Generation of Alb-Cre;CAG-Myc (ABC-Myc) mice, and Alb-Cre;CAG-Myc;TdTomato (ABC-Myc; TdTomato) mice
Albumin-Cre (Alb-Cre) (Strain #003574), R26StopFLMYC (CAG-MYC) (Strain #020458), and CAG-tdTomato (Strain #007914) mice were obtained from the Jackson Laboratory. ABC-Myc (Alb-Cre+/wt::CAG-MYCmyc/wt) mouse model was generated by crossbreeding Alb-Cre+/+ with CAG-MYCmyc/myc mouse, or Alb-Cre+/wt with CAG-MYCmyc/myc, or Alb-Cre+/wt with CAG-MYCmyc/wt. The littermates with genotypes of Alb-Crewt/wt::CAG-Mycmyc/wt, or Alb-Cre+/wt::CAG-MYCwt/wt, or Alb-Crewt/wt::CAG-MYCwt/wt were served as normal controls. In order to generate ABC-MYC;TdTomato mice, Alb-Cre+/+ mice were first bred with CAG-Tdtomatotd/td mice to obtain the mice with genotypes of Alb-cre+/wt::CAG-tdTomatotd/wt, which were then bred with CAG-Mycmyc/myc mice. For genotyping, the genomic DNA was extracted from tail biopsies, and PCR amplification assay was performed using KAPA Mouse Genotyping Kits (Roche Corporate, Cat#KK7352) according to The Jackson Laboratory genotyping PCR conditions for each mice strain. The primers 5’-TGC AAA CAT CAC ATG CAC AC, GAA GCA GAA GCT TAG GAA GAT GG-3’ and 5’-TTG GCC CCT TAC CAT AAC TG-3’ were used for Alb-Cre genotyping (AlbCre = 390bp and WT = 351bp). The primers 5’-CCA AAG TCG CTC TGA GTT GTT ATC-3’, 5’-GAG CGG GAG AAA TGG ATA TG, CCA AGA GGG TCA AGT TGG A-3’ and 5’-GCA ATA TGG TGG AAA ATA AC-3’ are used for CAG-Myc genotyping (MYC = 550bp and WT = 604bp). The primers 5’-AAG GGA GCT GCA GTG GAG TA, CCG AAA ATC TGT GGG AAG TC-3’, 5’-CTG TTC CTG TAC GGC ATG G-3’ and 5’-GGC ATT AAA GCA GCG TAT CC-3’ were used for CAG-tdTomato genotyping (tdTomato =196 bp and WT = 297bp). The genotyping PCR products were resolved in 2% agarose gel (Invitrogen, Cat#16500-500) and imaged with Alphaimager HP (ProteinSimple, Alphaimager HP) or Li-COR D-Digit (Li-COR, 3500). Mice were housed with temperature and 12h light /12h dark cycle controlled under specific-pathogen-free conditions (SPF) at the St Jude Children’s Research Hospital mouse facility. All experiments that involved the use of mice were performed in accordance with the guidelines outlined by the St Jude Children’s Research Hospital Institutional Animal Care and Use Committee (IACUC).
Generation of ABC-Myc-derived hepatoblastoma cell lines (NEJF1, NEJF2, NEJF4, NEJF5, NEJF6)
The livers from ABC-Myc mice were excised and placed in a sterile tube containing phosphate-buffered saline (PBS) on wet ice during transport from the animal research facility to the research laboratory. Tumor nodules were excised using a sterile scalpel and underwent an enzymatic digestion with collagenase IV (2 mg/ml; in 25 ml of RPMI medium) for 1 hour in a 37°C rotor (Robbins Scientific Corporation, model 2000). After digestion, cells were filtered using a 70-μm sterile strainer and cultured in ultra-nonadherent cell culture plate with DMEM medium with 10% FBS and 1% penicillin and streptomycin. Liver cancer cells form spheroids and are propagated in ultra-nonadherent cell culture plate. In parallel, the spheroids were transferred to adherent plates in standard DMEM media and adherent cell lines were derived. Notably, the adherent cell lines form spheroids when culturing in non-adherent cell culture plate.
Generation of NEJF10 from the ABC-Myc;TdTomato tumor
The liver from ABC-Myc; TdTomato mouse (#NEJF10) was excised and placed in a sterile tube containing cold phosphate-buffered saline (PBS). Tumor nodules were excised using a sterile scalpel and underwent an enzymatic digestion with collagenase IV (2 mg/ml; in 25 ml of DMEM medium) for 1 hour in a 37°C rotor (Robbins Scientific Corporation, model 2000). After digestion, cells were filtered using a 70-μm sterile strainer and cultured in 15cm culture dish (Fisher Scientific, FB012925) with DMEM medium with 10% FBS and 1% penicillin and streptomycin. Next day, the NEJF10 hepatoblastoma cells formed spheroids with tdTomato red color under EVOS M7000 Imaging System (Invitrogen, EVOS M7000). The spheroids were transferred and cultured in ultra-low attachment microplates for propagation (Corning, Cat#3471). The NEJF10 spheroids were transferred to adherent plates and the adherent NEJF10 cells were consequently derived. The adherent NEJF10 cells also form spheroids when cells were cultured back to ultra-low attachment microplates.
Human cell lines, reagents, and validation
HepG2 (ATCC, HB-8065) cells were cultured in 1X DMEM (Fisher Scientific, Cat#MT10013CM) supplemented with 10% FBS (Gibco, Cat#10437028), 1% Penicillin-Streptomycin solution (Gibco, Cat#15140122)) at 37 °C in 5% CO2 in a humidified incubator. All human-derived cell lines were validated by short tandem repeat (STR) profiling using PowerPlex® 16 HS System (Promega) once a month. Additionally, a polymerase chain reaction (PCR)-based method was used to screen for mycoplasma once a month employing the LookOut® Mycoplasma PCR Detection Kit (MP0035, Sigma-Aldrich) and JumpStart™ Taq DNA Polymerase (D9307, Sigma-Aldrich) to ensure cells were free of mycoplasma contamination.
Doxorubicin, Samuraciclib (ICEC0942), and EPZ015666 (GSK3235025) were purchased from Selleckchem. All other compounds used for screening were obtained from St Jude compounds deposit. AZD7648, Lot01, was purchased from Chemietek, and the quality was verified by Chemieteck by HPLC-MS nad NMR, with purity >99.5%. The purity of AZD7648 was further verified in house by using Waters UPLC-MS system (Acquity PDA detector, SQ detector and UPLC BEH-C18 column). The mass spectrometer was acquired using MassLynx v. 4.1. The chromatographic conditions are as follows: flow rate: 1.0 mL/min, sample injection volume: 2 µL, column temperature: 55 °C, mobile phase: 0.1% formic acid in CH3CN and H2O.
Pathological assessment of ABC-Myc hepatoblastomas
Liver tumors were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 4 μm, mounted on positive charged glass slides (Superfrost Plus; 12-550-15, Thermo Fisher Scientific, Waltham, MA) that were dried at 60°C for 20 minutes, and stained with hematoxylin and eosin (HE). The following immunohistochemistry protocols were used for the detection of AFP, ARG1, Beta-catenin, GS, and KRT19, respectively, on commercial autostainers: 1) anti-Alpha-1-fetoprotein, A0008, Agilent, 1:300, 32’ incubation. Heat-induced epitope retrieval, Cell conditioning media 2 (Ventana Medical Systems, Tucson, AZ), 32 minutes; Visualization with DISCOVERY OmniMap anti-Rb HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). 2) anti-Arginase-1 (H-52), sc-20150, Santa Cruz, 1:75, 60’ incubation. Heat-induced epitope retrieval, Cell conditioning media 2 (Ventana Medical Systems, Tucson, AZ), 48 minutes; Visualization with DISCOVERY OmniMap anti-Rb HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). 3) anti-Beta-catenin (Clone E247), RM-2101, ThermoFisher, 1:300, 60’ incubation, Heat-induced epitope retrieval, Cell conditioning media 1 (Ventana Medical Systems, Tucson, AZ), 48 minutes. Visualization with DISCOVERY OmniMap anti-Rb HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). 4) anti-Glutamine synthase, ab73593, Abcam, 1:1000, 60’ incubation, Heat-induced epitope retrieval, Cell conditioning media 1 (Ventana Medical Systems, Tucson, AZ), 32 minutes. Visualization with DISCOVERY OmniMap anti-Rb HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). 5) anti-Keratin19, TROMA-III, Developmental Studies Hybridoma bank, 1:1000, 15’ incubation. Heat-induced epitope retrieval, Epitope Retrieval solution 1 (ER2), 20 minutes. Visualization with rabbit anti-rat (712-4126; Rockland), Bond Polymer Refine Detection (DS9800, Leica Biosystems). Additionally CD3, Galectin-2 (Mac-2), and VEGF were detected by IHC using the following autostainer protocols: 1) anti-CD3, sc-1127, SantaCruz, 1:1000, 32’ incubation, heat-induced epitope retrieval with cell conditioning media 1 (Ventana Medical Systems, Tucson, AZ), 32 minutes; Visualization with Rabbit anti-goat IgG antibody (Vector labs, BA-5000) and DISCOVERY OmniMap anti-Rb HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). 2) anti-Galectin-2 (Mac-2), M3/38, ACL8942AP, Accurate Chemical and Scientific Corporation, 1:1000, 32’ incubation, heat-induced epitope retrieval with cell conditioning media 1 (Ventana Medical Systems, Tucson, AZ), 32 minutes; Visualization with DISCOVERY OmniMap anti-Rt HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). 3) anti-VEGF, EP1176Y, NB110-57642, Novus, 1:100, 32’ incubation, heat-induced epitope retrieval with cell conditioning media 1 (Ventana Medical Systems, Tucson, AZ), 32 minutes; Visualization with DISCOVERY OmniMap anti-Rb HRP (760-4311; Ventana Medical Systems), DISCOVERY ChromoMap DAB kit (760-159; Ventana Medical Systems). All HEs and IHCs were reviewed by light microscopy and interpreted by a board-certified veterinary pathologist (HT).
Clinical chemistry analysis
Once whole blood samples for chemistries are received in the Diagnostic Lab at St Jude, they are allowed to clot for 30 minutes, at which point the clot is removed and the serum separated by centrifuging at 5700 rpm for 10 minutes. Once separated, the serum is pipetted into a Horiba bio cup for processing. Data is processed on the Horiba Pentra 400 instrument and uploaded via the RSAS app directly into an excel spreadsheet for further analysis. The ABX Pentra chemistry panel reagents including ABX Pentra Albumin CP (REF# A11A01664), ABX Pentra ALP CP (REF# A11A01626), ABX Pentra ALT CP (REF# A11A01627), ABX Pentra Amylase CP (REF# A11A01628), ABX Pentra Urea CP (REF# A11A01641), ABX Pentra Calcium AS CP (REF# A11A01954), ABX Pentra Creatinine 120 CP (REF# A11A01933), ABX Pentra Glucose HK CP (REF# A11A01667), ABX Pentra Phosphorus CP (REF# A11A01665), ABX Pentra Potassium-E (REF# A11A01740), ABX Pentra Sodium-E (REF# A11A01738), ABX Pentra Bilirubin, Total CP (REF# A11A01639), ABX Pentra Total Protein 100 CP (REF# A11A01932), according to the manufacture’s instructions.
Complete blood counting
Once EDTA anti-coagulated samples for CBC’s are received in the lab, they are immediately organized by ID number and processed on the Oxford Science hematology analyzer. The results are automatically downloaded onto an excel spreadsheet, reviewed by lab personnel and sent to the investigator via email for further analysis. In brief, blood was collected in Eppendorf tubes containing 10 µl of 10% EDTA via retroorbital bleed using 200 µl heparinized capillary tubes (Cat# 22-362-566, Fisher brand). Blood samples were processed within 2 h to avoid hemolysis. The number of leucocytes (WBC), erythrocytes (RBC), lymphocytes (LY), neutrophil (NE), monocytes (MO), eosinophils (EO) and platelets (PLT) were counted. Proprietary lysing agent was added to liberate hemoglobin and ultimately convert it to cyanmethemoglobin to calculate the value.
Western blot and antibodies
For western blotting, samples from normal livers and tumors excised from livers were homogenized with calculated volume of 2X sample buffer (1M TRIS/HCl, 10% SDS, 0.1% bromophenol-blue, 10% β−mercaptoethanol, 10% glycerol) and heated for 15 minutes at 95°C. Proteins were resolved on protein gels (Bio-Rad, Cat#4568083) and transferred onto PVDF membrane (Bio-Rad, Cat#170-4272) with Trans-blot Turbo transfer system (Bio-Rad, Cat#1704150). After being incubated with the primary antibody, horseradish peroxidase-(HRP) conjugated secondary antibody (Novex, Life technologies) at 1: 5000 was used for 1 hour incubation. The signals were detected by chemiluminescence (ECL, Thermo scientific). Images were taken with Li-COR Odyssey FC (Li-COR, Cat#2800). Antibodies including TDO2 (Abclonal, A6766, RRID:AB_2767349, 1:1000), IDO2 (Mybiosource, MBS175296, 1:1000), C-MYC (Cell Signaling Technology, 5605S, RRID:AB_1903938, 1:1000), PRKDC (DNA-PK) )Novus, sc57-08, RRID: AB_2809479, 1:1000), b-actin (Sigma, A5441, RRID:AB_476744 , 1:5000) and GAPDH-HRP (Cell Signaling Technology, 3683S, RRID:AB_1642205, 1:1000) were used for western blot.
Small interfering RNA Transfection and doxorubicin treatment
Small interfering RNAs (siRNA) were transfected into NEJF10 cells using Lipofectamine RNAiMax (Invitrogen, Cat#13778150) according to manufacturer’s instructions. Non-Targeting siRNA#2 (Thermo Fisher Scientific, AM4637 https://www.thermofisher.com/order/catalog/product/AM4637?SID=srch-hj-AM4637 ) used as siRNA control. The siRNA oligos for Prkdc was ordered from Thermo Fisher Scientific (Thermo Fisher Scientific, AssayID151238, https://www.thermofisher.com/order/genome-database/details/sirna/151238?CID=&ICID=&subtype= , siRNA for mosue Prkdc:
Sense: 5-GGAAUAUACUAUAGAUCCUTT-3; Antisense: 5-AGGAUCUAUAGUAUAUUCCTG-3). 72h hour post transfection, cells were harvested for western blot. For doxorubicin treatment experiments, 24h post siPrkdc transfection, cells were treated with doxorubicin with concentrations of 0, 15nM for 4 days and fixed with formaldehyde for crystal violet staining.
Bulk RNA-seq and analysis
Total stranded RNA sequencing data were processed by the internal AutoMapper pipeline. Briefly the raw reads were firs trimmed (Trim-Galore version 0.60), mapped to mouse genome assembly (GRCm38, mm10) (STAR v2.7) and then the gene level values were quantified (RSEM v1.31) based on GENCODE annotation (VM22) . Low count genes were removed from analysis using a CPM cutoff corresponding to a count of 10 reads and only confidently annotated (level 1 and 2 gene annotation) and protein-coding genes are used for differential expression analysis. Normalization factors were generated using the TMM method, counts were then transformed using voom and transformed counts were analyzed using the lmFit and eBayes functions (R limma package version 3.42.2). The significantly up- and down- regulated genes were defined by at least 2-fold changes and adjusted p-value < 0.05. Then Gene set enrichment analysis (GSEA) was carried out using gene-level log2 fold changes from differential expression results against gene sets in the Molecular Signatures Database (MSigDB 6.2) (gsea2 version 2.2.3).
Single-cell RNA-seq and analysis
Library preparation and sequencing. The liver tumor was harvested from ABC-Myc mouse. Tumor mass was dissociated by using a modified two-step collagenase procedure 115. Briefly, the mouse was perfused with PBS containing 0.5mM EDTA and followed by perfusion with 2mg/ml of collagenase type IV (Worthington Biochemical Corporation, CLS-4) in DMEM (Dulbecco's Modified Eagle Medium) (Corning, 10-013-CM). The tumor from liver was chopped with razor and digested in 2mg/ml of collagenase type IV DMEM medium for 30 mins at 37°C. The cell suspension was filtered through a 70µm strainer and washed twice with DMEM. The dissociated cells were suspended in the DMEM medium. Before loaded into Chromium chips, cells were filtered again through a 40µm strainer and the single cells were counted by using a Luna cell counter, and then loaded into Chromium Chips V3 (10X Genomics) with a target capture of 8000 cells. The cDNA library construction and quality control were performed by following the manufacture’s protocol. The library was sequenced in Novaseq-V1 reagents. The sequenced data was processed by Cell Ranger Software (10X Genomics). Data preprocessing. UMIs mapped to genes encoding ribosomal/mitochondrial proteins were removed and cells with more than 40% of UMIs mapped to ribosomal/mitochondrial proteins were filtered. Cells with low (≤ 256, potentially dead cells with broken membrane) or high (≥ 32,768, potentially two or more cells in a single droplet) UMI counts were further filtered. A total of 11,405 cells were captured with an average of 6,320 mRNA molecules (UMIs, median: 5,639, range: 343 – 32,747). The expression level of each gene is normalized to 10,000 UMIs per cell and log transformed by adding 1 to the expression matrix. Clustering. The subpopulation structure of the whole dataset was inferred using Latent Cellular State Analysis (LCA), a novel clustering algorithm developed in house for analyzing large-scale scRNA-seq data116. Briefly, LCA first used singular value decomposition (SVD) to derive latent cellular states from the expression matrix for individual cells. Significant cellular states were determined using the Tracy-Widom test on eigenvalues. A modified version of spectral clustering was performed on the significant cellular states of individual cells (cellular states explained by potential technical variations including inter-sample difference and total UMIs were ignored) with different number of clusters (2-30). The optimal number of clusters was manually selected from top models determined by the silhouette measure for solutions with different number of clusters. Data visualization. Underneath cell variations were visualized in a 2D projection by t-distributed stochastic neighbor embedding (tSNE). Expression of individual genes or pathway scores were color coded (gray: not expressed; from low to high: blue-green-yellow-red) for each cell on tSNE plots. Differentially expressed gene analysis. Differentially expressed genes were analyzed by the negative binomial with independent dispersions117.
Tissue harverst. The ABC-Myc mouse was anesthetized with avertin (0.8 ml/20 g of mouse body weight). The mouse chest cavity was opened to expose the heart with needles, tweezers, and dissecting scissors. The right auricle was incised, and the needle filled with PBS containing 0.5mM EDTA was immediately inserted into the apex of the left ventricle for the perfusion, followed by perfusion with 2mg/ml of collagenase type IV (Worthington Biochemical Corporation, CLS-4) in Dulbecco's Modified Eagle Medium (Fisher Scientific, Cat#MT10013CM). The tumor tissue was immediately isolated in the ice cold DMEM medium.
Pathological assessment. Fresh frozen tissues were sectioned and mounted on the ST Library preparation slides, HE stained, and scanned with a Zeiss Axioscan slide scanner to generate 20x digital whole slide images. CZI files were imported into HALO (v3.2.1851.351, Indica Labs) to annotate and classify bulky tissue regions as neoplasia, non-neoplastic hepatocytes and stroma, extramedullary hematopoiesis, glass/clear space, or tissue folds/artifacts based on morphology and tinctorial staining characteristics. Tissue processing and data generation for spatial transcriptomics. Flash frozen samples were embedded in OCT (Tissue-Tek, Sakura) and cryosectioned as per Tissue preparation guide from Visium Spatial Gene expression Kit- 10X Genomics (Cat.1000184). Briefly, the tumor tissue was harvested from ABC-MYC mouse liver in the ice cold DMEM medium (Fisher Scientific, Cat#MT10013CM), excess liquid was removed from tissue and flash frozen immediately in the bath of Isopentane and Liquid nitrogen. The OCT embedded tissue block was sectioned (10µm) and placed on the capture area of Visium Gene Expression Slide and stored at -80ºC overnight. The tissue sections on the Visium slide were fixed with Methanol by incubating 30min at -20ºC. The tissue was H&E stained following Visium Gene expression kit procedure. The H&E stained sections were imaged using AxioScan Z.1. Whole slide scanner with standardized imaging protocol for Visium kit. After image acquisition, the slide sections were permeabilized for 18 min at 37ºC and cDNA, library was generated according to the Visium Spatial Gene Expression User Guide. The libraries were loaded and sequenced (R1-28cy, i7-10cy, i5-10cy and R2 -120 cycle) on Novaseq 6000 (Illumina) following recommendation of Visium Gene expression kit. The raw data was converted into FastQ and matrices of expression generated using the Space Range software V1.0 provided by 10X Genomics. Data analysis. Spatial transcriptome data were processed and visualized using R package Seurat. Data were normalized using sctransform recommended by Seurat. We also used Seurat’s dimension reduction and clustering functionality to explore the spatial structures and patterns of gene expression. Expression patterns for chosen genes were visualized using feature plot function implemented in Seurat and ggplot2.
Annexin V/DAPI staining
Cells were seeded at a density of 100, 000 cells in each well in 6 well plates. Next day, cells were pretreated with AZD7648 for 1 h, before adding doxorubicin for further 48 h. Cells were trypsinized (0.05% trypsin for NEJF10 and 0.25% trypsin for HepG2) for 4 min and centrifuged at 1000 rpm for 5 min at 4 °C. Apoptosis was detected by dual staining of Annexin V-FITC and DAPI using apoptosis assay kit (TONBO biosciences, CA, USA) according to manufacturer's instructions. Annexin V-FITC/DAPI positive cells were Collected using log amplification, and 10,000 events were recorded (BD LSR-II, BD Biosciences, NJ, USA), and data was analyzed using BD FACSDiva™ Software.
NAD and NADH quantification
The levels of NAD and NADH were determined by using commercially NAD/NADH Quantification Colorimetric Kit (BioVision, Catalog # K337-100) according to the manufacturer’s protocol. Briefly, the livers were excised from normal and ABC-Myc mice and placed in a sterile tube containing cold phosphate-buffered saline (PBS) in cold ice bucket during transport from the animal research facility to the research laboratory. About 20 mg of tumor nodules or normal liver tissues was homogenized in 400 μl of NAD/NADH extraction buffer in a micro-centrifuge tube and centrifuged at 14000 rpm at 4°C for 5 min. The extracted NAD/NADH supernatant was transferred into a new tube. Supernatants were divided to measure total NAD, (NADt = NAD + NADH) and NADH (by heating samples to 60 °C for 30 min to decompose NAD, while keeping NADH intact). Both NAD and NADH samples were mixed with NAD cycling enzyme and absorbance was measured at 450 nm with Synergy H1 microplate reader (BioTek, Synergy H1). Samples were always filtered with 10 K spin column (BioVision, Cat#1997) before performing the NAD/NADH quantification.
Crystal Violet Staining
ABC-Myc-derived hepatoblastoma cell lines (750 cells per well) and HepG2 (10,000 cells per well) cells were seeded in 6-well plates. After 24 hours, cells were treated with AZD7648 (0, 0.1, 0.33, 1 and 3.3 µM). Doxorubicin (0, 7.5,15,30 and 60 nM for NEJF10 cell line and 0, 5,25,125 and 625 nM for HepG2) was added post AZD7648 treatment 1 hour. NEJF10 cell was cultured with DMEM complete medium for 5 days and HepG2 was cultured with DMEM complete medium for 8 days. The culture medium and AZD7648 and doxorubicin were changed every 2-3 days. After removing media, cells were washed with Dulbecco’s phosphate buffered saline without calcium or magnesium (Lonza, Cat#17-516Q) and fixed with 4% formaldehyde in PBS for 20 minutes. Once formaldehyde was removed, cells were stained with 0.1% crystal violet (Sigma-Aldrich, Cat#HT90132-1L) for 1 hour. Plates were rinsed with water and imaged.
PrestoBlue assay and Bliss score calculation
PrestoBlue assay and Bliss score calculation was descripted as previous report (Alexandra et al., 2021) with minor modification. Briefly, NEJF10 (100 cells per well) and HepG2 (1000 cells per well) cells were seeded in 96-well plates. After 24 hours, cells were treated with AZD7648 (0, 0.1, 0.33, 1 and 3.3 µM) and doxorubicin (0, 7.5,15,30 and 60 nM for NEJF10 cell line and 0, 5,25,125 and 625 nM for HepG2) in an 8x5 matrix. Cells were treated for 5 days, and cell viability was determined using the PrestoBlue assay (Invitrogen, A-13262) according to manufacturer’s instructions. Cell viability for each treatment was normalized against the control group. A Bliss independence model was used to evaluate combination effects. Percentage over the Bliss score index was calculated with the equation (A+B)-AxB, in which A and B are the percentage of growth inhibitions induced by agents A and B at a given dose, respectively. The difference between the Bliss expectation and the observed growth inhibition induced by the combination of agent A and B at the same dose is the Bliss excess.
Cell viability assay for IC50 of CDK7 and AURKA inhibitors
Cell lines were plated in 384 well plates at 100 (NEJF1, NEJF10), 500 (NEJF2, NEJF4, NEJF6) or 1000 (CCLF_PEDS_0046_N), and treated with either samuraciclib or alisertib in technical quadruplicate at doses ranging from 2 nM to 20 uM using a Tecan D300e compound printer (Tecan Biosciences). All wells were normalized to 0.1% total DMSO input. Cells were incubated at 37C until timepoint of assay development, using the Cell-Titer Glo assay (Promega). Data was processed using Graphpad Prism 7.0. Cell Lines: CCLF_PEDS_0046_N normal fibroblasts were a kind gift of the Cancer Cell Line Factory (Broad Institute, Cambridge, MA).
CRISPR screening for cancer dependency gene and genetic modifiers to doxorubicin
The Mouse CRISPR Knockout Pooled Library (Brie, lentiCRISPRv2) was obtained from Addgene (Addgene#73632), which includes 1000 control gRNAs and 78,637 gRNAs targeting 19,674 genes. The plasmid library was amplified and validated in the Center for Advanced Genome Engineering at St. Jude Children’s Research Hospital as described in the Broad GPP protocol (https://portals.broadinstitute.org/gpp/public/resources/protocols) except EnduraTM DUOs (Lucigen) electrocompetent cells were used for the transformation step. The workflow of this whole genome genetic screen is illustrated in Figure 5A. We used NEJF-10 cells, a mouse hepatoblastoma cell line established in our laboratory by culturing dissociated liver mass cells from the ABC-Myc model. The cells were transduced with mouse CRISPR Knockout pooled library (Brie) which contains 78, 637 unique sgRNA sequences targeting 19,674 human genes (4 sgRNAs per gene, and 1000 non-targeting controls) at a low MOI (~0.3) to ensure effective barcoding of individual cells. Cells were replenished with fresh DMEM medium containing 2 μg/mL puromycin (Millipore Sigma) for 36 h. After puromycin selection, cells were washed to eliminate dead cell debris and maintained in complete DMEM medium, and 32x 106 cells were collected for genomic DNA extraction to ensure over 400× coverage of Brie library. The transduced cells were cultured for 5 days for CRISPR editing to generate a mutant cell pool, which was then treated with vehicle (DMSO) and doxorubicin (IC20~5 nM, 14 days: IC90~30nM 21 days), these concentrations were selected from colony formation assay that mimics similar experimental setup for actual experiment (IC20~7.8 n M, IC90~125 nM, for 3 days). During the experiment, at least 32x 106 cells were collected for genomic DNA extraction to ensure over 400× coverage of Brie library. The total genomic DNA was extracted using a DNeasy Blood & Tissue Kit (Qiagen) and quantified with a Nanodrop instrument. The sgRNA sequences were amplified using PCR method using NEB Q5 polymerase (New England Biolabs). PCR products were purified by AMPure XP SPRI beads (Beckman Coulter) and quantified by a Qubit dsDNA HS assay (Thermo Fisher Scientific). A total of 16 million reads were sequenced using an Illumina HiSeq sequencer, and the sequencing data were analyzed using MAGeCK-VISPR software. NGS sequencing was performed in the Hartwell Center Genome Sequencing Facility at St. Jude Children’s Research Hospital. Single-end, 100-cycle sequencing was performed on a NovaSeq 6000 (Illumina). Validation to check gRNA presence and representation was performed using calc_auc_v1.1.py (https://github.com/mhegde/) and count_spacers.py.
Drug response screen
For the screen, assay-ready plates were prepared by dispensing 50nl small molecules in empty white 384-well plates (Corning) using Echo 555 Liquid Handler (Labcyte). 50μl ABC-Myc cells per well were seeded in the assay-ready plates. The cells were incubated at 37°C, 5% CO2 in a humidified cell culture incubator (LiCONiC) for five days. Prior to the cytotoxicity assay, 25μl medium per well was removed by Apricot S2 (SPT Labtech). To quantify the cytotoxicity, the amount of intracellular ATP was measured by CellTiter Glo (Promega). Widget, an automated robot system in St. Jude Children’s Research Hospital was utilized for the cytotoxicity assay. 25μl CellTiter Glo reagent (Promega) was added to each well by Multidrop Combi (ThermoFisher). After shaking plates, plates were incubated for 20 minutes at room temperature. Then, the luminescent signal was measured by EnVision (PerkinElmer). Luminescent signal results were analyzed by Genedata Screener (Genedata). All the results were normalized by the negative control (DMSO) and the positive control (5μM 17-DMAG).
In vivo therapy
(1) Transgenic ABC-Myc mice mouse model. All the animals are procured at Animal Resource Center (ARC) at St. Jude Children’s Research Hospital and study was approved by Institutional Animal Care and Use Committee. Following genotyping, ABC-Myc mice were randomized and assigned to treatment groups. Inclusion criteria were the presence of the ABC-Myc allele, either in heterozygosity or both ABC-Myc alleles, age ranging 16-18 days after birth, both genders. Mice were treated with vehicle, doxorubicin (0.75 mg/kg, intraperitoneal, twice weekly) and AZD7648 (50 mg/kg/day, twice, oral gavage, everyday); either agent alone or in combination with doxorubicin and AZD7648 for three weeks. The mice weight and activity were monitored throughout the experiment. The humane end point was decided (notified by staff not directly involved in this study) to euthanize the mice. The livers from treatment groups of ABC-Myc mice and age matching normal mice were excised, weighed and imaged. (2) HepG2 xenograft study. 4–6-week-old female NSG mice (NOD.Cg-Prkdc scid Il2rg tm1Wjl /SzJ) were housed in pathogen-free conditions with food and water provided ad libitum. HepG2 cells (5 × 106/mouse) in 100 μl PBS were injected subcutaneously on the right flank of mice. When the tumor size reached up to ~100 mm3, the animals were randomized into four groups (n=5 mice per group). Mice were treated with vehicle, doxorubicin (1 mg/kg, intraperitoneal, twice weekly) and AZD7648 (50 mg/kg/day, oral gavage everyday); either agent alone or in combination with doxorubicin and AZD7648 for three weeks. The tumor volume and mice weight were measured twice in a week. All the mice were euthanized, and subcutaneous tumors were collected, imaged and weighed. The tumor volume and weight were presented as the means ± S.D (n = 5). In vivo studies were approved and conducted in accordance with Institutional Animal Care and Use Committee at St. Jude Children’s Research Hospital.