The binding site and mechanism of the RNA polymerase inhibitor tagetitoxin: an issue open to debate

Abstract

In a recent paper in the Journal of Biological Chemistry, Artsimovitch et al. report a major revision of a crystallographic model and proposed mechanism of the RNA polymerase inhibitor, tagetitoxin. This reassessment is based on theoretical modeling using molecular dynamics simulations. Here, we argue that this theoretical model contradicts experimental results and a published crystal structure cannot exclude several mechanistically distinct alternative models and does not support some major conclusions. We conclude that understanding the tagetitoxin mechanism is beyond the reach of currently available computational simulations and must await input from high-resolution crystal structures of tagetitoxin bound to elongation complex, extensive biochemical studies, or both.

Figure 1. Quality of the holo-RNAP/Tgt crystal structure (PDB ID: 2BE5). (A) The final, difference (|Fobs – Fcalc|) omit ED contoured at 3σ level (blue) for Tgt (yellow) and the active site Mg²⁺ ions (magenta). The map was calculated based on the coordinates and structure factors of the PDB entry 2BE5 using the twinning option of the CNS1.0 program. The catalytic triad of the T. thermophilus RNAP β’-subunit is shown as the gray balls-and-sticks model. (B) Ramachadran plot calculated by the PDB Validation Server (program PROCHECK; 12,424 citations according to the Web of Science Citation Index) for one of the two RNAP molecules in the asymmetric unit of the crystals of the RNAP/Tgt complex. The second RNAP molecule is characterized by essentially the same Ramachadran statistics.

Figure 2. Reconstruction of the theoretical model of Tgt docked in the insertion EC published by Artsimovitch et al. (2011). The overall (A) and (B) close-up views of the Tgt binding site/mode. This is a “semi-alternative” model in which the Tgt binding site and orientation closely resembles the docking model of Artsimovitch et al., while, in contrast to the results of Artsimovitch et al.,² the closed (folded) TL and the other crucial structures of the RNAP active catalytic remain unaltered as compared with the crystallographic structure of the insertion EC. In particular, the close-up view (B) emphasizes that the α-helical backbone of the folded TL and the conformation the two TL critical side chains (β’Met1238 and β’Arg1239) remain intact and retain experimentally observed interactions with NTP, while β’Arg1239 may form three additional contacts with docked Tgt. The views and color scheme are adjusted to that of the Figures 4 and 5A in the Artsmovitch et al.² article. The putative polar and experimental van der Waals stacking interactions are shown as the dashed blue and yellow lines, respectively. MgI, MgII – the catalytic Mg²⁺ ions. The T. thermophilus sequence numbering was used in these panels, the corresponding E. coli sequence numbers are shown in brackets. (C) Superposition of the experimental (dark yellow) and theoretical (magenta) Tgt models reveals dramatic reorientation (~160°) of Tgt that occurs in a course of modeling.

Figure 3. Analysis of the Tgt binding determinants. (A) Schematic diagram showing the major Tgt binding determinants in RNAP that have been predicted by the theoretical model of Artsimovitch et al.² (left panel) and/or observed in the X-ray structure (right panel), and whose substitutions have been tested and possessed the strongest (blue) and lesser but notable (cyan) resistance to Tgt. MgII – a second, low affinity catalytic metal. The percent of resistance shown in brackets corresponds to maximum concentration (~32μM) of Tgt used in ref. (the three mutants below MgII) and ref. (the two mutants above MgII). The putative (left panel) and observed (right panel) structural interactions with Tgt are shown by arrows. The absence/loss of the structural contacts with Tgt are marked by the red crosses and distances (in Å) where applicable. (B) Sequence alignment of the regions of E. coli (ECO) and T. thermophilus (TTH) RNAPs in vicinity to the Tgt determinants (emphasized by the respective colors) shown in (A).