Comparative transcriptome profiling identifies maize line specificity of fungal effectors in the maize–Ustilago maydis interaction

Abstract

The biotrophic pathogen Ustilago maydis causes smut disease on maize (Zea mays) and induces the formation of tumours on all aerial parts of the plant. Unlike in other biotrophic interactions, no gene-forgene interactions have been identified in the maize-U. maydis pathosystem. Thus, maize resistance to U. maydis is considered a polygenic, quantitative trait. Here, we study the molecular mechanisms of quantitative disease resistance (QDR) in maize, and how U. maydis interferes with its components. Based on quantitative scoring of disease symptoms in 26 maize lines, we performed an RNA sequencing (RNA-Seq) analysis of six U. maydis-infected maize lines of highly distinct resistance levels. The different maize lines showed specific responses of diverse cellular processes to U. maydis infection. For U. maydis, our analysis identified 406 genes being differentially expressed between maize lines, of which 102 encode predicted effector proteins. Based on this analysis, we generated U. maydis CRISPR/Cas9 knock-out mutants for selected candidate effector sets. After infections of different maize lines with the fungal mutants, RNA-Seq analysis identified effectors with quantitative, maize line-specific virulence functions, and revealed auxin-related processes as a possible target for one of them. Thus, we show that both transcriptional activity and virulence function of fungal effector genes are modified according to the infected maize line, providing insights into the molecular mechanisms underlying QDR in the maize-U. maydis interaction.

Dissection of the Complex Transcription and Metabolism Regulation Networks Associated with Maize Resistance to Ustilago maydis

“…Plant defense response to pathogens is usually tightly controlled at transcriptional, translational, and metabolism levels. The typical gene-for-gene interaction mechanism, usually identified in plant-biotrophic pathogen, has not been found during maize-Ustilago interaction, which is therefore considered to be a quantitative disease resistance (QDR) [20]. In this study, time-course transcriptome analysis using a pair of contrast lines showed that numerous maize genes were uniquely up-regulated or down-regulated during maize in response to U. maydis infection.…”

“…For example, U. maydis effector protein See1 induced maize cell cycle gene expression in leaves, leading to leaf cell division and eventually tumor formation [19]. Furthermore, maizeline-specific genes were also found to be involved in U. maydis and maize interaction [20]. Taken together, these findings indicate that the regulation of transcription reprogramming of host genes is critical for the biotrophic infection of U. maydis.…”

Many ways to TOPLESS – manipulation of plant auxin signalling by a cluster of fungal effectors

“…This study demonstrates that most of the cluster 6A effectors of the maize pathogenic fungus U. maydis are involved in auxin signalling induction and that five of the six auxin signalling inducers directly interact with TPL‐related proteins and play a role in virulence. The auxin signalling inducing effector Umag_11416, which did not interact with TPL in our experiments, has recently been linked with a maize‐line‐specific virulence function (Schurack et al ., 2021). Interestingly, the Tip effectors do not induce the same level of auxin signalling when overexpressed in planta .…”

WideEffHunter: An Algorithm to Predict Canonical and Non-Canonical Effectors in Fungi and Oomycetes

“…However, effector proteins that differ in one or more of these canonical criteria also exist and we will refer to them as “non-canonical effectors” (NCEs). Non-canonical effectors have been identified based on specific searches for motifs and domains that are associated with other characterized effectors [ 13 , 14 , 15 ], or because of overexpression data observed in transcriptomes of plant-pathogen interactions [ 8 , 16 ]. The effector Pi04314 (PexRD24) was identified while searching for the “RXLR” motif deduced from ESTs of the oomycete Phytophthora infestans during its interaction with potato and tomato.…”

“…Many recent reports continue to base their predictions of effectors on short amino acid lengths and cysteine richness [ 22 , 23 ], but others are searching by other means [ 8 , 13 , 14 , 15 , 16 ]. Available algorithms include the EffectorP machine learning (ML) series, among which the latest version, EffectorP 3.0, is able to classify effectors in the apoplast and cytoplast [ 24 ].…”