Medicinal chemistry of benzoisoxazole-4,7-dione analogues identifies the ribosomal small subunit as a target for specific mycobacterial translational fidelity

Abstract

A central dogma of molecular biology is the "speed-accuracy trade-off," where ribosomes must slow down to ensure accurate protein synthesis. In mycobacteria, a high basal level of mistranslation at glutamine and asparagine codons, caused by an indirect tRNA aminoacylation pathway, promotes tolerance to the antibiotic rifampicin. While pharmacologically increasing translational fidelity is a promising strategy to combat antibiotic tolerance, the underlying mechanisms remain poorly understood. Here, we screened 9,000 synthetic compounds and identified benzo[d]isoxazole-4,7-diones as a novel chemical class that reduces mycobacterial mistranslation. Medicinal chemistry optimization yielded a lead compound, 9787, with superior potency in decreasing mistranslation and reversing rifampicin tolerance. Using competitive chemical proteomics, we identified the 30S ribosomal protein S5 (RpS5) as the specific cellular target. Remarkably, compound 9787 enhances translational fidelity at concentrations that do not measurably impact the overall rate of protein synthesis. Our findings challenge the universality of the speed-accuracy trade-off, demonstrating that fidelity can be improved independently of translation speed. This work reveals that the ribosomal small subunit is a druggable target for modulating translational quality control and introduces a new strategy for combating antibiotic-tolerant bacteria without the associated fitness cost of slowed translation. ImportanceA fundamental principle in molecular biology holds that ribosomes face a trade-off between translation speed and accuracy: going faster means making more errors, while maintaining high fidelity requires slowing down. This study challenges that paradigm by identifying a small molecule that increases translational accuracy in mycobacteria without affecting the rate of protein synthesis. The compound targets ribosomal protein S5 and specifically reduce errors arising from physiologically mischarged tRNAs - a quality control problem distinct from the well-studied codon*anticodon mismatches. This form of mistranslation contributes to antibiotic tolerance in tuberculosis, making it a potential therapeutic target. Our findings reveal unexpected flexibility in how ribosomes maintain translation quality and suggest that pharmacologically increasing fidelity without the fitness cost of slowed protein synthesis may be an attractive strategy for combating antibiotic-tolerant bacteria.