Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The . gov means it’s official. Federal government websites often end in . gov or . mil. Before sharing sensitive information, make sure you’re on a federal government site VSports app下载. .

Https

The site is secure V体育官网. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely. .

. 2013 Jul;5(7):1051-66.
doi: 10.1002/emmm.201201823. Epub 2013 May 13.

Functional genomics identifies five distinct molecular subtypes with clinical relevance and pathways for growth control in epithelial ovarian cancer

Tuan Zea Tan  1 , Qing Hao Miow, Ruby Yun-Ju Huang, Meng Kang Wong, Jieru Ye, Jieying Amelia Lau, Meng Chu Wu, Luqman Hakim Bin Abdul Hadi, Richie Soong, Mahesh Choolani, Ben Davidson, Jahn M Nesland, Ling-Zhi Wang, Noriomi Matsumura, Masaki Mandai, Ikuo Konishi, Boon-Cher Goh, Jeffrey T Chang, Jean Paul Thiery, Seiichi Mori
Affiliations

Functional genomics identifies five distinct molecular subtypes with clinical relevance and pathways for growth control in epithelial ovarian cancer

"V体育安卓版" Tuan Zea Tan et al. EMBO Mol Med. 2013 Jul.

Abstract

Epithelial ovarian cancer (EOC) is hallmarked by a high degree of heterogeneity VSports手机版. To address this heterogeneity, a classification scheme was developed based on gene expression patterns of 1538 tumours. Five, biologically distinct subgroups - Epi-A, Epi-B, Mes, Stem-A and Stem-B - exhibited significantly distinct clinicopathological characteristics, deregulated pathways and patient prognoses, and were validated using independent datasets. To identify subtype-specific molecular targets, ovarian cancer cell lines representing these molecular subtypes were screened against a genome-wide shRNA library. Focusing on the poor-prognosis Stem-A subtype, we found that two genes involved in tubulin processing, TUBGCP4 and NAT10, were essential for cell growth, an observation supported by a pathway analysis that also predicted involvement of microtubule-related processes. Furthermore, we observed that Stem-A cell lines were indeed more sensitive to inhibitors of tubulin polymerization, vincristine and vinorelbine, than the other subtypes. This subtyping offers new insights into the development of novel diagnostic and personalized treatment for EOC patients. .

Keywords: cell line model for subtype; functional genomic screen; molecular subtype; ovarian cancer; tubulin V体育安卓版. .

PubMed Disclaimer

Figures

Figure 1
Figure 1. CC analysis revealed five subtypes of epithelial ovarian carcinoma

  1. Gene expression heatmap for the five tumour clusters (red = high; green = low expression). CC of 1538 samples identified five subtypes, designated by the associated gene components. Note the similarities between Epi-A/Stem-B subtype tumours, between Epi-A/Epi-B subtypes for epithelial genes, and the expression pattern of Epi-A/Stem genes V体育平台登录. Also note that none of cultured cell-line data was included in this analysis.

  2. Kaplan–Meier survival analysis for each subtype VSports注册入口. Among data for 1538 patient samples, survival information for 978 samples was available (GSE3149: 143, GSE9891: 277, TCGA: 400, GSE14764: 80, GSE18520: 53 and Oslo cohort: 25 samples) (Epi-A: 80, Epi-B: 264, Mes: 284, Stem-A: 220, Stem-B: 61 and others: 69 samples) and used for the Kaplan–Meier analysis.

  3. Subtype-specific pathway enrichment. Heatmap shows subtype-specific single sample gene set enrichment analysis (ss-GSEA) scores (false discovery rate (FDR) in significance analysis of microarrays (SAM) q = 0%, receiver operating characteristic (ROC) >0. 85) for 1538 ovarian cancer samples. Red = high; green = low enrichment scores. Gene sets are aligned in descending value of ROC. Samples are aligned by subtype classification and SW. Deep colour = positive SW (core samples); pale colour = samples classified, but negative SW V体育官网入口. “Others” indicates the unclassified samples not grouped in any of the five subtypes in the initial CC analysis in Fig 1. Arrows indicate positions of selected pathways.

  4. Ovarian cancer subtype predictors (BinReg). A heatmap is shown for the predicted probabilities of subtype status on 1413 clinical samples not used in the subtype predictor generation. Red = high; blue = low. Samples were aligned according to subtype classification by CC and SW VSports在线直播. Colour as for (C). “Others” is represented as for (C).

  5. Heatmap of Spearman correlation Rho between the subtype of training data (n = 1538) and the BinReg predicted subtype of samples in five independent datasets (GSE19829, GSE20565, GSE30311, GSE26712 and GSE27651; total n = 418). The validation samples are aligned horizontally according to the predicted subtype, whereas the training samples are aligned vertically according to the subtype. Yellow = high correlation; black = low correlation. Abbreviations: Epi-A, epithelial-A; Epi-B, epithelial-B; Mes, mesenchymal; Stem-A, stem-like-A; Stem-B, stem-like-B.

Figure 2
Figure 2. Identification of cell line subtype status

  1. Five subtypes in ovarian cancer cell line classification. Left panel. CC matrix of 142 ovarian cell lines. Red = high; white = low similarity. Middle panel. Gene expression heatmap of ovarian cell lines. Red = high; green = low expression. Right panel. Silhouette analysis for each subtype. Column to the right of silhouette plot is the SigClust (Liu et al, 2008b) p-value indicative of cluster significance for each subtype.

  2. Prediction of clinical samples by cell line predictors using BinReg. Upper panel. Gene expression heatmaps for subtype predictors based on cell line expression data. Red = high; blue = low expression. Middle panel. Predicted probability of core clinical samples for cell-line subtype predictor by BinReg. Each subtype signature detected the probability difference between the corresponding subtype from the remaining subtypes with statistical significance (p < 0.0001; Mann–Whitney U-test). Lower panel. Receiver operating characteristic (ROC) analyses of subtype predictors. Overall accuracy is shown by the area under the ROC curve (AUC) (Pejovic et al, 2009). Concordance (%) of the subtype status derived from CC with the prediction based on the cell line subtype predictors.

  3. Upper panel. Cell line subtype-specific pathway enrichment. Subtype-specific single sample gene set enrichment analysis (ss-GSEA) scores (false discovery rate (FDR) of the significance analysis of microarrays (SAM) q = 0%, ROC > 0.85 as overexpressed gene sets) for 142 ovarian cell lines are shown as a heatmap. Red = high; green = low enrichment scores. Gene sets aligned in descending value of ROC; samples are aligned according to the subtype classification by CC and the SW. Deep colour = positive SW (core samples); pale colour = samples classified to a subtype, but negative SW. Arrows indicate positions of selected pathways. Lower panel: Concordance (%) of the subtype status (from CC by genes) with the prediction result (from BinReg based on the subtype predictors by enrichment scores). The number in parentheses indicates the accuracy of the prediction against core samples.

  4. Characterization of in vitro phenotypes of cell lines in each subtype. Upper panel. Population doubling time of a cell line was measured with the MTS assay (Matsumura et al, 2011) and is shown as dot plots. Lower panel. Anchorage-independent cell growth ability for each cell line was measured using the methylcellulose assay (Mori et al, 2009). Log10-transformed colony number is shown. p-values were computed by Mann–Whitney U-test. Abbreviations: Epi-A, epithelial-A; Epi-B, epithelial-B; Mes, mesenchymal; Stem-A, stem-like-A; Stem-B, stem-like-B.

Figure 3
Figure 3. Subtype-specific functional relevance genes

  1. Schematic showing identification of functionally relevant genes for cell growth in a subtype-specific manner.

  2. Gene centred and normalized heatmap, compiled from two independent screens, shows hairpins selectively depleted or amplified in each subtype. The quadruplicates of three cell lines (OVCA433; Epi-A, HeyA8; Mes and PA-1; Stem-A) were assayed in the initial screen, while the second screen used one experimental replicate of 14 different cell lines (4 Epi-A: OVCA429, OVCAR-8, OVCA433, PEO1; 5 Mes: ovary1847, HEY, HeyA8, HeyC2, SKOV-3 and 5 Stem-A: A2780, CH1, PA-1, SKOV-4, SKOV-6). Using reads with a perfect match to the reference sequences (Sigma–Aldrich), the copy number of each hairpin was counted and normalized against the total number of reads in a sample and then rendered to RIGER analysis to find phenotype-specific, functionally relevant genes (Luo et al, 2008). Top panel. Subtype-specific depleted hairpins in Epi-A, followed by Mes and Stem-A subtypes. Each row represents shRNA hairpin copy number and is sorted according to the hairpin score identified in RIGER (Luo et al, 2008). Only hairpin scores ≥0.2 and genes significantly enriched in a subtype (q < 0.005) are shown. Bottom panel. Subtype-specific amplified hairpins arranged as in the top panel. Red = higher; green = lower copy number counts.

  3. Schematic of siRNA experiments validating the identified Stem-A-specific growth-promoting genes. This analysis led to the identification of two functionally relevant genes specific to Stem-A: TUBGCP4 and NAT10.

  4. Validation of subtype-selective effect of the genes on cell growth by siRNAs. Upper panel. Timeline of assay performed for the siRNA reverse-transfection experiment. Lower panel. Effect of gene knockdown on cell growth (bar plots) as a percentage ratio of growth suppression, normalized against the negative controls. Error bar indicates the SEM of three independent experiments. Stem-A-selective growth suppression effect is shown for the inhibition of the five validated PA-1 (Stem-A)-specific growth-promoting genes in OVCA433, HeyA8 and PA-1, respectively. Green = OVCA433 (Epi-A); red = HeyA8 (Mes); blue = PA-1 (Stem-A).

  5. Effect of silencing PA-1 (Stem-A)-selective genes on cell growth in other ovarian cancer cell lines. The five PA-1-selective genes were silenced individually by siRNA in non-Stem-A (OVCA433, OVCA429, PEO1, HeyA8, ovary1847, SKOV-3 and HEY) and Stem-A (PA-1, CH1, A2780 and OVCAR-3) cell lines in three independent experiments, and examined for their effect on cell growth relative to the negative control. Averaged percentages of growth suppression in each group are shown as a box plot and were statistically evaluated using Mann–Whitney U-test with GraphPad Prism. Bottom, middle and top lines of each box represent the 25th percentile, median and 75th percentile, respectively, and whiskers extend to the most extreme values of the group. Inhibition with siTUBGCP4 or siNAT10 significantly suppressed cell growth of Stem-A cell lines as compared to non-Stem-A cell lines. Grey = non-Stem-A cell lines; blue = Stem-A cell lines. Abbreviations: Epi-A, epithelial-A; Mes, mesenchymal; Stem-A, stem-like-A.

Figure 4
Figure 4. Susceptibility of Stem-A cells to microtubule assembly inhibitors

  1. Estimated microtubule activity in non-Stem-A and Stem-A subgroups of ovarian cancer. Microtubule activity in 1142 core samples of ovarian clinical tumours (Top panel) and in 129 core samples of ovarian cell lines (Bottom panel) was estimated based on the average single sample gene set enrichment analysis (ss-GSEA) enrichment score of 19 microtubule-related gene sets (Supporting Information Table 16) acquired from GSEA databases (Supporting Information Table 6). Differences in microtubule activity between non-Stem-A and Stem-A subgroups were statistically evaluated with Mann–Whitney U-test in Graphpad Prism. Grey = non-Stem-A subgroup; blue = Stem-A subgroup.

  2. Specificity of drug sensitivity in ovarian cancer cell lines. A panel of 18 ovarian cancer cell lines was classified into non-Stem-A (OVCA433, OVCA429, OVCAR-8, PEO1, OVCA432, OVCA420, HeyA8, HEY, HeyC2, SKOV-3, ovary1847 and DOV 13) or Stem-A (PA-1, CH1, A2780, OVCAR-3, SKOV-4 and SKOV-6) groups and analysed for their sensitivity to paclitaxel (Top panel), vincristine (Left bottom panel) and vinorelbine (Right bottom panel). GI50 values were calculated with the results from cell proliferation assays for each cell type in three independent experiments, and the mean GI50s are shown as dot plots. A non-parametric Mann–Whitney U-test in Graphpad Prism was used to evaluate the results statistically. A higher value along the y-axis indicates increased sensitivity to the drugs. Colour as for (A).

  3. Detection of apoptotic activity upon vincristine treatment. Six non-Stem-A (Upper panel) and four Stem-A (Lower panel) cell lines were subjected to increasing concentrations of vincristine (0 to 10 nM) for 48 h. The presence of apoptotic activity was determined by immunoblotting for cleaved PARP and Caspase-3, as indicated by arrows. Abbreviations: Stem-A, stem-like-A.

See this image and copyright information in PMC

References

    1. Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000;403:503–511. - PubMed
    1. Anglesio MS, Arnold JM, George J, Tinker AV, Tothill R, Waddell N, Simms L, Locandro B, Fereday S, Traficante N, et al. Mutation of ERBB2 provides a novel alternative mechanism for the ubiquitous activation of RAS-MAPK in ovarian serous low malignant potential tumors. Mol Cancer Res. 2008;6:1678–1690. - PMC - PubMed
    1. Armstrong DK, Bundy B, Wenzel L, Huang HQ, Baergen R, Lele S, Copeland LJ, Walker JL, Burger RA. Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med. 2006;354:34–43. - PubMed
    1. Barbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF, Schinzel AC, Sandy P, Meylan E, Scholl C, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature. 2009;462:108–112. - PMC - PubMed
    1. Bast RC, Jr, Hennessy B, Mills GB. The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer. 2009;9:415–428. - PMC - PubMed

MeSH terms (V体育2025版)

Substances