Dataset Information

In vitro gene expression profile of TGFbeta-regulated genes in MCF10A-based xenograft model of breast cancer progression

ABSTRACT: TGF-betas have complex roles in tumorigenesis, with context-dependent effects that can either suppress or promote tumor progression. Our goal was to use integrated genomic approaches in a model of human breast cancer progression to identify core TGF-beta-regulated genes that specifically reflect the tumor suppressor activity of TGF-beta. The model consisted of the non-tumorigenic MCF10A (“M1”), the premalignant MCF10AT1k.cl2 (“M2”), the early malignant MCF10Ca1h (“M3”) and the highly malignant, metastatic MCF10Ca1a.cl1 (“M4”) cell lines. We have previously shown that tumor suppressor activity of TGF-beta is lost in the highly malignant M4 cells. To determine how the spectrum of TGF-beta-regulated genes changes with cancer progression, we performed gene expression array analysis on four cell lines of the MCF10A-based model of breast cancer progression (M1-M4) cultured in vitro under serum-free conditions and treated with TGF-beta (5ng/ml plus condition) or vehicle (minus condition) for 1h or 6h.

ORGANISM(S): Homo sapiens

PROVIDER: GSE34270 | GEO | 2015/01/01

SECONDARY ACCESSION(S): PRJNA156533

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Purpose: The goal of the study is to compare NGS-derived transcriptome in an isogenic breast cancer progression model cell line system. By comparing the protein-coding and noncoding gene expression of normal versus tumorigenic breast cancer cell lines, we will be able to identify genes that show aberrant expression during breast cancer progression. Methods: We isolated poly A + RNA from four isogenic mammary epithelial cell lines showing various stages of breast cancer progression. The model system comprises of 4 isogenic cell lines, categorized as M1-M4. M1 represents the normal, non-tumorigenic, immortalized MCF10A cells. Transfection of MCF10A with activated T24-HRAS and selection by xenografting generated the M2 (MCF10AT1k.cl2) cell line, which is highly proliferative and gives rise to premalignant lesions with the potential for neoplastic progression. M3 (MCF10Ca1h) and M4 (MCF10CA1a.cl1) were derived from occasional carcinomas arising from xenografts of M2 cells. M3 gives predominantly well-differentiated low-grade carcinomas on xenografting, while M4 gives rise to relatively undifferentiated carcinomas and colonizes to the lung upon injection of these cells into the tail vein. We performed paired-end deep sequencing (190-260 million reads/sample) of poly A+ RNA isolated from these cells that were cultured as 3D acini in biological duplicates. Reads of the samples were trimmed for adapters and low-quality bases using Trimmomatic software before alignment with the reference genome (Human - hg19) and the annotated transcripts using STAR. The average mapping rate of all samples is 96%. Unique alignment is above 87%. There are 3.74 to 4.07% unmapped reads. The mapping statistics are calculated using Picard software. The samples have 0.59% ribosomal bases. Percent coding bases are between 67-72%. Percent UTR bases are 23-26%, and mRNA bases are between 94-96% for all the samples. Library complexity is measured in terms of unique fragments in the mapped reads using Picard’s MarkDuplicate utility. The samples have 31-52% non-duplicate reads. In addition, the gene expression quantification analysis was performed for all samples using STAR/RSEM tools. Both the normalized count and the raw count are provided as part of the data delivery. Results: Using an optimised data analysis workflow, we mapped ~190-250 million reads/sample and identified expression of 17396 protein-coding genes and 11509 long noncoding RNA genes. We initially compared gene expression between M1 and M4 cells. 4668 genes (2815 protein coding and 1853 lncRNAs) showed ~2 fold change in their expression between M1 and M4 cells in both biological repeats. 1159 out of the 1853 deregulated lncRNAs showed 2-fold upregulation in M4 cells in both repeats. On the other hand, 694 of lncRNAs displayed reduced levels in M4 compared to M1 cells. Further, we noticed that natural antisense transcripts (NATs) comprised one of the largest types of lncRNAs (504 out of 1853) that showed deregulation in M4 cells. Conclusion: Our study revealed differential expression of thousands of protein-coding and long noncoding RNAs during breast cancer progression using the isogenic cell line model system. This data set will act as a rich resource for downstream mechanistic studies to determine the role of these differentially expressed genes in breast cancer progression.

Project description:Epithelial to mesenchymal transition (EMT) is an extreme example of cell plasticity, important for normal development, injury repair, and malignant progression. Widespread epigenetic reprogramming occurs during stem cell differentiation and malignant transformation, but EMT-related epigenetic reprogramming is poorly understood. Here we investigated epigenetic modifications during TGF-β-mediated EMT. While DNA methylation was unchanged during EMT, we found global reduction of the heterochromatin mark H3-lys9 dimethylation (H3K9Me2), increase of the euchromatin mark H3-lys4 trimethylation (H3K4Me3), and increase of the transcriptional mark H3-lys36 trimethylation (H3K36Me3). These changes were largely dependent on lysine-specific deaminase-1 (LSD1), and LSD1 loss-of-function experiments showed marked effects on EMT-driven cell migration and chemoresistance. Genome-scale mapping revealed that chromatin changes were largely specific to large organized heterochromatin K9-modifications (LOCKs), suggesting that EMT is characterized by reprogramming of specific chromatin domains across the genome. Chromatin immunoprecipitation (ChIP) was performed with antibodies against H3K9Me2 (Abcam, ab1220), H3K36Me3 (ab9050), and H3K4Me3 (ab8580) on native (unfixed) chromatin isolated from fully differentiated mouse AML12 cells (confluent, serum starved for 48hrs) either treated with TGF-β for 36hrs to induce EMT or not treated with TGF-β (0hrs, differentiated AML12 cells). DNA purified from these samples was then either whole-genome amplified (H3K36Me3 and H3K4Me3) and hybridized or directly (H3K9Me2) hybridized to NimbleGen 2.1M economy whole-genome tiling arrays #2 (listed below as array 1) and #3 (listed below as array 2), which cover mouse chromosomes 4-14. For each sample, the immunoprecipitated DNA (IP) and the input (control) DNA were hybridized to the arrays, and the IP is normalized to the input. There are 16 total samples listed below. There are 8 sample for H3K9Me2 (TGF-β and no TGF-β for arrays 1 and 2, done in replicate for a total of 8). There are 4 samples for H3K36Me3 (TGF-β and no TGF-β done on array 1 and array 2). There are 4 total samples for H3K4Me3 (TGF-β and no TGF-β done on array 1 and array 2).

Dataset Information

In vitro gene expression profile of TGFbeta-regulated genes in MCF10A-based xenograft model of breast cancer progression

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure