Re-refinement of 4xan: hen egg white lysozyme with carboplatin in sodium bromide solution including details of solute, solvent, ion and split occupancy amino acid electron density evidence

TL;DR

This paper presents a detailed re-refinement of the hen egg white lysozyme structure with carboplatin, addressing previous interpretation issues and providing improved electron density models based on higher resolution data and multiple software analyses.

Contribution

It offers a revised, higher-resolution structure of 4xan with clarified electron density interpretations and addresses previous modeling critiques with new data processing.

Findings

Reprocessed diffraction data to 1.3 Å resolution.

Provided improved electron density models for solute and solvent.

Clarified the interpretation of ligand and ion electron densities.

Abstract

A re-refinement of 4xan, hen egg white lysozyme (HEWL) with carboplatin crystallised in NaBr solution, has been made (Tanley et al 2016). This follows our Response article (Tanley et al 2015) to the Critique article of Shabalin et al 2015, suggesting the need for corrections to some solute molecule interpretations of electron density in 4xan and removal of an organic moiety as a ligand to the platinum ion coordinated to His15. We note Shabalin et al (2015) model of a chlorine in that density and a close by bromine at partial occupancy to explain the shape. However, as the bromide concentration is in huge excess over chloride (by 20 fold), we think that the 4yem Shabalin et al 2015 interpretation highly unlikely, but nevertheless we still cannot offer an explanation for that shape, confirming our earlier analysis described in Tanley et al (2014). Following Shabalin et al (2015) reprocessing of the raw diffraction data for 4g4a, we also redid the diffraction data processing for 4xan to a higher resolution using EVAL (Schreurs et al 2010) concluding in favour of 1.3 Angstrom as the resolution limit and which is the basis for our revised PDB file for 4xan (5HMJ). It is very interesting that there is extra X-ray diffraction data from 1.47 to 1.30 Angstrom resolution e.g. with <I/sigma(I)> =0.39 and CC1/2 = 0.181 in the final shell (1.30 to 1.322 Angstrom). In this arXiv article we document in detail our different solvent and split occupancy side chain electron density interpretations as evidence for our statement of approach in our Response article (Tanley et al 2015). Our critical re-examination includes comparisons based on the 4xan diffraction data images reprocessing with three different software packages so as to evaluate the possibility of variations in electron density interpretations due to that. Overall our finalised model (PDB code 5HMJ) is now improved over 4xan.