Virtual Library
Start Your Search
E.A. Marshall
Author of
-
+
MA02 - RNA in Lung Cancer (ID 377)
- Event: WCLC 2016
- Type: Mini Oral Session
- Track: Biology/Pathology
- Presentations: 1
- Moderators:E. Brambilla, M. Noguchi
- Coordinates: 12/05/2016, 14:20 - 15:50, Stolz 2
-
+
MA02.09 - Long Non-Coding RNA Expression from Pseudogene Loci as a Novel Mechanism of Cancer Gene Regulation (ID 6287)
15:20 - 15:26 | Author(s): E.A. Marshall
- Abstract
- Presentation
Background:
The advent of next generation sequencing has lead to the discovery of the functional importance of non-coding RNAs (ncRNAs) in a wide variety of cellular processes, and these genes can be exploited by tumours to drive the hallmarks of cancer. Pseudogenes are DNA sequences that are defunct relatives of their functional parent genes but retain high sequence homology. Long non-coding RNAs (lncRNAs) have been shown to regulate protein-coding genes; however, complex folding patterns make lncRNA function difficult to predict. Several lncRNAs expressed from pseudogene loci have been shown to regulate the protein-coding parent genes of these pseudogenes in trans due to sequence complementarity. The biological impact of this mechanism has not been investigated in lung adenocarcinoma (LUAD). We hypothesize that expression changes in lncRNAs expressed from pseudogene loci can affect the expression of corresponding protein-coding parent genes in trans, and that these events provide an alternative mechanism of cancer gene deregulation in LUAD tumourigenesis.
Methods:
We analysed RNA-seq data from 50 LUAD with matched non-malignant tissue obtained from the TCGA for both protein-coding and non-coding gene expression. Significantly differentially expressed lncRNAs located within pseudogene loci were identified by sign-rank test (p<0.001). Mann Whitney U-tests were used to identify lncRNA-parent gene pairs which significantly correlated expression, and survival analysis was performed using a Cox proportional hazard model.
Results:
Our analysis has identified 172 lncRNAs expressed from pseudogene loci that were significantly deregulated in LUAD. Remarkably, many of these lncRNAs were expressed from the loci of pseudogenes related to known cancer genes. One of these lncRNAs, CTD-2583A14.8, was expressed from a pseudogene to ubiquitin-conjugating enzyme E2C (UBE2C), which regulates tumor growth, apoptosis, and angiogenesis through phospho-ERK1/2. We find CTD-2583A14.8 as well as the UBE2C parent gene to be significantly upregulated in LUAD tumours compared to matched normal tissue. Furthermore, tumours with higher levels of CTD-2583A14.8 have significantly higher levels of UBE2C expression than tumours with low levels of CTD-2583A14.8, indicating that CTD-2583A14.8 may positively regulate UBE2C in trans.
Conclusion:
Here we show expression of lncRNAs within pseudogene loci is deregulated in LUAD, and can correlate with the expression of their protein-coding counterparts. Many of these genes associated with this putative lncRNA-pseudogene-protein-coding axis have previously been implicated in cancer. Therefore, this represents an alternative mechanism of cancer gene deregulation, and may represent novel therapeutic intervention points for the treatment of LUAD.
Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.
-
+
P1.02 - Poster Session with Presenters Present (ID 454)
- Event: WCLC 2016
- Type: Poster Presenters Present
- Track: Biology/Pathology
- Presentations: 1
- Moderators:
- Coordinates: 12/05/2016, 14:30 - 15:45, Hall B (Poster Area)
-
+
P1.02-052 - Signal Regulatory Protein a (SIRPA): A Key Regulator of the EGFR Pathway Demonstrates Both Tumor Suppressive and Oncogenic Properties (ID 6061)
14:30 - 14:30 | Author(s): E.A. Marshall
- Abstract
Background:
The epidermal growth factor receptor (EGFR) signaling pathway is one of the most frequently deregulated pathways in non-small cell lung cancer. While targeted therapy prolongs survival in patients harbouring EGFR mutations, resistance to treatment eventually develops in all cases. As multiple genetic and epigenetic alterations are known to disrupt signaling pathways, the objective of this study is to perform a multidimensional analysis of signaling pathways to identify alterations essential to tumorigenesis that are overlooked when assessing a single genomic dimension.
Methods:
Multidimensional integrative analysis of copy number, DNA methylation, and gene expression profiles of 77 lung adenocarcinomas and matched non-malignant tissues identified Signal Regulatory Protein A (SIRPA) as a novel candidate tumor suppressor gene. Following validation of genomic findings in multiple external data sets, the tumor suppressive effects of SIRPA were assessed in vitro and in vivo with a panel of lung cancer cell lines.
Results:
SIRPA negatively regulates receptor tyrosine kinase signaling through activation of the protein phosphatases SHP1 and SHP2 and was found to be underexpressed in 70% of lung tumours, ranking it in the 95[th] percentile of altered genes within the EGFR pathway. Immunohistochemistry (IHC) confirmed reduced protein expression in tumors, which was found to correlate with EGFR mutation and adenocarcinoma histology. In vitro, SIRPA knockdown promoted migration while simultaneously inducing a dramatic senescent phenotype, suggesting SIRPA may act as a barrier to tumorigenesis. This phenotype is dependent upon upregulation of the CDK inhibitor p27, which hypophosphorylates RB leading to cell cycle blockade and reduced tumor growth in vivo. Importantly, increased expression of p27 resulted in mis-localization into the cytoplasm where it is known to promote an invasive phenotype. Inhibition of p27 confirmed previous findings and emphasized the importance of this pathway in lung tumorigenesis. Surprisingly, overexpression of SIRPA increased cell growth and migration, suggesting SIRPA may also possess oncogenic properties due to its regulation of multiple signaling pathways. Overexpression of SHP2 following ectopic expression of SIRPA promotes migration through the inhibition of focal adhesions. This phenotype is abrogated upon siRNA knockdown of SHP2.
Conclusion:
SIRPA is an important player in lung tumor biology, capable of acting as both an oncogene and tumor suppressor due to its ability to regulate multiple signaling pathways. Due to the complex nature in its signaling, future work should focus on elucidating how the timing of alterations to SIRPA affects tumorigenesis to design treatment strategy.
-
+
P2.01 - Poster Session with Presenters Present (ID 461)
- Event: WCLC 2016
- Type: Poster Presenters Present
- Track: Biology/Pathology
- Presentations: 4
- Moderators:
- Coordinates: 12/06/2016, 14:30 - 15:45, Hall B (Poster Area)
-
+
P2.01-022 - A PIWI-Interacting RNAs Co-Expression Networks as a Prognostic Factor in Lung Cancer (ID 5812)
14:30 - 14:30 | Author(s): E.A. Marshall
- Abstract
Background:
PIWI-interacting RNAs (piRNAs) are small (24-32 nucleotides) non-coding RNAs. Their functions, widely conserved across species, are associated to epigenetic control of gene expression and maintenance of genomic stability by the repression of mobile elements. In humans, >23,000 piRNAs are known, showing tissue-specific expression patterns. While the aberrant expression of individual piRNAs has been identified in some cancer types, the role of piRNA co-expression networks in the development of lung tumors and their utility as molecular markers remains unexplored. By analyzing over 7000 piRNA transcriptomes from human tumors and non-malignant tissues, we have identified lung cancer (LC) specific expression networks associated with clinically-relevant tumor features and patient prognosis.
Methods:
We developed a custom small-RNA sequence analysis pipeline to generate >7,000 human piRNA transcriptomes. piRNA expression baseline was deduced from 6,378 piRNA transcriptomes (non-malignant/tumors) from 11 organ sites. In lungs, we analyzed 1,082 tumors and 209 non-malignant samples from two cohorts: BC Cancer Agency (BCCA) and The Cancer Genome Atlas (TCGA). Network analysis was performed using the weighted gene co-expression network analysis (WCGNA). We evaluated tumour aggressiveness by considering correlation to several clinical parameters, including stage, number of mutations, nodal/distant metastasis, and overall/disease-free survival. piRNA survival signatures were identified using a Cox Proportional Hazard model.
Results:
A subset of piRNA showed robust expression in somatic tissues. Expressed piRNAs display organ-specific patterns and mainly map to coding transcripts, suggesting a role in regulation of gene expression. In lungs, 204 piRNAs were consistently expressed in both LC cohorts. Tumor piRNA expression profiles are markedly different from their non-malignant counterparts (133 piRNAs were differentially expressed). The patterns differ between the adenocarcinoma and squamous cell carcinoma, and were influenced by smoking status. Network-based analysis identified piRNA expression changes in two modules of piRNAs are associated with aggressiveness tumor features, such as increased number of mutations, tumor size and nodal metastasis. Finally, combined expression of piRNAs define signatures associated with patient overall and recurrence free survival.
Conclusion:
We provide evidence of somatic, tissue-specific human piRNA expression. In lungs, aberrant expression patterns are associated with well-established etiological factors of cancer and seem to contribute to lung cancer subtype-specific biology. We discover that specific piRNA-based expression patterns characterize aggressive lung tumors and also exhibit prognostic value. The unique expression patterns of piRNAs offer an opportunity to better understand lung cancer-specific biology as well as develop novel prognostic markers for clinical application.
-
+
P2.01-023 - Deregulation of Small Non-Coding RNAs at the DLK1-DIO3 Imprinted Locus Predicts Lung Adenocarcinoma Patient Outcome (ID 6142)
14:30 - 14:30 | Author(s): E.A. Marshall
- Abstract
Background:
Deregulation of small RNAs at the imprinted DLK1-DIO3 locus has been linked to lung adenocarcinoma (LUAD) patient outcome. While the contribution of microRNAs (miRNAs) is established, the role of Piwi-interacting RNAs (piRNAs), small RNAs involved in epigenetic regulation of gene transcription, is unexplored. We quantified expression of piRNAs and miRNAs mapping to this locus in two independent cohorts of LUAD and assessed the ability of a combined miRNA/piRNA signature to improve patient outcome stratification.
Methods:
Expression levels (RPKM) for miRNA/piRNA were determined from small RNA sequencing experiments from two cohorts (TCGA, n=154, 5-year follow up; BCCA, n=77, 8-year follow up). Associations with patient overall survival (OS) and recurrence free survival (RFS) were calculated by inputting miRNA and piRNA expression combinations into a Cox proportional hazard model. Risk scores were calculated by multiplying the expression value for each gene by its hazard coefficient, and summed per sample. Risk scores were ranked and divided into tertiles for log-rank survival analysis. DNA-level piRNA targets were predicted using MiRanda based on sequence complementarity in the region 3.5kb upstream of the transcription-start site of all human transcripts from ENSEMBL. Transcript-level miRNA targets were predicted using the miRDIP algorithm, which integrates 13 miRNA target prediction algorithms and six miRNA prediction databases.
Results:
Only 7 out of 138 piRNAs mapping to the locus were expressed. A combined miRNA/piRNA signature improved both OS and RFS predictions compared to signatures of miRNAs or piRNAs alone. In TCGA, log-rank analysis of risk groups indicated only the miRNA/piRNA signature significantly stratified patients (OS p=0.0038, RFS p=0.0229) into low, intermediate, and high risk groups compared to separated miRNA or piRNA signatures. Similarly, in the BCCA dataset, only the combined miRNA/piRNA signature significantly stratified high, intermediate, and low risk groups (p=0.0019). Target prediction of piRNAs and miRNAs from the signature indicated that 34 genes may be regulated at both the DNA (piRNA) and mRNA (miRNA) level.
Conclusion:
We find the combination of miRNA and piRNA expression derived from the DLK1-DIO3 locus produces a superior stratification of patient outcome than either metric alone. While the contribution of miRNAs to patient risk stratification is established, the improved model performance derived from the addition of piRNAs adds another layer of gene regulation at the DNA-level. Model performance is optimal when these two small RNA species are considered simultaneously; suggesting their coordinated biological effects as a result of deregulation at this locus in LUAD.
-
+
P2.01-024 - Expression of miR-106 Paralogs Improves Prognostic Value of Mesenchymal Signatures but Only miR-106b Promotes Invasiveness (ID 6250)
14:30 - 14:30 | Author(s): E.A. Marshall
- Abstract
Background:
Improved understanding of the molecular mechanisms driving lung cancer progression can lead to novel therapeutic strategies to improve the currently poor patient treatment outcome. Deregulation of microRNA (miRNA) expression in malignant cells activates molecular pathways that drive tumor progression such as epithelial-mesenchymal transition (EMT). We identify miRNA paralogs, miR-106a and miR-106b, to be elevated in metastatic lung adenocarcinoma (LUAD). We assess whether these two highly similar miRNAs share the same functions in vitro, and measure how their elevated expression increases invasiveness or induces EMT in LUAD tumor.
Methods:
MiRNA expression was obtained from small RNA sequencing data derived from clinical primary LUAD specimens and paired non-malignant tissues (60 localized, 27 with lymph node invasion). Non-invasive, epithelial LUAD cell lines with low endogenous miR-106a/b levels were transfected and co-transfected with overexpression vectors for miR-106a and miR-106b. Invasiveness of experimentally-modulated tumor cells was assessed in vitro by Boyden chamber assay and in vivo using a zebrafish model, and expression of EMT markers was determined by Western Blot. Predicted miRNA targets were identified using mirDIP portal. To identify putative genetic mechanisms of mir-106a/b overexpression, DNA copy number, methylation, and Gene Set Enrichment Analysis (GSEA) were performed. Clinical associations were computed in an independent cohort of TCGA LUAD samples.
Results:
Both miR-106 paralogs were significantly overexpressed in LUAD samples with lymph node invasion. However, increased expression of miR-106b alone or together with miR-106a, but not miR-106a alone, enhanced metastatic phenotypes, and correlated with increased mesenchymal and decreased epithelial marker expression. Predicted targets include EP300, a transcriptional activator of E-cadherin, and members of the TGFβ signaling pathway. Copy number and methylation status did not correlate with miRNA expression; however, GSEA analysis revealed enrichment of E2F transcription factor targets in LUAD with high expression of either miR-106 paralogs. Furthermore, expression of miR-106 paralogs was significantly positively correlated with E2F1 and E2F2, suggesting that upstream regulation by E2F is a potential mechanism. Interestingly, miR-106a and miR-106b expression was associated with poor survival and advanced stage when stratified by mesenchymal marker vimentin.
Conclusion:
Although both miR-106a and miR-106b are overexpressed in metastatic LUAD, the strongest prognostic association was found in LUAD with a mesenchymal expression signature and high expression of both miRNAs. Our cell models suggest that miR-106b may play a direct role in EMT, with miR-106a influencing tumor progression via alternative mechanisms. Inhibition of one or both of these miRNAs may provide a strategy for treating advanced stage disease.
-
+
P2.01-037 - Molecular Biology Underlying COPD and Lung Cancer Converge on FOXM1 Network (ID 5773)
14:30 - 14:30 | Author(s): E.A. Marshall
- Abstract
Background:
Chronic obstructive pulmonary disease (COPD) is a progressive, inflammatory lung disease associated with an up to 10-fold increased risk of lung cancer (LC). COPD and LC share common etiologies including genetic susceptibilities and risk factors, such as smoking. This study systematically characterizes the molecular overlap between COPD and LC.
Methods:
Small airway gene expression data was obtained from subjects with spirometry measures (n=267) (GSE37147). Genome-wide, multi-omics data for lung adenocarcinoma (LUAD) tumor and non-malignant lung tissues from two cohorts (TCGA, n=515; BCCA, n=90) was analyzed. Weighted correlation network analysis (WGCNA) was applied to identify clusters (modules) of highly correlated genes across airway expression profiles. Combined module expression (eigengene scores) were used to: 1) identify modules negatively associated with FEV~1~ and 2) calculate module preservation in lung tumors. Signaling network, pathway and gene ontology analyses were performed using IID, pathDIP, ClueGo and PARADIGM. Known and predicted protein-protein physical interactions (PPIs) were obtained from IID. Network analysis and visualization was performed in NAViGaTOR.
Results:
A module of 31 genes significantly co-expressed across small airways was negatively associated with FEV~1~ and preserved in LUAD tumors. Genes in this module were enriched in functions associated with cell cycle progression, and known and/or predicted to physically interact in the protein complex critical to mediating G2/M progression. The forkhead transcription factor FOXM1 network was the most highly perturbed entity across 515 LUAD tumors. FOXM1 is an essential mitotic protein, known to regulate expression of genes involved in cell cycle progression, as well as stress response to ROS and DNA damage, angiogenesis and metastasis. COPD-related airway mRNA changes and genes highly altered at the DNA and mRNA level in LUAD tumors directly converge on the FOXM1 regulated mitotic complex proteins and/or FOXM1 transcription factor network.
Conclusion:
FOXM1 is overexpressed in multiple cancer types where it is correlated with poor prognosis and oncogenic transformation of epithelia through induction of genomic instability. The convergence of COPD and LUAD changes on this network may underlie increased LC risk in COPD patients, warranting further exploration as a target for COPD treatment and/or LC prevention or treatment.
-
+
P3.01 - Poster Session with Presenters Present (ID 469)
- Event: WCLC 2016
- Type: Poster Presenters Present
- Track: Biology/Pathology
- Presentations: 1
- Moderators:
- Coordinates: 12/07/2016, 14:30 - 15:45, Hall B (Poster Area)
-
+
P3.01-049 - ELF3 Overexpression Leads to Oncogenic Reprogramming of Protein Interactions Exposing Therapeutically Actionable Targets (ID 5807)
14:30 - 14:30 | Author(s): E.A. Marshall
- Abstract
Background:
Emerging evidence has implicated ELF3 involvement in cancer signaling pathways. To determine the biological basis to pursue ELF3 as a novel therapeutic target, we investigated the role of ELF3 in lung adenocarcinoma (LUAD). Using a multi-omics approach in two independent cohorts of LUAD we (a) discover genetic mechanisms driving aberrant expression of this oncogene, (b) identify the protein-protein-interaction (PPI) partners of ELF3, and (c) determine the specific functions of ELF3 in LUAD using model systems.
Methods:
Comprehensive, multi-omic data was collected from the BC Cancer Research Centre (BCCRC), The Cancer Genome Atlas (TCGA), and several mouse models of LUAD tumourigenesis. ELF3 cellular localization was visualized by immunofluorescence. ELF3 knock-down and overexpression was achieved by lentiviral vector delivery for in vitro and in vivo assays. Physical protein-protein interaction (PPI) networks obtained from IID were overlaid onto cancer and non-malignant gene expression data from TCGA and 11 restructured datasets from Gene Expression Omnibus. PPIs were interrogated to investigate malignancy-associated ELF3 interactions. Pathway analysis was performed using pathDIP. Survival analysis was performed using the log-rank method.
Results:
ELF3 was significantly overexpressed in both cohorts, remarkably in >70% of cases (p=1.64E-21). However, mutation of known upstream regulators was not sufficient to explain the frequency of ELF3 overexpression. Instead, the ELF3 locus underwent frequent (>80%) genetic alteration including focal amplification and promoter hypomethylation, which corresponded with increased expression. ELF3 was predominantly localized to the nucleus, consistent with its transcription factor function. Analysis of PPI networks indicated highly LUAD-specific ELF3 interactions whereby loss and gain of interactions lead to reprogramming of LUAD transcriptional networks, including loss of TNFα pathway, and gain of TGFβ pathway, PI3K pathway, and translesion (DNA repair) pathway interactions. Furthermore, EGFR, KRAS, and MYC transgenic models of LUAD tumourigenesis all displayed a marked increase (6 to 8-fold) in ELF3 expression signifying its importance to LUAD of varied genetic backgrounds. In culture, ELF3 regulated proliferation, viability and anchorage-independent growth. In animal models, ELF3 knock-down cells underwent negative clonal selection, suggesting ELF3 expression is beneficial to tumour growth. Clinically, high expression of ELF3 was associated with poor survival regardless of tumour stage.
Conclusion:
Overexpression of ELF3 reprograms protein-protein-interactions in LUAD leading to the activation of cancer-specific pathways, and producing oncogenic phenotypes. Depletion of ELF3 with shRNAs reverses tumour cell growth, suggesting ELF3 is a promising therapeutic target. In addition to ELF3, interruption of cancer-specific PPIs also represents a therapeutically actionable strategy.
