Developing AI-Based Pipelines for Transporter Hijack and Inhibition

How do nutrients and drugs get into cells? For anything that is taken in orally (though the mouth), the chances are that it will get moved into a cell via a special kind of protein that lives in the membrane called a transporter. For example, protein from say a steak you might have eaten first gets broken down in the stomach into smaller units known as di- and tri-peptides. As these peptides work their way through the gut, they are recognized by peptide transporters

located in the membranes of the cells that line the gut wall. Similar transporters exist in the kidneys to maximise the re-uptake of amino acids that might otherwise escape in the urine. This sounds straight-forward, but in fact we don't understand how these transporters, which are membrane proteins, actually work beyond the level of quite basic cartoons.

Given there are 20 types of amino acid, the combinations of pairs (dipeptides) or triplets (tri-peptides) is vast - in fact we know about 8000 different di and tri-petides can be transported in this way. One aspect that is particularly intriguing is how these transporters recognize the different peptides that they transport and at the same time exclude other small molecules.

Understanding that would be extremely useful because it also turns out that it is possible to make drugs "look a bit like peptides" such that they can fool the transporter into taking them up into the human body. In other words, the drugs can be designed as a molecular Trojan Horse. However, we don't yet understand the rules to modify drugs to look like peptides, but if we could then we can make existing antibiotics much more effective because one of the current problems is that antibiotics don't actually get to their site of action in the concentrations that would be required to make them effective (this is known as bioavailability).

The pH of the gut is acidic, meaning there is a very high concentration of protons outside the cell compared to inside the cell. This sets up a chemical gradient (of protons) across the membrane, which is analogous to a battery - in this case the protons will want to move into the cell and this can be used to help provide the energy required to move the much larger peptides.

At present we don't know how that works either. Knowledge of that would also help us to how to understand how to modify antibiotics as well.

In this proposal, we are building on our previous work on the structural of the peptide transporter to develop and test and new state-of-the-art computational and experimental pipeline that will allow us to modify existing antibiotics to improve their uptake into cells and hence potentially, the bioavailability. Our team is comprised of world-leaders in computational methods, structural biology and chemical synthesis and by developing this pipeline we will not only be able to design modified antibiotics but we will have a level of understanding of these transporters that means we will be able to apply the pipeline in the future to develop inhibitors of other transporters in the body that may be associated with certain diseases.