Drug discovery on a chip

Scientists in the US have, for the first time, used microfluidics to discover drug leads. The team’s lab-on-a-chip device revealed inhibitors of a key membrane-bound protein in hepatitis C virus (HCV).

Stephen Quake, Jeffrey Glenn and colleagues at Stanford University investigated a protein called NS4B - which plays an essential role in the virus’s ability to cause infection - as a potential drug target. The team first used their microfluidic device to confirm that NS4B binds RNA as the virus replicates. They then screened a library of 1280 small molecules to identify inhibitors of this protein-RNA interaction.

’Hepatitis C is a big problem, because treatment requires a cocktail of drugs - like HIV - but we don’t have many good treatments, so we need to increase the repertoire of drug classes,’ says Glenn. The group had already identified NS4B as a potential new target, but the protein had evaded further investigation because it is so difficult to study in the lab. It spans the intracellular membrane of the virus several times, and without the membrane it doesn’t fold correctly. 

But Glenn and his team devised a way to make just enough of the protein to be tested. ’We’ve found that if you synthesise NS4B in vitro with microsomal membranes, we provide the conditions for it to fold properly - but the problem is that this process yields only small amounts of the protein,’ he says. ’We were lucky to have Stephen Quake across the hall - a world expert with a technique ideally suited to studying small amounts of protein.’

Having used microfluidics to confirm NS4B’s role in binding RNA, the team decided to take the process one step further and use it to screen for small molecule inhibitors - the first time the technology had been used to screen for drug leads. ’One of the hits - clemizole - is very potent in vivo, with significant antiviral activity and no toxicity, and has been used in humans before,’ adds Glenn. The team plan to use the device to further optimise the compound, and also to investigate a variety of other potential HCV targets.

Deformable device

Quake’s device consists of a flow layer of microchannels, where reaction and analysis takes place, which is separated from an overlayed control layer of channels by a flexible membrane. The control channel forms a ’valve’ in the flow channel wherever the two cross - the valve is closed by increasing the pressure in the control channel - forming a series of up to 2400 unit cells across the chip surface.

Each unit cell also features a button membrane, which when activated covers a circular section of the surface. The team attached labelled NS4B to the surface, then flowed labelled RNA into each cell. After activating the button membrane to trap the protein-RNA complex, unbound molecules are washed away - and the ratio of bound RNA to protein is measured by analysing the fluorescent signal from each.

’Quake has pioneered the use of deformable elastomer materials in microfluidics, which allows you to create lots of valves, and so isolated chambers, in a highly integrated device,’ says Andrew deMello, who develops microfluidic devices at Imperial College, London, UK. While many model studies have shown the potential of such devices, real applications are still rare, he adds. ’This is a lovely example of using microfluidics - which can give many parallel measurements with a very small amount of fluid - as a tool to answer a specific biological question, quickly and efficiently. I would expect this technology will be used increasingly in drug discovery.’

James Mitchell Crow

Enjoy this story? Spread the word using the ’tools’ menu on the left.