The helicase-like domain from "Thermotoga maritima" reverse gyrase : catalytic cycle and contribution to DNA supercoiling

Reverse gyrases are the only topoisomerases capable of introducing positive supercoils into circular DNA. Their exclusive presence in thermophilic and hyperthermophilic organisms indicates a DNA thermoprotective role in vivo. In spite of the efforts to improve our knowledge of reverse gyrase, modest progress has been made since its discovery. Currently, only one crystal structure of the enzyme is available, and the most widely accepted reaction mechanism is a hypothetical one, mostly derived from the functions of enzymes related to reverse gyrase domains.
In the present work we address mechanistic aspects of the reaction by exploiting the capabilities of a wide range of techniques, to elucidate the role of one module of T. maritima reverse gyrase. Reverse gyrase consists of an N-terminal helicase-like domain, fused to a C-terminal topoisomerase domain. We selected the helicase-like domain as a model of study due to its capacity to couple ATP binding and hydrolysis to DNA processing. Exploiting of these features by reverse gyrase turns this region into a key player at virtually every step of DNA supercoiling.
Steady-state ATPase assays and equilibrium binding titrations with the helicase-like domain and the full-length enzyme, enabled us to prove for the first time a harnessing effect of the topoisomerase over the helicase-like domain. We showed that properties intrinsic to the helicase-like domain, like DNA-stimulated ATP hydrolysis, nucleotide-dependent affinity switch for DNA, and thermodynamic coupling between DNA binding and ATP binding and hydrolysis, are strongly reduced in the context of reverse gyrase. At that time apparent contradictions arose, from reports stating that the isolated helicase-like domain is less active than within the context of the full-length enzyme. We reconciled these differences by demonstrating that the presence of the putative N-terminal Zn-finger in the helicase-like domain construct is the cause for the decreased activity. Furthermore, we have elucidated the thermodynamic and conformational cycle of the helicase-like domain, and predicted the stages fulfilling the requirements for interdomain communication, local duplex DNA unwinding, and the stages where DNA is in a suitable state to support the supercoiling reaction. Finally, besides the use of smFRET as a tool to investigate conformational changes in solution, we have also provided high-resolution snapshots of the helicase-like domain via X-ray crystallography. We have provided the most detailed structures of this region to this date, in the apo and ADP-bound forms. They also revealed high flexibility of the linker joining the RecA domains with relative orientations far from random, and local differences in secondary structure motifs that discard the assumption of all reverse gyrases having a “monolithic” build-up.
We also created a deletion mutant of the latch, region with a sui generis location, perfectly suited for interdomain communication. Previous reports stated that its deletion from reverse gyrase abolishes positive supercoiling. We demonstrated its strong involvement in DNA binding, DNA-stimulated ATP hydrolysis, and thermodynamic coupling between these processes in the isolated helicase-like domain. We also revealed its role in presenting the ssDNA to the topoisomerase domain and in guiding the strand passage and resealing, ensuring the directionality leading to the introduction of positive supercoils. Additionally, we also elucidated the nucleotide cycle and conformational transitions for this helicase-like domain mutant, which gave the first indications of why no positive supercoiling can be performed by the full-length reverse gyrase lacking the latch, and only DNA relaxation is allowed.
Finally, our pre steady-state kinetic studies allowed us to fully describe the unstimulated ATPase activity of the isolated helicase-like domain. We also demonstrated for the first time its DNA unwinding activity, shedding light on the rarely documented local B-DNA duplex destabilization of helicase-like modules, appended to bigger enzymes. Additionally, the sequence of ssDNA strand release, and identification of secondary structure motifs involved in ssDNA binding at different stages were determined. Together with the finding of new conformational states via smFRET, and “targeted” supercoiling assays with the full-length enzyme, we end up proposing a detailed catalytic mechanism, similar to the one derived from the reverse gyrase structure, only this time based on and supported by a combination of kinetic, thermodynamic, and structural data.