FLW TAKKEN, D SCHIPPER, HJJ NIJKAMP and J HILLE

Department of Genetics, IMBW, BioCentrum Amsterdam, Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands

Background and objectives
Leaf mould disease in tomato is caused by the biotrophic fungus Cladosporium fulvum. An Ac/Ds targeted transposon-tagging strategy was used to isolate the gene(s) conferring resistance to race 5 of C. fulvum, a strain expressing the avirulence gene Avr4.

Results and conclusions
An infection assay of a tagging population of 20,000 2-week-old seedlings yielded five susceptible mutants, of which two had a Ds element integrated in the same gene at different positions. The three other mutants exhibited a genomic deletion covering most of the Lycopersicon hirsutum introgressed segment including the Cf-4 locus. Southern blot analysis revealed that this deletion includes the tagged gene and five homologous sequences. To test whether the tagged gene confers resistance to C. fulvum via AVR4 recognition, the Avr4 gene was expressed in planta. Surprisingly, expression of the Avr4 gene still triggered a (delayed) specific necrotic response in the transposon-tagged plants, indicating that the tagged resistance gene is not, or not the only gene, involved in AVR4 recognition. Mutants harbouring the genomic deletion did not show this AVR4-specific response. The deleted segment apparently contains, in addition to the tagged gene, one or more other genes that play a role in the AVR4 responses. The tagged gene is present at the Cf-4 locus, but it does not necessarily recognise AVR4 and is therefore designated Cf-4A [1].

To isolate the gene involved in AVR4 recognition, a second transposon-tagging population was generated. A transposon-tagging population of plants expressing both the Cf-4 and Avr4 genes was screened for survivors. The surviving mutants, no longer showing a systemic HR, were expected to have a disrupted Cf-4 gene [2]. 1500 plants were screened and 22 survivors were found, of which two showed necrosis on their secondary leaves, later leading to leaf shed. These last two plants were shown to have a Ds tag in Cf-4A. The 20 other survivors showed a disruption of the Cf-4 gene which is closely homologous (>95%) to the Cf-4A gene present at the same locus. Six of the surviving plants also had, in addition to the disruption of Cf-4, a disruption of the Cf-4A gene. An inoculation assay of the survivors with race 5 of C. fulvum revealed that only plants having a disruption of both genes (Cf-4 and Cf-4A) become susceptible to infection.

Both identified genes are members of a gene family and show high sequence homology to the C. fulvum resistance genes Cf-9 and Cf-2. The genes are predicted to encode extracellular transmembrane proteins containing a divided domain of 22/25 leucine-rich repeats.

We propose a model in which the Cf-4 locus contains two active resistance genes (Cf-4 and Cf-4A) that both recognise different elicitors, AVR4 and AVR4A, respectively. In the first tagging experiment, susceptible plants were isolated as having only a tag in Cf-4A; this indicates 'cross-talk' between these two genes upon insertion of a transposon. This hypothesis is supported by the isolation of the two necrotic survivors obtained from the second tagging population having only a Ds tag in Cf-4A.

References
1. Takken FLW, Schipper D, Nijkamp HJJ, Hille J, 1998. Plant Journal, in press.
2. Thomas CM, Jones DA, Parniske M, Harrison K, Balint-Kurti PJ, Hatzixanthis K, Jones J, 1997. Plant Cell 9, 2209-2224.