Hepatocellular carcinoma (HCC) is the main form (90%) of primary liver cancer (1), which is the fourth most common cause of cancer-related death worldwide. Cancer stem cell (CSC) model has been attributed to create tumor heterogeneity, in which a sub-set of tumor cells capable of self-renewal and extensive proliferation result in high recurrence rates and chemoresistance acquisition (2). CSC-induced tumor heterogeneity is associated with poor prognosis in several cancers (3) and thus CSC-specific markers (extracellular and intracellular) have been reported as crucial therapeutic targets (4) to selectively eradicate cells with tumorigenic capabilities. As CSCs share their key properties such as self-renewal and proliferation with normal stem cells, there is an unsurprising large overlap between CSC and normal stem cell surface markers such as
ROR, HOTAIR and UCA1 are some well-established CSC-regulating long non-coding RNAs (lncRNAs) through various molecular mechanisms involving proliferation, self-renewal and metastasis promotion (8). lncRNAs are ideal target factors as their regulation span epigenetic, post-transcriptional and transcriptional levels (9) with reported function in various cancers (10) and cancer-type specific expression (11). Pseudogenes which were first defined as ‘junk’ genomic loci, have recently undergone resuscitation as a novel lncRNA class, namely pseudogene-derived lncRNAs with high relevance in cancer initiation and progression (12). Pseudogene-lncRNAs such as
In our study, we propose a user-friendly method of using public microarray data to identify lncRNA stemness-related factors in liver cancer, followed by preliminary experimental validation. Our candidate
Microarray datasets (GSE50206, GSE25097) were utilized for stemness-related candidate selection. HCC samples from GSE25097 were selected based on elevated expression of various stemness markers (
FANTOM CAT browser was accessed to define transcription start site (TSS) of
Fractionated RT-qPCR with
Outstanding issues despite the establishment of CSC model are i.) refining isolation and identification of CSCs through extracellular/intracellular factors and ii.) unraveling the complex mechanism of how such factors dysregulate self-renewal and extensive proliferation in CSCs. Extracellular markers such as
Here, we demonstrate the selection of stemness-related factors by identifying up-regulated genes in ESC and HCC samples respectively in microarray data via GEO2R. We reasoned that overlapping up-regulated genes from these two groups would shortlist factors conferring stemness properties by permitting continuous self-renewal and proliferative capabilities. Indeed, GO analysis of the shortlisted 139 factors displayed involvement in tumor-promoting signaling pathways and tumor cell growth. Moreover, previously characterized tumorigenesis-driving transcription factors such as
LncRNAs hold rich potential as therapeutic targets and prognostic markers as numerous studies indicate their contribution to self-renewal and proliferation of HCC CSCs – CUDR promotes malignant proliferation through multiple signaling pathways (22) while HOTAIR suppresses SETD2 in HCC CSCs (23). Thus, we selected the pseudogene-derived lncRNA
In conclusion, we have identified a stemness-related factor
GSE25097 microarray expression data was downloaded from HCCDB (28) to select HCC samples that displayed combinatorial elevated expression (> HCC average) of cancer stemness markers
Mean differential expression of selected genes (
Processed RNA-seq V2 for LIHC (HCC) was downloaded from public TCGA data portal, and the dataset contains 422 samples (50 normal samples, 372 tumor samples). Fragments Per Kilobase of exon per Million (FPKM) reads were used.
TIE scores for genomic region spanning from chromosome 6:27324485 to chromosome 6:27440849 (
Association between miRNAs and target genes (
GSEA software (version 4.1.0) was downloaded available at http://www.broadinstitute.org/gsea/downloads.jsp and expression data sets (.TXT format), phenotype labels (. CLS format) and gene sets (. GMX format) were created in accordance to GSEA specifications. Signature gene sets “Chiang Liver Cancer Subclass Proliferation Up” and “GOBP Regulation of Stem Cell Population Maintenance” were downloaded from Molecular Signatures Database (MSigDB v7.4) (34). Input for basic fields were: Enrichment statistic – weighted, Metric for ranking genes – Difference of class, Number of permutations – 1,000.
sgRNAs targeting the regions encompassing transcription start site (+/− 200 relative to TSS) of
HCC cell-line HepG2 was maintained in DMEM (Invitrogen, USA) supplemented with fetal bovine serum 10% (Cytiva, SH30084.03), 1% penicillin-streptomycin (Gibco, #15140122) in 37°C in 5% CO2 humidified incubator. Cells were passaged every 2-3 days and harvested using 0.25% Trypsin-EDTA (Invitrogen, #25200-056). For knockdown studies, 1 × 105 cells were seeded per well in a 24-well plate one day prior to transfection. sgRNA-encoding LRG (600 ng) and modified dCas9-KRAB-MeCP2 (300 ng) were mixed with 2 μl of Lipofectamine 3000 (Thermo Fisher Scientific, # L3000015) and incubated at room temperature for 5 minutes. Transfection mixture was added to each well, and cells were harvested for downstream analysis at 24th and 48th hour time-points.
100 μl of cell media from each well (duplicates) in 24-well plate was collected and transferred to a 96-well plate at the 24th and 48th hour post-perturbation. 10 μl of CCK-8 solution (Dojindo, CK04-13) was added to each well, and incubated for 2 hours at 37°C, 5% CO2. The absorbance was read at 450 nm using TECAN microplate reader (Infinite 200 PRO), and the average value of duplicates were taken for reading of each group.
100 μl of cell suspension (DMEM-1% penicillin/streptomycin with no serum) containing 5 × 103 cells subjected to CRISPRi for 24 hours were dispensed into the Transwell insert (Corning, 3422) and incubated for 10 minutes at 37°C and 5% CO2. 600 μl of media solution (DMEM-1% penicillin/streptomycin with fetal bovine serum) was added to the bottom of lower chamber. 48 hours post-incubation, inserts were removed from plate for further analysis. For image taking, remaining cells from the membrane top were wiped off using a cotton-tipped applicator. Inserts were subsequently fixed in ice-cold 100% methanol for 15 minutes, washed in PBS and stained with 1% crystal violet solution (Sigma Aldrich) for 30 minutes at room temperature. After rinsing in water, inserts were left to dry, and images were taken at 10× magnification using Olympus IX51 light microscope. For crystal violet quantification, cut-out insert membranes were solubilized in 250 μl of 1% sodium dodecyl sulfate solution for 10 minutes at room temperature. 50 μl (in triplicates) were transferred to 96-well plate for OD reading at 595 nm.
The same steps were repeated for invasion assays, with the exception of using inserts pre-coated with Matrigel (Corning, 354234) at diluted concentration of 200 ug/ml.
2 × 105 cells were seeded per well in 24-well plates pre-coated with 0.01% Poly-L-lysine (Sigma, P2636-25MG) one day prior to transfection. ‘0 hour’ time-point was established at 6 hours post-transfection, and each wound was made using a 200 ul pipette tip. Images were taken at 10x magnification using Olympus IX51 light microscope at stated time-points. Wound area was quantified through ImageJ available at https://imagej.nih.gov/ij/download.html using the polygon selection mode.
Transfected 2 × 103 cells per well were seeded in 6-well plates, and incubated for 10-14 days. All groups underwent PBS wash, 30 minute fixation with ice-cold 100% methanol, staining (0.5% crystal violet solution) for 2 hours at room temperature, and air-drying for at least 2 hours following washes with tap water. Images were taken at 10x magnification using Olympus IX51 light microscope at stated time points.
0.5-1 μg of RNA was used in synthesis of 1st strand comple-mentary DNA (cDNA) using Superscript II reverse transcriptase (Thermo Fisher Scientific, #18064014) and oligo (dT) primers (Macrogen). The resulting cDNA was diluted ×10 in nuclease-free water and stored at −20°C. Quantitative PCR (qPCR) was carried out in triplicates using KAPA SYBRⓇFAST Bio-Rad Icycler 2X Qpcr Master Mix (Roche, KK4608). For RNA fractionation, 2 × 106 HepG2 cells were harvested for fractionated RNA isolation using PARIS kit (Ambion, AM1921), as per manufacturer’s protocol. All qPCR primer sequences are included in Supplementary Table 3.
This work was supported by the Collaborative Genome Program for Fostering New Post-Genome Industry through the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT (NRF-2016M3C9A4921712) and by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education (NRF-2016R1D1A1B01015292).
J-H.H and J.L performed the experiments. JW.L and K.H.B.C analyzed the data. W.Y.C and M.J.K contributed to data interpretation and manuscript writing. J-H.H, L.K.K and Y-J.K conceived the study and wrote the manuscript. All authors read and approved the final manuscript.
The authors have no conflicting interests.