Napabucasin overcomes cisplatin resistance in ovarian germ cell tumor-derived cell line by inhibiting cancer stemness

Background Cisplatin resistance of ovarian yolk sac tumors (oYST) is a clinical challenge due to dismal patient prognosis, even though the disease is extremely rare. We investigated potential association between cisplatin resistance and cancer stem cell (CSC) markers in chemoresistant oYST cells and targeting strategies to overcome resistance in oYST. Methods Chemoresistant cells were derived from chemosensitive human oYST cells by cultivation in cisplatin in vitro. Derivative cells were characterized by chemoresistance, functional assays, flow cytometry, gene expression and protein arrays focused on CSC markers. RNAseq, methylation and microRNA profiling were performed. Quail chorioallantoic membranes (CAM) with implanted oYST cells were used to analyze the micro-tumor extent and interconnection with the CAM. Tumorigenicity in vivo was determined on immunodeficient mouse model. Chemoresistant cells were treated by inhibitors intefering with the CSC properties to examine the chemosensitization to cisplatin. Results Long-term cisplatin exposure resulted in seven-fold higher IC50 value in resistant cells, cross-resistance to oxaliplatin and carboplatin, and increased migratory capacity, invasiveness and tumorigenicity, associated with hypomethylation of differentially methylated genes/promotors. Resistant cells exhibited increased expression of prominin-1 (CD133), ATP binding cassette subfamily G member 2 (ABCG2), aldehyde dehydrogenase 3 isoform A1 (ALDH3A1), correlating with reduced gene and promoter methylation, as well as increased expression of ALDH1A3 and higher overall ALDH enzymatic activity, rendering them cross-resistant to DEAB, disulfiram and napabucasin. Salinomycin and tunicamycin were significantly more toxic to resistant cells. Pretreatment with napabucasin resensitized the cells to cisplatin and reduced their tumorigenicity in vivo. Conclusions The novel chemoresistant cells represent unique model of refractory oYST. CSC markers are associated with cisplatin resistance being possible targets in chemorefractory oYST.


Methylation profiling
Generation of methylation profiles of cell line genomic DNA isolated using ethanol precipitation was performed at the Erasmus MC Department of Pathology molecular diagnostics lab according to the Illumina protocols (EPIC). Copy number alterations were resolved using the Conumee package (Hovestadt V, Zapatka M. conumee: Enhanced copynumber variation analysis using Illumina DNA methylation arrays. R package version 1.6.0. http://bioconductororg/packages/conumee/. 2015). Differential methylation was identified using the RnBeads package (https://rnbeads.org) using "SWAN" for normalization [1].

miRNA profiling
Total RNA was prepared using Trizol (Thermo Fisher, USA). miRNAs were converted into cDNA using the specific megaplex primers (ThermoFisher, PN: 4399966) and the reverse transcription kit (ThermoFisher, PN: 3466596) and quantitated on TaqMan Low Density Arrays (384-well Microfluids TLDA card A, ThermoFisher, PN: 4398965) on a TaqMan 7900HT Fast Real-Time PCR Machine using the supplier protocols (ThermoFisher, PN: 4399721). TaqMan miRNA array output data (sds-files) were uploaded in the ThermoFisher Cloud App (https://www.thermofisher.com/nl/en/home/digital-science/thermo-fisher-connect/all-analysis-modules.html) and analyzed using defined threshold settings for each individual miRNA. Cq values were exported, and globally normalized in Excel.

RNAseq analysis
Total RNA was prepared using Trizol, DNase-treated (RNeasy Micro Kit, Qiagen, Germany), and quality verified using fragment analysis. The NEBNext Ultra Directional RNA Library Prep Kit for Illumina was used to process the samples. The sample preparation was performed according to the protocol "NEBNext Ultra Directional RNA Library Prep Kit for Illumina" (NEB #E7420). Briefly, rRNA was depleted from total RNA using the rRNA depletion kit (NEB# E6310). After fragmentation of the rRNA reduced RNA, a cDNA synthesis was performed. This was used for ligation with the sequencing adapters and PCR amplification of the resulting product. Clustering and DNA sequencing using the Illumina cBot and HiSeq 4000 was performed according to manufacturer's protocols. The experiments were performed at GenomeScan B.V., Plesmanlaan 1d, 2333 BZ, Leiden. Processing of RNA-seq data was performed using UCSC human genome build hg38 and GENCODE annotation release 28 (GRCh38). FASTQC (v0.11.5) [2] was applied on the paired-end FASTQ files for quality control, both before and after running trimmomatic (v0.36) [3], which removed TrueSeq adapter sequences. STAR (v2.5.3a) [4] was used as aligner, with 2-pass mapping for each sample separately. Mapping quality plot was generated and checked based on sambamba Flagstat (v0.6.7) statistics [5]. Count files, with the number of reads for each gene were created with subread FeatureCounts (v1.5.2) [6].

Migration assay
Thirty thousand of NOY-1 and NOY-1 CisR cells per well were plated in quadruplicates in ImageLock 96-well plates (Essen BioScience, UK) and let to adhere overnight. Confluent monolayers were wounded with wound making tool (Essen BioScience, UK), washed twice and supplemented with culture medium. Images were taken every two hours for next 48 hours in the IncuCyte ZOOM™ Kinetic Imaging System (Essen BioScience, UK). Cell migration was evaluated by IncuCyte ZOOM™ 2013A software (Essen BioScience, UK) based on the relative wound density measurements and expressed as means of three independent experiments run in quadruplicates ± SD.

Clonogenic assay
NOY-1 CisR cells were seeded (500 cells per well in 96-well culture plates) and let to adhere in standard culture medium (control) or in medium containing 0.06 μg/ml napabucasin. After 24 hours, cisplatin was added to the cells, and medium-only in control wells. Plates were incubated for 6 days, after which cells were washed and fresh medium was added for another 4 days. Then the cells were stained with May-Grünwald solution.