Thermodynamic Mapping of Antigen Antibody Interfaces

In contrast to the wealth of structural information on antigen-antibody and other protein-protein interfaces, the available data on the thermodynamics of the association reactions are far more limited. Indeed, our current view of the energetics of protein-protein association is largely based on detailed mutagenesis and binding studies of only a few complexes [15]. We have studied the binding of monoclonal antibody D1.3 to two structurally distinct ligands: its cognate antigen, hen egg white lysozyme (HEL), and the anti-D1.3 antibody E5.2. The crystal structure of the complex formed by D1.3 with HEL has been determined to a nominal resolution of 1.8 A [21]. In addition, the structure of the complex between D1.3 and E5.2 is known to 1.9-A resolution [12]. Surprisingly, D1.3 contacts HEL and E5.2 through essentially the same set of combining site residues (and most of the same atoms). Thus, of the 18 D1.3 residues that contact E5.2 and the 17 that contact HEL, 14 are in contact with both E5.2 and HEL. In this review, we will focus on the D1.3-HEL and D1.3-E5.2 complexes, as these currently represent the most extensively studied models for antigen-antibody recognition.

To evaluate the relative contribution of individual residues to stabilization in the D1.3-HEL and D1.3-E5.2 complexes, alanine-scanning mutagenesis was performed in the D1.3 combining site. In total, 16 single alanine substitutions were introduced and their effects on affinity for HEL and for E5.2 were measured using surface plasmon resonance detection, fluorescence quench titration, or sedimentation equilibrium [22]. Mutagenesis of D1.3 residues in contact with HEL in the crystal structure of the D1.3-HEL complex revealed that residues in VLCDR1 and VHCDR3 contribute more to binding than residues in VLCDR2, VlCDR3, VhCDR1, and VHCDR2. By far the greatest reductions in affinity (AGmutant - AGwild type > 2.5 kcal/mol) occurred on substituting three residues: VLTrp92, VHAsp100, and VHTyr101. By replacing VLTrp92 with residues bearing increasingly smaller side chains and determining the crystal structures and thermodynamic parameters of binding for each of the resulting mutant D1.3-HEL complexes, we demonstrated a correlation between the binding free energy and the apolar surface area that corresponds to 21 cal mol-1 A-2 [23]. This estimate of the hydrophobic effect in a protein-protein interface is in excellent agreement with predictions based on transfer free-energy values for small hydrophobic solutes.

Significant effects on HEL binding (1.0 to 2.0 kcal/mol) were also seen for substitutions at D1.3 positions VLTyr32 and VHGlu98, even though the latter is not involved in direct contacts with HEL. Mutations at nine other contact positions (V LHis30, VLTyr49, VLTyr50, VLSer93, VHTyr32, VHTrp52, VHAsp54, VHAsp58, and VHArg99) had little or no effect (<1.0 kcal/mol). Therefore, the binding of HEL by D1.3 is largely mediated by only 5 of the 14 residues tested.

For the interaction of D1.3 with E5.2, affinity measurements showed that VHCDR2, VHCDR3, and VLCDR1 of D1.3 are more important for binding E5.2 than VHCDR1, VLCDR2, and VlCDR3 [22]. Overall, D1.3 VH residues appear to contribute more to the free energy of binding than VL residues, as the most destabilizing alanine substitutions (>2.5 kcal/mol) are located in VHCDR2 (Trp52Ala and Asp54Ala) and VhCDR3 (Glu98Ala, Asp100Ala, and Tyr101Ala). Significant effects (1.0 to 2.0 kcal/mol) were also observed for the following contact residues: His30 and Tyr32 in VLCDR1, Tyr49 in VLCDR2, Tyr32 in VHCDR1, Asn56 and Asp58 in VhCDR2, and Aeg99 in VHCDR3. Mutations at positions VLTyr50, VLTrp92, and VHThr30 had little or no effect (<1.0 kcal/mol). Thus, of the 15 contact residues tested, 12 make significant contributions to binding E5.2.

On the basis of extensive mutational analysis of the complex between human growth hormone and its receptor, Wells and colleagues [24-26] proposed that the formation of specific protein-protein complexes is mediated by only a few productive interactions or "hot spots" that dominate the energetics of association. Consistent with this idea, our analysis of the D1.3-HEL interaction revealed that only a small subset of the total combining site residues of D1.3 appears to account for a large proportion of the binding energy; most residues (9 of 14) make little or no apparent net contribution (<1.0 kcal/mol). This contrasts with the interaction of D1.3 with E5.2 in which nearly all the contacting residues play a demonstrable role in binding ligand (>1.0 kcal/mol), even though a number of hot spots (AAG > 2.5 kcal/mol) are clearly present. Therefore, stabilization of the D1.3-E5.2 complex is achieved by the accumulation of many productive interactions of varying strengths over the entire interface between the two proteins.

The functional surfaces of D1.3 involved in binding HEL and E5.2 mapped onto its three-dimensional structure are shown in Figs. 1A and 1B, respectively. With the exception of VLTrp92, which lies at the periphery, the residues of D1.3 most important for binding HEL (VHTyr101, VHAsp100, VLTyr32, and VHGlu98) are located in a contiguous patch at the center of the combining site. Residues at the periphery make only minor contributions to the binding energy. A similar pattern is observed for the D1.3-E5.2 complex, with the most important residues (VLTyr32, VHTrp52, VHAsp54, VHGlu98, VHAsp100, and VHTyr101) forming a central band of key contacts. For the most part, however, the hot spots for the two interactions do not overlap. For instance, alanine substitution at position VLTrp92 of D1.3 produces a 100-fold decrease in affinity for HEL but does not appreciably affect binding to E5.2. Conversely, the VHTrp52Ala

Figure 1 Energetic maps of antigen-antibody interfaces. (A) Space-filling model of the surface of D1.3 (left) in contact with HEL and of the surface of HEL (right) in contact with D1.3. VL residues are marked with asterisks. The two proteins are oriented such that they may be docked by folding the page along a vertical axis between the components. Residues are color-coded according to the loss of binding free energy upon alanine substitution. (B) Model of the surface of D1.3 (left) in contact with E5.2 and of the surface of E5.2 (right) in contact with D1.3. Residues are colored as in (A).

Figure 1 Energetic maps of antigen-antibody interfaces. (A) Space-filling model of the surface of D1.3 (left) in contact with HEL and of the surface of HEL (right) in contact with D1.3. VL residues are marked with asterisks. The two proteins are oriented such that they may be docked by folding the page along a vertical axis between the components. Residues are color-coded according to the loss of binding free energy upon alanine substitution. (B) Model of the surface of D1.3 (left) in contact with E5.2 and of the surface of E5.2 (right) in contact with D1.3. Residues are colored as in (A).

substitution decreases affinity for E5.2 1000-fold but has virtually no effect on binding to HEL. Only substitutions VHAsp100Ala and VHTyr101Ala greatly affect the binding to both HEL and E5.2. We therefore conclude that a single set of contact residues on D1.3 binds HEL and FvE5.2 in energetically different ways. Thus, although D1.3 recognizes these two proteins in ways that are structurally very similar, this similarity extends only partially to the functional epitopes.

To probe the relative contribution to binding of HEL residues in contact with D1.3 in the crystal structure of the FvD1.3-HEL complex, 12 non-glycine HEL residues were individually mutated to alanine and their affinities for wildtype D1.3 measured [27]. Significant decreases in binding (AAG > 1 kcal/mol) were only observed for substitutions at four contact positions: Gln121, Ile124, Arg125, and Asp119. The most destabilizing mutation was at position Gln121 (AAG=2.9 kcal/mol). In the wild-type structure, Gln121 penetrates a hydrophobic pocket, where it is surrounded by the aromatic side chains of VLTyr32, VLTrp92, and VHTyr101 [16].

Mutations at the remaining eight contact positions (Asp18, Asn19, Tyr23, Ser24, Lys116, Thr118, Val120, and Leu129) had little or no effect (AAG < 1 kcal/mol). Therefore, for both the D1.3 and HEL sides of this interface, only small subsets of the total contacting residues appear to account for a large portion of the binding energy.

As shown in Fig. 1A, the residues of HEL most important for binding D1.3 (Asp119, Gln121, Ile124, and Arg125) form a contiguous patch located at the periphery of the surface contacted by the antibody [27]. Hot spot residues on the D1.3 side of the interface generally correspond to hot spot positions on the HEL side. For example, HEL hot spot residues Gln121 (AAG=2.9 kcal/mol) and Arg125 (1.8 kcal/mol) contact D1.3 hot spot residue VLTrp92 (3.3 kcal/mol); in addition, Gln121HEL contacts VLTyr32 (1.7 kcal/mol) and VHTyr101 (>4.0 kcal/mol). Similarly, functionally less important D1.3 and HEL residues tend to be juxtaposed in the antigen-antibody interface: Asp18HEL (AAG=0.3 kcal/mol) and Thr118 (0.8 kcal/mol) interact with D1.3 VLTyr50 (0.5 kcal/mol) and VHTrp52 (0.9 kcal/mol), respectively.

To investigate the apparent contribution of E5.2 residues to stabilization of the D1.3-E5.2 complex, single alanine substitutions were introduced at 9 of 21 positions in the combining site of E5.2 involved in contacts with D1.3, and the affinity of the mutants for wild-type D1.3 [28] was measured. The most destabilizing substitutions are located at positions VHTyr98 and VHArg100b (AAG > 4.0 kcal/mol). Substitutions at the other 7 positions tested (VLTyr49, VHLys30, VHHis33, VHAsp52, VHAsn54, VHIle97, and VHGln100) also resulted in significant effects on binding (1.2 to 2.8 kcal/mol). When the residues of D1.3 and E5.2 important in complex stabilization were mapped onto the three-dimensional structure of each antibody, we observed that hot spot positions on the E5.2 side of the interface generally corresponded to hot spots on the D1.3 side (Fig. 1B), as in the D1.3-HEL interface (Fig. 1A). This complementarity of functional epitopes is in agreement with the observation that energetically critical regions on human growth hormone match those on its corresponding receptor [24-26]. In the hormone receptor case, however, the functional epitopes pack together to form a hydrophobic core surrounded by hydrophilic residues, with substantial reductions in affinity occurring only on substitution of the nonpolar ones. In contrast, our analysis of the D1.3-E5.2 and D1.3-HEL systems shows that both polar (e.g., D1.3 residues VHAsp54, VHGlu98, and VHAsp100) and nonpolar residues (e.g., D1.3 residues VLTrp92 and VHTrp52) play a prominent role in complex stabilization and that there is not a clear segregation of polar residues at the periphery of the interface and nonpolar ones at the core (Fig. 1).

Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


Post a comment