The potential use of bacterial ice nuclei as a high sensitivity, low background detector for clinical immunoassays is proposed. Such a detection system could be adapted to current immunoassays, reducing sample sizes needed for accurate immunodetection. Ice nuclei could also be used to construct new assays for immunodetection of viruses, rickettsia, and bacteria, without previous biological amplification. The specific goals of this proposal are demonstration of coupling of ice nuclei to an immune recognition event, discovery of appropriate chemistries for covalent linkage of ice nuclei to antibodies and/or biotin, and construction of a working immunoassay for alpha-fetoprotein (an important marker for detection of certain cancers). The alpha- fetoprotein assay shall be compared in sensitivity and accuracy to existing assays based upon radioisotopic or enzymatic detection. The research makes use of standard immunological and biochemical techniques, as well as ice nuclei heterologously expressed and biochemical techniques, as well as ice nuclei heterologously expressed in E. coli, after cloning of an ice nucleation gene from P. syringae. If successful, this detection system could provide to many assay systems sensitivities previously attainable only via biological amplification.