An automated data processing and analysis pipeline for transmembrane proteins in detergent solutions

D. Molodenskiy; H. Mertens; D. Svergun

Unambiguous reconstruction of protein shape from a detergent solubilised MP is hampered by the strong scattering signal of bound detergent, typically forming a corona around the lipophilic trans-membrane region. The corona consists of hydrophobic and hydrophilic components, with significantly different scattering/electron densities. Thus the results of standard SAXS modelling procedures are, in such cases, non-realistic and ambiguous. Using complementary data from small-angle neutron scattering (SANS) including contrast matching approaches to isolate scattering from each solution component, the ambiguity of models calculated from detergent solubilised MPs can be reduced (21) (22) (23) (24) . It was recently shown that SANS data recorded at several contrast points together with an a priori estimate of detergent aggregation number, allow one to reliably reconstruct the shapes of MPs in detergent solutions (24) . In this publication it was suggested that a minimum of two SANS curves, collected under different contrast conditions, are required to obtain a stable solution. This requirement is necessary to isolate the signals from the two principal contributions to the scattering intensity, from the protein itself and from the hydrophillic part of the detergent corona. For this purpose the author developed an ab initio algorithm with a four-phase bead model for a protein: (1) protein, (2) detergent heads, (3) detergent tails, and (4) solvent (24) . An additional set of geometrical and symmetrical constraints was also required.