March 6, 2020, by mstalniceanu
Computational Neuroscience: GPCRs and Bioinformatics
In Conversation with Professor Dmitry Veprintsev
Written by María Ángeles Jiménez Sigstad
G-protein coupled receptors (GPCRs) are membrane proteins which are highly relevant due to a large number of drugs that activate them. The classical mechanism of activation of GPCRs is comprised by the binding of agonists, a type of ligand or drug, leading to the activation of heterotrimeric G proteins, which in turn modulates a specific signalling cascade. The implementation of computational methods has become a complement to model what cannot be achieved on experimental conditions yet.
To improve protein structure-based virtual screening, the main interest is in finding fingerprints or “features”. Biased signalling is the selectivity of an agonist to form a GPCR (G protein-coupled receptor) -active state, and activate specific cellular pathways. This is a new concept that has become a trend in drug discovery. Dmitry Veprintsev’s lab at the COMPARE (The Centre of Membrane Proteins and Receptors) group, a joint venture between the University of Nottingham and the University of Birmingham, has been working with Cannabinoid Receptors, a type of GPCR linked to different disorders, and there is a particular interest on them to develop painkiller drugs. Therefore, they are essential for Pain Research. His group studies GPCRs’ signalling via the application of mutations to specific areas related to normal GPCRs’ behaviour. Then they link those results to computational models to generate predictors of biased signalling, for example. To learn more about the main interests of his group and its link with Neuroscience Research, we spoke with Dmitry Veprintsev.
1. What was that made you interested in GPCRs?
GPCRs are fascinating proteins with very complex signalling properties. So, for me, it was the dream for a biophysicist, to study complex molecular machines which are responsible for the mechanism of action of many drugs.
2. How do you think that correlates with neuronal function?
We have many GPCRs in neurons. They are not the only molecules responsible for signalling transduction. However, they are responsible for a lot of the attenuation of signalling. We have many receptors in synapses and in near-synapses that control many receptors in the surrounding cells, like astrocytes, which make sure that neurons work. Mutations in GPCRs were linked to neurological diseases.
3. A specific example?
For example, CB1, Cannabinoid Receptor type 1, is the most expressed receptor in the human body and the human brain. If we do not have enough of it or if we have too much of it, it has severe consequences on normal functioning.
4. Would you correlate it to pain?
There is much research linking Cannabinoid Receptors to pain. Currently, there are no approved drugs to manage pain through cannabinoids, although there are plenty of recreational drugs to do that.
5. Do you think there has been a large cohort of researchers involved in the area, and what do you think about today’s trends in Molecular Neuroscience?
I think this has been active are since the 60s or 70s. Many people have been working on ion channels, back in the day, and GPCRs. Even before people knew what receptors are as biophysical entities, the concept of receptor did exist. There were about two hundred people per day in the field of GPCRs, which half of them were involved in GPCRs’ functional on the brain. Now, we understand how signalling happens in the brain bin depth, and how similar or different GPCRs are from each other. There is still a lot to do. There has been a good drive in the area from when I started in 2010.
6. Do you think that Bioinformaticians or “computational biologists” should come from an integrated background?
I, personally, think that the best people I have met have a diverse background. You have to be able to understand the biological problems to tackle them. On the other hand, you need computational skills.
7. How much do you think that Computational Neuroscience can help GPCRs’ research?
I can see a multi-scale modelling approach, from fundamental biochemical processes over to the whole organism level being developed. Nevertheless, I do not think they have been successful yet. Is a fascinating area of research, how to link the low biochemical reactions in cells to the individual signalling properties of a neuronal network.
8. Any take-home message?
I would like to see more collaboration between the different niches of the sciences. Can we connect to neuronal signalling?
Overall, from this interview, we can conclude that Computational Neuroscience is in demand. Drug discovery must integrate all biological mechanisms, from the internal chemical reactions to the overall neuronal functioning. To create a regulated alternative to recreational drugs, we need to develop a multi-scale analysis of brain function.
No comments yet, fill out a comment to be the first