**4.1.7 Driving forces for ligand binding by MUP and HBP**

Ligand binding to both HBP and MUP was strongly enthalpy-driven. The overall entropy of binding was the same within error for both MUP and HBP, yet the contributions from protein, ligand, and solvent are very different.

In the case of HBP, the dominant entropic contribution to binding arises from ligand desolvation, with a significant contribution from protein degrees of freedom. This contribution is favourable. However, the overall entropic contribution to binding is unfavourable, which indicates that the entropic contribution from desolvation of the protein binding pocket is strongly unfavourable.

In MUP, a favourable contribution to binding entropy is also derived from ligand desolvation, but is unable to overcome the unfavourable contribution from "freezing"

quantum coherence (HSQC) titrations of ABP with 1-deoxy-galactose, which is a fast-

From these assignments, a comparison was made of the chemical shifts of the backbone amide resonances of ABP in the unbound state and in the complex (Figure 7). Large chemical shift changes were observed in the binding site area and in the region linking the two domains of ABP. This suggested that ligand binding might be associated with a substantial conformational change in ABP (Figure 8). Such conformational change (domain reorientation) upon binding is observed in other members of the periplasmic-binding protein family, and have been proposed for ABP on the basis of the results of small-angle Xray scattering and theoretical studies (Mao *et al.*, 1982, Newcomer *et al.*, 1981). Smaller changes in chemical shift were also observed at sites distal to the binding site, which suggests that small conformational changes, resulting from protein dynamic behaviour,

Fig. 7. Chemical shifts for galactose binding to ABP. Changes in backbone amide chemical

Fig. 8. Conformational changes between apo-ABP (light blue) and galactose-ABP complex. The apo protein is much more 'open' than the complex. Large conformational changes are observed in the hinge region (coloured red), and the reorientation of N-domain (yellow) and C-domain (dark blue) towards each other is quite pronounced. Bound galactose is coloured cyan and displayed as VDW spheres. Side chains of several residues involved in direct

exchanging ligand (MacRaild *et al.*, 2007).

occur upon the binding event.

shifts are plotted against protein residue number.

interactions with ABP are showed and coloured dark cyan.

ligand degrees of freedom upon binding. The favourable entropic contribution from desolvation of the protein binding pocket that one would predict in a "classical" hydrophobic interaction is absent in MUP, since the occluded binding pocket is substantially desolvated prior to binding. This phenomenon has subsequently been observed in other proteins, such as streptavidin and HIV-protease receptors (Young *et al.*, 2007).

As suggested by MD simulations, there are around 6 water molecules (on average) occupying the binding site of HBP prior to ligand binding. As already stated, no significant change in entropic contribution should arise from displacing these waters upon ligand binding, since their dynamic behaviour inside the pocket is similar to the behaviour of the bulk water. However, four water molecules are sequestered in the binding pocket in the histamine-HBP complex. These waters are significantly more ordered than the bulk water (Syme *et al.*, 2010), which contributes negatively to the entropic term of binding free energy. This unfavourable entropic contribution can be estimated as about 30 to 40 kJ/mol, which is qualitatively consistent with the observed sign of *CP* on binding.

In conclusion, in the case of HBP, we found favourable entropic contributions to binding from desolvation of the ligand. However, the overall entropy of binding was unfavourable due to a dominant unfavourable contribution arising from the loss of ligand degrees of freedom, together with the sequestration of solvent water molecules into the binding pocket in the complex. This can be contrasted with MUP, where desolvation of the protein binding pocket made a minor contribution to the overall entropy of binding given that the pocket is substantially desolvated prior to binding.
