screen through top candidates from viability assays and identify specific inhibitors disrupting PRC2-EED interaction. Potential challenges to this method are optimization including adjusting buffer background conditions, titration of fluorescent probes, and fine tuning of the detection instrument settings. Regardless of the initial investment, fluorescence polarization assay is a powerful tool that should be explored to advance our study. Sorafenib resistance is one of the main challenges to treating HCC in the clinic and contribute to the high mortality rate. To strengthen the clinical relevance of this study, we cultured Huh7 and HepG2 cells with 10 µM of sorafenib for 5 passages to select for the sorafenib resistant population. We would first characterize the enriched sorafenib resistant population by flow cytometry for TIC markers (CD133 + ) and assess the level of critical TIC proteins such as NANOG and PRC2 subunits. We will evaluate the effects of EED knockdown on NANOG stability and drug resistance in this population. Next, we would treat the cells with 41 our top inhibitor compounds to see if the compounds can sensitize them to sorafenib which may reveal a powerful combination therapy strategy. Although this study focused on the novel interaction between EED and NANOG, it is crucial to check the consequences of EED knockdown on PRC2’s canonical epigenetic repressive functions (H3K27me3) in our models. EED knockdown may destabilize PRC2 and de-repress H3K27 which can significantly change TIC characteristics and disrupt TIC maintenance via alternative pathways besides NANOG destabilization. To address these questions, we can check the protein levels of other PRC2 subunits as well as H3K27 trimethylation mark in EED KD or KO Huh7 cells by western blot analysis. Most importantly, a broader gene expression study by RT-qPCR should be conducted in EED KD/KO cells to examine any changes to the known target genes of both PRC2 and NANOG. By doing these experiments, we can further inspect how EED knockdown stabilizes NANOG and explore potential accompanying effects. Other than the current top candidates, we will continue to screen through NCI available drug plates to identify more selective agents against TICs. While screening more compounds, the already identified top hit candidates should be screened for CC50 (cytotoxic concentration) to determine the appropriate dosage for future animal studies. Following the same protocol as the viability assay described above, the candidates will be serially diluted (0.1 to 20 µM) and added to the cells to determine the concentration of 50% growth inhibition. Additionally, toxicity to normal cells should also be considered by treating the same inhibitors at the CC50 concentration to normal primary hepatocytes to ensure the agents are not cytotoxic to normal liver cells. Lastly, Patient-derived xenograft (PDX) mice model would be used to include in vivo aspects to accompany the in vitro observations. Transplantation of tumorigenic cells would 42 contain either TICs, Huh7, or sorafenib-resistant Huh7 cells at 10,000 cells mixed with Matrigel (BD Biosciences) and injected iinto flanks of immunodeficient NSG (NOD.Cg-Prkdc scid Il2rg tm1Wjl /SzJ) mice (Jackson Laboratory). After tumor development, drug candidates or combinations will be administered into the tumor and observed until endpoint. Parameters should include survival as well as tumor volume reduction of treated mice group compared to vehicle treated groups. The tumor tissues collected from the animal studies should be further used to determine expression and protein levels of TIC markers such as CD133, NANOG, and PRC2 subunits. The mechanism of PRC2 and NANOG mediated self-renewal is multidimensional due to the many pathways and proteins involved. However, we have identified key interactions of these two well-studied factors which are both upregulated in TICs. Many novel EED-specific inhibitors have been identified and shown to have therapeutic benefits in many cancer types by inhibiting PRC2 or allosterically inhibit EZH2 (Zhao et al., 2022). Some studies have shown EED inhibitors sensitizes solid tumors to immunotherapy including HCC (Hong et al., 2019). Despite these advances, there are not many studies and currently no inhibitors targeting TICs which is the crucial population and therapeutic challenge. Therefore, it is critical for our efforts of searching a novel therapy specifically targeting TICs and understanding the mechanism of NANOG induced self-renewal in TICs. 43 References 1. Ammerpohl, O., Pratschke, J., Schafmayer, C., Haake, A., Faber, W., von Kampen, O., Brosch, M., Sipos, B., von Schonfels, W., Balschun, K., et al. (2012). Distinct DNA methylation patterns in cirrhotic liver and hepatocellular carcinoma. International Journal of Cancer 130, 1319-1328. 10.1002/ijc.26136. 2. Bezuidenhout, N., and Shoshan, M. (2019). A Shifty Target: Tumor-Initiating Cells and Their Metabolism. International Journal of Molecular Sciences 20, 5370. 10.3390/ijms20215370. 3. Cabral, L.K.D., Tiribelli, C., and Sukowati, C.H.C. (2020). Sorafenib Resistance in Hepatocellular Carcinoma: The Relevance of Genetic Heterogeneity. Cancers 12, 1576. 10.3390/cancers12061576. 4. Cao, Q., Wang, X.J., Zhao, M., Yang, R.D., Malik, R., Qiao, Y.Y., Poliakov, A., Yocum, A.K., Li, Y., Chen, W., et al. (2014). The central role of EED in the orchestration of polycomb group complexes. Nature Communications 5, 3127. 10.1038/ncomms4127. 5. Chen, C.L., Kumar, D.B.U., Punj, V., Xu, J., Sher, L., Tahara, S.M., Hess, S., and Machida, K. (2016). NANOG Metabolically Reprograms Tumor-Initiating Stem-like Cells through Tumorigenic Changes in Oxidative Phosphorylation and Fatty Acid Metabolism. Cell Metabolism 23, 206-219. 10.1016/j.cmet.2015.12.004. 6. Chen, C.L., Tsukamoto, H., Liu, J.C., Kashiwabara, C., Feldman, D., Sher, L., Dooley, S., French, S.W., Mishra, L., Petrovic, L., et al. (2013). Reciprocal regulation by TLR4 and TGF-beta in tumor-initiating stem-like cells. Journal of Clinical Investigation 123, 2832- 2849. 10.1172/jci65859. 7. Chen, L., Tong, Q.Q., Chen, X.W., Jiang, P.L., Yu, H., Zhao, Q.B., Sun, L.G., Liu, C., Gu, B., Zheng, Y.P., et al. (2021). PHC1 maintains pluripotency by organizing genome-wide chromatin interactions of the Nanog locus. Nature Communications 12, 2829. 10.1038/s41467-021-22871-0. 8. Chidambaranathan-Reghupaty, S., Fisher, P.B., and Sarkar, D. (2021). Hepatocellular carcinoma (HCC): Epidemiology, etiology and molecular classification. Mechanisms and Therapy of Liver Cancer 149, 1-61. 10.1016/bs.acr.2020.10.001. 9. Chu, L.P., Qu, Y.X., An, Y., Hou, L.J., Li, J.W., Li, W.J., Fan, G.F., Song, B.L., Li, E., Zhang, L.Y., and Qi, W. (2022). Induction of senescence-associated secretory phenotype underlies the therapeutic efficacy of PRC2 inhibition in cancer. Cell Death & Disease 13, 155. 10.1038/s41419-022-04601-6. 44 10. Hong, Y.K., Li, Y., Pandit, H., Li, S.P., Pulliam, Z., Zheng, Q.Q., Yu, Y.X., and Martin, R.C.G. (2019). Epigenetic modulation enhances immunotherapy for hepatocellular carcinoma. Cellular Immunology 336, 66-74. 10.1016/j.cellimm.2018.12.010. 11. Hong, Y.K., Li, Y., Pandit, H., Li, S.P., Pulliam, Z., Zheng, Q.Q., Yu, Y.X., and Martin, R.C.G. (2019). Epigenetic modulation enhances immunotherapy for hepatocellular carcinoma. Cellular Immunology 336, 66-74. 10.1016/j.cellimm.2018.12.010. 12. Hou, W., Wang, Z.Y., Lin, J., and Chen, W.M. (2020). Induction of differentiation of the acute myeloid leukemia cell line (HL-60) by a securinine dimer. Cell Death Discovery 6, 123. 10.1038/s41420-020-00354-3. 13. Iyamu, I.D., and Huang, R. (2020). Development of fluorescence polarization-based competition assay for nicotinamide N-methyltransferase. Analytical Biochemistry 604, 113833. 10.1016/j.ab.2020.113833. 14. Kasinath, V., Beck, C., Sauer, P., Poepsel, S., Kosmatka, J., Faini, M., Toso, D., Aebersold, R., and Nogales, E. (2021). JARID2 and AEBP2 regulate PRC2 in the presence of H2AK119ub1 and other histone modifications. Science 371, 362-+, eabc3393. 10.1126/science.abc3393. 15. Kim, S.H., Kim, M.O., Cho, Y.Y., Yao, K., Kim, D.J., Jeong, C.H., Yu, D.H., Bae, K.B., Cho, E.J., Jung, S.K., et al. (2014). ERK1 phosphorylates Nanog to regulate protein stability and stem cell self-renewal. Stem Cell Research 13, 1-11. 10.1016/j.scr.2014.04.001. 16. Kuciak, M., Mas, C., Borges, I., Sanchez-Gomez, P., and Altaba, A.R.I. (2019). Chimeric NANOG repressors inhibit glioblastoma growth in vivo in a context-dependent manner. Scientific Reports 9, 3891. 10.1038/s41598-019-39473-y. 17. Kusakabe, Y., Chiba, T., Oshima, M., Koide, S., Rizq, O., Aoyama, K., Ao, J.J., Kaneko, T., Kanzaki, H., Kanayama, K., et al. (2021). EZH1/2 inhibition augments the anti-tumor effects of sorafenib in hepatocellular carcinoma. Scientific Reports 11, 21396. 10.1038/s41598-021-00889-0. 18. Ling, G.Q., Chen, D.B., Wang, B.Q., and Zhang, L.S. (2012). Expression of the pluripotency markers Oct3/4, Nanog and Sox2 in human breast cancer cell lines. Oncology Letters 4, 1264-1268. 10.3892/ol.2012.916. 19. Liu, M.M., Zhang, L., Li, H.T., Hinoue, T., Zhou, W.D., Ohtani, H., El-Khoueiry, A., Daniels, J., O'Connell, C., Dorff, T.B., et al. (2018). Integrative Epigenetic Analysis Reveals Therapeutic Targets to the DNA Methyltransferase Inhibitor Guadecitabine (SGI-110) in Hepatocellular Carcinoma. Hepatology 68, 1412-1428. 10.1002/hep.30091. 20. Liu, X.Y., Yao, Y.Y., Ding, H.G., Han, C.C., Chen, Y.H., Zhang, Y., Wang, C.J., Zhang, X., Zhang, Y.L., Zhai, Y., et al. (2016). USP21 deubiquitylates Nanog to regulate protein 45 stability and stem cell pluripotency. Signal Transduction and Targeted Therapy 1, e16024. 10.1038/sigtrans.2016.24. 21. Matsushita, H., and Takaki, A. (2019). Alcohol and hepatocellular carcinoma. Bmj Open Gastroenterology 6, e000260. 10.1136/bmjgast-2018-000260. 22. Moretto-Zita, M., Jin, H., Shen, Z.X., Zhao, T.B., Briggs, S.P., and Xu, Y. (2010). Phosphorylation stabilizes Nanog by promoting its interaction with Pin1. Proceedings of the National Academy of Sciences of the United States of America 107, 13312-13317. 10.1073/pnas.1005847107. 23. Moussa, I., Day, R.S., Li, R.S., Du, X.L.L., Kaseb, A.O., Jalal, P.K., Daniel-MacDougall, C., Hatia, R.I., Abdelhakeem, A., Rashid, A., et al. (2021). Dietary Patterns and Hepatocellular Carcinoma Risk among US Adults. Nutrients 13, 2011. 10.3390/nu13062011. 24. Natsume, A., Ito, M., Katsushima, K., Ohka, F., Hatanaka, A., Shinjo, K., Sato, S., Takahashi, S., Ishikawa, Y., Takeuchi, I., et al. (2013). Chromatin Regulator PRC2 Is a Key Regulator of Epigenetic Plasticity in Glioblastoma. Cancer Research 73, 4559-4570. 10.1158/0008-5472.can-13-0109. 25. Nishimura, N., Kaji, K., Kitagawa, K., Sawada, Y., Furukawa, M., Ozutsumi, T., Fujinaga, Y., Tsuji, Y., Takaya, H., Kawaratani, H., et al. (2021). Intestinal Permeability Is a Mechanical Rheostat in the Pathogenesis of Liver Cirrhosis. International Journal of Molecular Sciences 22, 6921. 10.3390/ijms22136921. 26. Prevo, R., Pirovano, G., Puliyadi, R., Herbert, K.J., Rodriguez-Berriguete, G., O'Docherty, A., Greaves, W., McKenna, W.G., and Higgins, G.S. (2018). CDK1 inhibition sensitizes normal cells to DNA damage in a cell cycle dependent manner. Cell Cycle 17, 1513- 1523. 10.1080/15384101.2018.1491236. 27. Ramakrishna, S., Suresh, B., Lim, K.H., Cha, B.H., Lee, S.H., Kim, K.S., and Baek, K.H. (2011). PEST Motif Sequence Regulating Human NANOG for Proteasomal Degradation. Stem Cells and Development 20, 1512-1520. 10.1089/scd.2010.0410. 28. Reya, T., Morrison, S.J., Clarke, M.F., and Weissman, I.L. (2001). Stem cells, cancer, and cancer stem cells. Nature 414, 105-111. 10.1038/35102167. 29. Shi, Y., Fu, X.L., Hua, Y., Han, Y., Lu, Y., and Wang, J.C. (2012). The Side Population in Human Lung Cancer Cell Line NCI-H460 Is Enriched in Stem-Like Cancer Cells. Plos One 7, e33358. 10.1371/journal.pone.0033358. 30. Shi, Y., Wang, X.X., Zhuang, Y.W., Jiang, Y., Melcher, K., and Xu, H.E. (2017). Structure of the PRC2 complex and application to drug discovery. Acta Pharmacologica Sinica 38, 963-976. 10.1038/aps.2017.7. 46 31. Sung, H., Ferlay, J., Siegel, R.L., Laversanne, M., Soerjomataram, I., Jemal, A., and Bray, F. (2021). Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. Ca-a Cancer Journal for Clinicians 71, 209- 249. 10.3322/caac.21660. 32. Takeishi, S., and Nakayama, K.I. (2014). Role of Fbxw7 in the maintenance of normal stem cells and cancer-initiating cells. British Journal of Cancer 111, 1054-1059. 10.1038/bjc.2014.259. 33. Vassilev, L.T., Tovar, C., Chen, S.Q., Knezevic, D., Zhao, X.L., Sun, H.M., Heimbrook, D.C., and Chen, L. (2006). Selective small-molecule inhibitor reveals critical mitotic functions of human CDK1. Proceedings of the National Academy of Sciences of the United States of America 103, 10660-10665. 10.1073/pnas.0600447103. 34. Yang, Y., Niu, C.S., and Cheng, C.D. (2013). Pin1-Nanog expression in human glioma is correlated with advanced tumor progression. Oncology Reports 30, 560-566. 10.3892/or.2013.2481. 35. Yu, J.R., Lee, C.H., Oksuz, O., Stafford, J.M., and Reinberg, D. (2019). PRC2 is high maintenance. Genes & Development 33, 903-935. 10.1101/gad.325050.119. 36. Zhao, Y., Guan, Y.Y., Zhao, F., Yu, T., Zhang, S.J., Zhang, Y.Z., Duan, Y.C., and Zhou, X.L. (2022). Recent strategies targeting Embryonic Ectoderm Development (EED) for cancer therapy: Allosteric inhibitors, PPI inhibitors, and PROTACs. European Journal of Medicinal Chemistry 231, 114144. 10.1016/j.ejmech.2022.114144.

Abstract (if available)

Abstract Tumor-initiating stem like cells (TICs) contribute to chemoresistance and tumor relapses in hepatocellular carcinoma (HCC). Our group has previously reported the toll-like receptor 4 induced upregulation of stemness factor NANOG which generated CD133+ TICs in alcohol and high fat diet mice. Additionally, polycomb repressive complex 2 (PRC2) which has long been reported to be involved in cancer stem-like cell formation via histone 3 lysine 27 trimethylation was found to be associated with many NANOG target gene loci. This led to our question of whether PRC2 and NANOG cooperate to maintain TIC self-renewal and other phenotypes such as chemoresistance.
In this study, we discovered a unique binding of PRC2 core subunit, embryonic ectoderm development (EED), to NANOG N-terminus PEST domain which is essential to NANOG degradation through ubiquitin proteasome pathway. We further investigated the effects of EED knockdown to NANOG protein stability as well as TIC maintenance. Mechanistically, we examined known factors to NANOG degradation pathway and how EED may contribute to overabundance of NANOG in TICs. Lastly, we utilized cell viability assay screening to identify potential inhibitors of EED-NANOG interaction by examining the drug’s ability to selectively target TICs. Understanding the mechanism of TIC maintenance may present a novel therapeutic target and strategy to treat metastatic HCC. Altogether, this study showed a novel interaction between two known oncogenic proteins which contribute to cancer stem-like cells and identified new drugs that are selectively cytotoxic to TICs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Hepatitis B virus X protein regulation of β-catenin and NANOG and co-regulatory role with YAP1 in HCC malignancy

PDF

Tri-specific T cell engager immunotherapy targeting tumor initiating cells

PDF

Roles of epithelial-mesenchymal transition and niche in tumorigenesis of tumor-initiating cells

PDF

Investigating the effects of polycomb repressive complex inhibitors on cancer cell phenotypic plasticity

PDF

Targeting tumor-initiating stem-like cells through metabolic and epigenetic regulations

PDF

Targeting mitochondrion-nucleus PDH1 transfer to suppress self-renewal and epigenetic NANOG reprogramming of tumor-initiating cells

PDF

Tri-specific T cell engager immunotherapy targeting tumor initiating cells

PDF

Mechanisms of nuclear translocation of pyruvate dehydrogenase and its effects on tumorigenesis of hepatocellular carcinoma…

PDF

Identification of molecular mechanism for cell-fate decision in liver; &, SARS-CoV replicon inhibitor high throughput drug screening

PDF

Novel long non-coding RNA, LINC00824, contributes to stemness in hepatocytes

PDF

Musashi-2 promotes c-MYC expression through IRES-dependent translation and self-renewal ability in hepatocellular carcinoma

PDF

Molecular role of EZH2 overexpression in Colorectal Cancer progression

PDF

Hepatic differentiation in human naïve stem cells compared to human embryonic stem cells

PDF

Cell fate reprogramming of liver tumor-initiating stem-like cells via phosphorylated NUMB and TBC1D15

PDF

Steatosis induces Wnt/β-catenin pathway to stimulate proliferation of hepatic tumor initiating cells and promote liver cancer development

PDF

Developing a bioluminescence tracking system targeting on tumor cells and T cells for cancer immunotherapy

PDF

Capture and analysis of circulating tumor cells in patients with hepatocellular carcinoma: analysis of a pilot study

PDF

AKT exhibits isoform-specific function in hepatocyte transformation and response to cellular stress

PDF

SARS-CoV-2 suppression of CD1d expression and NKT cell function

PDF

Development of TCR-like antibody and novel chimeric antigen receptor for cancer immunotherapy

Asset Metadata

Creator Liu, Cheng (author)

Core Title Polycomb repressive complex 2 subunit stabilizes NANOG to maintain self-renewal in hepatocellular carcinoma tumor-initiating stem-like cells

School Keck School of Medicine

Degree Master of Science

Degree Program Molecular Microbiology and Immunology

Degree Conferral Date 2023-05

Publication Date 03/30/2023

Defense Date 03/30/2023

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag cancer stem cell,EED,HCC,hepatocellular carcinoma,NANOG,OAI-PMH Harvest,PRC2,self-renewal,TIC,tumor initiating cells,tumor initiating stem-like cells

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Machida, Keigo (committee chair), Bell, Oliver (committee member), Yuan, Weiming (committee member)

Creator Email cliu9284@usc.edu,jamescliu785@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC112857147

Unique identifier UC112857147

Identifier etd-LiuCheng-11542.pdf (filename)

Legacy Identifier etd-LiuCheng-11542

Document Type Thesis

Format theses (aat)

Rights Liu, Cheng

Internet Media Type application/pdf

Type texts

Source 20230331-usctheses-batch-1014 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu