Small in size, but with a large array of uses, microfluidic droplets could become the reaction vessels of choice for a significant fraction of biological research say Yolanda Schaerli and Florian Hollfelder of the University of Cambridge, UK 

Contemporary biology increasingly demands high-throughput experiments. Only these can provide the vast amounts of information needed to study cell populations, and DNA, protein and small molecule libraries appropriately. Practically, such a format should be highly economical, consuming minimal amounts of reagent and operating at the lowest cost per assay, which means running experiments on the microscale.

Compartmentalisation provides a system to address these requirements. This uses a stream of water-in-oil emulsion droplets as an extremely small equivalent of an array of test tubes: droplets can have femto to nanolitre volumes. Up to 10¹⁰ droplet reactors will fit into a millilitre and an equivalent number of experiments can be carried out simultaneously. 

Water-in-oil emulsion droplets are easily made by mixing oil and an aqueous phase using a stirrer, homogeniser or extruder. But researchers wanting the best quality droplets can use a microfluidic device to create drops of more uniform size, which allow the most precise quantitative experiments. Devices have recently been built that can generate up to 10 000 highly mono di sperse aqueous droplets per second (typically 10-200 micrometres in diameter). 

The microfluidic format can be used not only to form droplets but also to process them in a variety of ways. Droplets can be divided, fused, incubated, analysed, sorted and broken up. Integrating and automating these steps could potentially lead to systems for biological experimentation with a level of control akin to experiments on the macroscopic scale. 

A droplet compartment is not only small, but it can also be smart. It can combine the function of a molecule (such as the catalytic activity of an enzyme), with information on its identity (for example the DNA sequence encoding the protein) and a readout to assess how well the molecule performs its function. Thus, the droplet contains everything needed to assess and decode a particular experiment or profile a library member. 

The droplet principle is used in various applications. The most commercially successful use of compartmentalisation is in the emulsion polymerase chain reaction (ePCR), which uses a polymerase enzyme to amplify DNA held within droplets. For ePCR compartmentalisation provides monoclonality: a complex mixture can be divided into droplets containing a single DNA template. The DNA in each droplet is thus amplified bias-free, whereas in a mixture one DNA strand might be amplified to a greater extent than another. The result is a PCR reaction suitable for a number of applications, most importantly high-throughput DNA sequencing. 

Droplets can also be used to link genotype (DNA or RNA) and phenotype (an observable trait, such as binding or catalytic activity) just like a cell. In directed evolution a gene library is diluted so that, as in ePCR, each droplet contains no more than one copy of DNA. Genes are transcribed and translated to yield proteins of which improved variants are selected via a procedure tailored to their characteristic trait, such as turnover of a substrate to yield a fluorescent product. This phenotype-genotype linkage is essential to mimic natural selections in the laboratory to create proteins or nucleic acids with improved or new functions. 

But cells themselves can also be compartmentalised in droplets. By holding back any substance released by a captured cell, cell-based directed evolution experiments relying on detecting product formation have become possible. Additionally, being able to trap cells with additional external stimuli paves the way for studies into the mechanisms that control a cell's response to its environment at the single cell level. Resolving a mixture of cells into individuals will also allow access to other information that has been unavailable from conventional experiments with cell populations. 

Microfluidic droplets are becoming well-established tools in the lab, in approaches such as ePCR, for example. Directed evolution and cell-based assays are now in advanced stages of development and proof-of-principle experiments are appearing for a whole range of applications from diagnostics, cellomics, proteomics, to drug discovery and synthetic biology. Extrapolation of these approaches towards more highly integrated systems may change the way biological experiments are designed and carried out.

Read more in the review 'The potential of microfluidic water-in-oil droplets in experimental biology' in a themed issue on Computational and Systems Biology in Molecular BioSystems.