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

Molecular response to aromatase inhibitor treatment in primary breast cancer

Background

Aromatase inhibitors such as anastrozole and letrozole are highly effective suppressants of estrogen synthesis in postmenopausal women and are the most effective endocrine treatments for hormone receptor positive breast cancer in such women. Little is known of the molecular effects of these agents on human breast carcinomas in vivo.

Methods

We randomly assigned primary estrogen receptor positive breast cancer patients to treatment with anastrozole or letrozole for 2 weeks before surgery. Expression profiling using cDNA arrays was conducted on pretreatment and post-treatment biopsies. Sample pairs from 34 patients provided sufficient RNA for analysis.

Results

Profound changes in gene expression were seen with both aromatase inhibitors, including many classical estrogen-dependent genes such as TFF1, CCND1, PDZK1 and AGR2, but also many other genes that are likely to represent secondary responses; decrease in the expression of proliferation-related genes were particularly prominent. Many upregulated genes are involved in extracellular matrix remodelling, including collagens and members of the small leucine-rich proteoglycan family (LUM, DCN, and ASPN). No significant differences were seen between letrozole and anastrozole in terms of molecular effects. The gene changes were integrated into a Global Index of Dependence on Estrogen (GIDE), which enumerates the genes changing by at least twofold with therapy. The GIDE varied markedly between tumours and related significantly to pretreatment levels of HER2 and changes in immunohistochemically detected Ki67.

Conclusion

Our findings identify the transcriptional signatures associated with aromatase inhibitor treatment of primary breast tumours. Larger datasets using this approach should enable identification of estrogen-dependent molecular changes, which are the determinants of benefit or resistance to endocrine therapy.

Transformation of different human breast epithelial cell types leads to distinct tumor phenotypes

We investigated the influence of normal cell phenotype on the neoplastic phenotype by comparing tumors derived from two different normal human mammary epithelial cell populations, one of which was isolated using a new culture medium. Transformation of these two cell populations with the same set of genetic elements yielded cells that formed tumor xenografts exhibiting major differences in histopathology, tumorigenicity, and metastatic behavior. While one cell type (HMECs) yielded squamous cell carcinomas, the other cell type (BPECs) yielded tumors closely resembling human breast adenocarcinomas. Transformed BPECs gave rise to lung metastases and were up to 10(4)-fold more tumorigenic than transformed HMECs, which are nonmetastatic. Hence, the pre-existing differences between BPECs and HMECs strongly influence the phenotypes of their transformed derivatives.

Comparison of hybridization-based and sequencing-based gene expression technologies on biological replicates

Background

High-throughput systems for gene expression profiling have been developed and have matured rapidly through the past decade. Broadly, these can be divided into two categories: hybridization-based and sequencing-based approaches. With data from different technologies being accumulated, concerns and challenges are raised about the level of agreement across technologies. As part of an ongoing large-scale cross-platform data comparison framework, we report here a comparison based on identical samples between one-dye DNA microarray platforms and MPSS (Massively Parallel Signature Sequencing).

Results

The DNA microarray platforms generally provided highly correlated data, while moderate correlations between microarrays and MPSS were obtained. Disagreements between the two types of technologies can be attributed to limitations inherent to both technologies. The variation found between pooled biological replicates underlines the importance of exercising caution in identification of differential expression, especially for the purposes of biomarker discovery.

Conclusion

Based on different principles, hybridization-based and sequencing-based technologies should be considered complementary to each other, rather than competitive alternatives for measuring gene expression, and currently, both are important tools for transcriptome profiling.

Extracellular sulfatases, elements of the Wnt signaling pathway, positively regulate growth and tumorigenicity of human pancreatic cancer cells

Background

Heparan sulfate proteoglycans (HSPGs) are control elements in Wnt signaling, which bind extracellularly to Wnt ligands and regulate their ability to interact with signal transduction receptors on the cell surface. Sulf-1 and Sulf-2 are novel extracellular sulfatases that act on internal glucosamine-6-sulfate (6S) modifications within HSPGs and thereby modulate HSPG interactions with various signaling molecules, including Wnt ligands. Emerging evidence indicates the importance of reactivated Wnt signaling in a number of cancers, including pancreatic adenocarcinoma.

Principle Findings

Both Sulf proteins were upregulated in human pancreatic adenocarcinoma tumors and were broadly expressed in human pancreatic adenocarcinoma cell lines. Expression of human extracellular sulfatases Sulf-1 and Sulf-2 enhanced Wnt signaling in a reconstituted system. Three of four pancreatic adenocarcinoma cell lines tested exhibited autocrine Wnt signaling, in that extracellular Wnt ligands were required to initiate downstream Wnt signaling. Exposure of these pancreatic adenocarcinoma cells to a catalytically inactive form of Sulf-2 or siRNA-mediated silencing of endogenous Sulf-2 inhibited both Wnt signaling and cell growth. Sulf-2 silencing in two of these lines resulted in markedly reduced tumorigenesis in immunocompromised mice.

Conclusions/Significance

We have identified the Sulfs as potentiators of autocrine Wnt signaling in pancreatic cancer cells and have demonstrated their contribution to the growth and tumorigenicity of these cells. Since the Sulfs are extracellular enzymes, they would be attractive targets for therapy of pancreatic cancer. Our results run counter to the prevailing view in the literature that the Sulfs are negative regulators of tumorigenesis.

Novel markers for differentiation of lobular and ductal invasive breast carcinomas by laser microdissection and microarray analysis

Background

Invasive ductal and lobular carcinomas (IDC and ILC) are the most common histological types of breast cancer. Clinical follow-up data and metastatic patterns suggest that the development and progression of these tumors are different. The aim of our study was to identify gene expression profiles of IDC and ILC in relation to normal breast epithelial cells.

Methods

We examined 30 samples (normal ductal and lobular cells from 10 patients, IDC cells from 5 patients, ILC cells from 5 patients) microdissected from cryosections of ten mastectomy specimens from postmenopausal patients. Fifty nanograms of total RNA were amplified and labeled by PCR and in vitro transcription. Samples were analysed upon Affymetrix U133 Plus 2.0 Arrays. The expression of seven differentially expressed genes (CDH1, EMP1, DDR1, DVL1, KRT5, KRT6, KRT17) was verified by immunohistochemistry on tissue microarrays. Expression of ASPN mRNA was validated by in situ hybridization on frozen sections, and CTHRC1, ASPN and COL3A1 were tested by PCR.

Results

Using GCOS pairwise comparison algorithm and rank products we have identified 84 named genes common to ILC versus normal cell types, 74 named genes common to IDC versus normal cell types, 78 named genes differentially expressed between normal ductal and lobular cells, and 28 named genes between IDC and ILC. Genes distinguishing between IDC and ILC are involved in epithelial-mesenchymal transition, TGF-beta and Wnt signaling. These changes were present in both tumor types but appeared to be more prominent in ILC. Immunohistochemistry for several novel markers (EMP1, DVL1, DDR1) distinguished large sets of IDC from ILC.

Conclusion

IDC and ILC can be differentiated both at the gene and protein levels. In this study we report two candidate genes, asporin (ASPN) and collagen triple helix repeat containing 1 (CTHRC1) which might be significant in breast carcinogenesis. Besides E-cadherin, the proteins validated on tissue microarrays (EMP1, DVL1, DDR1) may represent novel immunohistochemical markers helpful in distinguishing between IDC and ILC. Further studies with larger sets of patients are needed to verify the gene expression profiles of various histological types of breast cancer in order to determine molecular subclassifications, prognosis and the optimum treatment strategies.