Targeting tRNA-Synthetase Interactions towards Novel Therapeutic Discovery Against Eukaryotic Pathogens

Abstract

43The development of chemotherapies against eukaryotic pathogens is especially 44 challenging because of both the evolutionary conservation of drug targets between host 45 and parasite, and the evolution of strain-dependent drug resistance. There is a strong 46 need for new nontoxic drugs with broad-spectrum activity against trypanosome 47 parasites such as Leishmania and Trypanosoma. A relatively untested approach is to 48 target macromolecular interactions in parasites rather than small molecular interactions, 49 under the hypothesis that the features specifying macromolecular interactions diverge 50 more rapidly through coevolution. We computed tRNA Class-Informative Features in 51 humans and eight clades of trypanosomes, identifying parasite-specific informative 52 features (including base-pairs and base mis-pairs) that are broadly conserved over 53 approximately 250 million years of trypanosome evolution. Validating these 54observations, we demonstrated biochemically that tRNA:aminoacyl-tRNA synthetase 55 interactions are a promising target for anti-trypanosomal drug discovery. From a marine 56 natural products extract library, we identified several fractions with inhibitory activity 57 toward Leishmania major alanyl-tRNA synthetase (AlaRS) but no activity against the 58 human homolog. These marine natural products extracts showed cross-reactivity 59 towards Trypanosoma cruzi AlaRS indicating the broad-spectrum potential of our 60 network predictions. These findings support a systems biology model in which 61 combination chemotherapies that target multiple tRNA-synthetase interactions should 62 be comparatively less prone to the emergence of resistance than conventional single 63 drug therapies. 64 65 Author Summary 66Trypanosome parasites pose a significant health risk worldwide. Conventional drug 67 development strategies have proven challenging given the high conservation between 68 humans and pathogens, with off-target toxicity being a common problem. Protein 69 synthesis inhibitors have historically been an attractive target for antimicrobial discovery 70 against bacteria, and more recently for eukaryotic pathogens. Here we propose that 71 exploiting pathogen-specific tRNA-synthetase interactions offers the potential for highly 72 targeted drug discovery. To this end, we improved tRNA gene annotations in 73 trypanosome genomes, identified functionally informative trypanosome-specific tRNA 74 features, and showed that these features are highly conserved over approximately 250 75 million years of trypanosome evolution. Highlighting the species-specific and broad-76 spectrum potential of our approach, we identified natural product inhibitors against the 77 parasite translational machinery that have no effect on the homologous human enzyme. 78 79 nontoxic drugs with broad-spectrum activity against different species of Leishmania and 89 other trypanosomes [6,7]. 90 Given their essential role in protein synthesis, aminoacyl-tRNA synthetases 91 (aaRSs) have been an attractive target for antimicrobial ...

Structural and Genetic Determinants of Convergence in theDrosophilatRNA Structure-Function Map

“…Furthermore, our tRNA CIF annotations could predict differential susceptibility to inhibition by chemical agents of homologous aminoacyl-tRNA synthetases (aaRSs) from trypanosomes and humans (Kelly et al 2020). We also found that tRNA gene clusters are conserved in trypanosomes and show clear evidence of rapid evolution by duplication, deletion, and rearrangements (Kelly et al 2020) consistent with findings from genome comparisons of diverse eukaryotic groups (Bermudez-Santana et al 2010, Velandia-Huerto et al 2016.…”

“…Our approach integrates sequence information across all tRNA functional gene families at once, using statistics on structure-conditioned functional information (Freyhult et al 2006). The relevance of our "information criterion" to predict the tRNA structurefunction map stems from the biophysics of translation, assuming promiscuous interactions across all species of tRNA-binding proteins and tRNAs co-expressed in the same cellular domain or compartment, with association rates that increase proportionally with concentrations (which we estimate for tRNAs by proxy from gene copy numbers, as in the tRNA Adaptation Index, dos Reis et al 2004;Sabi et al 2017) and with aminoacylation probabilities that depend on matching and mis-matching of structural and dynamic (motional) features across all interacting species operating in parallel (Collins-Hed and Ardell 2020). No matter how the phenotypic expression of a given base or base-pair contributes to the classification of tRNAs into substrates and non-substrates by tRNA-binding proteins, whether directly at the binding interface, as a substrate for a critical base modification, or indirectly, by contributing to recognition by base modifying enzymes, through indirect effects on the shape and motion of entire molecules, or via allosteric circuits connecting binding interfaces and active sites in complexes (Sethi et al 2009), the theoretical expectation under selection for translational accuracy is that the genetic bases of those identifying phenotypic traits will become increasingly restricted to the tRNA functional classes that rely on them for their identities (Collins-Hed and Ardell 2020).…”

“…Even though our tRNA structure-function maps are based on an information criterion rather than a conservation criterion, we recently showed that tRNA CIFs, including Class-Informative Base-Pairs (CIBPs) and Class-Informative Mis-Pairs (CIMPs), are highly conserved within trypanosomes and between trypanosomes and humans, even while showing evidence of co-evolutionary divergence (Kelly et al 2020). Furthermore, our tRNA CIF annotations could predict differential susceptibility to inhibition by chemical agents of homologous aminoacyl-tRNA synthetases (aaRSs) from trypanosomes and humans (Kelly et al 2020).…”

“…Even though our tRNA structure-function maps are based on an information criterion rather than a conservation criterion, we recently showed that tRNA CIFs, including Class-Informative Base-Pairs (CIBPs) and Class-Informative Mis-Pairs (CIMPs), are highly conserved within trypanosomes and between trypanosomes and humans, even while showing evidence of co-evolutionary divergence (Kelly et al 2020). Furthermore, our tRNA CIF annotations could predict differential susceptibility to inhibition by chemical agents of homologous aminoacyl-tRNA synthetases (aaRSs) from trypanosomes and humans (Kelly et al 2020). We also found that tRNA gene clusters are conserved in trypanosomes and show clear evidence of rapid evolution by duplication, deletion, and rearrangements (Kelly et al 2020) consistent with findings from genome comparisons of diverse eukaryotic groups (Bermudez-Santana et al 2010, Velandia-Huerto et al 2016.…”

“…As described inKelly et al (2020), we reimplemented our previously published algorithm to estimate structureconditioned functional information statistics fromFreyhult et al (2006) extending it to compute functional…”