Computing and cone snail create new painkiller

In this image, the cone snail, Conus consors, uses its harpoon tooth to capture and disable its prey before eating it. It injects venom in the fish that contains a cocktail of toxins called conotoxins, which target the nerve centers, bringing about a partial or total paralysis of the prey. Researchers are studying a protein within the venom called XEP-018, which shows great promise for pain control and anesthetic drugs for human brain disorders. Image courtesy EGI.

The toxic venom of the predatory underwater cone snail may help create better treatments for a rare and incurable brain disorder called dystonia.

This neurological disorder, which affects about 1% of the US population, is characterized by involuntary muscle contractions that cause slow repetitive movements or abnormal postures in its victims.

Disrupting pain pathways

Researchers at Utrecht University, in the Netherlands, are using advanced computing to understand how a useful painkiller can be re-engineered from a small protein, called XEP-018, discovered within the toxin of the paralyzing hypodermic-like sting of the Conus consors.

“All senses in our body are transmitted to and from the brain via neurons. The venom of the cone snail has peptides that can disrupt this circuitry,” said Henry Hocking, a researcher at Utrecht University. “The peptides do this by attaching themselves to the openings in neuron communication channels. This is sort of like a plug. Once attached, no signal can be transmitted to the brain and you stop feeling pain.”

Hocking is part of the CONCO project, an EU initiative to discover and develop new therapeutic molecules from the venomous marine cone snail species.

Mapping molecules with magnets

To understand how XEP-018 binds to a communication channel, Hocking needed to know the protein’s shape. He opted to recreate its 3-D structure using a technique called Nuclear Magnetic Resonance.

“NMR is a technique that people might know from hospitals, where Magnetic Resonance Imaging scanners are used. People are put into large magnetic fields and pictures are made of them. In NMR, we put protein molecules inside and bombard them with electromagnetic waves. Instead of making pictures, we measure distances between atoms. If you know all the distances between the atoms, you can try to reconstruct a 3-D object of the protein,” said Alexandre Bonvin, a computational structural biologist at Utrecht University.

From data to 3-D

A computer generated image showing a model of a neuron communication channel being blocked by the cone snail's toxin.

But, NMR doesn’t reveal the actual 3-D structure of these peptides. Computations are needed to transform NMR data into a 3-D protein structure.

Bonvin and his colleagues perform NMR analysis on the grid by combining gLite middleware with the WeNMR e-infrastructure, a European Commission FP7-funded multinational project within the European Grid Infrastructure.

“In order to calculate the 3-D structures of proteins, we have to repeat the process many times. We have to make thousands or tens of thousands of calculations,” said Bonvin.

To speed up this process, the work is split into small packages and sent out on the grid.

“You get your answer within a couple of hours. WeNMR, as a whole, had a submission volume of about 1.5 million jobs last year, corresponding to over 850 CPU years,” said Bonvin.

To expand their computational resources, Bonvin and his colleagues are now investigating if a dedicated desktop grid could be deployed within Utrecht University, to make good use of all PCs around campus.

Bonvin first presented this research in his keynote speech at the International Symposium on Grids and Clouds in Taipei on the 29 February 2012.

Now, research on XEP-018 is almost at the clinical trial stage.

“Many analogues have been designed and the product is currently in pre-clinical development for the treatment of dystonia. Our ultimate goal is to avoid injections and to develop a drug that everybody can use. Using specific devices, such as patches to facilitate the penetration of the peptide through the skin, we believe that XEP-018 has great chances of success,” said Reto Stöcklin, the scientific team leader for CONCO.

Bonvin and researchers within the CONCO project are also working on other therapies, ranging from DNA repair linked to HIV infection, neurodegenerative disease, and cancer.

Said Stöcklin, “We recently discovered cell-penetrating peptides that selectively enter some cancer cells without affecting other (healthy) cells or organs. This has led to the creation of a new company, Vector Lifesciences, that will develop these for improved cancer therapy.”

This work has wider implications for the structural biology community.

“The WeNMR program is a poster child for inter-laboratory coorperation and sharing of computational methods. This ensures that users access the most current versions of the software and minimizes the need for users to maintain the software themselves,” said Guy Montelione, a molecular biology and biochemistry researcher at Rutgers, The State University of New Jersey.

He said, “It also provides global and often instantaneous feedback to software developers, allowing them to improve the value of the software to the broad community of users. Some of the most important work in the field of structural biology is now being done using this infrastructure.  I believe that this model should be strongly supported by funding agencies, as it has high impact and high efficiency.”