Acute leukemia cells resistant to PI3K/mTOR inhibition display upregulation of P2RY14 expression

The PI3K/mTOR pathway is the second most frequently deregulated pathway in a majority of cancers such as breast cancer, lung cancer, and melanomas as well as leukemia. Mutations in the genes coding for receptor tyrosine kinases (RTKs) and G-protein-coupled receptors (GPCRs) are quite common in all forms of acute leukemia. This can be a major cause of deregulation of the PI3K-mTOR pathway. To understand how cells display resistance to the dual PI3K/mTOR inhibitor, we used a panel of 25 acute leukemia cell lines. We observed that while a number of cell lines displayed sensitivity to the dual PI3K/mTOR pathway inhibitor PKI-587, many cells displayed substantial resistance. Cells sensitive to PKI-587 also showed aberrant activation of PI3K/mTOR pathway components such as AKT and S6K and also displayed sensitivity to a panel of various other PI3K/mTOR inhibitors. Using RNA sequencing data, we observed that expression of a G protein-coupled receptor, P2RY14, was upregulated nine-fold in cells showing resistance to the PI3K/mTOR inhibitor. P2RY14 has not been much studied in hematologic malignancies. However, this receptor seems to have a role in the localization of hematopoietic stem cells (HSCs) and in promoting regenerative capabilities following injury. We observed that acute lymphoblastic leukemia (ALL) and FLT3-ITD-positive acute myeloid leukemia (AML) patients with higher expression of P2RY14 mRNA displayed relatively poor survival compared to patients carrying lower expression of P2RY14 suggesting a role of P2RY14 in patient survival. To understand the role of this receptor in cell signaling, we used phospho-protein arrays and observed activation of distinct signaling cascades. Furthermore, array data were verified using murine pro-B cell line Ba/F3 stably transfected with P2RY14. We observed that activation of P2RY14 by its ligand, UDP-glucose, resulted in selective induction of ERK1/2 phosphorylation. Taken together, our data suggest that acute leukemia cells resistant to PI3K/mTOR inhibition display upregulation of a GPCR, P2RY14, which has a role in patient survival and also couples to the activation of ERK signaling. Electronic supplementary material The online version of this article (10.1186/s13148-018-0516-x) contains supplementary material, which is available to authorized users.


SUP-B15
Immunoprecipitation and Western Blot: Cells were washed thrice with 1X DPBS or 1X RPMI1640 medium to remove serum and cytokines if any. The cells were then re-suspended in 1X RPMI1640 medium and kept for 4 hours at 37°C with 5% CO2 thus starving them of serum and cytokines. After incubation, Ba/F3 cells stably transfected with P2RY14 and Empty vector were stimulated with 100µM UDP-glucose at 37°C for the respective amounts of time. The stimulation was stopped by immediately placing the cells on ice and adding ice cold 1X PBS. Thereafter, these cells, as well as the leukemia cell lines, were lysed in the same manner in 1% Triton X-100 lysis buffer supplemented with 3 protease inhibitors (1mM Na3VO4, 1% Trasylol and 1mM PMSF) and incubated for 15 minutes on ice. Immunoprecipitation was performed using 1µg primary antibody for 1 ml cell lysates followed by incubation for 1 hour on ice. Immunoprecipitates were collected using magnetic beads which were washed thrice with the above-mentioned lysis buffer containing the protease inhibitors. The lysates were mixed with 1:1 volume of SDS sample loading buffer and boiled at 95°C for 5 minutes. The protein concentration of the total cell lysates was determined by BCA Protein Assay Kit (ThermoScientific, USA) and 20µg proteins from each lysate were separated on 8.5% SDS-PAGE gels followed by their transfer to PVDF membranes. These membranes were then blocked with 5% non-fat dry milk prepared in 0.2% PBS-T for 1 hour at room temperature. Prior to incubation with the respective primary antibodies overnight at 4°C, the membranes were washed thrice in 0.05% PBS-T for 5 minutes each. For immunodetection, the blots were incubated with the respective horseradish peroxidase-conjugated secondary antibodies for 1 hour at room temperature and were developed with the Luminata Forte Western HRP Substrate (Millipore) and Amersham Imager 600 (GE Healthcare, Sweden). The band intensities were quantified with the Multi-Gauge software from Fujifilm.

Inhibitor Assay/ Cell Viability Assay:
All the above mentioned 25 acute leukemia cell lines were tested for their viability against 10 different concentrations of the dual PI3K-mTOR inhibitor Gedatolisib (PF-05212384, PKI-587) ranging from 0.1nM to 5000nM. To perform this assay, all the cell lines were seeded at a density of 15,000 cells/well in a 96-well plate. Medium blank and cell blank (cells with no inhibitor) were used as controls. About 10µl Gedatolisib was added to 15,000 cells present in 90µl per well. Thus, each column in a 96-well plate represents one concentration of the inhibitor. The plate with the inhibitor was then incubated at 37°C -5% CO2 for 48 hours. After the incubation period, cell viability was determined by adding 10µl PrestoBlue in each well and then measuring the fluorescence with a 96well plate reader after 2-5 hours depending on the color change from blue to purple in the cell without inhibitor column. EC50 values (nM) were determined using GraphPad Prism software and thus, the relative viability curves were generated. Later three more inhibitors Apitolisib, BGT226 and Dactolisib were used to measure EC50 values using the same method.

Stable transfection of Ba/F3 cells:
The expression plasmid constructs pMSCVneo-P2RY14-Myc-DKK and pMSCVneo-empty vector were used for retroviral transfection of Ba/F3 cells. To achieve this, the packaging cell line for viral particles, EcoPack was transfected with both the vectors using Lipofectamine 2000. After incubating the transfected cells for 72 hours, the supernatant containing the viral particles was collected and were then used to infect Ba/F3 cells. Both the stably transfected Ba/F3 cells were selected for a period of 2 weeks against 0.8 mg/ml geneticin (G-418), an analog of neomycin. Flow cytometry and Western Blot were used to confirm the expression of P2RY14 using anti-Flag antibodies.
RNAseq, mutation and inhibitor data from COSMIC Database: COSMIC (Catalogue of Somatic Mutations In Cancer) is the world's largest database for containing the most comprehensive set of data for exploring the impact of somatic mutations in various human cancer cell lines. Expression Analysis in the form of RNA Sequence Data was downloaded from COSMIC for both PKI-587 sensitive and resistant sets of cell lines used in this project. This dataset was used to calculate the p-value and fold of expression changes for 16,445 genes in these 2 sets of cell lines. Inhibitor Data which gives a range of different inhibitors and their IC50 values against different cell lines was downloaded for our sensitive and resistant sets. Cancer Gene Census in COSMIC gives information about mutations in 616 genes and their implications in various cancers. Different genes mutated in our set of cell lines were identified from this Mutational Data downloaded from COSMIC.

Gene Set Enrichment Analysis (GSEA):
Gene Expression Data (RNA-Seq Data) downloaded from COSMIC Database was used to run Gene Set Enrichment Analysis (GSEA) using the GSEA software (v2.2.4) from Broad Institute with 1000 permutations to identify different pathways enriched by particular sets of genes in both PKI-587 sensitive and resistant groups of cell lines.

Survival analysis:
Gene expression data for acute lymphoblastic leukemia patients was obtained from TARGET database of National Cancer Institute (https://ocg.cancer.gov/programs/target/data-matrix). Within 207 patients 122 patient data had survival information. P2RY14 expression data from 122 patient samples were to calculate the Z-score. The Z-score 0.5 cut-off difference was used to divide two groups. Overall survival data was used to calculate the survival benefit for higher (n=47) and lower (n=57) P2RY14 gene expressing patients. Furthermore, gene expression data for acute myeloid leukemia patients carrying oncogenic FLT3-ITD mutations (GSE14468) was analyzed using similar Z-score cut-off.

Real-time Quantitative PCR (RT-qPCR):
Total RNA was isolated from cells using the RNeasy mini kit (Qiagen) following manufacturer's protocol. cDNAs were synthesized with High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific Catalog number: 4368814) according to the manufacturer's directions. Gene expression was assessed by RT-qPCR using an Applied Biosystems QuantStiduo 7 Flex detection system and gene-specific RT-qPCR primer assays (Thermo Fisher Scientific). β-Actin and GAPDH were used as endogenous controls to normalize the expression data. Each sample was analyzed in quadruplicate. The comparative Ct method was used to calculate the relative changes in gene expression.

Statistical Analysis:
GraphPad Prism 5.0 software was used to perform statistical analysis where data were expressed as mean ± SE. Where applicable, we used unpaired Student's t-test and one-way ANOVA with Bonferroni's post-test. P≤0.05 was considered as significant. Table S1: Inhibitor assay for Gedatolisib (PKI-587): Around 25 acute leukemia cell lines were tested for their sensitivity and viability against the dual PI3K-mTOR inhibitor, Gedatolisib (PKI-587). This assay was performed twice and the numbers 1 st and 2 nd denote the EC50 values obtained for the 1 st and 2 nd time respectively. Red-yellow-green is the color coding scheme used in the column of average EC50 (nM) which represents the values as low-middle-high. Thus, low and high EC50 values denote highly sensitive and resistant cells respectively.  Figure S3: Phospho-kinase array: Proteome Profiler Human Phospho-Kinase Array (ARY003B) Kits from R&D Systems was used to detect relative phosphorylation changes in UDP-glucose stimulated cells.