Lipopolysaccharide (LPS), the major causative agent of bacterial sepsis, has been used by many laboratories in genome-wide expression profiling of the LPS response. However, these studies have predominantly used in vitro cultured macrophages (Macs), which may not accurately reflect the LPS response of these innate immune cells in vivo. To overcome this limitation and to identify inflammatory genes in vivo, we have profiled genome-wide expression patterns in non-lymphoid, splenic myeloid cells extracted directly from LPS-treated mice. Genes encoding factors known to be involved in mediating or regulating inflammatory processes, such as cytokines and chemokines, as well as many genes whose immunological functions are not well known, were strongly induced by LPS
after 3 h or 8 h of treatment. Most of the highly LPSresponsive genes that we randomly selected from the microarray data were independently confirmed by quantitative RT-PCR, implying that our microarray data are quite reliable. When our in vivo data were compared to previously reported microarray data for in vitro LPS-treated Macs, a significant proportion (~20%) of the in vivo LPSresponsive genes defined in this study were specific to cells exposed to LPS in vivo, but a larger proportion of them (~60%) were influenced by LPS in both in vitro and in vivo settings. This result indicates that our in vivo LPSresponsive gene set includes not only previously identified in vitro LPS-responsive genes but also novel LPSresponsive genes. Both types of genes would be a valuable
resource in the future for understanding inflammatory responses in vivo.

Keywords: in vivo, LPS, microarray, splenic myeloid cells