Biology Department, St Joseph's University, 5600 City Ave, Philadelphia PA 19131, USA

Background and objectives
When compatible haploid cells of the fungus Ustilago maydis mate, they form an infectious dikaryon that can enter plant tissues and cause corn smut disease. Mating is a complex developmental phenomenon involving nutrient and pheromone sensing, a change from budding to filamentous growth, growth response to chemical gradients and cell fusion. Several genes involved in pheromone production and reception have been previously identified [1, 2], but we are using a novel screen for mating-deficient mutants in order to find additional genes involved in mating. Molecular and microscopic characterization of the mutants will provide new information about the mating process.

Materials and methods
Liquid cultures of a haploid U. maydis strain were exposed to UV radiation or to the chemical mutagen ethylmethanesulfonate. Colonies produced by surviving cells were picked into individual wells of 96-well microtitre plates containing medium and then grown on a platform shaker overnight. Cells were transferred in groups to solid medium using a 96-prong replicator. Isolates were tested for mating competency by transferring them to plates of medium containing charcoal, and then using the replicator to transfer drops of compatible tester strains onto the isolate drops. Two tester strains were used: a wild-type-compatible strain, and a mating-impaired mutant strain. Wild-type-compatible cells can mate with either of the testers, and mating results in the formation of macroscopically visible filaments on the plates. Any potential mutants that did not produce visible filaments with either or both of the testers are potential mim (mating impaired mutant) strains.

Results and conclusions
So far, 72 mim mutants have been isolated. Microscopic examination of these mutants revealed several different classes of mating defects. Some mutants failed to form normal mating filaments, although they induced filament formation in compatible cells and were able to fuse with nearby compatible cells. Other mutants formed filaments but were unable to fuse normally. Additional phenotypes have also been described, and several mutants have been examined in detail. mim1 induces normal mating filament formation in wild-type cells, but it produces few mating filaments of its own. In addition, mim1 mating filaments are slow to fuse with wild-type-compatible cells. mim1 cells paired with compatible wild-type cells cause normal infections of maize plants. mim31 makes normal pheromone and induces mating filaments in wild-type cells, but it is defective in cell fusion and responds to pheromone by making aerial filaments rather than mating filaments. mim129 is slow to form mating filaments and fuse with compatible wild-type cells. The resulting dikaryons can infect plants, but gall formation is reduced. These mutants have been complemented with a plasmid library of wild-type U. maydis DNA, and molecular characterization of the complementing plasmids is now under way.

1. Bolker M, Urban M, Kahmann R, 1992. Cell 68, 441-450.
2. Hartmann HA, Kahmann R, Bolker M, 1996. EMBO Journal 15, 1632-1641.