Opinion
Special Issue: Computation and Modeling
Advanced Modeling Reconciles Counterintuitive Decisions in Lead Optimization

https://doi.org/10.1016/j.tibtech.2016.12.003Get rights and content

Trends

As first described in maps of local dewetting propensities, the ease with which water is locally removed from around the target influences lead optimization (LO) decisions in drug design.

Common hotspots for water displacement often defy intuition, resulting in ‘counterintuitive”’ LO decisions.

We introduce biophysical advances on dielectric modulation down to single-water-molecule contributions to reconcile mismatches across drug–target interfaces resulting from counterintuitive LO decisions.

We incorporate three-body energy terms that account for the net stabilization of target structure on removal of interfacial water concurrent with drug–target association.

Unexplored drug-induced environmental changes affecting the target electrostatic interactions are validated against affinity data, yielding the computational accuracy required to improve drug design.

Lead optimization (LO) is essential to fulfill the efficacy and safety requirements of drug-based targeted therapy. The ease with which water may be locally removed from around the target protein crucially influences LO decisions. However, inferred binding sites often defy intuition and the resulting LO decisions are often counterintuitive, with nonpolar groups in the drug placed next to polar groups in the target. We first introduce biophysical advances to reconcile these apparent mismatches. We incorporate three-body energy terms that account for the net stabilization of preformed target structures upon removal of interfacial water concurrent with drug binding. These unexplored drug-induced environmental changes enhancing the target electrostatics are validated against drug–target affinity data, yielding superior computational accuracy required to improve drug design.

Section snippets

Counterintuitive Drug Design

We are concerned with molecular targeted therapy, specifically with designing small molecules that bind dysfunctional proteins that need to be blocked for therapeutic purposes 1, 2, 3, 4. Once a target has been validated, drug discovery commences with the identification of a lead. The lead is a compound with nanomolar or submicromolar target affinity and is typically found via high-throughput screening (see Glossary) against a proprietary compound library usually covering vast chemical

Reconciling Counterintuitive Drug Design by Incorporating Three-Body Energy Terms

A targetable backbone amide is known to occur typically in structured regions [7], where it is paired with a backbone carbonyl forming a hydrogen bond. Such water-exposed backbone hydrogen bonds are called dehydrons 4, 11, 12, 13. Due to confinement at subnanometer scales, water vicinal to such dehydrons is frustrated in its hydrogen bonding possibilities as it binds to the backbone carbonyl (Figure 1). This is consistent with the ‘high free-energy content’ (low entropy) of water around

Drug Design Guided by Three-Body Energy Contributions: Reworking Imatinib

To illustrate the overlooked effects of three-body contributions, we need to focus on a LO case that involves exclusively the incorporation of a nonpolar group to the lead scaffold with the express intent of removing water from the vicinity of a specific dehydron in the target protein. Such an example exists: the modification of the cancer drug imatinib by the incorporation of an extra methyl group to enhance the drug affinity and specificity, an optimization decision that resulted in the

Concluding Remarks and Future Perspectives

The end product of the drug discovery process is often far from what one would expect in terms of pairwise matching across the drug–target interface 4, 5, 6, 7, 8, 9. This tells us that LO does not follow a rational path, itself a symptom that the underlying physical principles governing drug–target affinity are not fully understood. LO is mainly guided by screening chemical combinatorial possibilities for lead modification and the net result is seldom intuitively appealing. This black-box

Glossary

Breakpoint cluster region (BCR)–Abelson murine leukemia (ABL)
results from a chromosomal translocation whereby the ABL viral oncogene homolog 1 (ABL1) gene from human chromosome 9 is juxtaposed onto the BCR gene from chromosome 22, encoding a hybrid constitutively active signaling protein, the BCR–ABL kinase, that causes cells to divide uncontrollably.
Chimera
hybridization and fusion of two or more gene products into a single molecule.
Chromosomal translocation
genetic abnormality caused by

References (26)

  • A. Heifetz

    Guiding lead optimization with GPCR structure modeling and molecular dynamics

    Curr. Opin. Pharmacol.

    (2016)
  • A. Fernández et al.

    Dehydron: a structurally encoded signal for protein interaction

    Biophys. J.

    (2003)
  • M. Fruber

    Lead generation

  • S. Ekins

    Thermodynamic proxies to compensate for biases in drug discovery methods

    Pharm. Res.

    (2016)
  • A. Fernández

    Physics at the Biomolecular Interface

    (2016)
  • A. Fernández

    An anticancer C-kit kinase inhibitor is re-engineered to make it more active and less cardiotoxic

    J. Clin. Invest.

    (2007)
  • D.D. Robinson

    Understanding kinase selectivity through energetic analysis of binding site waters

    ChemMedChem

    (2010)
  • T. Beuming

    Thermodynamic analysis of water molecules at the surface of proteins and applications to binding site prediction and characterization

    Proteins

    (2012)
  • L. Wang

    Accurate and reliable prediction of relative ligand binding potency in prospective drug discovery by way of a modern free-energy calculation protocol and force field

    J. Am. Chem. Soc.

    (2015)
  • A. Fernández

    Rational drug redesign to overcome drug resistance in cancer therapy: imatinib moving target

    Cancer Res.

    (2007)
  • D. Robinson

    Differential water thermodynamics determine PI3K-β/δ selectivity for solvent-exposed ligand modifications

    J. Chem. Inf. Model.

    (2016)
  • Fernández, A. et al. The Board of Regents of the University of Texas System, Rice University. Methods and compositions...
  • A. Fernández

    Epistructural tension promotes protein associations

    Phys. Rev. Lett.

    (2012)

    • Cited by (7)

    View all citing articles on Scopus
    View full text
    © 2016 Elsevier Ltd. All rights reserved.