Initial diffraction data were not sufficient to obtain a structure solution by molecular replacement approaches, most likely because at the time of our data analysis the closest five-bladed β-propeller homolog available had less than 13 % amino acid identity to our protein. Therefore, we successfully resorted to multi-crystal native SAD phasing (Liu et al., 2012), illustrating the robustness of using anomalous signal from light atoms such as sulfur for structure determination. Furthermore, we show how such signals can be critical to ascertain the nature of anomalous scatterers through the characteristic imaginary terms of their scattering factor (f″); the bound Ca2+ (P65) or Mg2+ (C2221) in our case. The crystal structure of the LPHN3-OLF shows that this domain has a five-blade β-propeller fold that is monomeric in solution. Despite the broad structural diversity of the β-propeller architecture, two features are often conserved amongst members of this fold: The four-stranded antiparallel β-sheets and the central pore where metal ions might also bind further stabilizing the structure (Chen et al., 2011).
