In silico, biochemical, and in vitro analysis of silvestrol binding to the DEAD box RNA helicase eIF4A reveals broad anti-pathogen potential of rocaglates across the eukaryotic tree of life

Abstract

Selective inhibition of eukaryotic initiation factor 4A (eIF4A), an RNA helicase, has been proposed as a strategy to fight pathogens. Plant-derived rocaglates exhibit some of the highest specificities among eIF4A inhibitors. Sensitivity to rocaglates is determined by key amino acid (aa) residues mediating reversible clamping of the eIF4A:RNA complex. To date, no comprehensive assessment of eIF4A sensitivity to rocaglates across the eukaryotic tree of life has been performed to determine their anti-pathogenic potential. We performed an in silico analysis of the substitution patterns of six aa residues in eIF4A1 critical to rocaglate binding (human positions 158, 159, 163, 192, 195, 199), uncovering 35 pattern variants among 365 eIF4As sequenced to date. In silico molecular docking analysis of the eIF4A:RNA:rocaglate complexes of the 35 variants, modeled in a human eIF4A environment, and in vitro thermal shift assays with recombinantly expressed human eIF4A mutants, representing select natural and artificial variants, revealed that sensitivity to a natural or one of two synthetic rocaglates--silvestrol, CR-1-31-B, or zotatifin--was associated with lower inferred binding energies and higher melting temperature shifts. Helicase activities were comparable across variants and independent of sensitivity to rocaglates. In vitro testing with silvestrol validated predicted resistance based on position 163 substitutions in Caenorhabditis elegans and Leishmania amazonensis and predicted sensitivity in Aedes sp., Schistosoma mansoni, Trypanosoma brucei, Plasmodium falciparum, and Toxoplasma gondii. Our analysis shows resistance to rocaglates emerging in disparate eukaryotic clades pointing to resistance being a selective neutral trait except in rocaglate-producing Aglaia plants and their fungal parasite Ophiocordyceps. The analysis further revealed the possibility of targeting important insect, plant, animal, and human pathogens including Galleria mellonella, Ustilago maydis, Babesia ovata, and Cryptosporidium sp., with rocaglates. Finally, combined docking and thermal shift analyses might help design novel synthetic rocaglate derivatives or alternative eIF4A inhibitors to fight pathogens. Author SummaryIn the ongoing search for novel ways to fight non-viral and non-bacterial pathogens, targeting translation--the universal process of protein synthesis--to inhibit growth and cell proliferation has emerged as an attractive strategy. Here, we focused on the potential of rocaglates, a group of plant-derived compounds, to inhibit an early step in translation mediated by a RNA helicase called eIF4A. We performed a comprehensive analysis of eIF4A sequence variants to determine their potential sensitivities to rocaglates, especially in pathogens of prokaryotic, fungal, or animal origin. We complemented this in silico analysis with enzyme-based in vitro and whole pathogen in vivo experiments to confirm the sensitivity or resistance to rocaglates of specific variants of eIF4A. Our analysis provides the first comprehensive picture of rocaglate sensitivity among pathogens and establishes targeting important insect, plant, animal, and human pathogens such as wax moth larvae, a major parasite of honey bees, corn smut, a widely distributed fungal disease, Babesia, a livestock parasite that causes anemia and babesiois, and Cryptosporidium, the causative organism of cryptosporidiosis in humans, with rocaglates as a viable anti-pathogen strategy.