The Ustilago maydis repetitive effector Rsp3 blocks the antifungal activity of mannose-binding maize proteins

Lay-Sun Ma,Bernd Zechmann,Gert Bange,Christine Trippel,Steffen Ullmann,Artemio Mendoza-Mendoza,Alexander Carsten,Lei Wang,Marino Moretti,Stefanie Reissmann,Jörg Kahnt,Regine Kahmann

doi:10.1038/s41467-018-04149-0

Abstract

To cause disease in maize, the biotrophic fungus Ustilago maydis secretes a large arsenal of effector proteins. Here, we functionally characterize the repetitive effector Rsp3 (repetitive secreted protein 3), which shows length polymorphisms in field isolates and is highly expressed during biotrophic stages. Rsp3 is required for virulence and anthocyanin accumulation. During biotrophic growth, Rsp3 decorates the hyphal surface and interacts with at least two secreted maize DUF26-domain family proteins (designated AFP1 and AFP2). AFP1 binds mannose and displays antifungal activity against the rsp3 mutant but not against a strain constitutively expressing rsp3. Maize plants silenced for AFP1 and AFP2 partially rescue the virulence defect of rsp3 mutants, suggesting that blocking the antifungal activity of AFP1 and AFP2 by the Rsp3 effector is an important virulence function. Rsp3 orthologs are present in all sequenced smut fungi, and the ortholog from Sporisorium reilianum can complement the rsp3 mutant of U. maydis, suggesting a novel widespread fungal protection mechanism.

Highlights

To cause disease in maize, the biotrophic fungus Ustilago maydis secretes a large arsenal of effector proteins
It is likely that the cysteine-rich region contributes to structural integrity of Rsp[3], which is needed for its shielding function
We speculate that when Rsp3HA is expressed by U. maydis, the protein could have a better chance to become incorporated in the cell wall already during secretion leading to the more uniform distribution

Summary

GDKNPSGDKNPNGDKNPSGDKNPNGDN

To determine the processing site in Rsp[3], Rsp3-HA purified from the supernatant of SG200Δrsp3-Potef-rsp3-HA was subjected to N-terminal sequencing by Edman degradation. Rsp39CA-HA and Rsp3Δ412-869-HA were still able to attach to the surface of fungal hyphae when constitutively expressed (Supplementary Fig. 4c), suggesting that binding is mediated by the N-terminal domain of Rsp[3]. An alignment of all secreted DUF26-domain proteins of maize revealed that amino acids putatively involved in mannose-binding are detected in several other family members beside AFP1 (Supplementary Fig. 8c). To elucidate whether Rsp[3] interferes with the presumed antifungal activity of AFP1, we compared the survival of SG200Δrsp[3] and SG200Δrsp3-Potef-rsp3-HA, a strain constitutively expressing Rsp3-HA and exposing the protein on its surface (Supplementary Fig. 4c). When incubated with AFP1, we observed a significantly higher plating efficiency of the strain constitutively expressing Rsp3-HA compared to the rsp[3] mutant (Fig. 6b, c), suggesting that Rsp[3] can protect hyphae against the antifungal activity of maize AFP1. In SG200Δrsp[3], the percentage of dead cells was about 83% in presence of AFP1-His versus 19% in the presence of AFP1**-His (Fig. 6d, e), confirming the antifungal activity of AFP1

Discussion

Findings

Methods