Only connect: The American Association for the Advancement of Science: mapping the brain

The first of four reports from the AAAS’s annual meeting looks at how brains are wired up
BIOLOGISTS used to suffer from physics envy. It was the physicists, not them, who got the big bucks for big science. That changed with the Human Genome Project, which spent about $3 billion over a decade reading (though not always understanding) the complete sequence of the genetic “letters” of the DNA that describes how to make and run a human being. The genome project, however, came to an end ten years ago. Ever since then, ambitious biologists have been looking for the Next Big Thing. And rumour has it that the NBT is about to be announced. In March the American government is expected to launch an attempt to solve biology’s most mysterious problem: how the brain works.
The Brain Activity Map, as this project seems likely to be called, will study how the brain is wired up at all levels, from the connections between individual nerve cells to the neuronal superhighways between its various lobes and ganglia. In so doing it will institutionalise the emerging science of connectomics—which was one of the centrepieces of this year’s meeting of the American Association for the Advancement of Science, held in Boston from February 14th-18th.
The connectomists, if they may be so called, are attempting to complete a project begun in 1543 when Andreas Vesalius published a book called “De humani corporis fabrica” (“On the fabric of the human body”), which was the first modern treatise on anatomy. The science of anatomy, with its mantra that form follows function, has illuminated understanding of every organ but the brain. Now, the connectomists hope, this last bastion will fall, and a comprehensive atlas of neuroanatomy will lead to a real knowledge of mental processes and mental disease.
Life in thin slices
Jeff Lichtman, of Harvard University, works at the lowest anatomical level of all, that of the links between individual nerve cells. He is using an industrial approach to build exquisite three-dimensional maps of how such cells connect to each other in the brains of mice. To do so, he and his colleagues embed brains in plastic and put them in a machine that cuts them into microscopically thin slices. The slices are automatically processed and stored on circular plates in a library. Tiny areas of these brains can then be photographed by an electron microscope, and clever software follows each cell from one slice to the next, to reconstruct, as the picture above shows, how those cells connect to each other.
The picture is of a piece of tissue only 100 microns across. Yet it required 2,000 slices to produce and about a terabyte of data to store, and it contains parts of thousands of nerve cells (only a few dozen of which are shown, in order to avoid confusion). A brain contains hundreds of billions. And that, in a nutshell, is the measure of the problem of brain anatomy. Yet you have to start somewhere, and Dr Lichtman’s hope is that when he has looked at enough small blocks of nerve cells, patterns will emerge that will cause someone to say “aha!”, and produce a testable theory of what is going on.
A comprehensive theory of how brains work will, however, require an understanding of their higher levels of organisation, as well. And the top end of that scale, where anatomical structures are measured in centimetres rather than microns, is the province of the Human Connectome Project (HCP). This, as its name suggests, was set up as a stepchild of the Human Genome Project—and if the rumours are correct, its fairy godmother is about to arrive and the $35m that it has to spend over the course of five years will soon be multiplied, albeit under a different name.
At the moment, the HCP’s largesse is split between two groups, one led by Harvard and one by Washington University, in St Louis. Steven Petersen, who works on the St Louis side of things, explained to the meeting what they were all up to.
Unlike Dr Lichtman’s project, the HCP studies still-living creatures, so rather than chopping brains up it uses the most up-to-date forms of scanning technology, including a technique that can follow the passage of water molecules around the brain and another that observes correlations between the metabolic activity of different brain areas, to map otherwise invisible pathways between various nerve centres.
The brains Dr Petersen and his colleagues are scanning belong to 1,200 volunteers, including 300 pairs of twins (some identical, some not), who each come in for a two-day assault course of questionnaires, cognitive tests and sessions lying in the scanner. This way, the HCP researchers hope to establish which features of brain architecture are common to all, how much they vary from person to person, how they relate to someone’s skills and behaviour, and (via the twins studies) how genetic variation changes them.
Co-ordinated action
The metabolic-activity approach has been particularly fruitful. Marcus Raichle, one of Dr Petersen’s colleagues in St Louis, explained to the meeting how it has revealed many previously invisible networks in the brain. These are formed of places all over the organ where metabolic activity ebbs and flows synchronously, even when the person in the scanner has been told to lie still and not to think about any particular task. And the discovery of these networks is not merely of academic interest. As William Seeley of the University of California, San Francisco, explained, they may also be the key to understanding some nasty neurological diseases.
Hitherto, doctors have found it convenient to distinguish between disorders of the mind, such as schizophrenia and clinical depression, which leave no obvious anatomical trace, and disorders of the brain, such as Parkinson’s and Alzheimer’s diseases, which do. But this is surely a false distinction; it is merely that the anatomical traces of psychiatric disorders have not yet been found—perhaps because they are actually caused by misconnections, known as “connectopathies” in the jargon, that current techniques are not clever enough to recognise. One of the aims of connectomics is to find these connectopathies. Conversely, it is not clear how Alzheimer’s and other dementias whose physical traces can easily be seen with existing techniques actually spread through the brain. But Dr Seeley thinks an important part of the answer has now been found, for in Alzheimer’s and four similar but rarer dementias the pattern of spread seems to match the networks of co-ordinated metabolism described by Dr Raichle. Somehow, the links that co-ordinate activity are also spreading the disease.
Exactly what is going on is unclear. But this discovery may be the clue needed to work out how to stop dementia in its tracks. If it is—given the burden that an ageing population threatens to impose on many countries over the next few decades—connectomics will have proved its worth even before the big bucks turn up.


0 yorum: