Journal:Acta Cryst D:S2059798319000214
From Proteopedia
In-house high energy remote SAD-phasing using the magic triangle: how to tackle the P1 low symmetry using multiple orientations on the same human IBA57 crystal to increase multiplicitySpyridon Gourdoupis, Veronica Nasta, Simone Ciofi-Baffoni, Lucia Banci and Vito Calderone [1] Molecular Tour This paper (DOI 10.1107/S2059798319000214) describes the approach used to solve in-house the structure of human IBA57 through 5-amino-2,4,6-triiodoisophthalic acid (I3C) high energy remote SAD-phasing. Multiple orientations (each of them corresponding to a different run) of the same P1 (triclinic) crystal have been exploited to acquire sufficient real data multiplicity for successful phasing and thus minimizing the difficulties of merging datasets coming from different crystals.  It is also described how the joint use of this I3C derivative and of an in-house native dataset through a SIRAS approach decreases the data multiplicity needed for phasing by almost 50%. Furthermore, it is illustrated that there is a clear data multiplicity threshold value for success and failure in phasing and how adding further data does not significantly affect substructure solution and model building. The multiplicity threshold for successful phasing appeared, in fact, to be around five, independent of the combination of datasets (runs) used. This value can be reduced to about half through SIRAS by exploiting the isomorphous differences with a second native dataset reaching a successful multiplicity value of less than three. To our knowledge, this is the only structure present in the PDB which has been solved in-house, through remote SAD, in space group P1 and using one crystal only. All the raw data used, deriving from the different orientations (runs), have been deposited to Zenodo (DOI 10.5281/zenodo.2531553)[8] both for educational purposes and to enable other crystallographers to improve methods for data processing and structure solution and thus to benefit from these findings. At the time of structure solution coordinates and structure factors were deposited and released in the Protein Data Bank under the accession codes 5oli (for the in-house I3C derivative) and 6esr (for the higher resolution synchrotron structure); at the time of writing the manuscript both entries have been re-refined in order to optimize model quality and statistics and so they have been superseded by 6qe4 and 6qe3 respectively. with the residues for which even the main chain electron density is very poor if not absent at all highlighted in red (53-59, 61, 88-92, 115-118, 138-147, 262, 296-300, 306-311). between 6esr (red) and 5oli (green) secondary structures. It appears that, in the case of 6esr, there is a slight loss in secondary structure elements (mainly β-strands in the N-terminus region) with respect to 5oli and there is the appearance of a very short 3/10 helix around residue 90. It must be pointed out anyway that those regions mostly correspond to the regions in which electron density is very weak and thus model tracing can be quite approximate.
PDB references: Re-refinement of 6ESR human IBA57 at 1.75 A resolution 6qe3; Re-refinement of 5OLI human IBA57-I3C 6qe4. References
| |||||||||||
