) To develop a complete index of genes expressed in breast tumors, the construction of full length representational cDNA libraries of human breast tissue is an essential first step. For the construction of initial relevant cDNA libraries, two important prerequisites must be met: (1) the availability of homogeneous cell populations representing various stages of malignant progression, such as, atypical hyperplasia, in situ carcinoma, primary invasive tumor, and metastatic disease which are not admixed with normal or interfering cell populations, and (2) the molecular methodology scaled down to employ small populations of human cells containing intact mRNA. We propose to focus on the standardization of Microquantity-cDNA (Mq-cDNA) library construction and in determining its applicability for expression analyses of human breast cells, and will be done in three phases. The first phase will be to optimize our present technologies for widespread usage of this methodology by construction of unidirectional Mq-cDNA libraries using the MCF7 and MDA 435 breast tumor cell lines. We will obtain full length sequences by identifying the most efficient and effective protocol for including 5' ends. Studies will be done to optimize and maintain transcript representation and frequency. The optimal conditions for amplification will be determined and the efficiency of repeated use of ds-cDNA templates will be tested. The second phase will be to utilize the Mq-cDNA protocol for establishing libraries of normal and neoplastic breast cells in order to determine the feasibility and validity of this approach. Mq-cDNA templates and libraries will be constructed that represent breast tissue-derived cells from non-cancerous breast cells, primary and metastatic breast cells and cultures. The third phase will be to utilize Mq-cDNA libraries in the rapid identification of tumor-associated gene expression. The resulting Mq-cDNA templates and libraries will be used to study gene expression in breast tumors. Comparisons will be done between 1) resting vs proliferative populations of non-malignant breast epithelium, 2) malignant and non-malignant breast epithelial cultures derived from the same individual, and 3) case matched tumors from lymph node negative (""""""""non-aggressive"""""""" tumors) versus lymph node positive (""""""""aggressive"""""""" tumors) patients. For these comparisons both differential display-revers transcriptase-PCR (DDRT-PCR) and suppressive subtractive hybridization (SSH) PCR will be used to identify differential gene expression.
