Establishment of the epithelial-specific transcriptome of normal and malignant human breast cells based on MPSS and array expression data

Abstract

Introduction: Diverse microarray and sequencing technologies have been widely used to characterise the molecular changes in malignant epithelial cells in breast cancers. Such gene expression studies to identify markers and targets in tumour cells are, however, compromised by the cellular heterogeneity of solid breast tumours and by the lack of appropriate counterparts representing normal breast epithelial cells.

Methods: Malignant neoplastic epithelial cells from primary breast cancers and luminal and myoepithelial cells isolated from normal human breast tissue were isolated by immunomagnetic separation methods. Pools of RNA from highly enriched preparations of these cell types were subjected to expression profiling using massively parallel signature sequencing (MPSS) and four different genome wide microarray platforms. Functional related transcripts of the differential tumour epithelial transcriptome were used for gene set enrichment analysis to identify enrichment of luminal and myoepithelial type genes. Clinical pathological validation of a small number of genes was performed on tissue microarrays.

Results: MPSS identified 6,553 differentially expressed genes between the pool of normal luminal cells and that of primary tumours substantially enriched for epithelial cells, of which 98% were represented and 60% were confirmed by microarray profiling. Significant expression level changes between these two samples detected only by microarray technology were shown by 4,149 transcripts, resulting in a combined differential tumour epithelial transcriptome of 8,051 genes. Microarray gene signatures identified a comprehensive list of 907 and 955 transcripts whose expression differed between luminal epithelial cells and myoepithelial cells, respectively. Functional annotation and gene set enrichment analysis highlighted a group of genes related to skeletal development that were associated with the myoepithelial/basal cells and upregulated in the tumour sample. One of the most highly overexpressed genes in this category, that encoding periostin, was analysed immunohistochemically on breast cancer tissue microarrays and its expression in neoplastic cells correlated with poor outcome in a cohort of poor prognosis estrogen receptor-positive tumours.

Conclusion: Using highly enriched cell populations in combination with multiplatform gene expression profiling studies, a comprehensive analysis of molecular changes between the normal and malignant breast tissue was established. This study provides a basis for the identification of novel and potentially important targets for diagnosis, prognosis and therapy in breast cancer.

Figure 1

Comparison of massively parallel signature sequencing (MPSS) data with microarray analysis. Differentially expressed gene profiles from MPSS (100%) were overlaid with each microarray platform individually. (a) Percentage of coverage (light grey) and concordance in differential expression between MPSS and individual arrays (dark grey) are shown together with the combined coverage and confirmation by at least one array (1 platform). (b) Enumeration of the differentially expressed transcripts detected by "MPSS-only", by "MPSS and array", and those transcripts reported as differential by at least two arrays, but not by MPSS ("Array only"). The results obtained by RT-PCR for these subgroups are shown below (see Additional file 6).

Figure 2

Functional classification of the differentially expressed epithelial tumour transcriptome. The top 15 biological processes showing overall (a) up-regulation and (b) down-regulation are shown. The biological processes are ranked from top to bottom according to their ascending P value as described in the Materials and methods. The numbers of genes within each process that are up-regulated or down-regulated for each category are also shown as black and grey bars, respectively.

Figure 3

Heatmap of the top 50 genes from the luminal-specific and myoepithelial-specific transcriptomes. Genes were ranked in order of fold change (myoepithelial over luminal) for each platform separately after which a median rank over all four platforms was determined. Genes are listed with their human transcriptome database (HTR) cluster, HUGO Name, description and UniGene and RefSeq identifiers. Green corresponds to luminal-type; red to myoepithelial-type; black indicates no corresponding microarray feature. Expression measurements obtained by: 1, Agilent; 2, 20 k brk; 3, CodeLink; 4, Affymetrix platform.

Figure 4

Enrichment of luminal and myoepithelial transcripts in the differentially expressed epithelial tumour transcriptome. (a) The top 20 deregulated biological processes identified by gene set enrichment analysis that are enriched in luminal (green) and myoepithelial (red) expression are shown. The definition of each Gene Ontology (GO) category is given in Additional file 8. (b) Heatmap of the skeletal developmental gene subset (GO:0001501) based on the Affymetrix expression data. L (luminal) and M (myoepithelial) show results from individual arrays. Genes are ranked according to their significance of enrichment as described in the Materials and methods.

Figure 5

Immunohistochemical analysis of periostin (POSTN), IL8 and cartilage oligomeric matrix protein (COMP). (a) POSTN-positive invasive ductal carcinoma (IDC; ×400), in which both epithelial and stromal cells show cytoplasmic expression. (b) POSTN-negative IDC in which only the spindle shaped stromal cells are stained (×400). (c) IL8 (×100), showing positive staining only in the malignant breast epithelial cells. (d) COMP expression in the epithelial and stromal cells of an IDC, showing strong expression in both stromal and epithelial cells (×100).

Figure 6

Cumulative Kaplan-Meier curves for epithelial expression of periostin (POSTN). A cohort of poor-prognosis estrogen receptor (ER)-positive tumours was analysed showing: (a) a significantly shorter overall survival (P = 0.0083); (b) a shorter disease free survival (P = 0.0136).