Gene expression clustering using self-organizing maps: analysis of the macrophage response to particulate biomaterials

The most common cause of total joint replacement failure is peri-implant bone loss causing pain and prosthesis loosening. This process, known as osteolysis or aseptic loosening, is characterized by macrophage phagocytosis of particulate implant wear debris. In an incompletely defined step, particulate biomaterial debris induces macrophages to release a variety of inflammatory mediators and signaling proteins that lead to bone loss. In an in vitro model of this process, we used microarray technology and data analysis techniques, including the use of self-organizing maps (SOMs), to understand the mRNA gene expression changes occurring in macrophages exposed to clinically relevant particles of ultra-high molecular weight polyethylene and TiAlV alloy. Earlier studies have been limited by technology that only allowed analysis of a few genes at a time, but the microarray techniques used in this paper generate the quantitative analysis of over a thousand genes simultaneously. Our microarray analysis utilized an SOM clustering to elucidate general patterns in the data, lists of top up- and down-regulated genes for each time point and genes with differential expression under different biomaterial exposures. The expression levels of the majority of genes (>95%) did not vary over time or with exposure to different biomaterials, but a few important genes, such as TNF-α, IL-1β, IL-6, and MIP1α, proved to be highly regulated in response to biomaterial exposure. We also uncovered a novel set of genes, which not only validates and logically extends the current model of the pathogenesis of osteolysis and aseptic loosening, but also provides new targets for further research and therapeutics.