STRATEGIES FOR HIGH-THROUGHPUT GENE EXPRESSION ANALYSIS OF THE BOVINE MAMMARY GLAND

¹ Department of Animal Sciences University of Arizona Tucson, AZ 85721

² Animal Agricultural Group Monsanto Company 700 Chesterfield Parkway North St. Louis, MO 63198

Several approaches have been described for high-throughput gene expression analysis. They include cDNA arrays, oligonucleotide arrays, differential display, SAGE and subtraction cDNA libraries. We have used human microarrays (UniGEM 1.0, Incyte Pharmaceuticals, CA) to determine the relative expression of genes in the bovine mammary gland. The gene expression profiles were then compared with phenotypic data such as milk production and cell proliferation index to identify genes whose regulation appears to be associated with these processes. Heterologous transcript profiling was used to determine the relative expression of genes in bovine mammary gland in two different experiments. A group of heifers (n=20) were hormonally treated to stimulate mammary epithelial cell proliferation. Mammary tissue was collected for determining thymidine incorporation into DNA and poly A+ RNA isolation for transcript profiling. In a second experiment, mammary tissue biopsies plus lactation records were collected from lactating dairy cows (n=15) on weeks 8, 22 and 44 of lactation and poly A+ RNA was isolated for expression profiling. Expression profiling data obtained from both of these experiments were analyzed with Mine Set software to identify genes with 1.8 fold or more absolute differential expression. Genes associated with proliferation and persistency (rate of decline of milk production) were identified by categorical correlation analysis. This analysis identified about 1200 genes associated with proliferation and 300 genes associated with persistency. Comparison of these two data sets was used to identify genes potentially associated with cell proliferation in the mammary gland of lactating dairy cows since this process will affect rate of decline of milk production. This analysis identified 49 genes that were cross-correlated to both phenotypes. Many of the genes identified by this approach were genes modulating cell proliferation and apoptosis. These genes will be selected for mapping either using a bovine radiation hybrid panel or by human-bovine comparative mapping. In addition, gene expression profile experiments such as these will facilitate our understanding of lactation biology during normal development and function.