| |
|
|
|
||||
 |
 |
||||||
 |
As part of the NCI supported CGAP-SAGE collaboration the Polyak lab has generated SAGE libraries from normal luminal mammary epithelium and intermediate and high-grade DCIS tumors. By comparing these SAGE libraries they were able to identify transcripts that are highly expressed in DCIS (High in DCIS-HID genes) and in normal (High in Normal-HIN genes) mammary epithelium. Further comparisons to other CGAP-SAGE libraries led to the identification of several transcripts that appear to be present only or most abundantly in DCIS or normal mammary epithelial SAGE libraries. One of these HID genes, HID-5, appeared to be particularly interesting, since it is one of the most highly abundant SAGE tags in high-grade DCIS (467 tags out of 41,000) and its expression is negligible in all other normal human adult tissue. These characteristics make HID-5 a potential high-grade DCIS/breast cancer tumor marker and implicate HID-5 in the pathobiology of high-grade comedo DCIS lesions most of which are ER-. HID-5 corresponds to S100A7 (also called psoriasin), a small calcium binding cytoplasmic protein with unknown function that is also detected extracellularly and is localized to chromosome 1q, a region showing frequent allelic imbalance in breast carcinomas. S100A7/psoriasin is also highly expressed in psoriatic and other hyperproliferative skin lesions, and in squamous carcinomas of the bladder and larynx. These diseases are all characterized by aberrant proliferation, apoptosis, and differentiation, up-regulation of HID-5 in these conditions may indicate its potential involvement in these cellular processes. OBJECTIVES To better understand the role of HID-5 in the initiation, progression and clinical behavior of high-grade ER- breast carcinomas, we propose to characterize its biochemical and biological functions. Specifically we will:
COLLABORATORS Judy E. Garber, M.D., M.P.H.-Dr. Garber is an Associate Professor of Medicine, Dana-Farber Cancer Institute and Harvard Medical School. She is the Director of the Cancer Risk and Prevention Clinic, Dana-Farber Cancer Institute. She is Chair of the CALGB Prevention Committee. She will serve as a co-investigator on P6 directing the clinical trials and a collaborator to Project 4. Email: Judy_Garber@dfci.harvard.edu Kornelia Polyak, M.D., Ph.D.-Dr. Polyak is an Assistant Professor of Medicine, Dana-Farber Cancer Institute and Harvard Medical School. She is a one of the developers of the use of SAGE to define genes involved in cancer progression. Her current research focus is to define the genetic changes involved in the progression from DCIS to invasive breast cancer. She will serve as a co-investigator on Project 4. Email: KPOLYAK@PARTNERS.ORG Stuart J. Schnitt, M.D.-Dr. Schnitt is Associate Professor of Pathology, Beth Israel Deaconess Medical Center and Harvard Medical School and Associate Director of Anatomic Pathology, Beth Israel Deaconess Medical Center. He is an expert in breast cancer pathology. He will serve as a co-investigator on Project 1 and a collaborator on Project 4. Email: Sschnitt@bidmc.harvard.edu DATA Analysis of HID-5 expression in primary breast carcinomas and breast cancer cell lines Comparison of SAGE libraries generated from normal mammary epithelium, and intermediate and high grade DCIS revealed that HID-5/psoriasin is among the most highly differentially expressed transcripts and is one of the most abundant mRNAs in high grade DCIS. However, this analysis included only two independent pairs of tumors. To analyze HID-5/psoriasin expression quantitatively in multiple DCIS samples, we performed real-time PCR analysis of 11 LCM (Laser Capture Microdissection) purified primary tumors and corresponding normal mammary epithelium (Fig. 6A). In most tumors, with the exception of two estrogen and progesterone receptor positive low and intermediate grade lesions (samples 57 and 65), HID-5/psoriasin levels were significantly (³10 fold) increased relative to corresponding normal mammary epithelium (Fig. 6A). To confirm HID-5/psoriasin expression in high grade DCIS epithelial cells at the cellular level we performed mRNA in situ hybridization of two low, two intermediate and two high grade DCIS tumors and corresponding normal epithelium, (Fig. 6B and data not shown). HID-5/psoriasin is highly and specifically expressed by the tumor cells of the two high-grade comedo DCIS (Fig. 6B). In contrast, no hybridization signal was detected in low and intermediate grade DCIS, and normal mammary epithelial cells (Fig. 6B and data not shown).
To analyze the expression of HID-5/psoriasin protein we have generated and characterized polyclonal and monoclonal antibodies and optimized their use for formalin fixed, paraffin embedded tissue. To analyze the in vivo expression of the HID-5/psoriasin protein we performed immunohistochemical analysis of two tissue arrays using monoclonal anti-HID-5/psoriasin antibodies. Samples were also analyzed for the expression of ERa. Among the 78 invasive ductal carcinomas examined, 38 were HIDÐ5/psoriasin positive (15 ERa+ and 23 ERa-) and 40 were HID-5/psoriasin negative (26 ER+ and 14 ERa-). Based on these results HID-5/psoriasin positive tumors are more likely to be ER- (P=0.04, Fisher exact test). Similar correlation between HID-5/psoriasin and ERa patterns was observed in breast cancer cell lines as well.
USEFUL LINKS
|
 |
|||||
