Laser microdissection and microarray analysis of breast tumors reveal ER-|[alpha]| related genes and pathways

doi:doi:10.1038/sj.onc.1209165

Oncogene (2006) 25, 1413–1419. doi:10.1038/sj.onc.1209165; published online 31 October 2005

Laser microdissection and microarray analysis of breast tumors reveal ER- related genes and pathways

F Yang¹, J A Foekens², J Yu¹, A M Sieuwerts², M Timmermans², J G M Klijn², D Atkins¹, Y Wang¹ and Y Jiang¹

¹Veridex LLC, a Johnson & Johnson Company, 3210 Merryfield Row, San Diego, CA, USA
²Department of Medical Oncology, Erasmus Medical Center – Daniel den Hoed Cancer Center, Rotterdam, The Netherlands

Correspondence: Dr Y Jiang, Veridex LLC, a Johnson & Johnson Company, 3210 Merryfield Row, San Diego, CA 92121, USA. E-mail: jjiang2@vrxus.jnj.com

Received 12 April 2005; Revised 12 August 2005; Accepted 7 September 2005; Published online 31 October 2005.

Top

Abstract

About 70–80% of breast cancers express estrogen receptor alpha (ER-), and estrogens play important roles in the development and growth of hormone-dependent tumors. Together with lymph node metastasis, tumor size, and histological grade, ER status is considered as one of the prognostic factors in breast cancer, and an indicator for hormonal treatment. To investigate genes and pathways that are associated with ER status and epithelial cells in breast tumor, we applied laser capture microdissection (LCM) technology to capture epithelial tumor cells from 28 lymph node-negative breast tumor samples, in which 17 patients had ER-+ tumors, and 11 patients have ER-- tumors. Gene expression profiles were analysed on Affymetrix Hu133A GeneChip. Meanwhile, gene profiles using total RNA isolated from bulk tumors of the same 28 patients were also generated. In total, 146 genes and 112 genes with significant P-value and having significant differential expression between ER-+ and ER-- tumors were identified from the LCM data set and bulk tissue data set, respectively. A total of 61 genes were found to be common in both data sets, while 85 genes were unique to the LCM data set and 51 genes were present only in the bulk tumor data set. Pathway analysis with the 85 genes using Gene Ontology suggested that genes involved in endocytosis, ceramide generation, Ras/ERK/Ark cascade, and JAT-STAT pathways may play roles related to ER. The gene profiling with LCM-captured tumor cells provides a unique approach to study epithelial tumor cells and to gain an insight into signaling pathways associated with ER.