5.3 Search for Different Chemical Structure Types
167
norm thinking (for such biases in chemical synthesis see Jia et al. 2019). AI also has
the potential to go beyond molecular docking for drug mode of action insights when
high quality data sets exist, which hopefully should be applicable in the targeted
antibacterial area (Rodrigues et al. 2018).
5.4 New Modes of Action/New Targets
Together with the search for different molecular structures the search for new bacterial
targets will continue to be of pressing importance. These two areas then interact in
the ligand design process with deliberate multi-action design in mind. New target
searching will necessarily involve cross disciplinary approaches perhaps in part via
the computer-based platforms like the commercial Discuva platform developed by
Summit Therapeutics Inc. This platform is aimed at the discovery of new bacterial
targets and the development of new antibiotics to attack these targets. The company
used this platform to identify a small molecule antibacterial SMT-571 with a new
mechanism of action (associated with cell division) for drug resistant strains of the
pathogenic bacterium Neisseria gonorrhoeae (Jacobsson et al. 2019).
The use of novel in-cell NMR techniques could herald new opportunities for drug
design and discovery through a direct assessment of detailed protein-drug interactions in living bacterial cells. The NMR structure determination of a heavy metal
binding protein overexpressed in Escherichia coli has been described (Sakakibara
et al. 2009; Ikeya et al. 2016), as well as other work on the NMR-based determination
of high resolution protein structures in living eukaryotic cells, including expressed
proteins originating from bacterial sources such as the Streptococcus protein G B1
immunoglobulin-binding domain (a cell surface protein). The structure of this protein
was the one determined most accurately under these physiological conditions (Tanaka
et al. 2019). Further extensions of this work could give new insights for the design and
assessment of selective antibacterials for example with three separate binding sites
on the one target protein, if concentration and resolution problems can be overcome.
Recently multiple e-pharmacophore modelling (an e-pharmacophore is a
structure-based energy optimised pharmacophore) has been used to identify novel
inhibitors with good binding capability to two active sites on the bacterial 30S ribosomal subunit (Anju et al. 2019). Potentially this could be expanded to three sites
or another target site affecting protein synthesis in addition to these two ribosomal
subunit sites.
It could also be worth pursuing a wider assessment of the neat bioaffinity approach
using native mass spectrometry (MS) methodology, in which the protein assembly
is retained in their native states in the gas phase, as applied by Quinn and collaborators at Griffith University in Queensland, Australia (Vu et al. 2018) for potential
protein targets from the malaria-causing protozoan, Plasmodium falciparum. This
is a very interesting way to locate new hits for these targets. Some 96 natural products of low molecular weight were found as binding partners of 32 potential Plasmodial targets, and 79 of the natural products showed promising growth inhibitory
167
norm thinking (for such biases in chemical synthesis see Jia et al. 2019). AI also has
the potential to go beyond molecular docking for drug mode of action insights when
high quality data sets exist, which hopefully should be applicable in the targeted
antibacterial area (Rodrigues et al. 2018).
5.4 New Modes of Action/New Targets
Together with the search for different molecular structures the search for new bacterial
targets will continue to be of pressing importance. These two areas then interact in
the ligand design process with deliberate multi-action design in mind. New target
searching will necessarily involve cross disciplinary approaches perhaps in part via
the computer-based platforms like the commercial Discuva platform developed by
Summit Therapeutics Inc. This platform is aimed at the discovery of new bacterial
targets and the development of new antibiotics to attack these targets. The company
used this platform to identify a small molecule antibacterial SMT-571 with a new
mechanism of action (associated with cell division) for drug resistant strains of the
pathogenic bacterium Neisseria gonorrhoeae (Jacobsson et al. 2019).
The use of novel in-cell NMR techniques could herald new opportunities for drug
design and discovery through a direct assessment of detailed protein-drug interactions in living bacterial cells. The NMR structure determination of a heavy metal
binding protein overexpressed in Escherichia coli has been described (Sakakibara
et al. 2009; Ikeya et al. 2016), as well as other work on the NMR-based determination
of high resolution protein structures in living eukaryotic cells, including expressed
proteins originating from bacterial sources such as the Streptococcus protein G B1
immunoglobulin-binding domain (a cell surface protein). The structure of this protein
was the one determined most accurately under these physiological conditions (Tanaka
et al. 2019). Further extensions of this work could give new insights for the design and
assessment of selective antibacterials for example with three separate binding sites
on the one target protein, if concentration and resolution problems can be overcome.
Recently multiple e-pharmacophore modelling (an e-pharmacophore is a
structure-based energy optimised pharmacophore) has been used to identify novel
inhibitors with good binding capability to two active sites on the bacterial 30S ribosomal subunit (Anju et al. 2019). Potentially this could be expanded to three sites
or another target site affecting protein synthesis in addition to these two ribosomal
subunit sites.
It could also be worth pursuing a wider assessment of the neat bioaffinity approach
using native mass spectrometry (MS) methodology, in which the protein assembly
is retained in their native states in the gas phase, as applied by Quinn and collaborators at Griffith University in Queensland, Australia (Vu et al. 2018) for potential
protein targets from the malaria-causing protozoan, Plasmodium falciparum. This
is a very interesting way to locate new hits for these targets. Some 96 natural products of low molecular weight were found as binding partners of 32 potential Plasmodial targets, and 79 of the natural products showed promising growth inhibitory