Deuterium

The programs CRYSOL (Svergun et al., 1995 ▶ ) for X-rays and CRYSON (Svergun et al., 1998 ▶ ) for neutrons evaluate the solution scattering from macromolecules with known atomic structure and fit a predicted curve to experimental scattering data by minimizing the discrepancy (Feigin & Svergun, 1987 ▶ ) where c is a scaling factor, N is the number of points and σ denotes the experimental errors. In the fitting process, the excess scattering density of the hydration shell, the average atomic group radius and the related total excluded volume can be adjusted. With the recent progress in high-resolution structure determination and advances in structure prediction and docking algorithms, tremendous numbers of structural models are becoming available. The screening of multiple models against experimental scattering data (typically SAXS) is often performed to select the best configuration in solution. The performance of these programs is crucial when applied to large numbers of structures and to deal with the increased number of angular data points in scattering profiles resulting from the improved resolution of detectors employed at the modern SAXS beamlines (e.g. PILATUS from DECTRIS; http://pilatus.web.psi.ch/pilatus.htm). In order to speed up CRYSOL calculations, experimental scattering intensities and associated errors are automatically remapped into a sparser grid for the search of the best fitting parameters. Depending on the number of experimental points, the regridding operation speeds up the fitting procedure by up to a factor of five. The final fits are recalculated for the optimum parameters for the original experimental data points.
Practice shows that in some cases (e.g. as a result of buffer mismatch) the higher-angle positions of the scattering data may contain systematic deviations, which can be accounted for by subtraction/addition of a constant term to the experimental data. An option of background constant adjustment has been added to CRYSOL to allow for the correction of such over- or under-subtracted buffer signal. A linear least-squares minimization with boundaries (Lawson & Hanson, 1995 ▶ ) is used to find the scaling coefficient and the background constant value when fitting a theoretical curve to experimental data.
Typically, CRYSOL and CRYSON skip all H atoms present in the PDB files and instead make an assignment of the number of bound H atoms for each atomic group based on the chemical compound library (ftp://ftp.wwpdb.org/pub/pdb/data/monomers/components.cif) in order to compute the scattering. If a full-atom model containing all H (or deuterium) atoms is available, the user has the option to take all the atoms ‘as is’, which can be specified in both interactive and batch modes [in the latter case the input parameters are specified on the command line (Konarev et al., 2006 ▶ )]. Since the remediation of the PDB archive in 2007–2008 (http://www.rcsb.org/pdb/static.do?p=general_information/news_publications/index.html) the nomenclature of many heteroatoms had been changed, and the assignments of bound hydrogen to an atom became ambiguous, such that the hydrogen assignment may be incorrect in some cases. To resolve this problem, both the new (after version 3.1) and the old (before version 3.0) PDB formats are now supported. By default, the new format is assumed, but the user can also enforce the old format by using the ‘/old’ key in the command line input.

Before 2 h of HDX analysis, the compound tutin (200 μM) was added into the sample, with the control sample adding an equal volume of tutin buffer. For deuterium labeling, CN (4 μM) in the buffer (20 mM Tris-HCl, 1 mM CaCl₂, 0.5 mM TCEP, and 150 mM NaCl, in H₂O, pH 7.5) in the presence or absence of 200 μM tutin was diluted 10-fold by the labeling buffer containing 20 mM Tris-HCl, 1 mM CaCl₂, 0.5 mM TCEP, and 150 mM NaCl, in 100% D₂O at pD 7.4. After incubation for 30, 90 or 300 seconds at 25 °C, the same volume of ice-cold quench buffer containing 4 M guanidine hydrochloride, 500 mM TECP and 200 mM citric acid in water solution at pH 1.8, 100% H₂O, was added to quench deuterium uptake. The sample was digested with pepsin (Promega) on ice for 5 min, and removed by centrifugation. An ACQUITY UPLC BEH C18 column (2.1 μm, 1.0 mm × 50 mm, Waters) equipped with an Ultimate 3000 UPLC system (Thermo Scientific) were used for the obtained peptides separation. A Q Exactive mass spectrometer was used for mass spectrometry analysis of the peptides. Mass spectrometry data were compared with Proteome Discoverer (Thermo Scientific) to match the corresponding peptide in CN. XCALIBUR (Thermo Scientific) was used to inspected peptide peaks. In order to estimate the max deuterium uptake of peptides, a repeated experiment was performed extending incubation in D₂O for 24 h. HDExaminer (Sierra Analytics) was used for calculating deuterium uptake levels. Deut % for different peptides were calculated as follows. ${\mathrm{Deut}}_i\%=\frac{\#{\mathrm{D}}_i/\left(\#{\left(-\mathrm{CO}-\mathrm{NH}-\right)}_i-\#{\mathrm{Pro}}_i-1\right)}{\mathrm{Max}{\mathrm{D}}_i}\times 100\%$ # D_i: deuterium numbers for peptide i at a certain hydrogen/deuterium exchange time; $\#{\left(-\mathrm{CO}--\mathrm{NH}-\right)}_i$ : amide bond numbers of each peptide; # Pro_i: the proline number for peptidei; Mxx D_i: maximum deuterium uptake for peptide i.