An open-source drug discovery platform enables ultra-large virtual screens.

On average, an approved drug currently costs US$2-3 billion and takes more than 10 years to develop1. In part, this is due to expensive and time-consuming wet-laboratory experiments, poor initial hit compounds and the high attrition rates in the (pre-)clinical phases. Structure-based virtual screening has the potential to mitigate these problems. With structure-based virtual screening, the quality of the hits improves with the number of compounds screened2. However, despite the fact that large databases of compounds exist, the ability to carry out large-scale structure-based virtual screening on computer clusters in an accessible, efficient and flexible manner has remained difficult. Here we describe VirtualFlow, a highly automated and versatile open-source platform with perfect scaling behaviour that is able to prepare and efficiently screen ultra-large libraries of compounds. VirtualFlow is able to use a variety of the most powerful docking programs. Using VirtualFlow, we prepared one of the largest and freely available ready-to-dock ligand libraries, with more than 1.4 billion commercially available molecules. To demonstrate the power of VirtualFlow, we screened more than 1 billion compounds and identified a set of structurally diverse molecules that bind to KEAP1 with submicromolar affinity. One of the lead inhibitors (iKeap1) engages KEAP1 with nanomolar affinity (dissociation constant (Kd) = 114 nM) and disrupts the interaction between KEAP1 and the transcription factor NRF2. This illustrates the potential of VirtualFlow to access vast regions of the chemical space and identify molecules that bind with high affinity to target proteins.

Nearest-neighbor NMR spectroscopy: categorizing spectral peaks by their adjacent nuclei.

Methyl-NMR enables atomic-resolution studies of structure and dynamics of large proteins in solution. However, resonance assignment remains challenging. The problem is to combine existing structural informational with sparse distance restraints and search for the most compatible assignment among the permutations. Prior classification of peaks as either from isoleucine, leucine, or valine reduces the search space by many orders of magnitude. However, this is hindered by overlapped leucine and valine frequencies. In contrast, the nearest-neighbor nuclei, coupled to the methyl carbons, resonate in distinct frequency bands. Here, we develop a framework to imprint additional information about passively coupled resonances onto the observed peaks. This depends on simultaneously orchestrating closely spaced bands of resonances along different magnetization trajectories, using principles from control theory. For methyl-NMR, the method is implemented as a modification to the standard fingerprint spectrum (the 2D-HMQC). The amino acid type is immediately apparent in the fingerprint spectrum. There is no additional relaxation loss or an increase in experimental time. The method is validated on biologically relevant proteins. The idea of generating new spectral information using passive, adjacent resonances is applicable to other contexts in NMR spectroscopy.

Exposing Small-Molecule Nanoentities by a Nuclear Magnetic Resonance Relaxation Assay.

Small molecules can self-assemble in aqueous solution into a wide range of nanoentity types and sizes (dimers, n-mers, micelles, colloids, etc.), each having their own unique properties. This has important consequences in the context of drug discovery including issues related to nonspecific binding, off-target effects, and false positives and negatives. Here, we demonstrate the use of the spin-spin relaxation Carr-Purcell-Meiboom-Gill NMR experiment, which is sensitive to molecular tumbling rates and can expose larger aggregate species that have slower rotational correlations. The strategy easily distinguishes lone-tumbling molecules versus nanoentities of various sizes. The technique is highly sensitive to chemical exchange between single-molecule and aggregate states and can therefore be used as a reporter when direct measurement of aggregates is not possible by NMR. Interestingly, we found differences in solution behavior for compounds within structurally related series, demonstrating structure-nanoentity relationships. This practical experiment is a valuable tool to support drug discovery efforts.

Mixed pyruvate labeling enables backbone resonance assignment of large proteins using a single experiment.

Backbone resonance assignment is a critical first step in the investigation of proteins by NMR. This is traditionally achieved with a standard set of experiments, most of which are not optimal for large proteins. Of these, HNCA is the most sensitive experiment that provides sequential correlations. However, this experiment suffers from chemical shift degeneracy problems during the assignment procedure. We present a strategy that increases the effective resolution of HNCA and enables near-complete resonance assignment using this single HNCA experiment. We utilize a combination of 2-13C and 3-13C pyruvate as the carbon source for isotope labeling, which suppresses the one bond (1Jαß) coupling providing enhanced resolution for the Cα resonance and amino acid-specific peak shapes that arise from the residual coupling. Using this approach, we can obtain near-complete (>85%) backbone resonance assignment of a 42 kDa protein using a single HNCA experiment.

Optimal control theory enables homonuclear decoupling without Bloch-Siegert shifts in NMR spectroscopy.

The Bloch-Siegert shift is a phenomenon in NMR spectroscopy and atomic physics in which the observed resonance frequency is changed by the presence of an off-resonance applied field. In NMR, it occurs especially in the context of homonuclear decoupling. Here we develop a practical method for homonuclear decoupling that avoids inducing Bloch-Siegert shifts. This approach enables accurate observation of the resonance frequencies of decoupled nuclear spins. We apply this method to increase the resolution of the HNCA experiment. We also observe a doubling in sensitivity for a 30 kDa protein. We demonstrate the use of band-selective Cß decoupling to produce amino acid-specific line shapes, which are valuable for assigning resonances to the protein sequence. Finally, we assign the backbone of a 30 kDa protein, Human Carbonic Anhydrase II, using only HNCA experiments acquired with band-selective decoupling schemes, and instrument time of one week.