In vertebrates, the recognition of pathogens is carried out by two distinct types of immune receptors. The adaptive immune system is characterized by the generation of antibodies and T-cell receptors which have the capacity to recognize an innumerable repertoire of allogenic molecules. On the other hand, innate immunity relies on a restricted number of genotypically encoded molecules such as TLRs, integrins, scavenger receptors etc. (known as pattern recognition receptors, PRRs) to recognize and respond to pathogens or their pathogen-associated molecular patterns (PAMPs). The innate immune receptors can only recognize a limited number of ligands. However, in contrast to T and B cell receptors, the activation of different PRRs by their respective ligands may activate distinct intracellular signaling pathways. This may result in specific changes in gene expression and production of cytokines and other immune mediators by leukocytes that will alarm the host about the presence of a particular type of pathogen. The vertebrate immune system has been studied mainly in mammals, and data on the immune response of non-mammalian vertebrates remains relatively scant. Therefore, a major goal of this dissertation research was to characterize the response of trout macrophages to PAMPs by analyzing the changes in the expression of known immune-relevant genes. In addition, a subtractive cloning approach was employed to isolate and identify additional genes that may be regulated upon activation of trout macrophages by lipopolysaccharide. The results indicate that although the PAMP-modulated gene expression is similar between mammalian and fish leukocytes, there also appear to be certain differences in the PAMPs recognition between mammalian and non-mammalian vertebrates. In particular, the activation of trout macrophages required treatments with LPS concentrations that were several orders of magnitude higher as compared to mammals. These observations, together with additional data from in vitro and in silico analyses, led to the hypothesis that the mechanisms for endotoxin recognition, that have been described in mammals, may not be present in non-mammalian vertebrates. It is possible that the TLR4-mediated recognition of endotoxin arose in mammals, or in their immediate predecessors, as a result of the integration of initially unrelated bacterial and viral recognition mechanisms.