Hae Ryung Chang

PATHOME: an algorithm for accurately detecting differentially expressed subpathways

The translation of high-throughput gene expression data into biologically meaningful information remains a bottleneck. We developed a novel computational algorithm, PATHOME, for detecting differentially expressed biological pathways. This algorithm employs straightforward statistical tests to evaluate the significance of differential expression patterns along subpathways. Applying it to gene expression data sets of gastric cancer (GC), we compared its performance with those of other leading programs. Based on a literature-driven reference set, PATHOME showed greater consistency in identifying known cancer-related pathways. For the WNT pathway uniquely identified by PATHOME, we validated its involvement in gastric carcinogenesis through experimental perturbation of both cell lines and animal models. We identified HNF4α-WNT5A regulation in the cross-talk between the AMPK metabolic pathway and the WNT signaling pathway, and further identified WNT5A as a potential therapeutic target for GC. We have demonstrated PATHOME to be a powerful tool, with improved sensitivity for identifying disease-related dysregulated pathways.

Detection of PIWI and piRNAs in the mitochondria of mammalian cancer cells.

Piwi-interacting RNAs (piRNAs) are 26-31 nt small noncoding RNAs that are processed from their longer precursor transcripts by Piwi proteins. Localization of Piwi and piRNA has been reported mostly in nucleus and cytoplasm of higher eukaryotes germ-line cells, where it is believed that known piRNA sequences are located in repeat regions of nuclear genome in germ-line cells. However, localization of PIWI and piRNA in mammalian somatic cell mitochondria yet remains largely unknown. We identified 29 piRNA sequence alignments from various regions of the human mitochondrial genome. Twelve out 29 piRNA sequences matched stem-loop fragment sequences of seven distinct tRNAs. We observed their actual expression in mitochondria subcellular fractions by inspecting mitochondrial-specific small RNA-Seq datasets. Of interest, the majority of the 29 piRNAs overlapped with multiple longer transcripts (expressed sequence tags) that are unique to the human mitochondrial genome. The presence of mature piRNAs in mitochondria was detected by qRT-PCR of mitochondrial subcellular RNAs. Further validation showed detection of Piwi by colocalization using anti-Piwil1 and mitochondria organelle-specific protein antibodies.

HNF4α is a therapeutic target that links AMPK to WNT signalling in early-stage gastric cancer

Background Worldwide, gastric cancer (GC) is the fourth most common malignancy and the most common cancer in East Asia. Development of targeted therapies for this disease has focused on a few known oncogenes but has had limited effects. Objective To determine oncogenic mechanisms and novel therapeutic targets specific for GC by identifying commonly dysregulated genes from the tumours of both Asian-Pacific and Caucasian patients. Methods We generated transcriptomic profiles of 22 Caucasian GC tumours and their matched non-cancerous samples and performed an integrative analysis across different GC gene expression datasets. We examined the inhibition of commonly overexpressed oncogenes and their constituent signalling pathways by RNAi and/or pharmacological inhibition. Results Hepatocyte nuclear factor-4α (HNF4α) upregulation was a key signalling event in gastric tumours from both Caucasian and Asian patients, and HNF4α antagonism was antineoplastic. Perturbation experiments in GC tumour cell lines and xenograft models further demonstrated that HNF4α is downregulated by AMPKα signalling and the AMPK agonist metformin; blockade of HNF4α activity resulted in cyclin downregulation, cell cycle arrest and tumour growth inhibition. HNF4α also regulated WNT signalling through its target gene WNT5A, a potential prognostic marker of diffuse type gastric tumours. Conclusions Our results indicate that HNF4α is a targetable oncoprotein in GC, is regulated by AMPK signalling through AMPKα and resides upstream of WNT signalling. HNF4α may regulate ‘metabolic switch’ characteristic of a general malignant phenotype and its target WNT5A has potential prognostic values. The AMPKα-HNF4α-WNT5A signalling cascade represents a potentially targetable pathway for drug development.

A pathway-based approach for identifying biomarkers of tumor progression to trastuzumab-resistant breast cancer

Although trastuzumab is a successful targeted therapy for breast cancer patients with tumors expressing HER2 (ERBB2), many patients eventually progress to drug resistance. Here, we identified subpathways differentially expressed between trastuzumab-resistant vs. -sensitive breast cancer cells, in conjunction with additional transcriptomic preclinical and clinical gene datasets, to rigorously identify overexpressed, resistance-associated genes. From this approach, we identified 32 genes reproducibly upregulated in trastuzumab resistance. 25 genes were upregulated in drug-resistant JIMT-1 cells, which also downregulated HER2 protein by >80% in the presence of trastuzumab. 24 genes were downregulated in trastuzumab-sensitive SKBR3 cells. Trastuzumab sensitivity was restored by siRNA knockdown of these genes in the resistant cells, and overexpression of 5 of the 25 genes was found in at least one of five refractory HER2 + breast cancer. In summary, our rigorous computational approach, followed by experimental validation, significantly implicate ATF4, CHEK2, ENAH, ICOSLG, and RAD51 as potential biomarkers of trastuzumab resistance. These results provide further proof-of-concept of our methodology for successfully identifying potential biomarkers and druggable signal pathways involved in tumor progression to drug resistance.

Abstract 5534: Metformin increases AMPKα activity by inhibition of AMPKα and cell cycle proliferation in Asian gastric cancer.

The LKB1/AMPK signaling pathway has been well elucidated and recent evidence suggests its involvement in cancer cell biology, demonstrating that the reinforcement of the tumor suppressive functions of LKB1/AMPK is a valuable therapeutic strategy for cancers. Interest in metformin as a novel anticancer agent for breast cancer and other solid tumors continue to grow, currently being investigated in several cancer types in both neoadjuvant and metastatic settings. The biological effect of metformin on cancer cells is driven by its ability to activate AMPK through upstream kinase LKB1, tumor suppressor gene in epithelial tissues. Metformin increases intracellular AMP level, which allosterically activates AMPK. We have previously identified the AMPKα as a modulator in gastric cancer (GC) and through experimental evidence. We show the impact of LKB1/AMPK modulation of HNF4α, a dramatic suppression of cancer cell growth. GC samples were collected and sequenced on SOLiD v 3.0 for both WT-seq and small RNA-seq. Computational analysis showed that 356 out of 18,890 genes were identified as GC related differentially expressed genes in the five-group comparison (normal, tumor stage I, II, III or IV). 28 genes were identified as stage-specific differentially expressed genes, and 13 out of the 28 genes were within the network between HNF4α and HNF1α. In order to test the anti-proliferation activity of metformin in GC cell lines associated with activation of PRKAA1, PRKAA2 and LKB1, and by HNF4α suppression, 4 GC cell lines (NCI-N87, AGS, HS 746T and MKN 45) were treated with metformin. Both PRKAA1/2 showed increased gene expression level when the cells were treated with 10mM of metformin. As for STK11 (LKB1 gene) and HNF4A gene expression level, LKB1 increased and HNF4A decreased with metformin treatment on all four cell lines. Metformin treated NCI-N87 and AGS show that it is involved in cell cycle arrest. Western-blot analysis shows, decreased protein expression of Cyclin A/B and D1 on metformin treated. Lastly, NCI-N87 xenograft study show metformin treated suppression of tumor progression compared to non-treated mouse. During the 28 day treatment of metformin, PRKAA1 and PRKAA2 expression level increased compared to the untreated with. Consistent with in vitro assay, LKB1 level was elevated in the metformin treated tumor compared to the non-treated, and HNF4α level decreased over time in the metformin treated tumor. Study shows that AMPK is a strong therapeutic tumor suppressor target and that metformin is a potential drug for Asian early gastric cancer patient. In our research in progress, we observe potential relationships between the Wnt pathway and AMPKα in light of WNT druggability with metformin. In conclusion LKB1/AMPK by HNF4α inhibition suggests metformin could be a candidate for gastric cancer treatment, probably in combination with conventional chemotherapy and/or as a maintenance therapy. Citation Format: Hae Rim Jung, Hae Ryung Chang, Hye-Hyun Seo, Robert Lemos, Hee Seo Park, Han Liang, Garth Powis, Yon Hui Kim. Metformin increases AMPKα activity by inhibition of AMPKα and cell cycle proliferation in Asian gastric 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 5534. doi:10.1158/1538-7445.AM2013-5534

CANcer-specific Evaluation System (CANES): a high-accuracy platform, for preclinical single/multi-biomarker discovery

The recent creation of enormous, cancer-related “Big Data” public depositories represents a powerful means for understanding tumorigenesis. However, a consistently accurate system for clinically evaluating single/multi-biomarkers remains lacking, and it has been asserted that oft-failed clinical advancement of biomarkers occurs within the very early stages of biomarker assessment. To address these challenges, we developed a clinically testable, web-based tool, CANcer-specific single/multi-biomarker Evaluation System (CANES), to evaluate biomarker effectiveness, across 2,134 whole transcriptome datasets, from 94,147 biological samples (from 18 tumor types). For user-provided single/multi-biomarkers, CANES evaluates the performance of single/multi-biomarker candidates, based on four classification methods, support vector machine, random forest, neural networks, and classification and regression trees. In addition, CANES offers several advantages over earlier analysis tools, including: 1) survival analysis; 2) evaluation of mature miRNAs as markers for user-defined diagnostic or prognostic purposes; and 3) provision of a “pan-cancer” summary view, based on each single marker. We believe that such “landscape” evaluation of single/multi-biomarkers, for diagnostic therapeutic/prognostic decision-making, will be highly valuable for the discovery and “repurposing” of existing biomarkers (and their specific targeted therapies), leading to improved patient therapeutic stratification, a key component of targeted therapy success for the avoidance of therapy resistance.

Systematic approach identifies RHOA as a potential biomarker therapeutic target for Asian gastric cancer

Gastric cancer (GC) is a highly heterogeneous disease, in dire need of specific, biomarker-driven cancer therapies. While the accumulation of cancer “Big Data” has propelled the search for novel molecular targets for GC, its specific subpathway and cellular functions vary from patient to patient. In particular, mutations in the small GTPase gene RHOA have been identified in recent genome-wide sequencing of GC tumors. Moreover, protein overexpression of RHOA was reported in Chinese populations, while RHOA mutations were found in Caucasian GC tumors. To develop evidence-based precision medicine for heterogeneous cancers, we established a systematic approach to integrate transcriptomic and genomic data. Predicted signaling subpathways were then laboratory-validated both in vitro and in vivo, resulting in the identification of new candidate therapeutic targets. Here, we show: i) differences in RHOA expression patterns, and its pathway activity, between Asian and Caucasian GC tumors; ii) in vitro and in vivo perturbed RHOA expression inhibits GC cell growth in high RHOA-expressing cell lines; iii) inverse correlation between RHOA and RHOB expression; and iv) an innovative small molecule design strategy for RHOA inhibitors. In summary, RHOA, and its oncogenic signaling pathway, represent a strong biomarker-driven therapeutic target for Asian GC. This comprehensive strategy represents a promising approach for the development of “hit” compounds.

Abstract 1781: Identification of focal adhesion and actin cytoskeleton regulation family genes as druggable target for gastric cancer

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Gastric cancer(GC) is the 4th most common malignancy in the world and the 2nd most common cause of cancer related death worldwide (1). The known causes of GC are diet, hygiene, Helicobacter pylori, Epstein-Barr virus and hereditary factors. Other genetic causes of GC are known to be heterogenic, partially due to differences in ethnic/genetic background. Surgical removal of the tumor is the first-line treatment. Targeted therapy agents and chemotherapeutic agents that show survival advantage in other cancer types are being evaluated in GC(2,3), which only target few known oncogenes. Targeted therapy for GC is still an unmet need, and in light of personalized medicine, identifying genes specific to GC is critical for its treatment. From our NGS data, we have discovered protein family members involved in focal adhesion and actin cytoskeleton regulation pathways, which are involved in mitosis and metastasis. We developed a subpathway network analysis method called PATHOME(unpublished), to identify pathways related to GC. For the first stage, two datasets (84 Korean GC: GSE13861 and 56 Japanese GC:GSE15081)(4,5). Two datasets (25 Korean GC: GSE36968 and 160 Chinese GC:GSE27342)(6,7,8) were used as validation sets. We identified top 10 common pathways. We have then categorized genes within these pathways according their functional category and druggability, either as drug transporter or druggable target. As a result, we identified genes involved in the focal adhesion and regulation of actin cytoskeleton pathways and cell junction, all critical for cell division and metastasis of tumor cells. Genes G1-9 are tight junction genes, and G10-20 are protein family group related to Ras-like GTP-binding protein, G10 - 20. G10, 11 and 12 are ras-like GTPases, and G13-20 are interacting proteins or downstream proteins. We then compared differential expression of these genes between the Asian and Caucasian dataset. G1-9 show co-regulation pattern among the two ethnic group. G9 in particular, is normally expressed in the stomach, and is downregulated (p=0.0045) in GC tumor samples. In contrast, G10 is upregulated in the Korean dataset and is shown to be upregulated in Japanese and Chinese GC sample (public) as well(9,10), but not in our Caucasian dataset(unpublished). The TCGA stomach cancer provisional data via cBioPortal(11) shows that most of the G10 alterations in GC cases are mutations (9% of total cases) with 2 known cases of downregulation. It is interesting that most of the G10 alterations were found in Caucasian patients (11/18 cases). In our study, we have identified genes that are significant in GC and are druggable targets/drug transporters. It is significant that the pathways we identified are detected in both ethnic groups, with different expression pattern. This finding is critical for personalized treatment and drug development for GC. Citation Format: Hae Ryung Chang, Seungyoon Nam, Myeong-Cherl Kook, Hee Seo Park, Hae Rim Jung, Youme Gim, Han Liang, Garth Powis, Yon Hui Kim. Identification of focal adhesion and actin cytoskeleton regulation family genes as druggable target for gastric 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 1781. doi:10.1158/1538-7445.AM2014-1781

Abstract 2783: Variability of HER2 expression between in vitro and in vivo models within gastric cancer cell lines

In the process of developing individualized cancer therapeutics, tumor derived cell line based preclinical (in vitro and in vivo) methods have been widely used. Patient derived genome analysis revealed significant cancer biomarkers for targeted drug development. Classic cancer biomarker, HER2, overexpression was discovered as a cancer biomarker and as a druggable target in breast cancer, which lead to the development of Trastuzumab. HER2 is overexpressed in 10-25% of gastric cancer, and treatment with Trastuzumab showed positive yet limited activity in advanced gastric cancer treatment in the ToGA study. These results present HER2 as an attractive target for gastric cancer. However, Gillet and colleagues reported the challenges in the use of in vitro and in vivo studies for drug validation and development. They reported variance in gene expression patterns between the in vitro and in vivo context within a single cancer cell line. In our study using gastric cancer derived cell lines, we observed differences in HER2 expression between in vitro and in vivo xenograft model. Also, we show the potential promising in vivo xenograft model in the development of HER2 targeting agents for gastric cancer. Eleven out of 25 gastric cancer cell lines showed HER2 expression, which was demonstrated using the IHC assay on a cell microarray. Sixteen out of 25 cell lines successfully produced xenograft models. Gastric cancer cell lines (SNU-5, SNU-16, SNU-638, SNU-1967, MKN-1, MKN-28, and NUGC-3) that had no HER2 expression when cultured in vitro showed HER2 expression in vivo. Two cell lines (NCC-19 and NCI-N87) that had HER2 expression in vitro showed a decrease in HER2 expression in vivo. Three cell lines (SNU-484, SNU-668, and SNU-719) showed various levels of increase in HER2 expression. Two cell lines showed minor to no changes in HER2 expression. NUGC-4 had consistent significant HER2 expression levels in both assays, while MKN-45 had consistent insignificant to no HER2 expression. Our study demonstrates inconsistency in HER2 gene expression between in vitro and in vivo cancer cell line models for the first time in gastric cancer. Citation Format: Youme Gim, Hee Seo Park, Hae Rim Jung, Hae Ryung Chang, Seungyoon Nam, Yong Doo Choi, Ja-Lok Ku, Yon Hui Kim. Variability of HER2 expression between in vitro and in vivo models within gastric cancer cell lines. [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 2783. doi:10.1158/1538-7445.AM2014-2783

Abstract 5540: Crosstalk between Wnt and AMPK pathway in Asian gastric cancer by metformin.

Stomach cancer is the most prevalent cancer in Eastern Asia(1). Given the diverse route to oncognesis, it is highly attractive to develop targeted therapy based on each patient9s biology(2). Several molecular targeting agents that show survival advantage in other cancer types are being under clinical investigation in gastric cancer (GC) (3-6). These agents focus on a few known oncogenic genes, yet its true efficacy of their biologically targeted therapies in gastric cancer is still unknown. To identify high-quality therapeutic targets, we utilized PATHOME (pathway and genome information), a computational method to identify therapeutic target through sub-pathway employing simple statistic-based approach (unpublished). PATHOME design had two stages: 1) discovery stage aimed at selecting target signaling pathway, and 2) validation stage aimed at identifying small number of reproducible genes from sub-pathways by using independent dataset to provide a novel theragnostic strategy in GC patients. Four sets of GEO database datasets were used: GSE13861 (84 Korean GC), GSE15081 (56 Japanese GC) for the discovery stage, and GSE36968 (25 Korean GC), and GSE27342 (160 Chinese GC) were added for validation. Using Vogelstein et al. as the gold standard, DAVID and GSEA vs. PATHOME were applied for comparison of potency in prediction of signaling pathways(7). Previous study showed that the energy sensing mechanism by AMPKα is involved in Asian GC(8). Metformin is a drug used to treat diabetes, which affects AMPKα in the cell. We treated GC cell line with metformin and saw the expression level of proteins that were selected by PATHOME to test its predictability of biological context. 13 up-regulated genes were identified using 3 datasets (GSE 13861, GSE 15081 and GSE36968), and 5 (GENE1, GENE2, GENE3, GENE4 and GENE5) from all 4 datasets using PATHOME. Transcription factor binding site analysis showed that two of the 5 genes contained GENE6 binding sites. Implications of AMPKα as a potential therapeutic target has been published(9). Metformin is shown to increase PRKAA2 level, which encodes AMPKα, also inhibits HNF4α(8). To see if metformin has any effect on the genes with HNF4α binding site, we observed the expression level of GENE4 and related proteins. Metformin treated GC cell lines (NCI-N87 and AGS) result in decreased viability compared to untreated cells. On Western-blot analysis, metformin treated cells show a decrease of GENE4, GENE6, but not GENE7 or GENE8. Utilizing PATHOME, which showed greater consistency compared to DAVID and GSEA in terms of cancer-related pathways, we showed Wnt pathway to have a potential impact on GC in East Asia. Therefore, our approach has simplified insight of multiplexed signals apply in molecular targeted therapies. We report a cross-talk between AMPKα and Wnt pathway in gastric cancer, which is known to be involved in other cancer types. This finding will lead to other potential drug targets for GC. Citation Format: Hae Ryung Chang, Seungyoon Nam, Hae Rim Jung, ChangHyuk Kwon, Hye-Hyun Seo, Frances S. Sung, Hee Seo Park, Taesung Park, Yon Hui Kim. Crosstalk between Wnt and AMPK pathway in Asian gastric cancer by metformin. [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 5540. doi:10.1158/1538-7445.AM2013-5540