Life Sciences Pulse

High Performance Computing: Redefining Our Understanding of Inherited Metabolic Disorders

 
June 25, 2018

Truth is often stranger than fiction. This became more apparent to me after I read about the illness of King George III of Britain – a dreadful disease called acute intermittent porphyria (AIP). This not only afflicted the monarch but played a pivotal role in altering the course of English history, portrayed excellently in the film The Madness of King George.

AIP is an inherited metabolic disorder characterised by severe abdominal pain, change in colour of urine, depression, and anxiety. Since it’s a rare disorder, affected patients are usually not aware of the symptoms till these manifest in an extreme form. Treatment includes controlling the symptoms via medication, proper diet, and fluid intake, and there is no permanent cure for the disease till date. This disease is caused by mutations in the enzyme hydroxymethylbilane synthase (HMBS), also known as porphobilinogen deaminase (PBGD). The enzyme catalyses the formation of hydroxymethylbilane from four molecules of porphobilinogen, an important step in the heme biosynthetic pathway. Although it is known that mutations in HMBS lead to AIP, we still do not know how these mutations affect the activity (or lack of activity) of the enzyme.

Decoding Disease Mechanics

In a recent study, published in Proceedings of the National Academy of Sciences of the United States of America (PNAS), we used a computational method called molecular dynamics (MD) simulation to show the role of different residues in the catalytic mechanism of the enzyme, and how mutations in these residues lead to AIP. Previous reports on HMBS had suggested that the catalytic mechanism is complicated: there are as many as six steps!

This particular study revealed details of the interactions occurring between the active site residues, the substrate molecules and active site water molecules during each step of the six-step catalysis process. We found that a total of twenty-seven residues in the active site plays a role in catalysis. Our study was further complemented by our collaborators at the Icahn School of Medicine at Mount Sinai, New York, led by Dr. Robert J. Desnick. They performed mutagenesisand in vitro studies on HMBS, which showed that mutations of the active site residues (identified in the MD simulations) could affect enzyme activity via (1) alteration of the binding site, (2) reduced binding of the enzyme cofactor, (3) decreased catalytic activity, or (4) decreased product release. These experiments further verified the residues identified from the MD simulations as critical to the enzyme’s activity. The findings have certainly shed light on the molecular origin of AIP.

Accelerating Drug Discovery with High Performance Computing

MD simulations can provide a detailed picture of the conformational dynamics of a biological molecule. On the other hand, structures from experimental techniques, such as X-ray crystallography, typically provide only a static picture of biomolecules. Such pictures do not tell us how certain conformations enable molecules to perform their functions. In MD simulations, you compute the forces between all the atoms in a molecular system, and then allow the system to evolve under the influence of these forces. The longer the simulations are performed, the better our understanding of the structure and conformational dynamics. MD simulations demand extensive use of high performance computing (HPC) resources, and typically use HPC clusters that can rapidly run parallel calculations across multiple processors.

MD simulations can lead to a better understanding of not only AIP, but also a number of other crippling inherited metabolic disorders that lack proper treatment strategies. With recent advancements in simulation methods and HPC technology, it will be possible to have a comprehensive understanding of the molecular origins of these disorders. In addition to providing fundamental insights into the molecular mechanisms of diseases, MD simulations can accelerate drug discovery programs. Lead molecules can be identified by estimating binding free energies between ligands and putative drug targets. Furthermore, insights gained into structure-function relationships in drug targets will aid the design of new chemical entities with desirable properties. This will minimise the chances of failure at later stages of drug discovery, and minimize the expenses and time taken.

Siladitya Padhi is a scientist working with the Life Sciences unit of TCS Innovation Labs, Hyderabad since March 2017. Through his research, he aims to find novel solutions for treating diseases by employing structure-based and systems-based computational biology methods. He has a PhD from IIIT Hyderabad.