BaseHunter demonstrates DNA sequence-dependent gyrase capture in 601 bp minicircle, indicating structural implications for supercoiling.
Found in all bacteria and a target of the widely used fluoroquinolone antibiotics, gyrase passes DNA strands through each other to modulate supercoiling. We recently published two single-particle electron cryo-microscopy (cryoEM) structures of gyrase bound to a negatively supercoiled DNA minicircle (Vayssières et al. Science 2024 384:227). Existing tools could not model individual DNA bases, however, so we were unable to determine which DNA sequence(s) the enzyme bound. I therefore co-developed a shape-based approach that we call “BaseHunter.” This approach ignores the sugar and phosphate to instead evaluate shape differences between purines and pyrimidines. Given high enough DNA resolution (∼3Å), BaseHunter enables an accurate sequence prediction from cryoEM density maps. Surprisingly, given that gyrase must act on any DNA and rapidly, we used our method to discover that gyrase binds the 601 bp negatively supercoiled minicircle at just one DNA sequence, a six-base palindrome. Gyrase binds this palindrome from either side, resulting in two different structures. Like different amino acid sequences forming the same protein fold, we uncovered that the gate-segments (the DNA to be cleaved by gyrase) in each structure are composed of different sequences yet form identical (RMSD ≤ 1Å) structures. In one orientation, the sharply bent gate segment is positioned below an unbent “transfer segment” (the DNA passed through the gate) with a tight wrap around the gyrase beta-pinwheel. Gyrase, in the second conformation, bound only the gate-segment. The sharp gate-segment bending and wrapping are only accommodated by certain DNA sequences, depending upon their deformability. In our model, instead of gyrase “selecting” sequence, supercoiled minicircle DNA contorts to “capture” gyrase. Our findings elucidate the structural basis of sequence-dependent gyrase capture and demonstrate the usefulness of BaseHunter for DNA modeling.
No takes yet. Share an insight, caveat, or question.
Johnson et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: