This article is reposted from the old Wordpress incarnation of Not Exactly Rocket Science. The blog is on holiday until the start of October, when I'll return with fresh material.
At Harvard University, a group of creative scientists have turned the brains of mice into beautiful tangles of colour. By mixing together a palette of fluorescent proteins, they have painted individual neurons with up to 90 different colours. Their technique, dubbed 'Brainbow', gives them an unprecedented vision of how the brain's cells are connected to each other.
The art of looking at neurons had much greyer beginnings. Over a century ago, a Spanish scientist called Santiago Ramón y Cajal, one of the founders of modern neuroscience, became the first person to get a clear look at the neural network that houses our thoughts. He found that neurons stood out among other cells when stained with a silver chromate salt.
These monochrome images told us what neurons were, but made it very difficult to work out how they joined up into a network. It would be like trying to make sense of London's famous tube map if all the lines were coloured with the same dull grey. Nowadays, neuroscientists can 'tag' neurons with fluorescent proteins, but even these are available in only a few shades.
Enter Brainbow, the brain-child of Jean Livet, Jeff Lichtman and colleagues from Harvard. It uses combinations of just four basic fluorescent proteins - which glow in either red, orange, yellow or blue - to paint neurons with a vast range of hues. It works like a TV, which combines red, green and blue light to form the entire colour spectrum.
The instructions for making the four glowing proteins come in four genes. These were crafted into a single 'transgene', which can then be loaded into the brain tissue under investigation.
The set is under the control of the 'Cre/Lox' system, a famous tool for shuffling genes. Livet and Lichtman arranged the four genes in such a way that, after they are shuffled, all four have an equal chance of being turned on, but only one ever is.
When the team tested their system in mice, they saw that individual neurons often took up multiple Brainbow transgenes, each one producing a fluorescent protein independently of the others.
These random combinations are the key to Brainbow's visual diversity. For example, a neuron with two blue proteins and a red one would look purple, while another neuron with two red proteins and one blue one would have a more magenta shade.
By eye, Livet and Lichtman managed to identify 89 distinct colours and the finer senses of a computer pinned down an even large palette of 166. The resulting images look like a cross between Fauvism, pointillism and a psychedelic acid trip. It's as far advanced over Ramon y Cajal's silver staining as modern high-definition televisions are over the grainy black-and-white sets of the past.
Livet and Lichtman also ascertained that neurons retain their colour over time, and are uniformly shaded across its entire length and throughout its many branches. That opens up a whole realm of possibilities for neuroscientists.
They can determine how a tangle of nerve cells connect with each other (and their supporting cells) on the basis of colour alone, and the team have already done this with over 400 cells in a small region of mouse brain. Brainbow's stability means that it can also be used to create colourful time-lapse videos of neural networks over time, to see how they change in response to new experiences or genetic switches.
Obviously, Brainbow is merely a tool, just like genome sequencing and other technologies of the modern biological revolution. Scientists will still need to interpret and understand what they see, but seeing anything in the first place is a massively important first step. And if what you see is colourful and pretty, that can't hurt either.
More on fluorescence:
Reference: Livet, Weissman, Kang, Draft, Bennis, Sanes & Lichtman. 2007. Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature doi:10.1038/nature06293.