Sara Kangaspeska

Transient cyclical methylation of promoter DNA.

Methylation of CpG dinucleotides is generally associated with epigenetic silencing of transcription and is maintained through cellular division. Multiple CpG sequences are rare in mammalian genomes, but frequently occur at the transcriptional start site of active genes, with most clusters of CpGs being hypomethylated. We reported previously that the proximal region of the trefoil factor 1 (TFF1, also known as pS2) and oestrogen receptor α (ERα) promoters could be partially methylated by treatment with deacetylase inhibitors, suggesting the possibility of dynamic changes in DNA methylation. Here we show that cyclical methylation and demethylation of CpG dinucleotides, with a periodicity of around 100 min, is characteristic for five selected promoters, including the oestrogen (E2)-responsive pS2 gene, in human cells. When the pS2 gene is actively transcribed, DNA methylation occurs after the cyclical occupancy of ERα and RNA polymerase II (polII). Moreover, we report conditions that provoke methylation cycling of the pS2 promoter in cell lines in which pS2 expression is quiescent and the proximal promoter is methylated. This coincides with a low-level re-expression of ERα and of pS2 transcripts.

FusionCatcher - a tool for finding somatic fusion genes in paired-end RNA-sequencing data

FusionCatcher is a software tool for finding somatic fusion genes in paired-end RNA-sequencing data from human or other vertebrates. FusionCatcher achieves competitive detection rates and real-time PCR validation rates in RNA-sequencing data from tumor cells. Fusion-Catcher is available at http://code.google.com/p/fusioncatcher/.

Epigenetic reprogramming of HOXC10 in endocrine-resistant breast cancer.

Genome-wide screen identifies methylation of the estrogen-repressed HOXC10 gene as a determinant of resistance to aromatase inhibitors in breast cancer. Playing Tug-of-War with HOXC10 Aromatase inhibitors are drugs that prevent androgens from being converted into estrogen, and they are frequently used to treat breast cancers that express the estrogen receptor. Unfortunately, some patients’ tumors never respond to these drugs, and others gradually become resistant over time. Although the development of resistance to aromatase inhibitors has been investigated in some previous studies and some potential mechanisms have been proposed, much about this process remains unknown. Pathiraja and colleagues began by performing a genome-wide methylation screen in breast cancer cells, which identified the developmental gene HOXC10 as a target of epigenetic silencing in the context of long-term estrogen withdrawal. When HOXC10 is active, it interferes with proliferation and can stimulate apoptosis, but estrogen suppresses its activity, thereby promoting tumor growth. By decreasing estrogen production, aromatase inhibitors up-regulate HOXC10, accounting for some of their antitumor activity. However, long-term estrogen deprivation eventually has the opposite effect, leading to methylation of HOXC10 and its long-term suppression even in the absence of estrogen. These findings suggest that a rational approach for overcoming aromatase resistance in breast cancer may involve the addition of demethylating drugs to overcome the methylation of HOXC10 and take advantage of its antitumor effects, although this remains to be demonstrated directly. Resistance to aromatase inhibitors (AIs) is a major clinical problem in the treatment of estrogen receptor (ER)–positive breast cancer. In two breast cancer cell line models of AI resistance, we identified widespread DNA hyper- and hypomethylation, with enrichment for promoter hypermethylation of developmental genes. For the homeobox gene HOXC10, methylation occurred in a CpG shore, which overlapped with a functional ER binding site, causing repression of HOXC10 expression. Although short-term blockade of ER signaling caused relief of HOXC10 repression in both cell lines and breast tumors, it also resulted in concurrent recruitment of EZH2 and increased H3K27me3, ultimately transitioning to increased DNA methylation and silencing of HOXC10. Reduced HOXC10 in vitro and in xenografts resulted in decreased apoptosis and caused antiestrogen resistance. Supporting this, we used paired primary and metastatic breast cancer specimens to show that HOXC10 was reduced in tumors that recurred during AI treatment. We propose a model in which estrogen represses apoptotic and growth-inhibitory genes such as HOXC10, contributing to tumor survival, whereas AIs induce these genes to cause apoptosis and therapeutic benefit, but long-term AI treatment results in permanent repression of these genes via methylation and confers resistance. Therapies aimed at inhibiting AI-induced histone and DNA methylation may be beneficial in blocking or delaying AI resistance.

E2-mediated cathepsin D (CTSD) activation involves looping of distal enhancer elements

Estrogen receptor alpha (ERα) is a ligand dependent transcription factor that regulates the expression of target genes through interacting with cis‐acting estrogen response elements (EREs). However, only a minority of ERα binding sites are located within the proximal promoter regions of responsive genes. Here we report the characterization of an ERE located 9kbp upstream of the TSS of the cathepsin D gene (CTSD) that up‐regulates CTSD expression upon estrogen stimulation in MCF‐7 cells. Using ChIP, we show recruitment of ERα and phosphorylated PolII at the CTSD distal enhancer region. Moreover, we determine the kinetics of transient CpG methylation on the promoter region of CTSD and for the first time, at a distal enhancer element. We show that ERα is crucial for long‐distance regulation of CTSD expression involving a looping mechanism.

Reanalysis of RNA-Sequencing Data Reveals Several Additional Fusion Genes with Multiple Isoforms

RNA-sequencing and tailored bioinformatic methodologies have paved the way for identification of expressed fusion genes from the chaotic genomes of solid tumors. We have recently successfully exploited RNA-sequencing for the discovery of 24 novel fusion genes in breast cancer. Here, we demonstrate the importance of continuous optimization of the bioinformatic methodology for this purpose, and report the discovery and experimental validation of 13 additional fusion genes from the same samples. Integration of copy number profiling with the RNA-sequencing results revealed that the majority of the gene fusions were promoter-donating events that occurred at copy number transition points or involved high-level DNA-amplifications. Sequencing of genomic fusion break points confirmed that DNA-level rearrangements underlie selected fusion transcripts. Furthermore, a significant portion (>60%) of the fusion genes were alternatively spliced. This illustrates the importance of reanalyzing sequencing data as gene definitions change and bioinformatic methods improve, and highlights the previously unforeseen isoform diversity among fusion transcripts.

Consistency in drug response profiling

The comparative analysis by Haibe-Kains et al.1 concluded that data from two large-scale studies of cancer cell lines2,3 showed highly discordant results for drug sensitivity measurements, whereas gene expression data were reasonably concordant. Here, we crosscompared the two original datasets2,3 against our own data of drug response profiles in overlapping cancer cell line panels. Our results indicate that it is possible to achieve concordance between different laboratories for drug response measurements by paying attention to the harmonization of assays and experimental procedures. There is a Reply to this Comment by Safikhani, Z. et al. Nature 540, http://dx.doi.org/10.1038/ nature20172 (2016). Haibe-Kains et al.1 reported on a comparative evaluation of two drug sensitivity and molecular profiling datasets, one from the Cancer Genome Project (CGP)2 and the other from the Cancer Cell Line Encyclopedia (CCLE)3. In their analyses, gene expression profiles between hundreds of common cancer cell lines across all genes showed high consistency between the two studies (median rank correlation (MRC) = 0.85), whereas the drug response data for 15 common compounds were highly discordant (MRC = 0.28 for halfmaximum inhibitory concentration (IC50) values). This report1 and the accompanying commentary4 suggested that differences in laboratory protocols, compounds and their tested concentration ranges, and computational methods may account for the differences, but these reports did not elaborate which of these factors are important and whether they can be controlled for. Here, we reanalysed the dose–response data from both CGP and CCLE using a standardized area under the curve (AUC) response metric, which we call the drug sensitivity score (DSS)5. We then compared the CGP and CCLE data with a new dataset of drug responses profiled using the Institute for Molecular Medicine Finland (FIMM) compound testing assay6, covering 308 drugs across 106 cancer cell lines. The FIMM data included 45 compounds in common with CGP and 14 with the CCLE in 50 cell lines (Supplementary Data 1). In the AUC calculation, we unified the drug concentration ranges across the CGP, CCLE and FIMM assays. We observed a significantly higher level of consistency (P = 4.2 × 10−5), especially between the CCLE and FIMM drug response data (MRC = 0.74), as compared to the consistency between FIMM and CGP data (MRC = 0.54) (Fig. 1a). Similar experimental protocols were applied at FIMM and CCLE, including the same readout (CellTiter-Glo, Promega), similar controls (vehicle as negative control and positive controls of toxic compounds 100 μ M benzethonium chloride or 1 μ M MG132). However, there were also differences, such as the plate format used (1,536 versus 384 wells). Importantly, there was no effort made to standardize cell numbers used or any other parameters between the three laboratories, such as the source, passage number and media used for cells, nor the origin and handling of drugs. Therefore, this observed level of drug response agreement could be substantially improved by further standardization of the laboratory protocols. The CGP experimental protocol differed from the two others in terms of the readout (fluorescent nucleic acid stain Syto 60, Life Technologies), in the use of controls (drug-free cells as negative and no cells as positive controls), and the plate format used (96or 384-well plates). We compared the drug response profiles between the same cell lines from different laboratories, in line with the approach of Haibe-Kains et al.1, in which they showed consistency in gene expression profiles from CGP and CCLE (MRC = 0.85)1. The Haibe-Kains et al.1 approach, in which the correlation is calculated for each drug separately across the cell lines, showed more variability (Fig. 1b), owing to the fact that some drugs show minimal efficacy in all the tested cell lines. Analogously, gene expression correlations vary more widely when analysed at the level of genes across cell lines (MRC = 0.58 between CGP and CCLE), as certain genes are not expressed above technical noise. Although both ways to compare the data are relevant to the overall goal of personalized

KRAS Oncogene Rearrangements and Gene fusions: Unexpected Rare encounters in Late-stage Prostate Cancers

Wang and colleagues identify a fusion between UBE2L3 and KRAS in a subset of metastatic prostate cancers.

Association of tamoxifen resistance and lipid reprogramming in breast cancer

BackgroundTamoxifen treatment of estrogen receptor (ER)-positive breast cancer reduces mortality by 31%. However, over half of advanced ER-positive breast cancers are intrinsically resistant to tamoxifen and about 40% will acquire the resistance during the treatment.MethodsIn order to explore mechanisms underlying endocrine therapy resistance in breast cancer and to identify new therapeutic opportunities, we created tamoxifen-resistant breast cancer cell lines that represent the luminal A or the luminal B. Gene expression patterns revealed by RNA-sequencing in seven tamoxifen-resistant variants were compared with their isogenic parental cells. We further examined those transcriptomic alterations in a publicly available patient cohort.ResultsWe show that tamoxifen resistance cannot simply be explained by altered expression of individual genes, common mechanism across all resistant variants, or the appearance of new fusion genes. Instead, the resistant cell lines shared altered gene expression patterns associated with cell cycle, protein modification and metabolism, especially with the cholesterol pathway. In the tamoxifen-resistant T-47D cell variants we observed a striking increase of neutral lipids in lipid droplets as well as an accumulation of free cholesterol in the lysosomes. Tamoxifen-resistant cells were also less prone to lysosomal membrane permeabilization (LMP) and not vulnerable to compounds targeting the lipid metabolism. However, the cells were sensitive to disulfiram, LCS-1, and dasatinib.ConclusionAltogether, our findings highlight a major role of LMP prevention in tamoxifen resistance, and suggest novel drug vulnerabilities associated with this phenotype.

Abstract 2935: Systematic drug testing and RNA-sequencing of tamoxifen resistant breast cancer cell lines

Tamoxifen, as a standard treatment of estrogen receptor (ER)-positive breast cancer, reduces breast cancer mortality by 31%. In 50% of advanced ER-positive cancers, however, de novo resistance exists and in 40% of patients with initial response acquired resistance frequently evolves. In order to explore mechanisms underlying endocrine therapy resistance in breast cancer and to identify new opportunities to treat patients, we created seven tamoxifen-resistant breast cancer cell line models that represent the luminal A subtype (MCF-7, T-47D, ZR-75-1), which expresses ER, and the luminal B subtype (BT-474), which additionally expresses the HER2 oncoprotein. We then performed drug sensitivity and resistance testing (DSRT), exome-sequencing and network analysis on all these cancer cell lines to determine the molecular and drug response profiles specific to tamoxifen resistance. As the cells became tamoxifen-resistant, we observed increasing sensitivity towards drugs like ERK1/2-, proteasome- and BCL-family inhibitors, but each of the isogenic cell line pairs had its distinct genomic and drug response profile. We then studied the molecular profiles of the 7 drug-resistant variants by RNA-sequencing in comparison to their 4 isogenic parental cells. We could not detect any common significantly differentially expressed genes (more than 2 fold change) across all the cell lines. However, the cell lines could be grouped into two different categories: The ones with low amount of differentially expressed genes, 1800 genes in T-47D and MCF-7 (with 7,3% overlap). Further analysis with the Ingenuity pathway analysis tool revealed an involvement of “Fatty Acid Activation” as well as “Stearate Biosynthesis I” in this group. Additionally, we discovered that genes involved in iron metabolism (TFRC, IREB2 and FTL) or iron-regulated genes like CP and NDRG1 are deregulated. These genes and pathways thereby provide avenues to identify new drug vulnerabilities for the tamoxifen resistant cancer cells, which we are now investigating at gene and protein level e.g. by image-based phenotyping. In summary, by combining drug testing data with the RNA-sequencing results, we hope to provide a number of potential drugs as well as matching biomarkers for planning clinical trials for patients with tamoxifen-resistant breast cancer. Citation Format: Susanne Hultsch, Sara Kangaspeska, Matti Kankainen, Vilja Pietiainen, Olli Kallioniemi. Systematic drug testing and RNA-sequencing of tamoxifen resistant breast cancer cell lines. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2935.

Abstract 4184: Drug set enrichment analysis : A computational approach to identify functional drug sets

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Most drug-testing approaches published so far focus on identifying a single drug that shows favorable response and is associated with a known cancer biomarker such as the drug Imatinib in BCR-ABL gene fusion positive cells. We developed and applied drug set enrichment analysis (DSEA) to find enriched patterns or statistically significant similarities (overlaps) between the drug responses of a test sample against a cohort of 182 previously screened cancer samples. The samples studied included established (ATCC) cancer cell lines, drug-resistant cancer cell models, ex-vivo patient cancer cells in primary cultures, including conditionally reprogrammed cancer cells from patients. DSEA is adopting Gene Set Enrichment Analysis statistics commonly used for gene expression analysis to high throughput drug testing data. Our drug screening (Pemovska et al., Cancer Discovery, 2013) was done with a panel of 306 established (FDA-approved) and emerging targeted cancer drugs such as tyrosine-kinase inhibitors (e.g. EGFR, PDGFR, BRAF, MET), S/T-type inhibitors, (e.g. MEK, Plk1, Akt, Aurora, Chk1), and inhibitors of other pathways (HDACs, Hh, BCL2, PI3K, PARP) and many others. The readout was based on viability of cells after a 72 hour culture. The DSEA approach is based on taking the top most sensitive drugs (above a defined sensitivity score cut-off) in an individual cancer sample and then identifying overlapping drug response profiles in previously screened reference samples. Our hypothesis is that the most sensitive drug sets in any given sample tend to show similar response profiles in a cohort of similar samples. We convey the correlations and drug set enrichment analysis results as dendrogram trees, plots and tables with enrichment and significance scores. Interestingly, our results show that clustering of drug sensitivity testing data does not place all cancer cell line samples within well-established subgroups based on biological features or histological origin. We find a similar tendency in ex vivo patient samples. Therefore, comprehensive drug response profiles seen may reveal novel biological data that reflect pharmacologically-relevant, phenotypic cancer cell states. DSEA could also provide novel means to subtype previously poorly characterized cancer samples based on their drug response profiles and thereby in the future facilitate the choice of therapies to patients whose cancers repond in an atypical way as compared to the expectations based on anatomical origin or genomic composition. Citation Format: John Patrick Mpindi, Dimitry Bychkov, Yadav Bhagwan, Disha Malani, Hirasawa Akira, Khalid Saeed, Susanne Hultsch, Sara Kangaspeska, Astrid Murumagi, Caroline A Heckman, Kimmo Porkka, Tero Aittokallio, Krister Wennerberg, Paivi Ostling, Olli Kallioniemi. Drug set enrichment analysis : A computational approach to identify functional drug sets. [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 4184. doi:10.1158/1538-7445.AM2014-4184