Institut de Genetique et Microbiologie, Universite Paris-sud, 91405 Orsay Cedex, France

Background and objectives
The ability of transposons to excise and re-insert into the genome makes them effective insertional mutagens. Transposon tagging has proved to be a powerful technique for the isolation of genes from maize and snapdragon where well characterized endogenous transposons exist, and this strategy has been successfully extended to a wider range of plant species that lack active transposons. The identification of active transposons in filamentous fungi now offers the opportunity to establish a transposon tagging system for gene cloning. This approach should be of great value in the molecular dissection of pathogenicity.

Results and conclusions
Activity of Fusarium oxysporum transposons: class II transposons were identified in the genome of the phytopathogenic fungus F. oxysporum through their transposition within the nia target gene. Two families have been well-characterized at the molecular level: Fot1, a member of the pogo family, and impala belonging to the Tc 1-mariner superfamily. They are short elements with inverted repeats, always inserting at a TA site. By introducing different copies of Fot1 and impala in free genetic backgrounds, autonomous elements have been identified [2]. A number of findings about their transposition are pertinent when designing gene tagging experiments and they have been exploited to identify novel transposon-induced mutations.
Transposition of impala as a mutagen for the recovery of path- mutants: from an impala copy inserted within the nia gene, somatic excisions of impala are easily detected through the restoration of nia expression. In 50-70% of the revertants, impala has re-inserted at a new genomic position. Such a system allows many independent insertions to be isolated and may serve to recover mutations in genes involved in pathogenicity. From a strain pathogenic on melon, hundreds of revertants have been inoculated on plants. Upon impala transposition, a high proportion of path- mutants has already been obtained. To determine the molecular basis of impala integration events, different flanking regions of transposed impala elements were recovered by IPCR. The impala-flanking sequences showed no significant similarity with any other sequence in the databases. IPCR probes have been used to screen a cosmid library and complementation of the mutant defect is being performed.
Transposition in heterologous species: the diversity of families including Fot1 and impala suggests that these elements can transpose in heterologous fungal species. The activity of autonomous elements was studied in two species, F. moniliforme and Aspergillus nidulans, using the nia phenotypic assay. With the success of these experiments, the potential of transposons as genetic tools for filamentous fungi appears promising. The availability of nia mutants in many species is a great advantage to test their ability to transpose in a wide range of fungal species.
Prospects: the activity of autonomous elements has to be improved before these elements could be routinely used as mutagens. The delineation of cis requirements would help in developing more active elements and the construction of non-autonomous elements carrying dominant markers. Various modifications of existing elements are being made to improve the usefulness of these transposon systems. A two-element system, consisting of a non-mobile transposase source under the control of strong promoters, and a mobile marked element as insertion mutagen, is being tested.

1. Daboussi MJ, 1996. Journal of Genetics 75, 325-339.
2. Migheli Q, Lauge R, Daviere JM et al., 1998. Genetics (in press).