Feature: Building better drugs

This feature appeared in the July/August 2011 issue of Australian Life Scientist. To subscribe to the magazine, go here.

Finely tuned biological signalling systems are instrumental to the functional activities in and amongst our cells on a scale of complexity that makes the London tube map look like a chidls’s jigsaw puzzle.

Like the tube stations, cell surface receptors that transform messages outside the cell into functional outputs are an integral and often incompletely understood part of the total complexity.

Associate Professor Kevin Pfleger is focusing on one such receptor type – G protein-coupled receptors (GPCRs) – with the hope of simplifying that map. Or at least finding some of the tunnels. Pfleger is head of Molecular Endocrinology – GPCRs at the Western Australian Institute for Medical Research (WAIMR) and the Centre for Medical Research, University of Western Australia (UWA).

The GPCR receptor family is involved in numerous diseases and, indeed, are the target of up to half of all currently produced drugs. In studying these signalling ‘gateways,’ Pfleger hopes that his work will ultimately improve our understanding of drug responses, leading to better drugs and fewer side effects for patients.

As well as his academic research role, Pfleger is also involved in commercialising his research, seeing this as one of the ways to realise his research goals. He is the Chief Scientific Officer of Dimerix, a biotech company spun out of WAIMR/UWA in 2004.

“Dimerix gives us an interface with the pharmaceutical industry so that the work we do here at the bench can be translated, in time, to the patient,” he says. Pfleger started working with the company around 2006 having co-invented, with colleagues at WAIMR, the current platform technology to investigate interactions of GPCRs, with multiple granted and pending patent applications arising from the project so far.

HIT technology

This technology is called GPCR Heteromer Identification Technology, or GPCR-HIT, and its main function is to better characterise GPCR heteromers and the particular signalling pathways activated through them, profiling both known and novel receptor combinations.

The aim of the technology is enabling the development of safer and more effective drugs for targeting GPCRs. GPCR-HIT has already generated a lot of interest, particularly since the group published the technology details this year in Assay and Drug Development Technologies, where it scored the front cover. The article is the most read original article for the journal in the last 12 months, and it only came out in February.

The novelty of GPCR-HIT lies in the assay configuration. As with many assays that measure the interactions of signalling partners in a pathway, GPCR-HIT is a proximity-based technology.

---PB---

Such approaches work on the principle of tagging or labelling the components of interest in such a way that the reporter signal only occurs when the targets are in close enough proximity to represent a biochemical association, like during the formation of a signalling protein complex.

“There are a number of proximity-based assays around, and our technology can use any of them, but we tend to use one called bioluminescence resonance energy transfer, or BRET, because that is the area of my expertise,” says Pfleger.

Indeed, Pfleger and his colleagues at WAIMR/UWA are considered world leaders in BRET technology, publishing several major technological developments in recent years.

BRET involves linking the proteins of interest to a bioluminescent donor enzyme, luciferase, or acceptor, fluorophore. Complex formation is detected following energy transfer between these reporter molecules, resulting in a measurable signal from the acceptor fluorophore.

A substantial range of protein-protein interactions can be readily monitored in real time using BRET, which is actually based on a naturally-occurring phenomenon perfected by some sea creatures such as jellyfish.

Getting the lowdown

Conventionally, assays to investigate protein-protein interactions involve putting half of the reporter component system on one protein of interest, which in Pfleger’s case is one of the receptors, and the other half on the second protein of interest. Then the signal is measured to signify the level of interaction.

“The problem with that,” says Pfleger, “is that you don’t know if you have a specific signal or whether you have just overexpressed too much protein. Consequently, people can end up doing some quite laborious and complicated assays to demonstrate specificity.

“So, we have actually taken a step back and looked at it from a drug discovery point of view, and asked how could we make it more straightforward and ligand-dependent to increase specificity.”

What the team came up with is actually deceptively simple: in essence they tag one receptor, for instance with a donor for BRET, and leave the other receptor or partner untagged. They then put the acceptor part of the reporter system on something else that is recruited specifically to the untagged receptor when it is activated, like beta-arrestin or a G protein.

“The idea is if you add an agonist for the untagged receptor, it will recruit the tagged arrestin or G protein, but no real interaction between the receptors means no signal because the two reporter tags are not coming together.

---PB---

Whereas, if a complex forms involving the two receptors, the recruited arrestin or G protein with the acceptor tag will now be close enough to the donor-tagged receptor to produce a signal.

“By doing that we get ligand-dependent reporting of the receptor complex. In other words, we will only detect a signal when the receptors of interest are in the heteromer complex. This gets around the specificity problems to a large extent and cleans up the overall signal enormously. And compound screening becomes a lot more straightforward.”

Back in his own non-commercial lab, Pfleger focuses more on the beta-arrestin side of the GPCR signalling story. “We are interested in a number of receptors with regard to beta-arrestin recruitment, including the orexin, vasopressin and angiotensin receptors.”

The recently discovered orexin family of GPCRs (OxRs) are an interesting example. The orexin system was characterised about 12 years ago with the discovery of two peptide ligands in the central nervous system, orexin A and orexin B.

Orexin signalling has been implicated in a number of disorders ranging from narcolepsy to addiction, but much remains unknown about the ins and outs of their signalling complexes and complexity compared with some other GPCR systems.

In the brain, orexin ligands have specificity for two G protein-coupled receptors that display significant protein sequence similarity, orexin receptors 1 and 2 (OxR1 and OxR2).

“This is interesting because evolution doesn’t tend to maintain two receptors in the body just for the hell of it. There needs to be a reason why both evolved and continue to occur together, often in the same tissues. So, using a specialised approach called eBRET (extended BRET), we have been able to tease apart different and novel profiles for these two particular receptors.

“This approach, which we pioneered and published on five years ago, enables the analysis of long timescales. This gives us more context and texture from these assays than looking just at single time points.

“For example, it might relate more to function. Consequently, we are now pulling out kinetic profiles and dose-response relationships for our specific receptor systems, seeing dynamic information that provides us with more in-depth pharmacological profiling and differences that just wouldn’t be seen with other technologies.”

In functional terms, the team is trying to decipher why this particular orexin system uses two receptors and what that may mean for a drug response. Also what are the molecular aspects governing the particular profiles of the two receptors?

Pfleger refers to some current unpublished work as an example of where they are heading. “There are particular sites in the two receptor structures that may influence phosphorylation and therefore arrestin recruitment, and looking at these sites we might have actually worked out at the molecular level why these receptors are different. Now, we need to tie those differences back to receptor function.”

---PB---

Biased communication

Another focus for Pfleger’s group at the moment is based around the concept of biased signalling. It was thought not that long ago that a cell receptor, regardless of which agonist binds to it, would set off the same signalling cascade, and therefore downstream function in a cell.

However, it is now understood that depending on which agonist is bound, the receptor can actually assume different structural shapes or conformations that can each potentially elicit a different signalling output, so: a biased response.

“In extreme cases, such biased signalling can result in completely different functional outputs from the one receptor type,” explains Pfleger. “So again with our technology approach, the use of BRET and our very sensitive arrestin-recruitment assay, we can actually tease out a lot of this bias.

“For example, there is a lot of interest at the moment in differentiating G protein-mediated signalling from beta-arrestin-mediated signalling and looking for compounds that do one thing or another in these different signalling contexts. This is a case of biased signalling.”

Distinguishing one pathway from another is important for the pharmacological field because arrestin is involved not only in desensitisation of G protein signalling, but activating distinct signalling in its own right.

“So, if you get a certain agonist that activates G proteins, but that doesn’t recruit arrestin you may get overstimulation of G protein signalling (that is not desensitised), without activation of arrestin-mediated signalling. Vice versa, if you get activation of arrestin without the G protein involved you may only get arrestin-mediated signalling.

All these nuances have relevance pharmacologically to drugs like morphine, for example, which act differently in different situations and desensitisation is a big issue. Getting to the bottom of why and how this happens could have some important clinical outcomes.”

In an ideal world, drugs would go in and turn on some signals, turn off others, and leave the rest alone, rather than the all-or-nothing effect seen with many compounds now used to target GPCR systems. Besides working more effectively, the ideal drug would also avoid the common and often serious side effects.

“Many of these side effects come from the drug affecting signalling pathways that are not the intended target. So if we can develop compounds that don’t do that, suddenly you have a much better drug.

“The analogy I always use is that we want these compounds to play the piano, whereas a lot of the drugs in the market are just hitting the keyboard with a hammer, while others are just shutting the lid.”

The ‘holy grail’ for scientists like Pfleger is working out the biology of these systems such that they could design or select the compounds with the desired pharmacological effect. “I guess we are working towards that at two levels: n an academic sense principally with the arrestin work; and at the commercial level by looking more directly at the heteromer complexes.”

Having a foot in each of these camps suits Pfleger. “I like the innovation and discovery aspect of science and this interface between academia and biotech is fabulous for that. Particularly now when big pharma are outsourcing a lot of their innovation to small biotechs like ours.It is a very exciting place to be.”