Animals. All animal experiments were approved by the Institutional Animal Care Committees at Fox Chase Cancer Center and Robert H Lurie Comprehensive Cancer Center. To generate A novel KP-MSI2 model (129S/Sv-Krastm3Tyj/J;Trp53tm1Brn/J; Msi2−/−), we crossed 129S/Sv-Krastm3Tyj/J;Trp53tm1Brn/J (KP) mice[34] with C57BL/6 floxed Msi2−/−Cre transgenic mice [33]. Following lung-specific induction with Adeno-Cre, each 6-week-old mouse get 2.5 x106 units of Cre recombinase adenovirus (Vector Biolabs, Malvern, PA), the 129S/Sv-KP mice develop lung disease similar to that of humans; large papillary adenomas are seen after 6 weeks and adenocarcinoma after ~ 10 weeks, in 4-month-old animals [34]. Mice were euthanized after ~ 5 month of activation and their lungs, heart, liver, spleen were collected for paraffin blocks, and lung cancer cell lines were established from fresh lung mouse tumors.
Mouse genotyping. Small segments of mouse tails (0.5 ~ 1.0 cm) were collected and digested in 150 µL of DNA-lysis buffer with 0.2 mg/ml of proteinase K for overnight at 56 ºC. Then, DNA samples were diluted 50 times and mixed with DreamTaq master mix (2X) (Thermofisher) following vendor protocol for genotyping PCR. Then, DNA electrophoresis was performed in 2% agarose gel. Bands detection was performed using imaging system ChemiDoc (Bio-Rad). All primers used in this procedure are noted in Supp Table S1.
Assessment of in vivo tumor growth. For in vivo studies 2 x 106 of KP-1 and KPM2-2 cells were injected subcutaneously (s.c.) in the right flank of 6-week-old mice C57BL/6 (total: 20 males, 20 females) from The Jackson Laboratory (ME, USA). All mice were over 18g at the start of live-phase study. When tumor volumes reached 100 mm3 animals were randomly assigned to the control or experimental groups (n = 10 mice/group, 5 males, 5 females). The mice were treated with 5% DMSO, 40% PEG 300, 5% Tween-80, 50% sterile H2O (vehicle) or olaparib 50 mg/kg in 5% DMSO, 40% PEG 300, 5% Tween-80, 50% sterile H2O. Treatment was perfomed by daily oral gavage. Tumors were measured twice weekly, and their volumes were calculated with the formula: [volume = 0.52 × (width)2 × length]. Mice were euthanized when tumor volume exceeded 1500 mm3 or on 41st day of experiment.
Immunohistochemistry of mouse NSCLC samples. Scientific review committee (SRC) and IRB approvals were obtained for de-identified samples from Fox Chase Cancer Center, both at Fox Chase Cancer Center and at Robert H Lurie Comprehensive Cancer Center. Tissue samples were stained for MSI2, MSI1, ATM and CyclinB1, Ki-67 and phospho-H2AX via immunohistochemical (IHC) approach and hematoxylin and eosin (H&E) stained sections were used for morphological evaluation purposes, and unstained sections were used for IHC staining using standard methods. Briefly, 5 µm formalin-fixed, paraffin-embedded sections were deparaffinized and hydrated. Sections were then subjected to heat-induced epitope retrieval with 0.01M citrate buffer (pH 6.0) (MSI2, MSI1, ATM) or EDTA buffer (CyclinB1, Ki-67, phospho-H2AX). Endogenous peroxidases were quenched by the immersion of the slides in 3% H2O2 solution. The sections were incubated overnight with primary antibodies at 4°C in a humidified slide chamber. As a negative control, the primary antibody was replaced with normal mouse/rabbit IgG to confirm absence of specific staining. Immunodetection was performed using the Dako Envision + polymer system and immunostaining was visualized with the chromogen 3, 3′-diaminobenzidine. All slides were viewed with a Nikon Eclipse 50i microscope and photo- micrographs were taken with an attached Nikon DS-Fi1 camera (Melville, NY, USA). For IHC quantification, each spot was examined by board-certified pathologists (ED and NK) who assigned a score of 0 (no staining), 1+ (weak staining), 2+ (moderate staining), and 3+ (strong staining) within carcinomatous areas. The score for each of the two tumor spots was averaged for statistical analysis. The H-score, which ranges from 0 to 300, was calculated using the following formula: [1x(% cells 1+) + 2x(% cells 2+) + 3x(% cells 3+)], which reflects staining intensity as well as percentage of positive cells. A sum of p2 and p3 represents a sum of 2 + and 3 + cells (2x (% cells 2+) + 3x(% cells 3+), which excludes 1 + cells
Establishment of cell lines. Murine lung tumors were collected and used to generate. murine NSCLC cell lines by standard approaches. Briefly, the lines 129S/Sv-Krastm3Tyj/J;Trp53tm1Brn/J; Msi2−/− (KPM2-1, KPM2-2, KPM2-3) and Krastm3Tyj/J;Trp53tm1Brn/J (KP-1, KP-2, KP-3) were prepared from freshly isolated cancer cells. Initial stocks were cryopreserved, and at every 6-month interval, a fresh aliquot of frozen cells was used for the experiments. The tumor pieces were incubated with collagenase (400 u/ml) at 37 ºC for 2 hours, then we used cell strainer (70 µm) for cells separation. After that, the cells were incubated with ACK lysis buffer at room temperature for 1–2 minutes, then the cells were washed and transferred to T25 flask with B27 cell culture media for propagation.
Cell culture of human cancer cell lines. Human lung cancer cell lines with KRAS-mutations (A549 and H441) were obtained from the American Type Culture Collection (ATCC). No additional authentication was performed. All cells were cultured in RPMI 1640 (Gibco, Gaithersburg, MD) supplemented with 10% FBS (Hyclone, Logan, UT), penicillin (100U/ml), streptomycin (100µg/ml), sodium pyruvate (1 mM) and non-essential amino acids (0.1 mM) under conditions indicated in the figure legends.
Antibodies and drugs. Anti-MSI2 (#ab76148), anti-MSI1 (#ab21628), anti-phCHK2 T68 (#ab278548) were obtained from Abcam (Cambridge, UK). Anti-phATM S1981 (#GTX132146), anti-ATM (#GTX70103), anti-ATR (#GTX128146), anti-phCDC25A (#GTX55131), anti-CDC25A (#GTX102308) were obtained from GeneTex, Inc. (Irvine, CA). Anti-phH2A.X S139 (#9718), anti-ATM (#2873), anti-CHK2 (#2662), anti-phCHK1 S345 (#2348), anti-CHK1 (#2360), anti-β-actin (#3700), anti-rabbit HRP-linked (#7074), anti-mouse HRP-linked (#7076) were obtained from Cell signaling (Danvers, MA). Anti-MSI1 (#AF2628) was obtained from R&D systems (Minneapolis, MN) Anti-ATM (#27156-1-AP) was obtained from Proteintech (Rosemont, IL). Olaparib (#HY-10162), cisplatin (#HY-17395), doxycycline (#HY-N0565) were obtained from MedChemExpress (Monmouth Junction, NJ).
Vector construction and lentivirus production. To generate stable human cell lines with inducible MSI2 knockdowns, self-complementary single-stranded DNA oligos (Supp Table S2) were annealed and cloned into AgeI/EcoR1 sites of Tet-pLKO-puro vector (Addgene plasmid # 21915). All constructs were validated by direct sequencing. All generated cell lines used in the study are noted in Supp Table S3, and have been previously described [21, 32].
Western blot analysis. Cell lysates preparation and Western blot analysis were performed using standard methods as previously described[21]. Signals were detected by X-ray films and digitized by photo scanner. Image analysis was done using ImageJ (version 1.53e, National Institutes of Health, Bethesda, MD), with signal intensity normalized to β-actin; 3–4 repeats were used for each quantitative analysis. Final data was analyzed in GraphPad Prism by unpaired t-test to determine statistical significance.
Reverse transcription and qPCR. RNA was extracted using a phenol-chloroform based method. RNA concentration and quantity were measured using NanoDrop Lite (cat# ND-LITE ThermoFisher Scientific). First strand cDNA synthesis was performed with iScript cDNA synthesis kit (cat#1708841, Biorad, California, USA) according to manufacturer’s instructions. The generated cDNA was diluted tenfold and used as a template for qPCR, which was performed with Applied Biosystems QuantStudio 3 system using PowerTrack™ SYBR Green Master Mix (Applied Biosystems). Relative quantification of genes expression was performed using 2−ΔΔCt method, Ct greater than 35 cycles was considered as undetected, using primers indicated in Supp Table S4.
Cell viability assay. To analyze the effects of MSI2 depletion or compound treatment on cell proliferation, cells were plated (500 cells/well) in 96-well cell culture plates in complete media. We used increasing concentrations of compounds to calculate IC50 values for each cell line. After 72 hours incubation with compounds, we added CellTiter-Blue® reagent (Promega, Fitchburg, WI) to cells and incubated them for 2 hours at 37 ºC. After that we recorded fluorescence (560Ex/590Em). Data was analyzed using GraphPad Prism software.
Clonogenic assay. Murine KP, KPM2 cells and and human A549 cells (500 cells per well) and H441 cells (2000 cells per well) were plated in 12-well pates and incubated in complete media. After 24 hours compounds were added, after which cells were incubated for 7 days. Cells were fixed in 10% acetic acid/ 10% methanol solution and stained with 0.5% (w/v) crystal violet as previously described[35]. A colony was defined as consisting of > 50 cells and counted digitally using ImageJ software as described previously[36].
RNA-IP assays. RNA was immunoprecipitated from cell lysates (2 × 107 cells per IP) using either a control normal rabbit IgG or rabbit monoclonal anti-MSI2 antibody and the Magna RIP RNA-binding Protein Immunoprecipitation kit (cat#17–700, Millipore, Burlington, MA). The manufacturer’s instructions were followed with the exception that RNeasy MinElute Cleanup kit (cat#74202, Qiagen, Venlo, Netherlands) was used to prepare RNA. Immunoprecipitated RNAs were quantified by quantitative PCR (qPCR) using primers indicated in Supp Table S4, using PTP4A1 as a normalization (positive control) and GAPDH as a negative control.
RPPA. The murine cell lines were lysed and prepared according to MD Anderson Core Facility instructions[37–39], and RPPA was performed at MD Anderson facility. Heatmap was generated using the GraphPad Prism.
In silico evaluation of MSI2 binding to mRNAs. Human and murine genome sequences for ATM and other mRNAs from the NCBI (ATM human (NCBI Reference Sequence: NM_000051.4), ATM mouse (NCBI Reference Sequence: NM_007499.3), CHK2 human (NCBI Reference Sequence: NM_001005735.2), CHK2 mouse (NCBI Reference Sequence: NM_001363308.1) were scanned for Musashi binding motifs previously defined by Bennett et al. (15 motifs with highest p values) [23], Wang et al. (8 motifs with highest p values) [24] and Nguyen et al. (12 motifs with highest p values) [25] (Supp Tables S5, S6, S7).
Statistical analysis. All statistical analyses, including unpaired two-tailed t-test, ANOVA analysis, Spearman correlation, were performed in GraphPad Prizm 9 (San Diego, CA).