HPC-Derived Affinity Enhancement of Antiviral Drugs

Project reference: 2222

Pharmaceutical intervention is always a compromise between unwanted side effects and anticipated interference with a pathological condition. A particular class of well tolerated drugs is formed by small peptides because their building blocks are simply the 20 naturally occurring amino acids – the most fundamental constitutents of any living cell. In order to transform such peptidic compounds into effective medical drugs high affinity binding to the target must be accomplished (and is usually the outcome of a tedious and long-lasting optimization process).

In a recent study the peptide TAT-I24 made of 22 amino acids was shown to exhibit strong antiviral activity  against double-stranded DNA viruses, e.g. herpes simplex, cytomegalovirus, adenovirus type 5, SV40 polyomavirus and vaccinia virus.  (doi:10.3390/biologics1010003). At the molecular level, direct binding of the peptide to viral ds-DNA could be  observed. In addition, computer simulation studies of  the binding affinity

between peptide and ds-DNA were found to agree very well with experimentally derived affinity constants. Consequently, a straightforward next step in further increasing drug efficacy was to change any of the 22 amino acids and study the effect in terms of improved or deteriorated binding strength. Doing so repeatedly for various positions in the 22mer can help to identify new peptidic variants with significantly enhanced binding affinity. Moreover,  key residues can be discovered, both with respect to sequence position as well as amino acid type. As the theoretical number of such variants is 20²² an exhaustive search will be entirely out of reach and a cleverly devised “mutation strategy” will have to be developed. In any case, the more such a screening process can be outsourced to some in-silico approach, the more cost-effective will be the entire drug discovery process.

Regarding methods, the MM-PBSA technique can be applied (doi:10.1021/ar000033j). This is a compute-demanding approach that allows semi-quantitative predictions of binding free energies, ∆G, using standard tools in contemporary biophysical/biochemical research. One major enabling factor has been the recent boost in molecular dynamics simulations (MD) faciltated by the availability and standard employment of GPUs.

This SoHPC project aims to carry out HPC based computational screening for novel variants of the TAT-I24 peptide with improved binding affinity to ds-DNA. All previously established protocols are readily available and can serve as a template. However, no prior domain knowledge is required and interested participants can easily join from scratch. SoHPC fellows will  (i) get acquainted with fundamentals in structural biology,  (ii) master basic handling of HPC workloads,  (iii) obtain familiarity with AMBER, a package of widespread use in the biophysics/biochemistry/computational biology community,  (iv) design, test, optimize, verify, run, evaluate GPU workloads on HPC architectures,  (v) critically analyze results and advise on progressive modifications of the TAT-I24 peptide to achieve affinity enhancement. Ideally, one would obtain a novel compound with significantly improved binding free energy to the target ds-DNA. However, equally important were identification and characterization of key-positions in the TAT-I24 peptide.

Structural representation of the complex between ds-DNA (purple) and the TAT-I24 peptide (white, magenta) as revealed from MM/PBSA analysis with derived binding free energy of -36 kcal/mol.

Project Mentor: Siegfried Hoefinger

Project Co-mentor :Markus Hickel

Site Co-ordinator: Claudia Blaas-Schenner

Learning Outcomes:
Routine operation and interaction with HPC environments. A broader understanding and appreciation of the impact HPC can have on contemporary science and research.

 Student Prerequisites (compulsory):
Just a positive attitude towards HPC for scientific applications and the readiness for critical and analytical thinking.

Student Prerequisites (desirable):
Familiarity with Linux, basic understanding of formal methods and their translation into scientific applications, basic scripting skills, experience with GPUs;

Training Materials:
Public domain materials and some web information about AMBER and MM/PBSA;

Workplan:

• Week 1: Basic HPC training; familiarization with local HPC system;
• Week 2: Introduction into basic concepts; AMBER tutorials;
• Week 3: Workplan formulation;
• Weeks 4-7: Actual project implementation, verification and re-evaluation of previous workloads, adaptations, fine-tuning and upscaling;
• Week 8: Write up a final report and submit it;

Final Product Description:
Ideally we get a novel variant of the TAT-I24 peptide with significantly enhanced binding affinity and clear-cut knowledge of key positions in the sequence, but even more important were SoHPC fellows having gained good experience with practical work in HPC environments.

Adapting the Project: Increasing the Difficulty:
Making acceptable affinity gains more stringent or becoming more selective in terms of key-residue specifications;

Adapting the Project: Decreasing the Difficulty:
Various optional subtasks can either be dropped or carried out in greater detail.

Resources:

Basic access to the local HPC infrastructure (including various GPU architectures) will be granted. AMBER-20 is available on VSC systems. All previously established compute protocols can readily be used.

Organisation:
VSC Research Center, TU Wien