(10g) Application of Microarrays to Reverse Engineer Attachment Dependence in Mammalian Cell Lines

With the advent of high quality glass slides and precision robotics, microarrays have gained widespread recognition as powerful bioinformatics tools. Increasingly, microarray-based experiments have focused on exploring metabolic pathways, biological mechanisms, and genome-wide differences between tumor cells and normal/healthy cells. We have used both cDNA and oligonucleotide microarrays to identify variations in gene expression between similarly derived, but different phenotypic HeLa cell lines. We concentrated on comparing attachment-dependent with attachment-independent HeLa cells; both derived from a common ancestral cell line. Initial work focused on growth characterization in controlled environments, enabling us to prepare samples with cells properly aligned in the same phase of growth; a key prerequisite for meaningful microarray analysis. The microarray data was first normalized using statistical software and then filtered to scour the genome for likely candidates for further investigation. A combination of algorithms such as principal components analysis and k-means clustering were employed for this purpose; organizing genes into distinct groups. Using this approach, six genes were identified as potential targets for further study. Expression ratios obtained by microarray analysis were verified using Real time RT-PCR for the six selected genes previously indicated. Focusing first on a type II membrane protein, we were able to alter the physiology of the attachment-independent HeLa cells using siRNA. Numerous size characterization studies indicated that blocking the expression of this membrane protein in HeLa cells dramatically increased the relative number of multi-cellular aggregate clusters. In addition, we over-expressed a protein associated with attachment, in attachment-independent HeLa cells in order to alter the adhesion properties. We believe our strategy of applying bio-informatics techniques to characterize and manipulate phenotypic behaviors will be used to alter the properties of numerous cell lines for desired biotechnology objectives.