Could Fragment-Based Screening Using NMR Tackle Antibiotic Resistance?

The World Health Organization names antibiotic resistance as one of the biggest threats to global health, food security and development and this year, published a list of antibiotic-resistant ‘priority pathogens’ - naming 12 families of bacteria that they see as being the most serious threat to our health, with the aim of directing increased research and development towards the pathogens posing the highest risk.

Pseudomonas aeruginosa - A priority pathogen

Attention is turning to innovative strategies in antimicrobial drug development pathways, and one avenue focuses on anti-virulence, which targets the capacity of the bacterium to cause disease rather than focusing on the survival of the bacterium itself.

A recent study by Mohanty et al (2017) explored the anti-virulence approach using the bacterium Pseudomonas aeruginosa, a pervasive variety of Gram-negative bacteria, known for its inherent antibiotic resistance.

P. aeruginosa features in the ‘critical’ category of the WHO ‘priority pathogens’ and can cause a wide range of severe infections in patients with serious underlying medical conditions (Gellatly and Hancock 2013). P. aeruginosa is considered to be at high risk of becoming resistant to all antibiotics currently in use.

Within the pathogen, disulfide bond protein A (PaDsbA1) plays a central role in the normal formation of disulfide bonds, the inhibition of which may be therapeutically exploited in the creation of antimicrobials which act on virulence factors.

By screening their in-house fragment library, the researchers set out to identify molecular compounds that bind to PaDsbA1 to potentially identify inhibitors anticipated to disrupt normal P. aeruginosa protein folding.
Fragment-based screening to the fore

Mohanty and his team assessed fragment binding by recording saturation transfer difference (STD) in nuclear magnetic resonance (NMR) experiments using a Bruker Avance spectrometer equipped with CryoProbe, giving them the precise digital control, fast NMR, high speed spectroscopy, flexibility and pure NMR frequency generation they needed to get robust results.

Fragment-based screening (FBS) is becoming an indispensable tool for the identification of potentially significant compounds in drug discovery and results in a higher hit rate than traditional screening methods. NMR is one of the best methods for performing primary screens as it is able to detect low affinity ligands and it permits vigorous screening library quality control.

Bruker Avance instruments can be used in combination with TopSpin software for improved spectrometer control and data analysis. An FBS workflow solution is built into TopSpin, which streamlines FBS data handling and analysis, providing an easier way to screen compounds to find potential new drugs.
Eight fragments with potential identified

The fragment screening conducted in the study identified eight fragments with measurable KD (dissociation constant) values of <5mM. The three highest-affinity hits were consistent with an unexpected mode of binding of PaDsbA1, at the non-catalytic face, which to the authors’ knowledge is a previously unreported site of specific small molecule binding to DsbA proteins.

Similar work with the Escherichia coli DsbA (EcDsbA) protein determined that molecules bound in the active site-adjacent groove, rather than at the non-catalytic face (Adams et al 2015). This indicates that targeting the non-catalytic face of PaDsbA1 may be a viable means of inhibiting PaDsbA1 function and is therefore a potential avenue for future research.

The disparate binding site locations of the fragments when comparing E. coli and P. aeruginosa was borne out by the fact that there was little overlap found between compounds that bound to PaDsbA1 and those that showed affinity with EcDsbA.
Implications for the antibiotic drug discovery pipeline

The data revealed in this study regarding: the interaction between the fragment library hits and the protein; the structure of the molecules that produced the hits; and the selectivity of the binding of the fragments to PaDsbA1 rather than the DsbA protein from E. coli, represent consequential discoveries.

Together, these results constitute a foundation for the development of improved compounds that could potentially selectively inhibit PaDsbA1 and interfere with P. aeruginosa virulence, in what could be a welcome boost for the flagging antimicrobial drug discovery pipeline.

The authors point out that such anti-virulence approaches confer an additional benefit, as removing selection pressure imparted by bactericidal agents may reduce the tendency for antibiotic resistance to develop.