Abstract IA21: Tumor heterogeneity and cell-of-origin of mouse small cell and non-small cell lung cancer

Small cell lung cancer (SCLC) is one of the most lethal human malignancies, due to its high metastatic potential and chemo-resistance upon relapse. Using the Rbf/f;p53f/f mouse model for SCLC, we found that the tumors are often composed of phenotypically different cells, characterized by mesenchymal and neuroendocrine markers. These cells often share a common origin. Crosstalk between these cells can endow the neuroendocrine component with metastatic capacity, illustrating the potential relevance of tumor cell heterogeneity in dictating functional tumor properties. Also specific genetic lesions appear to be associated with metastatic potential.

Interestingly, the two cell types can also interconvert raising the question of their cell-of-origin. To investigate this in more detail, we inactivated Trp53 and Rb1 in distinct cell types in the adult lung by targeting Cre-recombinase expression to Clara, neuroendocrine, and alveolar type II cells using adenoviral vectors. We could show that neuroendocrine cells serve as the predominant although not the exclusive cell of origin of SCLC.

In contrast, mutant Kras-driven adenocarcinomas (one of the NSCLC subtypes) originates primarily from alveolar type II cells. However, in LSL-mutant-Kras;p53flox/flox mice also other cell types gave effectively rise to adenomas and adenocarcinomas. Our data indicate that both cell type specific features and the nature of the oncogenic lesion(s) are critical factors in determining the tumor initiating capacity of lung (progenitor) cells. Furthermore, the cell-of-origin appears to influence the malignant properties of the resulting tumors.

Citation Format: Anton Berns. Tumor heterogeneity and cell-of-origin of mouse small cell and non-small cell lung cancer. [abstract]. In: Proceedings of the AACR Special Conference: The Translational Impact of Model Organisms in Cancer; Nov 5-8, 2013; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2014;12(11 Suppl):Abstract nr IA21.

Improved pregnancy and birth rates with routine application of nonsurgical embryo transfer

Nonsurgical embryo transfer (NSET) of blastocysts to pseudopregnant female recipients provides many benefits over surgical implantation with less distress for the mice, no anesthesia or analgesia required and a considerable reduction in implantation time per mouse. Although a disposable device to perform NSET is on the market since 2009, it is not generally used in transgenic facilities, most likely because surgical implantation is efficient and inexpensive. Here, we report that with several refinements to the original protocol, the NSET method becomes very attractive and outperforms the traditional surgical transfer on basis of pregnancy rate, birth rate and implantation-related discomfort. Furthermore, repeated use of the same NSET device on several recipient females reduces the costs to a reasonable level. The data presented covers all embryo transfers over the last 5 years at the transgenic facility of the Netherlands Cancer Institute, of which the last 2 years were performed exclusively with NSET.

Chromatin landscapes of retroviral and transposon integration profiles

The ability of retroviruses and transposons to insert their genetic material into host DNA makes them widely used tools in molecular biology, cancer research and gene therapy. However, these systems have biases that may strongly affect research outcomes. To address this issue, we generated very large datasets consisting of to unselected integrations in the mouse genome for the Sleeping Beauty (SB) and piggyBac (PB) transposons, and the Mouse Mammary Tumor Virus (MMTV). We analyzed (epi)genomic features to generate bias maps at both local and genome-wide scales. MMTV showed a remarkably uniform distribution of integrations across the genome. More distinct preferences were observed for the two transposons, with PB showing remarkable resemblance to bias profiles of the Murine Leukemia Virus. Furthermore, we present a model where target site selection is directed at multiple scales. At a large scale, target site selection is similar across systems, and defined by domain-oriented features, namely expression of proximal genes, proximity to CpG islands and to genic features, chromatin compaction and replication timing. Notable differences between the systems are mainly observed at smaller scales, and are directed by a diverse range of features. To study the effect of these biases on integration sites occupied under selective pressure, we turned to insertional mutagenesis (IM) screens. In IM screens, putative cancer genes are identified by finding frequently targeted genomic regions, or Common Integration Sites (CISs). Within three recently completed IM screens, we identified 7%–33% putative false positive CISs, which are likely not the result of the oncogenic selection process. Moreover, results indicate that PB, compared to SB, is more suited to tag oncogenes.

Retroviruses and transposons are widely used in cancer research and gene therapy. However, these systems show integration biases that may strongly affect results. To address this issue, we generated very large datasets consisting of to unselected integrations for the Sleeping Beauty and piggyBac transposons, and the Mouse Mammary Tumor Virus (MMTV). We analyzed (epi)genomic features to generate bias maps at local and genome-wide scales. MMTV showed a remarkably uniform distribution of integrations across the genome, and a striking similarity was observed between piggyBac and the Murine Leukemia Virus. Moreover, we find that target site selection is directed at multiple scales. At larger scales, it is similar across systems, and directed by a set of domain-oriented features, including chromatin compaction, replication timing, and CpG islands. Notable differences between systems are defined at smaller scales by a diverse range of epigenetic features. As a practical application of our findings, we determined that three recent insertional mutagenesis screens - commonly used for cancer gene discovery - contained 7%–33% putative false positive integration hotspots.

Rapid target gene validation in complex cancer mouse models using re-derived embryonic stem cells

Human cancers modeled in Genetically Engineered Mouse Models (GEMMs) can provide important mechanistic insights into the molecular basis of tumor development and enable testing of new intervention strategies. The inherent complexity of these models, with often multiple modified tumor suppressor genes and oncogenes, has hampered their use as preclinical models for validating cancer genes and drug targets. In our newly developed approach for the fast generation of tumor cohorts we have overcome this obstacle, as exemplified for three GEMMs; two lung cancer models and one mesothelioma model. Three elements are central for this system; (i) The efficient derivation of authentic Embryonic Stem Cells (ESCs) from established GEMMs, (ii) the routine introduction of transgenes of choice in these GEMM-ESCs by Flp recombinase-mediated integration and (iii) the direct use of the chimeric animals in tumor cohorts. By applying stringent quality controls, the GEMM-ESC approach proofs to be a reliable and effective method to speed up cancer gene assessment and target validation. As proof-of-principle, we demonstrate that MycL1 is a key driver gene in Small Cell Lung Cancer.

Chromatin position effects assayed by thousands of reporters integrated in parallel

Reporter genes integrated into the genome are a powerful tool to reveal effects of regulatory elements and local chromatin context on gene expression. However, so far such reporter assays have been of low throughput. Here, we describe a multiplexing approach for the parallel monitoring of transcriptional activity of thousands of randomly integrated reporters. More than 27,000 distinct reporter integrations in mouse embryonic stem cells, obtained with two different promoters, show ∼1,000-fold variation in expression levels. Data analysis indicates that lamina-associated domains act as attenuators of transcription, likely by reducing access of transcription factors to binding sites. Furthermore, chromatin compaction is predictive of reporter activity. We also found evidence for crosstalk between neighboring genes and estimate that enhancers can influence gene expression on average over ∼20 kb. The multiplexed reporter assay is highly flexible in design and can be modified to query a wide range of aspects of gene regulation.

Genomic characterization and high-throughput therapeutic screening of malignant mesothelioma to reveal novel tumor dependencies

7581

Background: Clinical trials of targeted therapeutics in mesothelioma have demonstrated limited efficacy. It has been suggested that combinations of targeted agents may be more effective in this disease, but given the unknown status of tractable oncogenic mutations it has been difficult to ascertain which key signalling nodes drive these tumours that may lend themselves to therapeutic intervention. We therefore aimed to characterise a representative panel of mesothelioma cell lines at the genomic level, and then determine critical “nodes” utilising a high throughput screen of targeted agents. Methods: 19 mesothelioma cell lines and 6 mesothelioma primary tumour early passage lines underwent Illumina whole exome sequencing and copy number analysis. In parallel a high throughput screen was performed utilising a panel of targeted therapeutics enabling each mesothelioma to be screened across 48 compounds. Efficacy was confirmed in 3D spheroid culture. Results: Exome sequencing of the mesothelioma panel revealed mutations in tumour suppressor genes previously described in mesothelioma including NF2, as well as previously unreported mutations in this disease affecting histone modifying genes MLL2 and SETD2. Notably an absence of mutated “driver” oncogenes was observed. High throughput screening demonstrated limited activity for most small molecule inhibitors as single agents. However, despite an absence of mutations in PIK3CAor related genes, PI3K/mTORC inhibition had the broadest single agent efficacy. Conclusions: We demonstrate that mutations in cell lines are similar to those found in patient-derived tumours implying fidelity at the genomic level. Mesotheliomas harbour multiple mutations in tumour suppressors, but lack commonly tractable oncogenic mutations which may explain poor efficacy seen in clinical trials to date. We further demonstrate that PI3K may represent a critical node therapeutically that may be useful in combinatorial approaches.

Abstract 2488: Novel candidate protein biomarkers for cisplatin response prediction in non-small cell lung cancer

The five-year survival rate for non-small cell lung cancer (NSCLC) is still less than twenty percent in part due to treatment failure and lack of biomarkers for personalized therapy. Most NSCLC patients are treated by chemotherapy and cisplatin is still the major component of most chemotherapy regimens. However, resistance to cisplatin is a common phenomenon. Aim of this study was to identify protein biomarkers that can be used for cisplatin response prediction.

The proteomes and secretomes1 of a series of human NSCLC cell lines with a range of IC50-values for cisplatin (1.5 - 15 μM) were analyzed by label-free proteomics based on GeLC-MS/MS and quantified by spectral counting. Significant differential expression of proteins was determined by the in-house developed beta-binomial statistical test2. Network and gene ontology analysis was performed, with STRING and DAVID respectively, to reveal cellular and molecular functions associated with the differential proteins.

In total, 2885 and 2342 proteins were detected in the cell lysates and secretomes, respectively. The proteins with significantly higher expression in the resistant cell lines were associated RNA processing and vesicle transport, while the proteins with higher expression in the sensitive cell lines were known to be involved in RNA splicing, DNA damage repair and extracellular matrix organization. The most promising candidates are being validated in tissues and sputum of patients with known response to cisplatin.

1 Piersma SR et al. J Proteome Res. 2010 Apr 5;9(4):1913-22.

2 Pham TV et al. Bioinformatics. 2010 Feb 1;26(3):363-9.

Citation Format: Tieneke B M Schaaij-Visser, Remco Nagel, Inge de Reus, Sander R. Piersma, Thang V. Pham, Egbert F. Smit, Frederik B. Thunnissen, Ruud H. Brakenhoff, Anton Berns, Connie R. Jimenez. Novel candidate protein biomarkers for cisplatin response prediction in non-small cell lung cancer. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2488. doi:10.1158/1538-7445.AM2013-2488

Mouse models for lung cancer

Lung cancer is a devastating disease and a major therapeutic burden with poor survival rates. It is responsible for 30% of all cancer deaths. Lung cancer is strongly associated with smoking, although some subtypes are also seen in non-smokers. Tumors in the latter group are mostly adenocarcinomas with many carrying mutations in the epidermal growth factor receptor (EGFR). Survival statistics of lung cancer are grim because of its late detection and frequent local and distal metastases. Although DNA sequence information from tumors has revealed a number of frequently occurring mutations, affecting well-known tumor suppressor genes and proto-oncogenes, many of the driver mutations remain ill defined. This is likely due to the involvement of numerous rather infrequently occurring driver mutations that are difficult to distinguish from the very large number of passenger mutations detected in smoking-related lung cancers. Therefore, experimental model systems are indispensable to validate putative driver lesions and to gain insight into their mechanisms of action. Whereas a large fraction of these analyzes can be performed in cell cultures in vitro, in many cases the consequences of the mutations have to be assessed in the context of an intact organism, as this is the context in which the Mendelian selection process of the tumorigenic process took place and the advantages of particular mutations become apparent. Current mouse models for cancer are very suitable for this as they permit mimicking many of the salient features of human tumors. The capacity to swiftly re-engineer complex sets of lesions found in human tumors in mice enables us to assess the contribution of defined combinations of lesions to distinct tumor characteristics such as metastatic behavior and response to therapy. In this review we will describe mouse models of lung cancer and how they are used to better understand the disease and how they are exploited to develop better intervention strategies.

Abstract SY10-02: Mouse models for (non) small cell lung cancer: Heterogeneity and cell of origin

Small cell lung cancer (SCLC) is one of the most lethal human malignancies, due to its high metastatic potential and chemo-resistance upon relapse. Using the Rb1flox/flox;Trp53flox/flox mouse model for SCLC, we found that the tumors are often composed of phenotypically different cells, characterized by mesenchymal and neuroendocrine markers. These cells had a common origin as they shared specific genomic aberrations. Crosstalk between mesenchymal and neuroendocrine cells can endow the neuroendocrine cells with metastatic capacity, illustrating the potential relevance of tumor cell heterogeneity in dictating functional tumor properties. We have further studied which genes might be critical for this process. We could show that in particular the activation of the transcription factor PEA3 in neuroendocrine cells was instrumental although not sufficient to confer full metastatic potential to the neuroendocrine tumor cells. Interestingly, these neuroendocrine cells can convert into the mesenchymal component by Ras pathway activation, suggesting that these cell types might interconvert and mediate paracrine signaling. This interconversion also raises the question of the cell-of-origin of this tumor.

To investigate this in more detail, we inactivated Trp53 and Rb1 in distinct subsets of cells in the adult lung by targeting Cre-recombinase expression to Clara (CC10 positive), neuroendocrine (CGRP positive), and alveolar type 2 (SPC positive) cells using adenoviral vectors harboring cell-type specific promoters driving Cre recombinase. Using these cell-type-specific Cre expression adenoviruses we could show that inactivation of Trp53 and Rb1 can efficiently transform neuroendocrine (CGRP-positive) and to a lesser extent alveolar type II (SPC-positive) cells leading to SCLC. In contrast CC10-expressing clara cells were largely resistant to transformation. The results clearly indicate that neuroendocrine cells serve as the predominant cell of origin of SCLC. A different cell type specificity was found when a Lox-Stop-Lox (LSL-)mutant-Kras driven non-small-cell lung cancer (NSCLC) model was used to reveal the cell-of-origin of NSCLC. In this case we noted a significant difference whether LSL-mutant-Kras mice or LSL-mutant-Kras;Trp53flox/flox mice were infected with the various virus constructs. In LSL-mutant-Kras mice alveolar type II cells appeared the most effective target for inducing adenomas whereas in LSL-mutant-Kras;Trp53flox/flox mice multiple cell types gave effectively rise to adenomas and adenocarcinomas. Our data indicate that both cell type specific features and the nature of the oncogenic lesion(s) are critical factors in determining the tumor initiating capacity of lung (progenitor) cells to give rise to the various lung cancer subtypes.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr SY10-02. doi:1538-7445.AM2012-SY10-02

Abstract 3295: Systematic in vivo analysis of PI3K pathway aberrations in a mouse model for invasive lobular carcinoma

Invasive lobular carcinoma (ILC) is a form of breast cancer that develops in the milk-producing glands (lobules) of the breast. ILC is an aggressive form of breast cancer due to its ability to spread to other parts of the body, both by bloodstream as well as lymphatic system. One of the key features found in ILC is loss of E-cadherin, encoding CDH1. Mammary-gland specific deletion of E-cadherin does not lead to tumour formation indicating that other factors are involved in the induction of ILC. Indeed, combined tissue-specific inactivation of E-cadherin and p53 induces development of invasive and metastatic mammary tumours resembling human ILC. However, a limitation of the current conditional mouse models is that human ILCs rarely contain p53 mutations. In contrast, PI3K pathway mutations are frequently found in human ILCs. Currently we are developing novel mouse ILC models to systematically examine the contribution of PI3K pathway components such as PIK3CA, AKT and PTEN to ILC development. To reduce time spent on generating a genetically engineered mouse (GEM) tumour cohort expressing three or more transgenes we set-up embryonic stem (ES) cell derivation and targeting of ES cells from validated GEM models, the so-called GEMM-ESC strategy. WapCre-Cdh1F/F ES cells targeted with PIK3CA or AKT mutants are now being used to generate the elaborate PI3K pathway tumour cohort. These advanced models will allow us to study the role of the PI3K pathway in the pathogenesis of ILC. Moreover, we can use these advanced mouse models which recapitulate key hallmarks of human ILC to perform refined and realistic preclinical prevention and intervention studies for ILC.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3295. doi:1538-7445.AM2012-3295

Insertional mutagenesis identifies multiple networks of cooperating genes driving intestinal tumorigenesis

The evolution of colorectal cancer suggests the involvement of many genes. To identify new drivers of intestinal cancer, we performed insertional mutagenesis using the Sleeping Beauty transposon system in mice carrying germline or somatic Apc mutations. By analyzing common insertion sites (CISs) isolated from 446 tumors, we identified many hundreds of candidate cancer drivers. Comparison to human data sets suggested that 234 CIS-targeted genes are also dysregulated in human colorectal cancers. In addition, we found 183 CIS-containing genes that are candidate Wnt targets and showed that 20 CISs-containing genes are newly discovered modifiers of canonical Wnt signaling. We also identified mutations associated with a subset of tumors containing an expanded number of Paneth cells, a hallmark of deregulated Wnt signaling, and genes associated with more severe dysplasia included those encoding members of the FGF signaling cascade. Some 70 genes had co-occurrence of CIS pairs, clustering into 38 sub-networks that may regulate tumor development.

Rapid validation of cancer genes in chimeras derived from established genetically engineered mouse models

Recent technological advances have opened the door for the fast and cost-effective generation of genetically engineered mouse models (GEMMs) to study cancer. We describe here a conceptually novel approach for the generation of chimeric GEMMs based on the controlled introduction of various genetic elements in embryonic stem cells (ESCs) that are derived from existing mouse strains with a predisposition for cancer. The isolation of GEMM-derived ESC lines is greatly facilitated by the availability of the newly defined culture media containing inhibitors that effectively preserve ESC pluripotency. The feasibility of the GEMM-ESC approach is discussed in light of current literature and placed into the context of existing models. This approach will allow for fast and flexible validation of candidate cancer genes and drug targets and will result in a repository of GEMM-ESC lines and corresponding vector collections that enable easy distribution and use of preclinical models to the wider scientific community.

Cell of origin of small cell lung cancer: inactivation of Trp53 and Rb1 in distinct cell types of adult mouse lung

Small cell lung cancer (SCLC) is one of the most lethal human malignancies. To investigate the cellular origin(s) of this cancer, we assessed the effect of Trp53 and Rb1 inactivation in distinct cell types in the adult lung using adenoviral vectors that target Cre recombinase to Clara, neuroendocrine (NE), and alveolar type 2 (SPC-expressing) cells. Using these cell type-restricted Adeno-Cre viruses, we show that loss of Trp53 and Rb1 can efficiently transform NE and SPC-expressing cells leading to SCLC, albeit SPC-expressing cells at a lesser efficiency. In contrast, Clara cells were largely resistant to transformation. The results indicate that although NE cells serve as the predominant cell of origin of SCLC a subset of SPC-expressing cells are also endowed with this ability.

Computational identification of insertional mutagenesis targets for cancer gene discovery

Insertional mutagenesis is a potent forward genetic screening technique used to identify candidate cancer genes in mouse model systems. An important, yet unresolved issue in the analysis of these screens, is the identification of the genes affected by the insertions. To address this, we developed Kernel Convolved Rule Based Mapping (KC-RBM). KC-RBM exploits distance, orientation and insertion density across tumors to automatically map integration sites to target genes. We perform the first genome-wide evaluation of the association of insertion occurrences with aberrant gene expression of the predicted targets in both retroviral and transposon data sets. We demonstrate the efficiency of KC-RBM by showing its superior performance over existing approaches in recovering true positives from a list of independently, manually curated cancer genes. The results of this work will significantly enhance the accuracy and speed of cancer gene discovery in forward genetic screens. KC-RBM is available as R-package.

Abstract 5100: Secretome proteomics to identify indicators for lung cancer treatment response prediction and monitoring

Lung cancer is currently the number one cause of cancer-related deaths worldwide. Five-year survival rates for both non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) are still less than twenty percent due to late stage of presentation, treatment failure and lack of biomarkers for personalized therapy. Aim of this study is to identify protein biomarkers that can be used for treatment response prediction and therapy monitoring.

Rather than following a broad strategy by analyzing the proteome of whole cell or tissue lysates, we chose to use a more specific approach based on the assumption that the best tumor markers are shed or secreted by the tumor and can be detected in blood. Secretome (i.e. all proteins shed from or secreted by a cell or tissue) proteomics (Piersma SR et al. J Proteome Res. 2010 Apr 5;9(4):1913-22.) was performed on a set of tumors from conditional mouse models for SCLC and NSCLC (Meuwissen R et al. Oncogene. 2001 Oct 4;20(45):6551-8; Meuwissen R et al. Cancer Cell. 2003 Sep;4(3):181-9.), a set of human and mouse SCLC and NSCLC cell lines and a series of human NSCLC cell lines with a range of IC50-values for cisplatin (1.5 – 15 µM). Secretomes were obtained by collecting serum-free media from the cell lines or collecting PBS in which a tumor tissue piece had been incubated.

In the secretome samples, up to 2,000 proteins can be identified with 80% reproducibility by label-free, shotgun nanoLC tandem mass spectrometry and quantified by spectral counting. The beta-binomial test (Pham TV et al. Bioinformatics. 2010 Feb 1;26(3):363-9.) is used to find significant differences in protein expression between the different secretomes and network analysis is performed to provide insight into the underlying cellular mechanisms. The secretomes from the mouse models are used to find proteins specifically secreted by either NSCLC or SCLC, while the comparison between the secretomes of the several human and mouse cell lines allows extrapolation of the results from the mouse models to the human situation. Furthermore, the analysis of the cell lines with various cisplatin IC50-values permits selection of the protein markers that are indicative of sensitivity or resistance. The results of the different proteomics analyses will be presented along with the most promising biomarker candidates and an overview of the functional network analysis.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 5100. doi:10.1158/1538-7445.AM2011-5100

A functional role for tumor cell heterogeneity in a mouse model of small cell lung cancer

Small cell lung cancer (SCLC) is the lung neoplasia with the poorest prognosis, due to its high metastatic potential and chemoresistance upon relapse. Using the previously described mouse model for SCLC, we found that the tumors are often composed of phenotypically different cells with either a neuroendocrine or a mesenchymal marker profile. These cells had a common origin because they shared specific genomic aberrations. The transition from neuroendocrine to mesenchymal phenotype could be achieved by the ectopic expression of oncogenic Ras(V12). Crosstalk between mesenchymal and neuroendocrine cells strongly influenced their behavior. When engrafted as a mixed population, the mesenchymal cells endowed the neuroendocrine cells with metastatic capacity, illustrating the potential relevance of tumor cell heterogeneity in dictating tumor properties.