Limits...
Free Energy Perturbation Calculation of Relative Binding Free Energy between Broadly Neutralizing Antibodies and the gp120 Glycoprotein of HIV-1

View Article: PubMed Central - PubMed

ABSTRACT

Direct calculation of relative binding affinities between antibodies and antigens is a long-sought goal. However, despite substantial efforts, no generally applicable computational method has been described. Here, we describe a systematic free energy perturbation (FEP) protocol and calculate the binding affinities between the gp120 envelope glycoprotein of HIV-1 and three broadly neutralizing antibodies (bNAbs) of the VRC01 class. The protocol has been adapted from successful studies of small molecules to address the challenges associated with modeling protein–protein interactions. Specifically, we built homology models of the three antibody–gp120 complexes, extended the sampling times for large bulky residues, incorporated the modeling of glycans on the surface of gp120, and utilized continuum solvent-based loop prediction protocols to improve sampling. We present three experimental surface plasmon resonance data sets, in which antibody residues in the antibody/gp120 interface were systematically mutated to alanine. The RMS error in the large set (55 total cases) of FEP tests as compared to these experiments, 0.68 kcal/mol, is near experimental accuracy, and it compares favorably with the results obtained from a simpler, empirical methodology. The correlation coefficient for the combined data set including residues with glycan contacts, R2 = 0.49, should be sufficient to guide the choice of residues for antibody optimization projects, assuming that this level of accuracy can be realized in prospective prediction. More generally, these results are encouraging with regard to the possibility of using an FEP approach to calculate the magnitude of protein–protein binding affinities.

No MeSH data available.


Top: Loss of the hydrogen bond of W100B (center) with gp120 residue A141 (dashed yellow line) and the stabilizing network of pi–pi stacking interactions (dashed light blue lines) result in the large unfavorable effect upon mutation to alanine. Bottom: Five frames from the end of the wild-type end (green) and five frames from the end of the mutant end trajectory (blue) with light chain backbones aligned to the first wild-type frame are shown with the H3 loops highlighted in yellow for the wild-type frames and red for the blue frames.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5383735&req=5

f0035: Top: Loss of the hydrogen bond of W100B (center) with gp120 residue A141 (dashed yellow line) and the stabilizing network of pi–pi stacking interactions (dashed light blue lines) result in the large unfavorable effect upon mutation to alanine. Bottom: Five frames from the end of the wild-type end (green) and five frames from the end of the mutant end trajectory (blue) with light chain backbones aligned to the first wild-type frame are shown with the H3 loops highlighted in yellow for the wild-type frames and red for the blue frames.

Mentions: The first case, W100B, was the most unfavorable mutation in the alanine scan cases. Its indole nitrogen formed a highly stable hydrogen bond with a gp120 side chain (glu 204 in the RSC3 homology model) and also formed very stable pi-stacking interactions with two adjacent residues (Y100 and W47) that appear important in stabilizing the overall binding mode of VRC01 with gp120 (see the top panel of Fig. 7). The latter effectively anchored the long flexible CDR H3 loop to the sheet that W47 is on. In the bottom panel of Fig. 7, several frames from the wild type and mutant replicas are shown together, aligned to the backbone of the first wild-type frame. With W100B present, W100B (and Y100, not shown) is aligned well with the light chain, constraining the CDR H3 loop configuration. In the mutant end trajectory frames, in contrast, the much smaller alanine is essentially uncorrelated with the alignment of the light chain backbone, and the CDR H3 loop is able to assume a more favorable backbone configuration, explaining at least a part of the slow relaxation. The mutant trajectory frames also show a noticeable shift in the overall alignment between the heavy and light chains and the bNAb and the gp120. In addition to explaining the high unfavorability of the W100B to ALA mutation, this also explains why W47, which appears to have no direct contacts with the gp120, gave an unfavorable result upon mutation to ALA. The very stable contact with W100B also likely explains why Y100 on VRC01 showed little sensitivity to the presence of the glycan fragment NAG776, despite being in direct contact with it, as it could not shift into the space of the missing glycan when no fragment was included, due to the interaction with W100B. From the standpoint of future binding affinity optimization efforts, W100B and the residues contacting it are likely poor candidates for modification, as there is a large cost to destabilizing this network of interactions.


Free Energy Perturbation Calculation of Relative Binding Free Energy between Broadly Neutralizing Antibodies and the gp120 Glycoprotein of HIV-1
Top: Loss of the hydrogen bond of W100B (center) with gp120 residue A141 (dashed yellow line) and the stabilizing network of pi–pi stacking interactions (dashed light blue lines) result in the large unfavorable effect upon mutation to alanine. Bottom: Five frames from the end of the wild-type end (green) and five frames from the end of the mutant end trajectory (blue) with light chain backbones aligned to the first wild-type frame are shown with the H3 loops highlighted in yellow for the wild-type frames and red for the blue frames.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5383735&req=5

f0035: Top: Loss of the hydrogen bond of W100B (center) with gp120 residue A141 (dashed yellow line) and the stabilizing network of pi–pi stacking interactions (dashed light blue lines) result in the large unfavorable effect upon mutation to alanine. Bottom: Five frames from the end of the wild-type end (green) and five frames from the end of the mutant end trajectory (blue) with light chain backbones aligned to the first wild-type frame are shown with the H3 loops highlighted in yellow for the wild-type frames and red for the blue frames.
Mentions: The first case, W100B, was the most unfavorable mutation in the alanine scan cases. Its indole nitrogen formed a highly stable hydrogen bond with a gp120 side chain (glu 204 in the RSC3 homology model) and also formed very stable pi-stacking interactions with two adjacent residues (Y100 and W47) that appear important in stabilizing the overall binding mode of VRC01 with gp120 (see the top panel of Fig. 7). The latter effectively anchored the long flexible CDR H3 loop to the sheet that W47 is on. In the bottom panel of Fig. 7, several frames from the wild type and mutant replicas are shown together, aligned to the backbone of the first wild-type frame. With W100B present, W100B (and Y100, not shown) is aligned well with the light chain, constraining the CDR H3 loop configuration. In the mutant end trajectory frames, in contrast, the much smaller alanine is essentially uncorrelated with the alignment of the light chain backbone, and the CDR H3 loop is able to assume a more favorable backbone configuration, explaining at least a part of the slow relaxation. The mutant trajectory frames also show a noticeable shift in the overall alignment between the heavy and light chains and the bNAb and the gp120. In addition to explaining the high unfavorability of the W100B to ALA mutation, this also explains why W47, which appears to have no direct contacts with the gp120, gave an unfavorable result upon mutation to ALA. The very stable contact with W100B also likely explains why Y100 on VRC01 showed little sensitivity to the presence of the glycan fragment NAG776, despite being in direct contact with it, as it could not shift into the space of the missing glycan when no fragment was included, due to the interaction with W100B. From the standpoint of future binding affinity optimization efforts, W100B and the residues contacting it are likely poor candidates for modification, as there is a large cost to destabilizing this network of interactions.

View Article: PubMed Central - PubMed

ABSTRACT

Direct calculation of relative binding affinities between antibodies and antigens is a long-sought goal. However, despite substantial efforts, no generally applicable computational method has been described. Here, we describe a systematic free energy perturbation (FEP) protocol and calculate the binding affinities between the gp120 envelope glycoprotein of HIV-1 and three broadly neutralizing antibodies (bNAbs) of the VRC01 class. The protocol has been adapted from successful studies of small molecules to address the challenges associated with modeling protein–protein interactions. Specifically, we built homology models of the three antibody–gp120 complexes, extended the sampling times for large bulky residues, incorporated the modeling of glycans on the surface of gp120, and utilized continuum solvent-based loop prediction protocols to improve sampling. We present three experimental surface plasmon resonance data sets, in which antibody residues in the antibody/gp120 interface were systematically mutated to alanine. The RMS error in the large set (55 total cases) of FEP tests as compared to these experiments, 0.68 kcal/mol, is near experimental accuracy, and it compares favorably with the results obtained from a simpler, empirical methodology. The correlation coefficient for the combined data set including residues with glycan contacts, R2 = 0.49, should be sufficient to guide the choice of residues for antibody optimization projects, assuming that this level of accuracy can be realized in prospective prediction. More generally, these results are encouraging with regard to the possibility of using an FEP approach to calculate the magnitude of protein–protein binding affinities.

No MeSH data available.