JOIN USSlice by slice, building brain's search engine
As the square container, fixed to a moving platform, inched toward a steel blade mounted level with its surface, the group held its collective breath. The blade peeled off the top layer, rolling it up in slow motion like a slice of pale prosciutto. “Almost there,” someone said.
Off came another layer, another, and another. And then there it was: a pink spot at first, now a smudge, now growing with every slice like spilled rose wine on a cream carpet — a human brain. Not just any brain, either, but the one that had belonged to Henry Molaison, known worldwide as HM, an amnesiac who collaborated on hundreds of studies of memory and died last year at age 82. (Molaison agreed to donate his brain years ago, in consultation with a relative.)
“You can see why everyone’s so nervous,” said Jacopo Annese, an assistant professor of radiology at the University of California, San Diego, as he delicately removed a slice with an artist’s paintbrush and placed it in a labelled tray of saline solution. “I feel like the world is watching over my shoulder.”
And so it was: Thousands logged on to view the procedure via live webcast. The dissection marked a culmination, for one thing, of HM’s remarkable life, and of more than a year of preparation for just this moment, orchestrated by Suzanne Corkin, a memory researcher at the Massachusetts Institute of Technology who had worked with Molaison for the last five decades of his life.
But it was also a beginning of something much larger, Annese and many other scientists hope. “The advent of brain imaging opened up so much,” said Sandra Witelson, a neuroscientist with the Michael G DeGroote School of Medicine at McMaster University in Canada, who manages a bank of 125 brains, including Albert Einstein’s. “But I think in all the excitement people have forgotten how important the anatomical study of brain tissue still is, and this is the sort of project that could really restart interest in this area.”
The University of California project — called the Brain Observatory, set up to accept many donated brains — is an effort to bridge past and future. Brain dissection is a craft that goes back centuries and has helped scientists to understand where functions like language processing and vision are clustered, to compare gray and white matter and cell concentrations across different populations and to understand the damage done in ailments like Alzheimer’s disease and stroke.
Yet there is no single standard for cutting up a brain. Some researchers slice from the crown of the head down, parallel to the plane that runs through the nose and ears; others cut the organ into several chunks, and proceed to section areas of interest. No method is perfect, and any cutting can make it difficult, if not impossible, to reconstruct circuits that connect cells in disparate areas of the brain and somehow create a thinking, feeling mind.
To create as complete a picture as possible, Annese cuts very thin slices — 70 microns each, paper-thin — from the whole brain, roughly parallel with the plane of the forehead, moving from front to back. Perhaps the best-known pioneer of such whole-brain sectioning is Dr Paul Ivan Yakovlev, who built a collection of slices from hundreds of brains now kept at a facility in Washington.
Reproduction
But Annese has something Yakovlev did not: advanced computer technology that tracks and digitally reproduces each slice. An entire brain produces some 2,500 slices, and the amount of information in each one, once microscopic detail is added, will fill about a terabyte of computer storage. Computers at UCSD are now fitting all those pieces together for Molaison’s brain, to create what Annese calls a ‘Google Earthlike search engine’, the first entirely reconstructed, whole-brain atlas available to anyone who wants to log on.
“We’re going to get the kind of resolution, all the way down to the level of single cells, that we have not had widely available before,” said Donna Simmons, a visiting scholar at the Brain Architecture Centre at the University of Southern California. The thin whole-brain slicing “will allow much better opportunities to study the connection between cells, the circuits themselves, which we have so much more to learn about”.
Experts estimate that there are about 50 brain banks in the world, many with organs from medical patients with neurological or psychiatric problems, and some with a stock donated by people without disorders. “Ideally, anyone with the technology could do the same with their own specimens,” Corkin said.
The technical challenges, however, are not trivial. To prepare a brain for dissection, Annese first freezes it in a formaldehyde and sucrose solution, to about minus 40 degrees Celsius. The freezing in the case of HM was done over four hours, a few degrees at a time: The brain, like most things, becomes more brittle when frozen. It can crack.
Molaison lost his ability to form new memories after an operation that removed a slug-size chunk of tissue from deep in each hemisphere of his brain, making it more delicate than most.
“A crack would have been a disaster,” Annese said. It did not happen.
With the help of David Malmberg, a mechanical engineer at UCSD who had designed equipment for use in the Antarctic, the laboratory fashioned a metal collar to keep the suspended brain at just the right temperature. A few degrees too cold and the blade would chatter instead of cutting cleanly; too warm, and the blade wants to dip into the tissue. Malmberg held the temperature steady by pumping ethanol through the collars continually, at minus 40 degrees. He suspended the hoses using surfboard leashes picked up days before the dissection.
After the slicing and storing, a process that took some 53 hours, Annese’s laboratory will soon begin the equally painstaking process of mounting each slice in a glass slide. The lab will stain slides at regular intervals, to illustrate the features of the reconstructed organ. And it plans to provide slides for study. Outside researchers can request samples and use their own methods to stain and analyse the composition of specific high-interest areas.
“For the work I do, looking at which genes are preferentially expressed in different areas of the brain, this will be an enormous resource,” Simmons said.