PETRA: Drug Engineering via Rigidity Analysis

Abstract

Rational drug design aims to develop pharmaceutical agents that impart maximal therapeutic benefits via their interaction with their intended biological targets. In the past several decades, advances in computational tools that inform wet-lab techniques have aided the development of a wide variety of new medicines with high efficacies. Nonetheless, drug development remains a time and cost intensive process. In this work, we have developed a computational pipeline for assessing how individual atoms contribute to a ligand's effect on the structural stability of a biological target. Our approach takes as input a protein-ligand resolved PDB structure file and systematically generates all possible ligand variants. We assess how the atomic-level edits to the ligand alter the drug's effect via a graph theoretic rigidity analysis approach. We demonstrate, via four case studies of common drugs, the utility of our pipeline and corroborate our analyses with known biophysical properties of the medicines, as reported in the literature.

Keywords: ligand engineering; rigidity analysis.

Figure 1

The rigidity analysis of a protein (PDB 1CRN, (left)) identifies rigid clusters of atoms (right).

Figure 2

Protein-Ligand complex Engineering Through Rigidity Analysis (PETRA) compute pipeline. Dotted lines specify data; solid lines specify control flow.

Figure 3

Mock ligand and tree structure paths for generating an example set of 14 ligand variants. The oxygen is the root of the ligand because it engages in a stabilizing interaction with the protein; consequently, the oxygen is a member of each ligand variant. A depth-first traversal of the tree represents successive removals of atoms. Variants (b), (c), (d), for example, are all variants that have one atom less than the ligand that is the root of that tree (a).

Figure 4

Box plot for $SingelCu{t}_i$, Equation (1). The x-axis represents a singular cut atom and y-axis values are the average change in rigidity distance of the protein when the singular atom is cut from a ligand variant. ‡ designates involvement in a hydrophobic interaction.

Figure 5

Heat map for PairCut${}_{i,j}$, Equation (2). The x and y axes represent all the atoms in a given ligand, while the colors represent the rigidity distance of a protein when each pairwise set of atoms (i.e., C5${}^{\prime }$ and O2) is contained in a ligand variant. † = Hydrogen Bond ‡ = Hydrophobic Interaction.

Figure 6

A protein-ligand interaction map generated from the third party software, OpenEye, visualizing the interaction between PDB 2JJ8 and ligand AZZ.

Figure 7

Protein-ligand map (left), and schematic (right) of Ciprofloxacin, CPF, from PDB 4KRA.

Figure 8

$SingleCut$ metric for PDB 4KRA in complex with CPF. ring = $C2,C1,N1,C10,C9,C8,C7,C6,$$C5,C4$, ring${}^{\prime }$ = $N2,C17,C16,N3,C15,C14$, ring${}^{″}$ = $C11,C12,C13$. † and ‡ designate atoms engaging in hydrogen bonds and hydrophobic interactions, respectively.

Figure 9

$PairCu{t}_{i,j}$ heatmaps for Myristic Acid in complex with Human serum albumin (HSA) ((left), PDB 3B9L), and in complex with Decanoic Acid ((right), PDB 1TF0). ‡ designates an atom engaging in hydrophobic interactions.

Figure 10

Salicyclic Acid (left), and schematic (right) of AZT and SAL in complex with HSA, PDB 3B9M.

Figure 11

Box plot of $SingleCut$ metrics for atoms in Salycylic Acid, in complex with HSA, PDB 3B9M. The flat quartile boxes designate no variance when cut.

Figure 12

Heat maps for $PairCut$ metrics for HSA in complex with AZT and Myristic Acid (MYR) and SAL ((left), PDB 3B9M), and in complex with just AZT and MYR ((right), PDB 3B9L). ‡ designates hydrophobic interactions.

Figure 13

Ibuprofen (IBP) schematic and sample interaction map, and box plots for SingleCut metrics for 5 protein-IBP complexes. † and ‡ identify atoms involved in hydrogen bond and hydrophobic interactions, respectively.

Figure 14

PDB 3P6D-ZGB contains a variant of Ibuprofin with its aromatic ring containing a methyl group (SingleCut in (a), PairCut in (c)), while PDB 3P6E-ZGC does not include the methyl group (SingleCut in (b), PairCut in (d)).