Individual Cell Lines Research Articles

Abstract B21: Enhanced understanding of drug responses and drug-drug interactions in triple negative breast cancer cells with a multiplexed cell viability and cell death readout

Abstract Triple negative breast cancers (TNBC) are a heterogeneous group of cancers that remain a considerable clinical challenge with no known effective targeted therapy, limiting the therapies to cytotoxic chemotherapy, radiation and surgery. Molecular sub-classifications of TNBC have been established but none of these define therapy. With an aim to establish potential novel stratified therapeutic strategies to target TNBC, we applied the Drug Sensitivity and Resistance Testing (DSRT) (Pemovska et al., Cancer Discov. 2013) chemo-sensitivity profiling approach in which the responses of cells to 304 approved and investigational oncology drugs were explored. We studied a diverse panel of 15 TNBC cell lines in an attempt to functionally classify the TNBCs. To go beyond standard growth inhibition drug sensitivity testing we utilized a multiplexed cell viability and cytotoxicity readout allowing us to differentiate between cytostatic and cytotoxic drug responses. Evaluating both types of drug responses, several discoveries were made: First, the multiplexed viability and cytotoxicity readout identified several drug classes that previously had been assumed to have cytotoxic effects based on their strong effects on cell viability but in fact only showed cytostatic effects in most cells. Drug classes exhibiting this type of response included anti-mitotics, mTOR, CDK, metabolic inhibitors, as well as many targeted agents selectively inhibiting oncogenically activated pathways in individual cell lines. Second, within the responses to these broad cytostatic-acting classes of drugs, there were subsets of the cell lines that responded by cell death, suggesting that these cell lines may represent the TNBC subtypes that might best respond to the drugs in the clinic. To explore this further, we tested whether the cytostatic vs. cytotoxic responses could be linked to expression of protein biomarkers using a published TNBC cell line reverse-phase protein array data set (Daemen et al, Genome Biol. 2013). This led to the identification of potential predictive biomarkers such as PAI-1, MAPK phosphatase and Notch-3 levels linking to toxic responses to mitotic and proteasome inhibitors, suggesting that these could be explored as predictors for clinical responses for the drugs. Third, drug response based clustering of cell lines was very distinctive to transcriptomics based TNBC subgroups and recurrent mutation patterns, highlighting the considerable challenges in converting genetic and transcriptomic data to therapeutic responses. Fourth, the cytotoxicity readout highlighted effective synergistic drug combinations that were not apparent using cell viability readouts. For instance, we identified selective synergistic combinations of the MEK inhibitor trametinib with PARP inhibitors or the tyrosine kinase inhibitor ponatinib in DU4475 cells; as well as of the rapamycin analog everolimus with the ATP-competitive mTOR inhibitor dactolisib in several mTOR addicted cell lines. In conclusion, we showed that tracking drug-induced toxic responses in drug sensitivity testing and drug combination evaluation provides novel critical information on drug vulnerabilities. This argues that high throughput chemo-sensitivity profiling of cancer need to go beyond the current standard cell viability testing. Citation Format: Prson Gautam, Leena Karhinen, Agnieszka Szwajda, Sawan Kumar Jha, Bhagwan Yadav, Tero Aittokallio, Krister Wennerberg. Enhanced understanding of drug responses and drug-drug interactions in triple negative breast cancer cells with a multiplexed cell viability and cell death readout. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr B21.

Abstract B20: Transcriptional landscape of drug response guides the design of specific and potent drug combinations

Abstract Understanding responses to targeted agents is a key step toward the design of new therapeutic strategies that improve clinical cancer care. Here, we profiled the effects of a collection of kinase inhibitors using the L1000 transcriptional assay and combined these results with both phenotypic and biochemical response measurements to gain a more complete understanding of drug response. Using algorithms that reconstruct which signaling pathways are perturbed by specific kinase inhibitors, we identified potentially synergistic drug combinations and validated them experimentally. We treated six breast cancer cell lines with more than 100 targeted inhibitors at six doses and measured their transcriptional response at two time points. We focused on inhibitors targeting key the PI3K and MAPK signaling pathways, as well as receptor tyrosine kinases (RTKs) and cyclin-dependent kinases (CDKs); many of them are currently studied in clinical trials. We identified that 37% of the perturbations induce a significant difference in their gene expression profile based on the characteristic direction of the response. Clustering of signatures revealed they are time point specific: 3 hour signatures differ from the 24 hour ones. Some clusters contain perturbations from multiple cell lines, like CDK inhibitors that down regulate genes related to the cell cycle in all six lines. In contrast, clusters comprising inhibitors of the PI3K/AKT and MAPK pathways are specific to each cell line and pathway. The perturbations induced by RTK and non-RTK inhibitors cluster with either the PI3K or the MAPK inhibitors depending on the cell line. Thus, the transcriptional response allow us to identify differences in pathway usage between cell lines, in particular which RTK signals predominantly to the PI3K or the MAPK pathway. When we related transcriptional response to the growth inhibition after three days, we found that the strength of the transcriptional signature is not necessarily related to growth inhibition. In particular, we identified cases where inhibitors have little effect on growth, yet induce a significant transcriptional signature. The most striking case is the inhibition of MEK and EGFR in BT-20 that induces strong transcriptional and biochemical responses but only 20-30% of growth inhibition. Based on the transcriptional signature we inferred and validated experimentally that FoxO, which is generally regulated by the PI3K pathway, is partially activated following MEK or EGFR inhibition. This suggests that EGFR inhibitors and PI3K inhibitors act synergistically in BT-20, which we validated experimentally both at the level of FoxO activation and growth inhibition. We are currently verifying the most promising drug pair in xenografts. We have shown how we can use measurements of expression signatures and cellular phenotypes following single drug perturbations to identify drug combinations that are potent and specific to individual cell lines. This approach is a step toward the rational design of co-drugging strategies with differential effect and larger therapeutic windows. Citation Format: Marc Hafner, Mario Niepel, Qiaonan Duan, Evan Paull, Josh Stuart, Aravind Subramanian, Avi Ma'ayan, Peter Sorger. Transcriptional landscape of drug response guides the design of specific and potent drug combinations. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr B20.

Generation of Induced Pluripotent Stem Cells from Human Peripheral T Cells Using Sendai Virus in Feeder-free Conditions

Recently, iPSCs have attracted attention as a new source of cells for regenerative therapies. Although the initial method for generating iPSCs relied on dermal fibroblasts obtained by invasive biopsy and retroviral genomic insertion of transgenes, there have been many efforts to avoid these disadvantages. Human peripheral T cells are a unique cell source for generating iPSCs. iPSCs derived from T cells contain rearrangements of the T cell receptor (TCR) genes and are a source of antigen-specific T cells. Additionally, T cell receptor rearrangement in the genome has the potential to label individual cell lines and distinguish between transplanted and donor cells. For safe clinical application of iPSCs, it is important to minimize the risk of exposing newly generated iPSCs to harmful agents. Although fetal bovine serum and feeder cells have been essential for pluripotent stem cell culture, it is preferable to remove them from the culture system to reduce the risk of unpredictable pathogenicity. To address this, we have established a protocol for generating iPSCs from human peripheral T cells using Sendai virus to reduce the risk of exposing iPSCs to undefined pathogens. Although handling Sendai virus requires equipment with the appropriate biosafety level, Sendai virus infects activated T cells without genome insertion, yet with high efficiency. In this protocol, we demonstrate the generation of iPSCs from human peripheral T cells in feeder-free conditions using a combination of activated T cell culture and Sendai virus.

Generation of Induced Pluripotent Stem Cells from Human Peripheral T Cells Using Sendai Virus in Feeder-free Conditions.

Recently, iPSCs have attracted attention as a new source of cells for regenerative therapies. Although the initial method for generating iPSCs relied on dermal fibroblasts obtained by invasive biopsy and retroviral genomic insertion of transgenes, there have been many efforts to avoid these disadvantages. Human peripheral T cells are a unique cell source for generating iPSCs. iPSCs derived from T cells contain rearrangements of the T cell receptor (TCR) genes and are a source of antigen-specific T cells. Additionally, T cell receptor rearrangement in the genome has the potential to label individual cell lines and distinguish between transplanted and donor cells. For safe clinical application of iPSCs, it is important to minimize the risk of exposing newly generated iPSCs to harmful agents. Although fetal bovine serum and feeder cells have been essential for pluripotent stem cell culture, it is preferable to remove them from the culture system to reduce the risk of unpredictable pathogenicity. To address this, we have established a protocol for generating iPSCs from human peripheral T cells using Sendai virus to reduce the risk of exposing iPSCs to undefined pathogens. Although handling Sendai virus requires equipment with the appropriate biosafety level, Sendai virus infects activated T cells without genome insertion, yet with high efficiency. In this protocol, we demonstrate the generation of iPSCs from human peripheral T cells in feeder-free conditions using a combination of activated T cell culture and Sendai virus.

Inhibition of TNF-α protects in vitro brain barrier from ischaemic damage.

Cerebral ischaemia, associated with neuroinflammation and oxidative stress, is known to perturb blood-brain barrier (BBB) integrity and promote brain oedema formation. Using an in vitro model of human BBB composed of brain microvascular endothelial cells and astrocytes, this study examined whether suppression of TNF-α, a potent pro-inflammatory cytokine, might attenuate ischaemia-mediated cerebral barrier damage. Radical decreases in transendothelial electrical resistance and concomitant increases in paracellular flux across co-cultures exposed to increasing periods of oxygen-glucose deprivation alone (0.5-20 h) or followed by 20 h of reperfusion (OGD ± R) confirmed the deleterious effects of ischaemic injury on cerebral barrier integrity and function which concurred with reductions in tight junction protein (claudin-5 and occludin) expressions. OGD ± R elevated TNF-α secretion, NADPH oxidase activity, O2(-) production, actin stress fibre formation, MMP-2/9 activities and apoptosis in both endothelial cells and astrocytes. Increases in MMP-2 activity were confined to its extracellular isoform and treatments with OGD+R in astrocytes where MMP-9 could not be detected at all. Co-exposure of individual cell lines or co-cultures to an anti-TNF-α antibody dramatically diminished the extent of OGD ± R-evoked oxidative stress, morphological changes, apoptosis, MMP-2/9 activities while improving the barrier function through upregulation of tight junction protein expressions. In conclusion, vitiation of the exaggerated release of TNF-α may be an important therapeutic strategy in preserving cerebral integrity and function during and following a cerebral ischaemic attack.

Evaluation of Glucose Uptake in Normal and Cancer Cell Lines by Positron Emission Tomography

To date, there is no definitive demonstration of the utility of positron emission tomography (PET) in studying glucose metabolism in cultured cell lines. Thus, this study was designed to compare PET to more standardized methods for the quantitative assessment of glucose uptake in nontransformed and transformed living cells and to validate PET for metabolic studies in vitro. Human colon and breast carcinoma cell lines and mouse embryo fibroblasts were evaluated for [(18)F]fluorodeoxyglucose ([(18)F]FDG) uptake by PET and autoradiography and 2-deoxyglucose (2-DG) incorporation by colorimetric assay and analyzed for the radiotoxic effects of [(18)F]FDG and the expression levels of glucose transporters. Indeed, [(18)F]FDG incorporation on PET was comparable to [(18)F]FDG uptake by autoradiography and 2-DG incorporation by colorimetric assay, although radiotracer-based methods exhibited more pronounced differences between individual cell lines. As expected, these data correlated with glucose transporters 1 to 4 and hexokinase II expression in tumor cell lines and mouse fibroblasts. Notably, [(18)F]FDG incorporation resulted in low apoptotic rates, with fibroblasts being slightly more sensitive to radiotracer-induced cell death. The quantitative analysis of [(18)F]FDG uptake in living cells by PET represents a valuable and reproducible method to study tumor cell metabolism in vitro, being representative of the differences in the molecular profile of normal and tumor cell lines.

Lipid analysis of eight human breast cancer cell lines with ToF-SIMS.

In this work, four triple negative (TN) cell lines, three ER+ and PR+ receptor positive (RP) cell lines, and one ER+, PR+, and HER2+ cell line were chemically distinguished from one another using time-of-flight secondary ion mass spectrometry (ToF-SIMS) and principal component analysis (PCA). PCA scores separation was observed between the individual cell lines within a given classification (TN and RP) and there were distinctly different trends found in the fatty acid and lipid compositions of the two different classifications. These trends indicated that the RP cell lines separated out based on the carbon chain length of the lipids while the TN cell lines showed separation based on cholesterol-related peaks (in the positive ion data). Both cell types separated out by trends in fatty acid chain length and saturation in the negative ions. These chemical differences may be manifestations of unique metabolic processes within each of the different cell lines. Additionally, the HER2+ cell line was distinguished from three other RP cell types as having a unique distribution of fatty acids including anticorrelation to 18-carbon chain fatty acids. As these cell lines could not be grown in the same growth media, a combination of chemical fixation, rinsing, C60 (+) presputtering, and selection of cellular regions-of-interest is also presented as a successful method to acquire ToF-SIMS data from cell lines grown in different media.

Abstract POSTER-THER-1407: Cytotoxic natural human MAb against ovarian carcinoma

Abstract MAb216 is a human IgM mAb encoded in germline configuration by the immunoglobulin heavy chain VH4-34 gene. This antibody and many others derived from the VH4-34 heavy chain gene bind to long chain poly n-acetyl lactosamine (P-NAL) on the cell surface, referred to as the i/I antigen and found on normal fetal red blood cells and mature B cells. We have shown that VH4-34 IgM mAbs that are strong B cell binders also exhibit cytotoxicity to normal and malignant B cells. The mechanism of cytoxicity employed in the absence of complement involves crosslinking of the carbohydrate ligand leading to membrane disruption and loss of cell integrity accompanied by formation of unusual giant pore-like structures on the cell membranes as demonstrated by scanning electron and confocal microscopy. The IgM isotype also activates complement mediated cytotoxicity. Interestingly mAb cytoxicity is greater in vitro at 4 degrees without complement than at room temperature perhaps due to lack of membrane repair. MAb216 has been studied in a phase I trial of acute B cell leukemia, showing some efficacy and no grade 3 toxicity at the doses tested. Recently we found that mAb216 and several similar VH4-34 encoded IgM mAbs bound to the surface of a subpopulation of serous ovarian carcinoma (EOC) cells from individual tumors, ascites and cell lines. Reactivity was assayed using immunofluorescence of live cells or flow cytometry analysis or both. mAb216 was labeled with biotin or directly labeled with Alexa488. The punctate pattern was similar to that seen on B lymphocytes. Of 33 tumor or ascites samples assayed, mAb216 bound to a cell population on 29 specimens (87%). Reactivity ranged from 5-80% of the tumor cells depending on the individual patient. Endometroid (n=3) and mucinous (n=2) ovarian cancer were negative for staining. Three EOC cell lines, OVCAR3, OVCAR5, and Kuramochi show a percent of cells that expressed the P-NAL antigen. OVCAR4, OVCAR8 and SKOV3 did not bind mAb216. In vitro cytotoxicity experiments with ascites and OVCAR3 cell line showed that mAb alone could kill the percent of cells expressing antigen, and the cytotoxicity effects dramatically increased with human complement. Cell death was detected using the CCK-sk assay or flow cytometry counting with propidium iodide to detect dead cells. Like mAb216 cytotoxicity on B lymphocytes, 4 degree temperature enhanced cytotoxicity of OVCAR3 cells. Further, endo-beta-galactosidase treatment removed mAb216 binding and scanning electron microscopy of OVCAR3 showed the same membrane “pores” seen on B lymphocytes after mAb216 treatment. Fluorescence cell sorting of the bright positive population and negative population after mAb216 staining showed that the negative mAb216 population re-expressed the P-NAL antigen by 72 hours indicating that epitope expression is in someway cell cycle or proliferation related. Human VH4-34 IgM mAbs could therefore be used as a new therapy and a new target antigen in treatment of EOC. Preclinical studies are planned. Citation Format: Yi Chen, Marcia Bieber, Neelima Bhat, Nelson N.H. Teng. Cytotoxic natural human MAb against ovarian carcinoma [abstract]. In: Proceedings of the 10th Biennial Ovarian Cancer Research Symposium; Sep 8-9, 2014; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2015;21(16 Suppl):Abstract nr POSTER-THER-1407.

Abstract 5013: Toward better selection of PD-1 blockade responders through direct markers of adaptive immune resistance

Abstract Introduction. The role of PD-L1 as a predictive biomarker for response to PD-1 blockade is unclear. Emerging evidence, including our findings reported here, suggests that PD-L1 expression is important only when induced by the presence of T cells producing interferons, in what is termed adaptive immune resistance (Tumeh, Nature 2014). Therefore, studying PD-L1 expression in response or not to interferons may define molecular mechanisms to select patients whose tumors are incapable to respond to T cell infiltration by upregulating PD-L1. Experimental procedures. After the optimization process, we exposed 50 established melanoma cell lines to interferons (alpha, beta and gamma) and measured PD-L1, pSTAT1 (Y701), pSTAT3, pSTAT5, pSTAT6, NF-kB, pERK (1/2), pAKT, pP38, cleaved PARP, CyD3 expression level by flow cytometry (FACS LSRII) and PD-L1 as well as STAT1 transcript level was checked by quantitative real time PCR. Western blots were performed to further analyze interferon receptor signaling in selected cells lines. Results. Of 50 cell lines tested, 44 responded to at least one interferon ranging from 2 to 10-fold increase in PD-L1 compared to baseline expression, and 5 cell lines failed to respond to interferons with PD-L1 upregulation. Measurement of PD-L1 transcript level by qPCR as well as surface expression level by flow cytometry demonstrated that PD-L1 transcript level was up-regulated close to 20 fold by IFNγ in 30 minutes for two representatives of good PD-L1 responders. Phosphoflow experiments confirmed that 3 representative good responding cell lines had significant up-regulation of pSTAT1 compared to the baseline. However, among non-PD-L1 responding cell lines M412B showed significant change in pSTAT1, but WM1366 and M395 did not both, by phosphoflow and western blot analyses. M368 is a cell line that did not up-regulate PD-L1 upon exposure to interferon gamma, but it did up-regulate PD-L1 when exposed to type I interferons (alpha and beta). This cell line showed loss of JAK2 expression. WM1366 showed low JAK2 expression with loss of pJAK2 and SOCS1 overexpression, which is consistent with non-significant up-regulation of PD-L1 upon interferon treatment. Conclusions. PD-L1 is inducible in the majority of melanoma cell lines tested, but approximately 10% of them are not able to up-regulate PD-L1 expression upon interferon exposure. JAK-STAT pathway is critically important in regulation of inducible expression of PD-L1, and we have defined molecular defects in individual cell lines, which did not up-regulate PD-L1 expression. This information may be important to assess clinical data from PD-1/L1 blockade trials as to whether intact interferon receptor pathway inducible response may better predict the response of anti-PD-1/L1 immunotherapy than baseline PD-L1 levels. Citation Format: Daniel S. Shin, Angel Garcia-Diaz, Helena Escuin-Ordinas, Nicolaos J. Palaskas, Sara M. Komenan, Thomas G. Graeber, Begonya Comin-Anduix, Antoni Ribas. Toward better selection of PD-1 blockade responders through direct markers of adaptive immune resistance. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 5013. doi:10.1158/1538-7445.AM2015-5013

Abstract 1116: Oncogene signatures identify novel therapeutic targets and combinatorial strategies in breast cancer

Abstract During the course of tumorigenesis, cancer cells acquire a large number of genetic and epigenetic alterations. A subset of these alterations results in the generation of driver oncogenes that play critical roles in initiating and maintaining the transformed phenotype. We hypothesize that most solid tumors are simultaneously driven by multiple driver oncogenes that together form an oncogene network and that durable therapies for solid tumors will need to target multiple oncogenes. Identifying multiple driver oncogenes from among the large number of mutated/amplified genes present in most solid tumors is a nontrivial task that will require novel approaches. In this work, we have combined genome-wide, RNAi-based dropout screens with copy number, expression and exome sequencing analysis to identify the full complement of driver oncogenes that function in a panel of breast cancer cell lines. The panel of cell lines used for this study is the SUM panel of breast cancer cells which contains 10 cell lines and includes representatives of HER2 positive, ER positive, and triple negative breast cancer subtypes. For each line we have identified multiple activated driver oncogenes that when combined define a complete oncogene signature for the cell line. All of the oncogene signatures contain both well-characterized oncogenes that are known to play a role in breast cancer as well as novel driver genes that do not have a well-characterized role in breast cancer. Subsequent work has confirmed the role of several of the novel driver oncogenes including the epigenetic modifier KAT6A and the apoptotic regulator BCL2L1. Additionally, we show that the cell line oncogene signatures can be used to successfully predict sensitivity of individual cell lines to small molecule inhibitors. Predictions of sensitivity to inhibitors were confirmed through the integration of oncogene signatures with published data from large-scale drug screens across numerous cancer cell lines. Finally, we demonstrate the utility of oncogene signatures through their use in the design of novel therapeutic strategies that effectively reduce cancer cell viability. Citation Format: Stephen Guest, Zachary Kratche, Jonathan Irish, Aliccia Bollig-Fischer, Elizabeth Garrett-Mayer, Stephen Ethier. Oncogene signatures identify novel therapeutic targets and combinatorial strategies in breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1116. doi:10.1158/1538-7445.AM2015-1116

Metabolic rescue in pluripotent cells from patients with mtDNA disease.

Mitochondria have a major role in energy production via oxidative phosphorylation, which is dependent on the expression of critical genes encoded by mitochondrial (mt)DNA. Mutations in mtDNA can cause fatal or severely debilitating disorders with limited treatment options. Clinical manifestations vary based on mutation type and heteroplasmy (that is, the relative levels of mutant and wild-type mtDNA within each cell). Here we generated genetically corrected pluripotent stem cells (PSCs) from patients with mtDNA disease. Multiple induced pluripotent stem (iPS) cell lines were derived from patients with common heteroplasmic mutations including 3243A>G, causing mitochondrial encephalomyopathy and stroke-like episodes (MELAS), and 8993T>G and 13513G>A, implicated in Leigh syndrome. Isogenic MELAS and Leigh syndrome iPS cell lines were generated containing exclusively wild-type or mutant mtDNA through spontaneous segregation of heteroplasmic mtDNA in proliferating fibroblasts. Furthermore, somatic cell nuclear transfer (SCNT) enabled replacement of mutant mtDNA from homoplasmic 8993T>G fibroblasts to generate corrected Leigh-NT1 PSCs. Although Leigh-NT1 PSCs contained donor oocyte wild-type mtDNA (human haplotype D4a) that differed from Leigh syndrome patient haplotype (F1a) at a total of 47 nucleotide sites, Leigh-NT1 cells displayed transcriptomic profiles similar to those in embryo-derived PSCs carrying wild-type mtDNA, indicative of normal nuclear-to-mitochondrial interactions. Moreover, genetically rescued patient PSCs displayed normal metabolic function compared to impaired oxygen consumption and ATP production observed in mutant cells. We conclude that both reprogramming approaches offer complementary strategies for derivation of PSCs containing exclusively wild-type mtDNA, through spontaneous segregation of heteroplasmic mtDNA in individual iPS cell lines or mitochondrial replacement by SCNT in homoplasmic mtDNA-based disease.

Optimal drug prediction from personal genomics profiles.

Cancer patients often show heterogeneous drug responses such that only a small subset of patients is sensitive to a given anticancer drug. With the availability of large-scale genomic profiling via next-generation sequencing, it is now economically feasible to profile the whole transcriptome and genome of individual patients in order to identify their unique genetic mutations and differentially expressed genes, which are believed to be responsible for heterogeneous drug responses. Although subtyping analysis has identified patient subgroups sharing common biomarkers, there is no effective method to predict the drug response of individual patients precisely and reliably. Herein, we propose a novel computational algorithm to predict the drug response of individual patients based on personal genomic profiles, as well as pharmacogenomic and drug sensitivity data. Specifically, more than 600 cancer cell lines (viewed as individual patients) across over 50 types of cancers and their responses to 75 drugs were obtained from the genomics of drug sensitivity in cancer database. The drug-specific sensitivity signatures were determined from the changes in genomic profiles of individual cell lines in response to a specific drug. The optimal drugs for individual cell lines were predicted by integrating the votes from other cell lines. The experimental results show that the proposed drug prediction algorithm can be used to improve greatly the reliability of finding optimal drugs for individual patients and will, thus, form a key component in the precision medicine infrastructure for oncology care.

Synthesis of 3,3,3-trifluoroethyl isocyanate, carbamate and ureas. Anticancer activity evaluation of N-(3,3,3-trifluoroethyl)-N'-substituted ureas.

A new method is described for producing 3,3,3-trifluoroethyl isocyanate from perfluoroisobutene (PFIB). Isocyanate was used for synthesis of carbamates and ureas. A series of trifluoroethyl-substituted ureas has been tested in the National Cancer Institute (NCI, Bethesda, USA) by the NCI-60 DTP Human Tumor Cell Line Screening Program at a single high dose (10-5 M). The moderate anticancer activity was shown against some types of cancer on the individual human cell lines for leukemia, non-small cell lung cancer and renal cancer.

Abstract P4-07-05: Exosome-mediated trafficking of microRNAs by breast cancer cells

Abstract Introduction Cellular communication in the primary tumour micro-environment is known to play a key role in tumour development and progression. Exosomes are membrane-derived nanovesicles that are actively secreted by cells. Exosomes have been implicated in cell-to-cell communication through the transfer of genetic material including messenger RNA (mRNA), and more recently microRNA (miRNA). miRNAs are small non-coding RNAs approximately 22 nucleotides in length. They play an important role in posttranscriptional regulation of gene expression. Recent reports suggest exosome-mediated trafficking of microRNAs between cells. The aim of this study was to identify the panel of exosome-encapsulated microRNAs secreted by breast cancer cells in vitro. Methods Four breast cancer cell lines, MDA-MB-231, BT-20, Sk-Br-3 and T47D were cultured in exosome-depleted media for 48 hours. Exosomes secreted by the cells were isolated through a process of differential centrifugation, microfiltration and ultracentrifugation. The presence of exosomes was first confirmed by Transmission Electron Microscopy (TEM) and Western Blot analysis. Global miRNA array analysis of RNA extracted from exosomes was performed to identify the panel of miRNAs secreted by these cells. miRNA targets of interest were further validated using relative quantification PCR (RQ-PCR). Transfer of Red fluorescent protein (RFP)-labelled exosomes between cell populations was visualized using Confocal microscopy. Results TEM analysis of secreted exosomes revealed vesicular bodies of 40-100nm in size. Immunoblotting confirmed the presence of the exosome-associated protein CD63. MicroRNA array analysis of exosome fractions targeting 2089 miRNAs, revealed secretion of between 324 and 394 miRNAs by the cell lines. The miRNAs appeared to cluster in a biologically relevant fashion. 282 miRNAs were common to exosomes from all 4 cell lines, while a small selection were specific to individual cell lines. Exosome-mediated trafficking of miRNA targets of interest including miR-451a and miR-744-5p was successfully validated using RQ-PCR. Further validated targets included miR-10b, miR-145, miR-492 and miR-498, all of which have established roles in regulation of proliferation and apoptosis in cancer. Confocal microscopy supported visualization of miRNA-enriched exosome release from donor cells and subsequent uptake of the RFP-labelled exosomes by recipient cells. Conclusions A distinct panel of miRNAs are actively and selectively packaged into exosomes and secreted by breast cancer cells. This transfer of functional miRNAs between cells may play an important role in intercellular communication in the primary tumour microenvironment. The data presented also has potentially important implications in the identification of a circulating miRNA signature for breast cancer detection. Citation Format: Doireann P Joyce, Claire L Glynn, James Brown, Emma Holian, Peter Dockery, Michael J Kerin, Roisin M Dwyer. Exosome-mediated trafficking of microRNAs by breast cancer cells [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-07-05.

Non-viral generation of marmoset monkey iPS cells by a six-factor-in-one-vector approach.

Groundbreaking studies showed that differentiated somatic cells of mouse and human origin could be reverted to a stable pluripotent state by the ectopic expression of only four proteins. The resulting pluripotent cells, called induced pluripotent stem (iPS) cells, could be an alternative to embryonic stem cells, which are under continuous ethical debate. Hence, iPS cell-derived functional cells such as neurons may become the key for an effective treatment of currently incurable degenerative diseases. However, besides the requirement of efficacy testing of the therapy also its long-term safety needs to be carefully evaluated in settings mirroring the clinical situation in an optimal way. In this context, we chose the long-lived common marmoset monkey (Callithrix jacchus) as a non-human primate species to generate iPS cells. The marmoset monkey is frequently used in biomedical research and is gaining more and more preclinical relevance due to the increasing number of disease models. Here, we describe, to our knowledge, the first-time generation of marmoset monkey iPS cells from postnatal skin fibroblasts by non-viral means. We used the transposon-based, fully reversible piggyback system. We cloned the marmoset monkey reprogramming factors and established robust and reproducible reprogramming protocols with a six-factor-in-one-construct approach. We generated six individual iPS cell lines and characterized them in comparison with marmoset monkey embryonic stem cells. The generated iPS cells are morphologically indistinguishable from marmoset ES cells. The iPS cells are fully reprogrammed as demonstrated by differentiation assays, pluripotency marker expression and transcriptome analysis. They are stable for numerous passages (more than 80) and exhibit euploidy. In summary, we have established efficient non-viral reprogramming protocols for the derivation of stable marmoset monkey iPS cells, which can be used to develop and test cell replacement therapies in preclinical settings.

Heterogeneous tumor stromal microenvironments of oral squamous cell carcinoma cells in tongue and nodal metastatic lesions in a xenograft mouse model.

Oral squamous cell carcinoma exhibits a poor prognosis, caused by aggressive progression and early-stage metastasis to cervical lymph nodes. Here, we developed a xenograft mouse model to explore the heterogeneity of the tumor microenvironment that may govern local invasion and nodal metastasis of tumor cells. We transplanted five oral carcinoma cell lines into the tongues of nude mice and determined tongue tumor growth and micrometastatic dissemination by serially sectioning the tongue and lymph node lesions in combination with immunohistochemistry and computer-assisted image analysis. Our morphometric analysis enabled a quantitative assessment of blood and lymphatic endothelial densities in the intratumoral and host stromal regions. All cell lines tested were tumorigenic in mouse tongue. The metastatic lesion-derived carcinoma cell lines (OSC19, OSC20, and HSC2) yielded a 100% nodal metastasis rate, whereas the primary tumor-derived cell lines (KOSC2 and HO-1-u-1) showed <40% metastatic potential. Immunohistochemistry showed that the individual cell lines gave rise to heterogeneous tumor architecture and phenotypes and that their micrometastatic lesions assimilated the immunophenotypic properties of the corresponding tongue tumors. Notably, OSC19 and OSC20 cells shared similar aggressive tumorigenicity in both the tongue and lymph node environments but displayed markedly diverse immunophenotypes and gene expression profiles. Our model facilitated comparing the tumor microenvironments in tongue and lymph node lesions. The results support that tumorigenicity and tumor architecture in the host tongue environment depend on the origin and properties of the carcinoma cell lines and that metastatic progression may take place through heterogeneous tumor-host interactions.

Mechanisms of hormonal therapy resistance in breast cancer.

Whilst estrogen receptor (ER)-positive breast cancers are preferentially treated with hormone therapy, approximately one-third of them relapse. The mechanisms of refractoriness have been investigated by numerous studies but have not been fully clarified. Hormonal therapy resistance, particularly aromatase inhibitor (AI) resistance, may be related to the acquisition of alternative intracellular ER signaling. We have been investing the mechanisms using cancer specimens and cell lines by monitoring the transcription activity of ERs. AI refractory specimens showed diverse ER activity in the adenovirus estrogen receptor element-green fluorescent protein (ERE-GFP) assay and varied sensitivity to anti-estrogens, indicating the existence of multiple resistant mechanisms. We established six different types of cell lines mimicking AI resistance from ERE-GFP-introduced ER-positive cell lines. They revealed that multiple and alternative ER activating pathways were involved in the resistance, such as phosphorylation-dependent or androgen metabolite-dependent mechanisms. The response to fulvestrant and mammalian target of rapamycin inhibitor also varied among individual resistant cell lines. These results indicate that further subclassification of ER-positive breast cancer is extremely important to decide the therapeutic management of not only hormonal therapy but also new molecular target therapy.

MACE: mutation-oriented profiling of chemical response and gene expression in cancers

The mutational status of specific cancer lineages can affect the sensitivity to or resistance against cancer drugs. The MACE database provides web-based interactive tools for interpreting large chemical screening and gene expression datasets of cancer cell lines in terms of mutation and lineage categories. GI50 data of chemicals against individual NCI60 cell lines were normalized and organized to statistically identify mutation- or lineage-specific chemical responses. Similarly, DNA microarray data on NCI60 cell lines were processed to analyze mutation- or lineage-specific gene expression signatures. A combined analysis of GI50 and gene expression data to find potential associations between chemicals and genes is also a capability of this system. This database will provide extensive, systematic information to identify lineage- or mutation-specific anticancer agents and related gene targets. The MACE web database is available at http://mace.sookmyung.ac.kr/. Supplementary data are available at Bioinformatics online. yoonsj@sookmyung.ac.kr.

The Novel Plk Inhibitor Volasertib Overcomes Cytarabine Resistance in Acute Myeloid Leukemia

Drug resistance is a major cause of treatment failure for patients with acute myeloid leukemia (AML) and novel approaches that circumvent or overcome resistance mechanisms are critically needed. Polo-like kinase 1 (PLK1) is a serine/threonine kinase that plays an important role in regulating cell cycle progression in M phase. It is overexpressed in a broad range of solid and hematological malignancies. Volasertib (BI 6727) is a novel PLK inhibitor that is currently under investigation in a Phase III clinical trial for patients with AML. Despite its reported clinical anti-AML activity, the efficacy and pharmacodynamics (PD) of volasertib have not been rigorously investigated in AML models in the preclinical setting. Here, we studied the activity and PD effects of volasertib in a panel of human AML cell lines and primary patient specimens. Low nanomolar concentrations of volasertib potently disrupted the viability and clonogenic survival of AML cells. Notably, the specific cytogenetic background of individual cell lines did not significantly impact cellular sensitivity to volasertib. Similarly, primary AML blasts from patients demonstrated ex vivo sensitivity to volasertib regardless of their cytogenetic and molecular features or prior treatment history. This in vitro data set is in line with recent Phase II clinical data in AML patients showing efficacy of volasertib across all cytogenetic risk categories. Volasertib triggered caspase-dependent apoptosis in AML cells in a manner that was preceded by disruption of the spindle assembly checkpoint. Notably, analysis of the activity of volasertib in three different paired cytarabine-sensitive and cytarabine-resistant AML cell line models revealed that volasertib retained its ability to diminish viability and stimulate apoptosis in cells with three log-fold increased cytarabine resistance. These findings suggest that volasertib may be effective for patients that have relapsed or are refractory to cytarabine-based regimens. Combination studies showed that volasertib and cytarabine yielded additive effects in cytarabine-sensitive cells. Gene expression profiling studies demonstrated that the chemokine CXCL10 was the most significantly upregulated gene following short-term treatment with low concentrations of volasertib. Subsequent qRT-PCR and ELISA experiments confirmed drug-induced CXCL10 expression and elevated CXCL10 secretion by AML cells. To our knowledge, this is the first report linking PLK1 inhibition to alterations in CXCL10 expression and secretion. We are currently investigating the specific contribution of CXCL10 to the anti-leukemic activity of volasertib and are elucidating the molecular mechanisms that mediate this effect. Collectively, our findings demonstrate that inhibition of PLK1 with volasertib is a promising new strategy for the treatment of AML that is able to overcome cytarabine resistance. DisclosuresNo relevant conflicts of interest to declare.

Abstract 2099: Cellular and molecular determinants of retinoic acid sensitivity in breast cancer

Abstract All-trans-retinoic acid (ATRA) is the primary metabolite of vitamin A and a promising agent in the treatment/prevention of solid tumors, including breast cancer. Breast cancer is a very heterogeneous disease that can be classified according to the global gene-expression profiles into various subgroups. A rational use of ATRA in the clinics requires definition of the breast cancer subtypes characterized by sensitivity/natural-resistance to this compound and derived retinoids. In addition, it is fundamental to define the cellular and molecular mechanisms underlying the sensitivity/resistance to ATRA. As a first step in this direction, 44 breast cancer cell lines, belonging to the Cancer Cell Line Encyclopedia (CCLE) of the Broad Institute and recapitulating the heterogeneity of this tumor, were evaluated for their sensitivity to the anti-proliferative activity of ATRA. Individual cell lines were challenged with increasing concentrations of the retinoid for 3, 6 and 9 days. The concentrations of ATRA causing 50% of the maximal growth inhibition (IC50) and the maximal effect observed were calculated. The two parameters, along with the duplication time of the cell lines, were used to define a novel algorithm (ATRA SCORE). The ATRA score is a continuous parameter which defines the sensitivity of the cell lines to ATRA in a quantitative fashion. The lower the ATRA score the higher the sensitivity of the cell to the anti-proliferative action of ATRA. Using this score, an enrichment in cell lines characterized by a luminal phenotype and/or estrogen receptor (ER)-positivity was observed in the group showing sensitivity to ATRA. In contrast an enrichment in cell lines with a basal phenotype and/or ER-negativity was evident in the group characterized by natural resistance to the retinoid. Measurement of the mRNAs corresponding to various isoforms of the retinoid receptors belonging to the RAR and RXR family in the 44 cell lines allowed us to establish correlations between the expression of these transcripts and ATRA sensitivity. With respect to this, we found a significant and direct correlation with the expression levels of RARalpha1. This indicates that the receptor represents a determinant and a predictor of ATRA- sensitivity in breast cancer cells. The correlation was confirmed also at the protein level using specific RARalpha antibodies. Most of these findings were validated with the use of short-term cultures of breast cancer tissue slices obtained from surgical specimens. The availability of the global gene-expression profiles of each cell line permitted the construction of a gene fingerprint consisting of approximately 130 elements that has the potential to predict sensitivity to ATRA not only in breast cancer cell lines, but also in tumor samples. In addition, the identified gene fingerprint allowed the identification of various and unexpected molecular determinants of sensitivity which do not belong to the classic retinoid pathway. Citation Format: Enrico Garattini, Floriana Centritto, Silvio K. Garattini, Gabriela Paroni, Marco Bolis, Maddalena Fratelli, Mineko Terao. Cellular and molecular determinants of retinoic acid sensitivity in breast cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2099. doi:10.1158/1538-7445.AM2014-2099