Supplementary Information. ATM and MET kinases are synthetic lethal with. non-genotoxic activation of p53

Transcription

1 Supplementary Information ATM and MET kinases are synthetic lethal with non-genotoxic activation of p53 Kelly D. Sullivan 1, Nuria Padilla-Just 1, Ryan E. Henry 1, Christopher C. Porter 2, Jihye Kim 3, John J. Tentler 3, S. Gail Eckhardt 3, Aik Choon Tan 3, James DeGregori 2,4, Joaquín M. Espinosa 1* 1 Howard Hughes Medical Institute & Department of Molecular, Cellular and Developmental Biology, University of Colorado at Boulder 2 Department of Pediatrics, University of Colorado at Denver Anschutz Medical Campus 3 Department of Medicine/Medical Oncology, University of Colorado at Denver Anschutz Medical Campus 4 Department of Biochemistry and Molecular Genetics, University of Colorado at Denver Anschutz Medical Campus * To whom correspondence should be addressed: joaquin.espinosa@colorado.edu; FAX: (303)

2 Supplementary Results Supplementary Figure 1. A Synthetic Lethal with Nutlin-3 Screen. (a) Flow chart of wet and in silico steps of shrna analysis after cell harvest. (b) Hierarchical clustering analysis of raw sequence counts of shrnas from A549 cells. Left, heatmap of 21,067 shrnas where the median count of one treatment is greater than the maximum count of the other treatment. Right, actual read counts of top 10 candidate SLN shrnas. (c) Distribution of fold change for each shrna with a p-value >0.05 (total of 6,004) from A549 cells. Red diamonds indicate the position of the 30 shrnas chosen for validation in Fig. 2e. Arrows indicate the position of 5 shrnas highlighted throughout Fig. 2.

3 Supplementary Figure 2. ATM is synthetic lethal with p53 activation by Nutlin-3. (a) ATM knockdown reduces cell viability upon Nutlin-3 treatment. Top, a panel of cell lines stably expressing shrnas targeting ATM or a non-targeting shrna (Control) was treated with DMSO or 20 µm Nutlin-3 for 72 h prior to SRB

4 staining. The ratio of Nutlin-3 to DMSO for each line was compared to the nontargeting Control cell line. Bottom, lysates were prepared from each cell line and Western Blots carried out for ATM and Nucleolin to demonstrate the degree of ATM knockdown for each cell line. (b) Combined treatment with Nutlin-3 (20 µm) and ATMi (10 µm) affects long-term viability of HCT116 cells. Cells were treated for one week before crystal violet staining. (c) [Nutlin-3+ATMi] synthetic lethality is dependent on caspases and p53. HCT116 (p53 +/+) cells were pretreated with 10µM Z-VAD-FMK for 1 h prior to addition of drugs and HCT116 p53-/- cells were treated with Nutlin-3 and ATMi. After 24 h, all cells were analyzed for Annexin V and PI staining by flow cytometry. (d) ATMi and Nutlin-3 synergize to decrease cell viability at multiple dosages. Cells were treated with the indicated doses of drug for 24 h prior to SRB viability assay. (e) ATMi inhibits ATM autophosphorylation in a dose-dependent manner. Cells were pre-treated with ATMi at the indicated concentrations for 1 h prior to exposure to 10 gy IR and 30 min recovery. Western Blots were then performed to determine levels of ps1981- ATM and Nucleolin.

5 Supplementary Figure 3. Knockdown of ATM or MET sensitize A549 cells to apoptosis upon Nutlin-3 treatment. A549 cells expressing individual hairpins targeting ATM, MET or a Control were treated with DMSO or 20 µm Nutlin-3 for 24 h and Annexin V and PI levels measured by flow cytometry.

6 Supplementary Figure 4. MET is synthetic lethal with p53 activation by Nutlin-3. (a) MET knockdown results in decreased viability upon Nutlin-3 treatment. Cell lines expressing individual shrnas targeting MET were treated

7 with 20 µm Nutlin-3 for 72 h prior to analysis by SRB assay. A non-targeting shrna serves as a control. (b) Rescue of MET knockdown synthetic lethal phenotype. Cells stably expressing MET shrna 6, which targets the 3 UTR, were transiently transfected with 2 µg of PCDNA3-MET-V5 plasmid for 24 h prior to 24 h 20 µm Nutlin-3 treatment. Apoptotic index was determined by measuring Annexin-V-FITC and PI levels by flow cytometry. All bars represent the mean of 4 replicates ± standard deviation. P-values were calculated using Student s t-test (upaired, two-tailed). (c) Crizotinib-Nutlin-3 synthetic lethality is dependent on caspases and p53. HCT116 (p53 +/+) cells were pretreated with 10µM Z-VAD- FMK for one hour prior to addition drugs and HCT116 p53-/- cells were treated with Nutlin-3 and Crizotinib. After 24 h, all cells were analyzed for Annexin V and PI staining by flow cytometry. (d) Crizotinib and Nutlin-3 synergize to decrease viability at multiple dosages. Cells were treated with the indicated doses of drug for 24 h prior to SRB viability assay. (e) A second inhibitor of MET, SU11274, leads to increases apoptosis upon Nutlin-3 treatment. Cells were treated for 24 h with 10 µm SU11274 and 20 µm Nutlin-3 before Annexin V and PI levels were measured by flow cytometry. (f) Crizotinib and SU11274 inhibit MET autophosphorylation. HCT116 cells were pre-treated with the 7µM Crizotinib, 10µM SU11274 or DMSO for 3 hours prior to addition of 30 ng/ml HGF for 30 min. Western Blots were performed to determine the amount of (pt1234/1235) MET, total MET and Nucleolin.

8 Supplementary Figure 5. Parallel Action of Synthetic Lethal with Nutlin-3 Pathways. (a) Schematic of hypothetical regulation of SLN mrna expression by p53 or other 5FU-activated factors (dubbed X) during differential cell fate choice upon Nutlin-3 and 5FU treatment. (b) SLN mrna expression is largely unaffected by p53 activating agents causing arrest versus apoptosis. Cells were treated with 20 µm Nutlin-3 or 375 µm 5FU for 12 h and RNA prepared for microarray analysis. Heatmap showing relative expression levels of the top 505 SLNs (ranked by p(wz)) in each treatment condition. (c) Not a single SLN fulfills the criteria shown in model depicted in (a). Venn diagram of top 505 SLNs, genes repressed only by 5FU and genes induced only by Nutlin-3.

9 Supplementary Figure 6. (a) Nutlin-3 treatment does not affect MET levels. HCT116 cells were treated with DMSO or 20 µm Nutlin-3 for 24 h before Q-RT- PCR analysis of MET mrna expression. p21 mrna serves as a positive control. (b) Lysates were prepared from cell lines treated as in a and Western blots

10 performed for MET and actin. (c) MET expression correlates with Nutlin-3 sensitivity across cell types. Q-RT-PCR was performed on RNA from HCT116, A549 and BV173 cells. (d) Lysates were prepared from cell lines treated as in c and Western blots performed for MET and actin. (e) ATM expression does not correlate with Nutlin-3 sensitivity. Q-RT-PCR was performed on RNA from HCT116, A549 and BV173 cells. (f) Lysates were prepared from cell lines shown in e and Western blots performed for ATM and actin.

11 Supplementary Figure 7. Western Blots. Full Blots from main text Figures 1, 4, 5, 6 and 7.

12 Supplementary Table 1. Oligonucleotide primer sequences used in this study. This table contains primer sequences used both for cloning of shrnas for deep sequencing as well as those used for Q-RT-PCR. Deep Sequencing cdna Synthesis GNH Fwd-GNH Rev-GNH nfwd-gnh-iss nrev-gnh-iss CSP-GNH qrt-pcr Genbank Name and ID 18S ribosomal RNA 18S ribosomal RNA CDKN1A ( CDKN1A) CDKN1A ( CDKN1A) MDM2 (HDMX; hdm2) MDM2 (HDMX; hdm2) PUMA (BBC3) PUMA (BBC3) ATM ATM MET MET 5'-ACACACTACTTGAAGCACTCAAGGCAA-3' 5'-TGCATGTCGCTATGTGTTCTGGGA-3' 5'-CTCCCAGGCTCAGATCTGGTCTAA-3' 5'- CAAGCAGAAGACGGCATACGAAGAAGCAAAA AGCAGAATCGAAGAA-3' 5'- AATGATACGGCGACCACCGAGATCTACACTC TTTCCCTACACGACGCTTCCTGTCAGA-3' 5'ACACTCTTTCCCTACACGACGCTTCCTGTCA GA-3' Sequence GCCGCTAGAGGTGAAATTCTTG CTTTCGCTCTGGTCCGTCTT CTGGAGACTCTCAGGGTCGAAA GATTAGGGCTTCCTCTTGGAGAA GGCGATTGGAGGGTAGACCT CACATTTGCCTGGATCAGCA TCAACGCACAGTACGAGCG AAGGGCAGGAGTCCCATGAT CCGTGATGACCTGAGACAAG AACACCACTTCGCTGAGAGAG AGATGTGTGGTCCTTTGGCGT AGGATGGGCGCATTTCGGCT

13 Supplementary Table 2. Antibodies used in this study. Antibody Vendor Cat. Nº P53 Calbiochem OP43 P21 Santa Cruz Biotechnologies σ (C-18) Santa Cruz Biotechnologies 7683 BAX Technology 2772 BCL2 Technology 2872 PUMA Santa Cruz Biotechnologies Caspase3 Technology 9661 Caspase8 Technology 9746 tbid Technology 2002 perk1/2 Technology 9101 piκbα (Ser 32) 14D4 Technology 2859 MET Technology 3148 pmet Technology 3077 ATM Abcam ab78 patm Technology 4526 Actin Santa Cruz Biotechnologies 1616R Nucleolin Santa Cruz Biotechnologies 8031

14 Supplementary Dataset 1. HCT116 screen data. Table contains data from significant shrnas sequenced in HCT116 SLN screen. Columns are as follows: A- SBI shrna ID, indicates which shrna from the library was sequenced B- HUGO Gene Symbol C- shrna sequence, actual mapped sequence D- NCBI ID, indicates gene product targeted by a given shrna E- shrna/gene, indicates the number of unique shrna sequences identified in the screen (DMSO or Nutlin treatment) which target a given gene product F- DMSO Replicate #1 G- DMSO Replicate #2 H- DMSO Replicate #3 I- Nutlin-3 Replicate #1 J- Nutlin-3 Replicate #2 K- Nutlin-3 Replicate #3 L- Nutlin-3 Replicate #4 M- Adjusted p-value, calculated from p-value and q-value of false discovery rate N- Log fold change, indicates average fold change of each shrna from one treatment to another, calculated in edger O- p-value, calculated from negative binomial model in edger P- p(wz), indicates p-value of the weighted Z score for each gene Supplementary Dataset 2. Raw sequence counts from HCT116 screen. Table contains raw read counts prior to analysis. Columns are as follows: A- SBI shrna ID, indicates which shrna from the library was sequenced B- shrna sequence, actual mapped sequence C- DMSO Replicate #1 D- DMSO Replicate #2 E- DMSO Replicate #3 F- Nutlin-3 Replicate #1 G- Nutlin-3 Replicate #2 H- Nutlin-3 Replicate #3 I- Nutlin-3 Replicate #4 Supplementary Dataset 3. A549 screen data. Table contains data from significant shrnas sequenced in A549 SLN screen. Columns are as follows: J- SBI shrna ID, indicates which shrna from the library was sequenced

15 K- HUGO Gene Symbol L- shrna sequence, actual mapped sequence M- NCBI ID, indicates gene product targeted by a given shrna N- shrna/gene, indicates the number of unique shrna sequences identified in the screen (DMSO or Nutlin treatment) which target a given gene product O- DMSO Replicate #1 P- DMSO Replicate #2 Q- DMSO Replicate #3 R- DMSO Replicate #4 S- Nutlin-3 Replicate #1 T- Nutlin-3 Replicate #2 U- Nutlin-3 Replicate #3 V- Nutlin-3 Replicate #4 W- Adjusted p-value, calculated from p-value and q-value of false discovery rate X- Log fold change, indicates average fold change of each shrna from one treatment to another, calculated in edger Y- p-value, calculated from negative binomial model in edger Z- p(wz), indicates p-value of the weighted Z score for each gene Supplementary Dataset 4. Raw sequence counts from A549 screen. A- SBI shrna ID, indicates which shrna from the library was sequenced B- shrna sequence, actual mapped sequence C- DMSO Replicate #1 D- DMSO Replicate #2 E- DMSO Replicate #3 F- DMSO Replicate #4 G- Nutlin-3 Replicate #1 H- Nutlin-3 Replicate #2 I- Nutlin-3 Replicate #3 J- Nutlin-3 Replicate #4 Supplementary Dataset 5. shrna sequences used in this study.