Using quantum chemistry to estimate chemical shifts in biomolecules

Abstract

An automated fragmentation quantum mechanics/molecular mechanics approach (AFNMR) has shown promising results in chemical shift calculations for biomolecules. Sample results for ubiquitin, and an RNA hairpin and helix are presented, and used to recent directions in quantum calculations. Trends in chemical shift are stable with regards to change in density functional or basis sets, and the use of the small "pcSseg-0" basis, which was optimized for chemical shift prediction [1], opens the way to more extensive conformational averaging, which can often be necessary, even for fairly well-defined structures.

Keywords: Chemical shift NMR conformational averaging.

Figure 1:

AFNMR model for an RNA hairpin. The solute (RNA) molecule is illustrated in the center. A local fragment is constructed centered on each nucleotide (or amino acid for proteins); four such fragments are illustrated at on the left and right. The atoms shown for each fragment form the quantum region, and point charges surrounding each fragment are used to represent the electrostatic effects arising from the remaining RNA atoms (which result from Amber charges embedded in the dielectric of ε = 4), and effects of solvent (represented by a dielectric ε = 78 and a Debye-Huckel ionic strength taken as 0.1M by default).

Figure 2:

Average shifts in the 10 NMR structures from pdb ID 2koc. Calculations carried out with the deMon 5.0 code, using the OLYP functional and either the pcSseg-0 or pcSseg-1 basis sets.

Figure 3:

Average shifts in the 10 NMR structures from pdb ID 2koc. Calculations carried out at the OLYP/pcSseg-1 level; experimental shifts from BMRB entry bmr5705.

Figure 4:

Calculated vs. observed shifts for ubiquitin. Calculated values are the average shift for the ten models in PDB ID 1d3z, at the OLYP/pcSseg-0 level. Observed shifts are from BMRB entry bmr17769. Solid line is y = x ; dashed line is a least-squares best-fit line,whose slope is shown. R is the Pearson correlation coefficient.

Figure 5:

Effects of conformational averaging on the ¹³C shifts in 2gbh. Snapshots (including solvent waters and ions) are taken from Ref. [44]. Calculated shifts at the OLYP/pcSseg-0 level are shown for a single snapshot (left) and for the average over 20 snapshots spaced 0.5 nsec apart (right). Experimental shifts from Ref. [43].