Gels are being formed on-chip for easier and cheaper protein separation. Amy Herr, at the University of California, Berkeley, US, and colleagues designed the miniature customisable gels to sort proteins by size. 

Protein mixtures are often separated on slabs of polyacrylamide gel. By applying a potential difference across the gel, the negatively-charged proteins travel through the pores in the gel towards the anode - with larger proteins having more difficulty moving through the gel pores and so taking longer to migrate. For higher resolution protein sizing, particularly in proteomics, biologists can use gels with pore-size gradients. Now Herr and co-workers have made these gradient gels reproducibly inside a chip. 

The researchers used two solutions of different acrylamide concentration to create the gradient. Using a high concentration solution they fill a channel in a microfluidic device and shine light on one end to photopolymerise the acrylamide, plugging the end with a small pore-size gel. They then flush the other end of the channel with a low concentration solution, which will form a larger pore-size gel. They allow the two solutions to diffuse into each other and a second exposure to light results in a reproducible pore-size gradient gel.

Gradient gels allow different sized proteins to be separated, with larger proteins taking longer to migrate through the gel

This is the first time such gels have been prepared on-chip and the team sees the method as a means to optimise chip systems for biomolecule separation. 'The gels' planar geometry makes integrating them with sample processing, analysis, and collection achievable,' says Herr, 'something that is possible, but can be cumbersome, in capillary systems.' 

Herr's team can separate protein mixtures using gels as short as 0.3cm and can change both the gel length and pore-size gradient to customise the separation for different mixtures. The ultra-short channel lengths also reduce the electric potential required for the separation. 'This eliminates the need for high voltage power supplies - making the system more amenable to portability and non-laboratory settings,' say the scientists.

Rustem Ismagilov, an expert on microfluidics and proteomics from the University of Chicago in the US, is impressed with the work. 'This is a good idea, and has a wide range of potential applications in proteomics,' he says. 

Freya Mearns