The axonal connections that link brain regions form a surprisingly regular grid that may someday yield insights into the pathology underlying schizophrenia, autism, and other disorders, according to researchers involved with the Human Connectome Project.
“Now we see there's a level of organization to cerebral connectivity that is learnable and communicable,” said Van J. Wedeen, MD, lead author of an article in the March 30 issue of Science on the geometric structure of brain fiber pathways.
“This knowledge will enable us to make better measurements and describe normal anatomy more systematically. The clinical goal is to be able to ask, does this particular brain fall into the normal range, or does it fit better in a diagnostic category? I think the medical implications of this finding are among the most exciting.”
DIFFUSION SPECTRUM IMAGING
The grid structure of neural connections was revealed through diffusion spectrum imaging (DSI), a refinement of diffusion tensor imaging that uses fMRI technology to infer the presence of axonal fibers from the motion of surrounding water molecules. Although DSI can be performed with any MRI scanner, the new DSI scanner, known as the Siemens 3T Connectom, has a magnetic gradient field four to eight times more powerful than conventional scanners. With it the researchers were able to produce two-millimeter voxels of the living human brain and voxels a half a millimeter across of brains taken from four species of non-human primate — rhesus and owl monkeys, marmosets, and prosimian galagos, also known as bushbabies.
Although DSI reveals fiber pathways in striking detail, linking those pathways to specific functions remains a tantalizing goal being pursued by several labs, according to Dr. Wedeen, professor of radiology at Massachusetts General Hospital and director of Connectomics at the Athinoula A. Martinos Center for Biomedical Imaging at Harvard Medical School, where the new scanner is located. However, certain key pathways whose functions are fairly well known have been have been clearly documented, such as the cingulum bundle, which connects the frontal lobe with the hippocampus, and which is severed surgically in some patients with intractable obsessive-compulsive disorder.
“With our new scanner, we can show connectivity far beyond what can be achieved by the current generation of scanners,” he said. “It pulls 24 megawatts and has a ton of copper in it. It's also the quietest scanner ever built, and we can do 20-minute scans instead of two-hour scans. It's a real engineering achievement.”
Despite the large increase in power, the DSI scanner did not produce any detectible side effects, such as magnetophosphenes — bursts of light produced in some people during an MRI when the magnetic field stimulates the retina or visual cortex.
“Not one of the 50-odd subjects we scanned said they saw flashes of light,” Dr. Wedeen said. “We found no peripheral stimulation either.”
The study was funded in part by the National Institute of Mental Health (NIMH) and the Human Connectome Project, a five-year, $40-million NIH effort funded to map out the structural and functional connectivity of the brain. It is one of two Connectome Project initiatives. The other involves a five-year, $30-million effort that uses a resting-state functional MRI (rs-MRI) to map functional connectivity in the brain.
“That one, which is actually much larger, involves looking at the functional connections of more than 1,000 individuals,” said NIMH Director Thomas R. Insel, MD. “That's being done by groups at Washington University and at the University of Minnesota.”
That effort is led by David Van Essen, PhD, professor of and department head of anatomy and neurobiology at Washington University in St. Louis, and Kamil Ugurbil, PhD, the McKnight Presidential Endowed Chair of Radiology at the University of Minnesota, and director of the Center for Magnetic Resonance Research. Their group also developed a new scanner that doubles the resolution of standard MRI.
Together these initiatives are expected to provide clues to schizophrenia, autism, depression, epilepsy, dementia, post-traumatic stress disorder, and a host of other brain disorders, according to Dr. Insel.
“We think of those as circuitry disorders,” he said. “We also recognize that schizophrenia and autism in particular are probably many disorders, but we haven't had way to separate them out. By looking at the underlying circuitry we'll be able to see subgroups (of such disorders) with greater precision than we can through clinical symptoms or treatment responses.”
The Science paper that describes the grid structure of brain wiring provides valuable insight into communication across the brain, according to Arthur Toga, PhD, who co-directed the effort with Dr. Wedeen and Bruce R. Rosen, MD, PhD, professor in radiology at Harvard Medical School and Director of the Martinos Center.
“This is a great leap forward and an important milestone for connectomics,” said Dr. Toga, professor of neurology, co-director of the Division of Brain Mapping, and Director of the Laboratory of Neuro Imaging at the UCLA School of Medicine. “There clearly are disease processes that interfere with how the brain utilizes those connections, but you first have to understand how they are wired up in normal subjects. From there, one can see how they are miswired or malfunctioning in these disease cohorts. The clinical relevance is clear to me, but we need to improve our fundamental understanding of normal connectomes before we can understand disorders affecting them.”
According to Dr. Toga, a better understanding of neural connections may also improve the diagnosis and treatment of several neurological and psychiatric disorders such as MS and schizophrenia, provide better guidance for surgical interventions that must pass near or through white matter, and illuminate how stroke affects fibers of passage.
“In a number of diseases we look at gray matter atrophy,” he said. “What we don't always do is look at how these gray matter changes relate to changes in white matter. The hope is that the research we're conducting in connectomics will improve studies that don't currently include measurements of white matter and connections between brain regions.”
David Eidelberg, MD, director of the Center for Neurosciences at the Feinstein Institute for Medical Research in Manhasset, NY, called the mapping of the grid structure of neural fibers “a technical tour de force” that verified what people have suspected for a long time.
“What is new is that they have convincingly demonstrated that this (grid structure) is a fixed primate characteristic, a universal feature of the primate brain,” he said. “You can move across species and this architecture is going to be maintained. The size of the grid simply becomes larger and more intricate in higher-order species. This is how the brain organizes to handle traffic to and from different structures.”
But the grid only shows the roads, not the constantly changing flow of traffic, said Cornelia I. Bargmann, PhD, of Rockefeller University, who just published a paper in Bioessays that emphasizes the importance of neuromodulators in shaping the function of neural circuits.
“There's all this excitement about large-scale neural mapping, which is necessary but not sufficient for understanding the brain,” said Dr. Bargmann, Torsten N. Wiesel Professor and Head of the Laboratory of Neural Circuits and Behavior at the Rockefeller University. “I wanted to update ideas about dynamic regulation of these systems.”
Her paper points out that the simple neural circuits of crustaceans, C. elegans, and Drosophila reveal the extent to which modulators add complexity to the processing of information along relatively primitive maps. “For the scientist examining neuronal connectivity,” she writes, “neuromodulators are a hidden challenge.”
Even if it were possible to know all the synapses activated in a given moment, neuromodulators such as hormones would be exerting an influence on the flow of information – redirecting the traffic, so to speak, Dr. Bargmann explained. The 302 neurons of C. elegans, for example, provide ample alternatives for traffic flow because “it is possible to chart a path from virtually any neuron to any other neuron in three synapses,” she noted. And neuromodulators can alter neuronal traffic. Even the simple avoidance of a repulsive odor activates different circuits depending on whether the worm is well fed or hungry.
“Anatomical connectivity is a fixed picture of a very dynamic process,” Dr. Bargmann said. “Functional connectivity shows that while there are strong routes where the traffic will travel very reliably, there's also a lot of modification of brain pathways that makes information flow differently under different circumstances. In stress-induced analgesia, for example, the wave of adrenalin completely blocks the propagation of pain, and for a brief period you can't tell you've been hurt. The stress just sort of snaps the brain into a different mode so you don't even detect the painful stimuli.”