**4.1.2 Calorimetric studies of MUP**

Given that the binding pocket of MUP is very hydrophobic, an entropy-driven binding signature might have been expected for ligand-MUP interactions. Surprisingly, global

Typical ITC isotherms for the binding of histamine to the HBP-D24R mutant (top panel) and wild-type HBP (middle panel), and the resulting thermodynamic parameters (bottom panel) are shown in Figure 5. Consistent with structural data, histamine bound to wild-type HBP with 2:1 stoichiometry (both binding sites occupied), while to mutant with 1:1 stoichiometry. In both cases, histamine bound with high affinity (Kd in nanomolar range), but the binding enthalpies and entropies were somewhat different, even though binding was largely enthalpy-driven in both cases. These differences in thermodynamic details are a manifestation of the enthalpy-entropy compensation, introduced in the previous sections of this chapter: wild-type HBP binds histamine with similar affinity than D24R mutant, but its enthalpic contribution is more favourable than that of D24R, at the expense of the entropic

A major concern regarding the data shown in Figure 5 was the possibility of proton exchange (release or binding) during the binding event. In such case, the observed enthalpy

explained previously. To assess this effect, titrations between histamine and HBP-D24R mutant were performed in two different buffers, characterised by very different values of

contribution from proton exchange should be easily detectable on the basis of the

kJ/mol, respectively) showed that there were no significant protonation effects associated

In order to gain deeper insight into the entropic contribution to binding of ligands to HBP and MUP, 15N NMR relaxation measurements were employed to probe per-residue conformational entropies for backbone amides for the free (apo) protein and for the ligandprotein complexes. Backbone 15N longitudinal and transverse relaxation rates (*R1* = 1/*T1* and *R2* = 1/*T2*, respectively) were determined for the free protein and the complexes with ligands. Amide 15N and 1H-15N resonance assignments in the apo-HBP, apo-MUP, and the ligand-protein complexes were determined by use of conventional three-dimensional triple-

Fig. 6. Backbone amide entropy changes for HBP-D24R, quantified as differences between entropy of holo and apo protein for each protein residue. The plot shows histamine binding

– induced changes in entropy assessed by 15N relaxation measurements.

*Hb* would contain contributions from the ionisation of the buffer ( )

*Hion* 47.45 kJ/mol and 3.6 kJ/mol, respectively). Thus, any

*Hb* for the histamine-HBP interaction. Obtained enthalpies (-58 and -61

*Hion* , as

contribution, which is less favourable in wild-type HBP than in D24R mutant.

**4.1.3 Calorimetric studies of HBP** 

change

differences in

the ionisation enthalpy (

resonance experiments.

with histamine-HBPD24R binding.

**4.1.4 NMR relaxation measurements** 

thermodynamics data obtained for pyrazine ligands (Barratt *et al.* 2004) and alcohols (Barratt *et al.*, 2006) showed that binding is driven by favourable enthalpic contributions, rather than the classical hydrophobic effect. The only hydrogen bond that could be formed between a ligand and the protein binding site involved the hydroxyl group of tyrosine Y120. Barratt *et al.* (2004) reported that ITC measurements on the binding of isobutyl-methoxypyrazine (IBMP) to the Y120F (phenylalanine side chain lacks hydroxyl group) mutant showed slightly reduced enthalpy of binding compared to wild-type MUP, but the binding was nonetheless enthalpy-driven.

Binding of long-chain alcohols, such as n-octanol, n-nonanol, and 1,8 octan-diol was characterised by similar thermodynamic signature. Contrary to expectations, binding was enthalpy-driven (Barratt *et al.*, 2006). Each complex was characterised by a bridging water molecule between the hydroxyl group of Y120 and the hydroxyl group of ligand. The thermodynamic penalty to binding derived from the unfavourable desolvation of 1,8 octandiol (+21 kJ/mol with respect to n-octanol, which came from an additional hydroxyl group facing a hydrophobic pocket) was partially offset by a favourable intrinsic contribution.

Fig. 5. ITC data for obtained for HBP. Binding curves for HBPD24R mutant and wild-type BP are displayed in left and right panel, respectively. Bottom panel shows thermodynamic parameters for mutated and wild-type HBP, obtained from ITC measurements.

### **4.1.3 Calorimetric studies of HBP**

28 Thermodynamics – Interaction Studies – Solids, Liquids and Gases

thermodynamics data obtained for pyrazine ligands (Barratt *et al.* 2004) and alcohols (Barratt *et al.*, 2006) showed that binding is driven by favourable enthalpic contributions, rather than the classical hydrophobic effect. The only hydrogen bond that could be formed between a ligand and the protein binding site involved the hydroxyl group of tyrosine Y120. Barratt *et al.* (2004) reported that ITC measurements on the binding of isobutyl-methoxypyrazine (IBMP) to the Y120F (phenylalanine side chain lacks hydroxyl group) mutant showed slightly reduced enthalpy of binding compared to wild-type MUP, but the binding was

Binding of long-chain alcohols, such as n-octanol, n-nonanol, and 1,8 octan-diol was characterised by similar thermodynamic signature. Contrary to expectations, binding was enthalpy-driven (Barratt *et al.*, 2006). Each complex was characterised by a bridging water molecule between the hydroxyl group of Y120 and the hydroxyl group of ligand. The thermodynamic penalty to binding derived from the unfavourable desolvation of 1,8 octandiol (+21 kJ/mol with respect to n-octanol, which came from an additional hydroxyl group facing a hydrophobic pocket) was partially offset by a favourable intrinsic contribution.

Fig. 5. ITC data for obtained for HBP. Binding curves for HBPD24R mutant and wild-type BP are displayed in left and right panel, respectively. Bottom panel shows thermodynamic

parameters for mutated and wild-type HBP, obtained from ITC measurements.

nonetheless enthalpy-driven.

Typical ITC isotherms for the binding of histamine to the HBP-D24R mutant (top panel) and wild-type HBP (middle panel), and the resulting thermodynamic parameters (bottom panel) are shown in Figure 5. Consistent with structural data, histamine bound to wild-type HBP with 2:1 stoichiometry (both binding sites occupied), while to mutant with 1:1 stoichiometry. In both cases, histamine bound with high affinity (Kd in nanomolar range), but the binding enthalpies and entropies were somewhat different, even though binding was largely enthalpy-driven in both cases. These differences in thermodynamic details are a manifestation of the enthalpy-entropy compensation, introduced in the previous sections of this chapter: wild-type HBP binds histamine with similar affinity than D24R mutant, but its enthalpic contribution is more favourable than that of D24R, at the expense of the entropic contribution, which is less favourable in wild-type HBP than in D24R mutant.

A major concern regarding the data shown in Figure 5 was the possibility of proton exchange (release or binding) during the binding event. In such case, the observed enthalpy change *Hb* would contain contributions from the ionisation of the buffer ( ) *Hion* , as explained previously. To assess this effect, titrations between histamine and HBP-D24R mutant were performed in two different buffers, characterised by very different values of the ionisation enthalpy ( *Hion* 47.45 kJ/mol and 3.6 kJ/mol, respectively). Thus, any contribution from proton exchange should be easily detectable on the basis of the differences in *Hb* for the histamine-HBP interaction. Obtained enthalpies (-58 and -61 kJ/mol, respectively) showed that there were no significant protonation effects associated with histamine-HBPD24R binding.

#### **4.1.4 NMR relaxation measurements**

In order to gain deeper insight into the entropic contribution to binding of ligands to HBP and MUP, 15N NMR relaxation measurements were employed to probe per-residue conformational entropies for backbone amides for the free (apo) protein and for the ligandprotein complexes. Backbone 15N longitudinal and transverse relaxation rates (*R1* = 1/*T1* and *R2* = 1/*T2*, respectively) were determined for the free protein and the complexes with ligands. Amide 15N and 1H-15N resonance assignments in the apo-HBP, apo-MUP, and the ligand-protein complexes were determined by use of conventional three-dimensional tripleresonance experiments.

Fig. 6. Backbone amide entropy changes for HBP-D24R, quantified as differences between entropy of holo and apo protein for each protein residue. The plot shows histamine binding – induced changes in entropy assessed by 15N relaxation measurements.

subjected to COSMO calculations at three different temperature settings (270, 300, and 330 K) in order to extract the enthalpic and entropic contributions to the free energy of solvation, using the finite-difference approach. The approximation used therein was based on the assumption that the heat capacity is constant over a certain range of temperatures near the target temperature, *T*. In the case of solute molecules solvated in water, this approximation

K. Using the finite-difference approximation, the entropy can be approximated at the target

 ( ) 2 *G T T GT T*

*T* is the temperature difference. For the calculations presented here

Obviously, it was very difficult to assess the accuracy of such calculations, but the experimental solvation thermodynamics of two related "fragments" of histamine have been reported (Cabani *et al.*, 1981). Comparison (Table 1) shows that the solvation free energies are reproduced very well, and the solvation entropies and enthalpies reasonably well (for *n*propylamine) compared with experiment, which lends some confidence in the computed

*G* **(kJ/mol)** <sup>0</sup>

Histamine -70.1 -137.2 -67.1 n-propylamine -16.5 -65.5 -49 n-propylamine (exp) -18.4 -55.8 -37.4 2-methyl imidazole -43 -96.3 -53.3 2-methyl imidazole (exp) -42.9 n.a. n.a.

Table 1. Solvation parameters: free energy, enthalpy, and entropy at temperature 300 K for histamine and related ligands. All values are given in kJ/mol. Experimental values (exp) are

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

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

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

 

*T*

 

*H* **(kJ/mol)** <sup>0</sup> *T S*

*G T*( ) is the free energy of solvation

(11)

*T* was 30

 **(kJ/mol)** 

*T*) as wide as 50

usually holds near room temperature for temperature ranges (denoted as

*S T*

*S T*( ) denotes entropy at the target temperature,

**Ligand** <sup>0</sup>

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

protein, ligand, and solvent are very different.

binding pocket is strongly unfavourable.

temperature as in equation (11):

values for histamine.

taken from Cabani *et al.* (1981).

where

K.

energy, and

For both HBP and MUP, both positive and negative changes in local backbone entropy were observed. Entropy changes were not restricted to the binding pocket but were dispersed over the protein (Figure 6). For HBP as well as MUP, *T S* summed over backbone amides was close to the error value, i.e., an overall change in backbone entropy that was not statistically different from zero (Syme *et al.*, 2010).
