The Department of Biomedicine

BBB seminar: Gerhard Klebe

Addressing protein targets with small molecule fragments and self-optimized inhibitors by crystallography

Gerhard Klebe
Institute of Pharmaceutical Chemistry, Philipps University Marburg, Germany

In recent years, fragment-based lead discovery has become increasingly popular to identify first lead candidates for relevant drug targets. One advantage of the usage of fragments is their low molecular weight (< 250 g/mol), which leaves sufficient space for subsequent chemical optimization. However, the inherent low-binding affinity of fragment-like molecules (typically mM to µM range) poses a challenge to current biophysical and biochemical screening methods, particularly whether typically applied pre-screening methods are able to reliably detect all putative fragments that can subsequently be crystallized with the target protein, or whether a direct fragment screening on protein crystals is the superior strategy. We therefore started to design a 361-entry fragment library, and tested this collection against the aspartic protease endothiapepsin. Initially, we screened all compounds by six biophysical methods to identify putative binders. Disappointingly, each individual screening method revealed varying hit rates with only a minimal overlap of commonly shared hits. We then performed a crystallographic screening of the entire fragment library directly on crystals. Remarkably, 71 crystal structures could be determined out of the 361 entries and nearly half of the hits were missed by any of the biophysical screening methods. We therefore strongly advocate to screen fragments directly on crystals instead of applying elaborated pre-filters, as the determination of fragment-bound crystal structures is pivotal for any subsequent design project to evolve a fragment into a drug candidate. Remarkably, we detected two structures, in which the fragments had reacted under the crystallization conditions to novel products. We therefore used this information for dynamic combinatorial chemistry to discover novel leads for endothiapepsin. Via two cycles of subsequent dynamic library design, a 54 nM binding hit was identified.

Selected references:

J. Schiebel et al. (2015). One question, multiple answers: Biochemical and biophysical screening methods retrieve deviating fragment hit lists. Chem. Med. Chem. 10:1511-21 http://dx.doi.org/10.1002/cmdc.201500267

J. Schiebel et al. (2016). Six biophysical screening methods miss a large proportion of crystallographically discovered fragment hits: A case study. ACS Chem. Biol. 11:1693-1701 http://dx.doi.org/10.1021/acschembio.5b01034

M. Mondal et al. (2014). Structure-based design of inhibitors of the aspartic protease endothiapepsin by exploiting dynamic combinatorial chemistry. Angew. Chem. 53:3259-63 http://dx.doi.org/10.1002/anie.201309682

M. Mondal et al. (2016). Fragment-linking and optimization of inhibitors of the aspartic protease endothiapepsin: Fragment-based drug design facilitated by dynamic combinatorial chemistry. Angew. Chem. 55:9422-26 http://dx.doi.org/10.1002/anie.201603074

J. Cramer et al. Watch out for the red herring: Surprising reactivity of a fragment results in biological activity. Angew. Chem. in press http://dx.doi.org/10.1002/anie.201609824R1

Chairperson: Ruth Brenk <ruth.brenk@uib.no>, Department of Biomedicine