Scaffold Hopping for Drug Discovery

Designing new drugs is expensive. Using virtual drug design software to produce novel medicines circumvents some of the considerable costs of developing drugs using traditional approaches.

Many potential new drugs fail to progress through rigorous regulation requirements and clinical trials. The recent steady decline in the number of new drugs making it to market has resulted in use of non-traditional routes to discovery.

A lot of compounds can be synthesised and then screened, but it is a very labour intensive and expensive process. The more economical way – in terms of cost and time – to do this is virtually on a computer. The aim is to take compounds and make virtual molecules. Then to treat them in ways that approximate reality, and recreate in a virtual format the physiological state.

Using a chemical approach and exploiting virtual computational techniques – scaffold hopping – helps reduce expensive pitfalls. It offers the potential to develop novel compounds, different to those identified using traditional approaches.

Ligand Approach

Virtual drug discovery exploits biological action as a driver of investigation. This involves looking at biologically relevant receptor-ligand interactions. The starting point for the development of a new drug is screening for similar biological activity to that of an existing medication. Then the structure is manipulated to improve certain specific properties of the molecule, such as cost of production, improved action, limiting side-effects and increased bioavailability.

The ‘lock and key’ approach leads to making a virtual protein model and using it to try and find out what specific ligands fit inside it. Ligands only have a certain number of conformations, and molecules usually adopt a structure that is low energy, so this helps.

When assessing a protein-ligand interaction, a docking programme computationally reinserts the molecule over and over again in slightly different conformations. The big challenge is if there is a big hole and small molecules, so that they fit in lots of different directions and conformations. Then it is hard to find the orientation that fits best.

Various modifications are made to these compounds to improve efficacy as long as ligand efficiency is preserved.

Chemical Structure Approach

Looking specifically at the chemical structure is an alternative approach to computational drug discovery. This involves using pharmacophores – molecular elements of the compound that makes it active. A minimum number of structural chemical features are required in three-dimensional space to have activity.

Using molecular scaffolds to create molecules that look completely different involves mining drug space containing molecules with biological activity for novel pharmacophores that might make new drugs. Candidate molecules may sit outside of the traditional small medicinal chemistry space.

Pharmacophores are selected from large compound libraries, and then those with undesirable physicochemical properties or moieties are eliminated. The researchers filter through the virtual molecular space to look for subsets of molecules with similar properties. They might have the same kind of polarity, a similar ionisation potential, make a certain hydrogen bond, or have similar molecular weight. After filtering the number of pharmacophores, a number of end candidate drugs are tested.

Combination Approach

It is best to combine use of both structure-based and ligand-based approaches.  The two pathways are different, but the interphase may include a collection of similar molecules that are putatively suggested to be active. This will lead to the accumulation of a group of potential candidates for future research both in vitro and, if successful, in vivo studies.

Looking at Cancer

The ideal cancer drug is a compound available as a once a day oral tablet. The drug needs to obey Lipinski’s rule of 5s – a set of physical properties that govern absorption, distribution, metabolism and excretion. This means the candidate drug will have a molecular weight of less than 500, be lipophilic, and have a limited number of H-bond donors or acceptors.

When a candidate drug is identified and tested in vitro and found to behave as predicted from a computational point of view and this translates into efficacy at a cellular level, the next step is to test it in vivo first in a non-human model and then in humans.

Computational approaches are valid for developing new drugs and an effective alternative to traditional methods. It is target and biology informed.


No comments yet... Be the first to leave a reply!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: