ABSTRACT: The normal growth and function of mammary epithelial cells depend on interactions with the supportive stroma. Alterations in this communication can lead to the progression or expansion of malignant growth. The human mammary gland contains two distinctive types of fibroblasts within the stroma. The epithelial cells are surrounded by loosely connected intralobular fibroblasts, which are subsequently surrounded by the more compacted interlobular fibroblasts. The different proximity of these fibroblasts to the epithelial cells suggests distinctive functions for these two subtypes. In this report, we compared the gene expression profiles between the two stromal subtypes. No statistically significant difference was detected between the gene expression profiles of the interlobular and intralobular fibroblasts by microarray analysis and RT-PCR. However, for some of the genes tested, the protein expression patterns between the two subtypes of fibroblasts were significantly different. This study is the first to report the gene expression profiles of the two distinct fibroblast populations within the human mammary gland. While there was no significant difference in the gene expression profiles between the groups, there was an obvious difference in the expression pattern of several proteins tested. This report also highlights the importance of studying gene regulation at both the transcriptional and post-translational level. Experiment Overall Design: Sample Collection Experiment Overall Design: This study was performed in accordance with the guidelines of the National Cancer Institute Review Board, protocol 02-C-0144. All patients provided written informed consent. Fresh human mammary tissue was collected from four (two Caucasian, one African-American, and one Hispanic) female, pre-menopausal, reduction mammoplasty patients, ages ranging from 18 to 40 years old. The tissue was embedded in Tissue-Tek O.C.T. embedding medium (Sankura Finetek Inc., Torrance CA) and frozen on dry ice immediately after surgery. Eight - 10 micron sections of tissue were cut using a Leica 2800 Frigocut-E cryostat (Bannockburn, IL). Every tenth section was subjected to hematoxylin and eosin staining. For each patient sample, sections with distinctive intralobular and interlobular regions were selected for laser capture and microarray analysis. Experiment Overall Design: Laser capture microdissection and microarrays Experiment Overall Design: Laser capture microdissection (LCM) and microarrays were performed by Cogenics, Inc. (Morrisville, NC). Briefly, selected intralobular and interlobular stroma sections of frozen tissue were subjected to an AutoPixTM automated LCM system from Arcturus, using static image settings. RNA was isolated from each specimen, pooled, and then evaluated by spectrophotometry and by using an Agilent Bioanalyzer before proceeding to sample amplification. For each sample, 50 ng of total RNA was amplified using Affymetrix Two-Cycle Target Labeling kit (Santa Clara, CA). Ten micrograms of biotinylated cRNA spiked with bioB, bioC, bioD, and cre as a control was hybridized to the Affymetrix Human Genome U133 GeneChip® for 16 h at 45°C. Following hybridization, arrays were washed and stained with Affymetrix GeneChip Fluidics Station. Stained arrays were scanned with an Affymetrix GeneChip Scanner 3000. Quality check and preliminary data analysis were carried out using Affymetrix GeneChip Operating Software and Quality Reporter. Experiment Overall Design: Microarray analysis Experiment Overall Design: Microarray data were analyzed using the Affy package available at the Bioconductor website (www.bioconductor.org). The raw data were first background-corrected by the Robust Multichip Average (RMA) method and then normalized by an invariant set method. Unsupervised hierarchical clustering analysis was performed on 1,115 most variable genes. The difference of gene expression between the inter- and intra-stromal samples was analyzed by the Limma package available at the Bioconductor website. P-values obtained from the multiple comparison tests were corrected by false discovery rates.
