Bristol University uses Oracle Cloud Infrastructure to speed up smoking cessation drug discovery

Scientists at the University of Bristol have published research showing how nicotine affects receptors in the brain as part of an effort to design drugs that will help smokers to quit.

They have done so using Oracle Cloud Infrastructure donated by the supplier and in collaboration with Achieve Life Sciences, a Seattle-based pharmaceutical company focused on the commercialisation of Cytisinicline, a plant-based alkaloid with a high binding affinity to the nicotinic acetylcholine receptor in the human brain.

According to the US National Institute on Drug Abuse, the majority of smokers would like to stop smoking, and each year around half try to quit permanently. Yet, only about 6% of smokers are able to quit in any given year. Smoking is the second most common cause of death worldwide.

The paper that issued from the Bristol research, A general mechanism for signal propagation in the nicotinic acetylcholine receptor family, was published in the Journal of the American Chemical Society in December 2019.

Two of the authors are from Oracle’s Cloud Development Centre in Bristol, Phil Bates and Gerardo Viedma Nunez.

Adrian Mulholland from the University of Bristol’s Centre for Computational Chemistry was co-lead author on the paper, along with Richard Sessions, senior research fellow at the School of Biochemistry at Bristol.

Mulholland told Computer Weekly: “Our work shows how nicotine exerts its effects on nicotinic acetylcholine receptors. Understanding this should help us design better smoking cessation aids.”

The study, led by led University of Bristol scientists but involving academics from other institutions, used Oracle’s cloud infrastructure. The researchers used new computational simulation methods to conduct 450 assessments of the biochemistry associated with the binding of nicotine to a subtype of nicotinic acetylcholine receptors, a mechanism believed to be responsible for the highly addictive nature of the drug.

“Each simulation takes eight hours to run on a single cloud node”, said Mulholland. “If we had used our own high-performance computing facility, it would have taken 90 days to do what we did in five.

“We are lucky at Bristol to have pretty good HPC resources, but what the Oracle Cloud enabled us to do was to run a new class of simulation – ‘non-equilibrium’ simulations, of which there are hundreds that have to be done in parallel. The Oracle Cloud enabled us to run them in a matter of weeks, whereas it would otherwise have taken us a year.

“To understand why nicotine is so addictive, and to develop molecules to help people quit smoking, we need to understand how nicotine affects the nervous system. By harnessing the power of cloud computing, we can quickly observe how nicotine exerts its effects at the molecular level. This information can inform future drug development of new treatments for companies like Achieve.”

According to a press statement from Achieve Life Sciences, Oracle and Bristol, the university and the pharmaceutical firm have teamed up to “formulate molecules and potential treatments to combat addiction and neurological disorders based on … smoking cessation compound in development, cytisinicline”.

Cytisinicline is, according to the statement, a “plant-based alkaloid with a high binding affinity to the nicotinic acetylcholine receptor. It is believed to aid in smoking cessation by interacting with nicotine receptors in the brain by reducing the severity of nicotine withdrawal symptoms and by reducing the reward and satisfaction associated with smoking.”

The drug has been approved in Central and Eastern Europe for more than two decades, and has been used by “more than 20 million people”, according to the press statement.

The paper is one output of research originally funded by the EPSRC in 2016, with £724,000.

Mulholland said the beauty of being able to use cloud computing for this sort of scientific research lies in its capacity to enable collaboration. “I’m a great believer in different sorts of scientists working together to get the best results. And that’s not about computation in its own right, but as part of a product development programme,” he said. 

“It’s helping to inform what sort of molecules people might make to test as potential medicines. Being able to do the computational simulations fast enough so that scientists can design and adapt their experiments quickly should accelerate drug development. We couldn’t have done this two years ago.”

The work brought together computational chemists, biochemists and research software engineers, working together to deploy the simulations of nicotine receptors.

The computer simulations methodology used in this particular area of neuroscience could also, said Mulholland, be applied to the study of schizophrenia and Alzheimer’s.