Fungal adaptation to plant defenses through convergent assembly of metabolic modules

Abstract

25The ongoing diversification of plant defense compounds exerts dynamic selection pressures on 26 the microorganisms that colonize plant tissues. Evolutionary processes that generate resistance 27 towards these compounds increase microbial fitness by giving access to plant resources and 28 increasing pathogen virulence. These processes entail sequence-based mechanisms that result in 29 adaptive gene functions, and combinatorial mechanisms that result in novel syntheses of existing 30 gene functions. However, the priority and interactions among these processes in adaptive 31 resistance remains poorly understood. Using a combination of molecular genetic and 32 computational approaches, we investigated the contributions of sequence-based and 33 combinatorial processes to the evolution of fungal metabolic gene clusters encoding stilbene 34 cleavage oxygenases (SCOs), which catalyze the degradation of biphenolic plant defense 35 compounds known as stilbenes into monophenolic molecules. We present phylogenetic evidence 36 of convergent assembly among three distinct types of SCO gene clusters containing alternate 37 combinations of phenolic catabolism. Multiple evolutionary transitions between different cluster 38 types suggest recurrent selection for distinct gene assemblages. By comparison, we found that 39 the substrate specificities of heterologously expressed SCO enzymes encoded in different 40 clusters types were all limited to stilbenes and related molecules with a 4'-OH group, and 41 differed modestly in substrate range and activity under the experimental conditions. Together, 42 this work suggests a primary role for genome structural rearrangement, and the importance of 43 enzyme modularity, in promoting fungal metabolic adaptation to plant defense chemistry. 44 45 106 remodeling can operate independently of sequence-based evolution. We analyzed the substrate 107 specificities of four heterologously expressed SCOs from different cluster types and unclustered 108 genomic regions, and found evidence for broad conservation of substrate specificity in both 109 unclustered and clustered SCOs, with only modest variation in the substrate specificity of SCOs 110 found in alternate cluster types. Conversely, we found evidence for multiple independent origins 111 of and repeated transitions between different types of sco clusters, consistent with selection 112 acting upon recurrent combinations of enzyme-encoding genes. These results suggest that the 113 6 reorganization of sco clusters may underpin adaptive shifts in stilbene degradation pathways, 114 while changes in enzymatic function may play a secondary role. 115 116 Methods 117 sco cluster retrieval and annotation 118 sco clusters were identified in publicly available genomes of 550 fungi (Table S1) as previously 119 described (Gluck-Thaler & Slot 2018). Briefly, all hits to a previously functionally characterized 120 SCO amino acid query (accession: EAA32528.1) were retrieved from predicted proteomes using 121 BLASTp from the BLAST suite v.2.2.25+ (Altschul et al....