In compatible interactions, biotrophic microbial phytopathogens rely on the supply of assimilates by the colonized host tissue. It has been found in rice that phloem localized SWEET sucrose transporters can be reprogrammed by bacterial effectors to establish compatibility. We observed that sweet11/sweet12 double mutants, but not single mutants, exhibited increased resistance toward the fungal hemibiotroph Colletotrichum higginsianum (Ch), both in the biotrophic and the necrotrophic colonization phase. We therefore investigated if the phloem localized transporters AtSWEET11 and AtSWEET12 represent additive susceptibility factors in the interaction of Arabidopsis with Ch. AtSWEET12-YFP fusion protein driven by the endogenous promoter strongly accumulated at Ch infection sites and in the vasculature upon challenge with Ch. However, susceptibility of sweet12 single mutants to Ch was comparable to wild type, indicating that the accumulation of AtSWEET12 at Ch infection sites does not play a major role for compatibility. AtSWEET12-YFP reporter protein was not detectable at the plant–pathogen interface, suggesting that AtSWEET12 is not targeted by Ch effectors. AtSWEET11-YFP accumulation in pAtSWEET11:AtSWEET11-YFP plants were similar in Ch infected and mock control leaves. A close inspection of major carbohydrate metabolism in non-infected control plants revealed that soluble sugar and starch content were substantially elevated in sweet11/sweet12 double mutants during the entire diurnal cycle, that diurnal soluble sugar turnover was increased more than twofold in sweet11/sweet12, and that accumulation of free hexoses and sucrose was strongly expedited in double mutant leaves compared to wild type and both single mutants during the course of Ch infection. After 2 days of treatment, free and conjugated SA levels were significantly increased in infected and mock control leaves of sweet11/sweet12 relative to all other genotypes, respectively. Induced genes in mock treated sweet11/sweet12 leaves were highly significantly enriched for several GO terms associated with SA signaling and response compared to mock treated wild-type leaves, indicating sugar-mediated priming of the SA pathway in the double mutant. Infection assays with salicylic acid deficient sweet11/sweet12/sid2 triple mutants demonstrated that reduced susceptibility observed in sweet11/sweet12 was entirely dependent on the SA pathway. We suggest a model how defects in phloem loading of sucrose can influence SA priming and hence, compatibility.

We used insertional mutagenesis by Agrobacterium tumefaciens mediated transformation (ATMT) to isolate pathogenicity mutants of Colletotrichum higginsianum. From a collection of 7200 insertion mutants we isolated 75 mutants with reduced symptoms. 19 of these were affected in host penetration, while 17 were affected in later stages of infection, like switching to necrotrophic growth. For 16 mutants the location of T-DNA insertions could be identified by PCR. A potential plasma membrane H+-ATPase Pma2 was targeted in five independent insertion mutants. We genetically inactivated the Ku80 component of the non-homologous end-joining pathway in C. higginsianum to establish an efficient gene knockout protocol. Chpma2 deletion mutants generated by homologous recombination in the ΔChku80 background form fully melanized appressoria but entirely fail to penetrate the host tissue and are non-pathogenic. The ChPMA2 gene is induced upon appressoria formation and infection of A. thaliana. Pma2 activity is not important for vegetative growth of saprophytically growing mycelium, since the mutant shows no growth penalty under these conditions. Colletotrichum higginsianum codes for a closely related gene (ChPMA1), which is highly expressed under most growth conditions. ChPMA1 is more similar to the homologous yeast genes for plasma membrane pumps. We propose that expression of a specific proton pump early during infection may be common to many appressoria forming fungal pathogens as we found ChPMA2 orthologs in several plant pathogenic fungi.