The exquisite possibilities of Drosophila genetics, the ease of transgenesis, and the recent sequencing of the genome all have contributed to the rapid development of our basic knowledge on the mechanisms by which the fruitfly responds to fungal and bacterial infections and which are reviewed in the article. We have now gained a basic understanding on how the fly senses infectious organisms and discriminates between various classes of infecting agents (e.g. fungi, Gram-positive or Gram-negative bacteria). Further we have learned how this sensing may lead to activation of signaling pathways in immune responsive cells which ultimately control expression of effector genes through members of the NF-k B family of inducible transactivators. Of particular interest in the field was the discovery that sensing of fungal and Gram-positive bacterial infection in Drosophila induces NF-k B dependent gene expression via a transmembrane receptor named Toll. This result led to the search for mammalian homologues of Toll and has contributed to the discovery of a family of Toll-like receptors (TLR) in mice and humans. TLRs now appear as essential sentinels of the innate immune system in mammals and their activation by diverse microbial ligands further significantly contributes to stimulation of adaptive immune responses. The paper reviews these data and compares our present knowledge on antimicrobial defenses in the fly and innate defenses in mammals, highlighting the impressive similarities, but also outlining some crucial differences.

Many of the significant results on fruitfly immunity were obtained over the last decade in the laboratory of the author of the review. This group had a long tradition of studies on endocrine control of development and reproduction in insects and had noted, like many entomologists, that transplantation and excision experiments performed under septic conditions never led to wound infection. This prompted the conclusion that insects must have strong antimicrobial defenses, which indeed had already been perceived by workers in the late 19^th and early 20^th centuries. Furthermore, the role of inducible antimicrobial peptides had been described in 1980 by the Boman group in Sweden, and the time appeared ripe around 1990 to use Drosophila molecular genetics to address the problems of sensing, signaling, and gene transcription during the host defense.