Estimation of binding energies

The difference, 4G b is very much greater than the observed free 

A comparison of dG with (dG B — dGg) provides a lower limit to the intrinsic loss of entropy upon binding A (Eqn. 73). 

At a standard state of 1 M the maximum loss of entropy upon binding is about — 150 J/K/mole and therefore a comparison of TAS, obtained as in Eqn. 73, with this value allows an estimation of the tightness of binding for A. may be up to 45 kJ/mole (a factor of 10 M in binding constants) more favourable than the sum of AGj and AG.  

The most interesting problem to arise from the estimation of intrinsic binding energies is the apparently large values associated with small substituents. Examples are 9-16 kJ/mole for CHj [35] 21 kJ/mole for SCH3 [36] 34 kJ/mole for OH [37] and 23-38 kJ/mole for SH [38], These intermolecular interactions are much greater than those generally observed between solute and solvent or between molecules within a crystal. 

If the interior of a protein is analogous to a liquid the transfer of a group from water to a non-polar liquid may be taken as a model for the transfer of the group from water to the enzyme (Eqns. 74-75). 

---

HJ point out that in the detailed work on H2 by Kolos and Wolniewicz,114 the first excited state 3 2 was found to have a very weak minimum at a large separation. This binding presumably arises from a van der Waals force which is not included in the density functional theory when a local approximation to exchange and correlation is employed. Nevertheless, as HJ point out, their study of the corresponding state of the dimers Li2-Cs2 revealed a weak, but definite maximum in each case. Rough estimates of binding energy and equilibrium separation are shown in Table 16. It is, of course, possible that these results are a consequence of the local spin-density approximation, so that further work will... 

Table 16 Estimates of binding energy and equilibrium separation of alkali dimers,...

There are several experimental techniques available to study the chemical properties of clusters on surfaces and when used in combination, a detailed picture can be obtained. A classical approach is the use of temperature programmed desorption (TPD) and temperature programmed reaction (TPR) spectroscopy [39, 239,240] in combination with temperature-dependent and eventually time-dependent Fourier transformed infrared spectroscopy (FTIR) [39,192,241-244]. This combination allows for obtaining information on estimates of binding energies of reactant molecules (TPD), for characterizing... 

A number of principles are clear from this experiment. The first is that a straightforward approach to obtaining easy-to-synfhesize inhibitors may be successful, so is worth trying. A second is that empirical estimates of binding energy such as implemented in Ludi are capable of giving useful results. A third is that the protein may adapt to the inhibitor rather than vice versa. 

There have been a number of DPT calculations of Ci species on metal surfaces (see Further Reading section), producing quantitative estimates of binding energies, but few surprises. 

In comparison. Eex/E is = 0 for mass spectrometric detectors. Even after correcting for these relative efficiencies there remains a disturbing discrepancy between the two types of measurement. Huisken estimates a one-photon background of only 5%, whereas the measurements of Bizzarri et al show a cold- dimer-excitation to warm-dimer-dissociation signal ratio of 13%, under similar source conditions. This discrepancy has got to be cleared up experimentally. It does not, however, invalidate the above estimate of binding energy. 

Hermans, J., Wang, L. Inclusion of loss of translational and rotational freedom in theoretical estimates of free energies of binding. Application to a complex of benzene and mutant T4-lysozyme. J. Am. Chem. Soc. 119 (1997) 2707-2714... 

Hansson, T., Aqvist, J. Estimation of binding free energies for HIV proteinase inhibitors by molecular dynamics simulations. Prot. Eng. 8 (1995) 1137-1144... 

Hansson T and J Aqvist 1995. Estimation of Binding Free Energies for HIV Proteinase Inhibitors b Molecular Dynamics Simulations. Protein Engineering 8 1137-1144. 

The same technique can be used in some cases to obtain accurate estimates of binding free energy differences for a set of ligands of interest [25, 31-34]. The molecule taken as the reference need not be a real molecule. Indeed, the reference molecule could be intermediate between a large set of molecules of interest, so that conformations that are sufficiently representative of them all are sampled in the reference simulation. The justification for this approach is discussed in detail in Chap. 6. To achieve this for a variety of substituted phenols, Liu et al. [25] added dummy atoms to the ring at the sites they wished to substitute. Such dummy atoms can be softer ... 

T. Hansson and J. Aqvist, Estimation of binding free energies for HIV proteinase... 

---