3.4 Multiple factors govern the interaction of hotspot residues
To examine the predominant forces governing the interaction for the
hotspot residues, it would be worthwhile to correlate ∆∆G values of
hotspot residues with respect to changes in various types of
interactions upon mutation. Multiple linear regression analysis was
carried out to examine the correlation between ∆∆G values with energy
contributions from Coulomb, lipophilic, hydrogen bonding, van der Waal’s
interactions, packing desolvation, entropy and surface
complementarity. Strong correlation exists (in the range of 0.9
-1, with p-value close to 0) between the binding affinity ∆∆G and its
components (∆∆G of Coulomb, lipophilic, hydrogen bonding, van der Waal’s
interactions and desolvation). Notably, these interactions contribute in
equal proportion in the making of hotspot residues and no single factor
predominantly governs over other in the hotspot characteristics across
all three weak, moderate and strong types. As peptide has binding
characteristic similar to protein-ligand interactions, some degree of
correlation (~0.1) was also observed between ∆∆G and its
component ∆∆G (surface complementarity) in PPepI.
CONCLUSION
Peptides and peptidomimetics are straightforward alternatives to
protein-based biologics due to multiple advantages of larger shelf life,
feasibility of oral delivery, flexibility of optimization, screening and
synthesis. The knowledge of subtle differences between protein-protein
and protein-peptide interactions should aid in the effective design of
peptide-based biologics. In the present study, we have focussed on two
important class of residues, namely, hotspot and anchor residues, to
characterize differences between protein-protein and protein-peptide
interactions.
Using implicit solvation-based free energy calculations, alanine
scanning has been extensively performed on benchmarking datasets and
hotspot and anchor residues were identified, which has revealed many
interesting findings. The presence of sizable population (about 65%) of
hotspot residues at the interface of the complex suggest that nature has
remarkably optimized a great majority of the interface residues
responsible for protein-protein interactions during evolution. It turned
out that the differences in the two categories – PPI and PPepI are
readily apparent, once we group the hotspot data into three distinct
types, namely - weak hotspots (having binding free energy loss upon Ala
mutation, ΔΔG in 2-10 kcal/mol range), moderate hotspots (ΔΔG, 10-20
kcal/mol) and strong hotspots (ΔΔG, 20 kcal/mol and higher). Correlation
studies using MLR suggest that calculated free energy of binding of
hotspot directly correlate with coulomb, lipophilic, hydrogen bonding,
van der Waal interactions and desolvation penalty and no specific
preference of any of the factor(s) over other was observed across all
three types of hotspots. The analysis suggests that for PPI the
preference is charged and polar followed by hydrophobic residues while
for PPepI it is polar and hydrophobic followed by charged residues. In
PPI, weak hotspots are predominantly populated by polar and hydrophobic
residues. The distribution shifts towards charged and polar residues for
moderate type, and charged residue (Arg) is overwhelmingly present in
the strong type. In contrast, in the protein-peptide dataset, the
distribution shifts from predominantly hydrophobic & polar (in the weak
type) to more or less similar preference for polar, hydrophobic and
charged residues and finally the charged residue (Arg) and Trp are
mostly occupied in the strong type. Similar trend has been observed for
anchor residues in both categories. The present work is an attempt to
characterize and distinguish PPI and PPepI, focussing on two important
class of residues. Further work is required to facilitate the discovery
of new generations of peptide and peptidomimetic modulators. which can
be utilized in the effective design of biologics.
Acknowledgement Kiran Kumar thanks UGC, India for PhD fellowship.