Gene Defects Discovered that Illuminate Development of Brain and Heart

Genetic errors in a human HOX gene derail normal development, triggering flawed communication between the brain and its sensory outposts and unexpected malformations of the cardiovascular system. In the October issue of Nature Genetics, Elizabeth Engle, HMS associate professor of neurology at Children’s Hospital Boston, and her colleagues identify homozygous HOXA1 mutations in a pair of human syndromes distinguished by defective eye movement. This work—the first example of heritable HOX mutations that disrupt human brain development—provides a view into the molecular genetics of these conditions and may also steer the course of basic developmental studies of the brain, heart, and vasculature. “These human mutations raise interesting questions for scientists in the HOX field,” Engle said.

Courtesy Elizabeth Engle

Blood vessels redrawn. Abnormalities in the head vasculature are among the features recently found in individuals with HOXA1 mutations. Magnetic resonance angiograms show that humans typically have a pair of carotid arteries (left), which are the chief conduits for blood traveling to the brain, but that both are missing in an affected individual (right). Neighboring vessels, such as the basilar artery (open arrow) and the posterior communicating arteries (white arrows), swell to offset this vascular deficiency.

Accumulating evidence suggests the root of inherited gaze disorders lies in faulty cranial nerves, the wires that join the brainstem to the eye muscles and other muscles and to sensory organs in the head. Engle seeks to unearth the genetic culprits that fray these connections, resulting in hindered movement or sensation. While pursuing this goal, her research team identified individuals with deafness and mental impairment, in addition to abnormal eye movement. Along with nerve defects, their imaging studies revealed malformations of the inner ear, the head vasculature, and the outflow tract of the heart. Genetic linkage analyses implicated a region that encompasses the HOXA family, one of four clusters of multiple HOX genes. Previous reports demonstrated that targeted deletion of mouse Hoxa1 caused defects remarkably similar to those seen in the human cohorts, which helped Engle’s team pinpoint a candidate gene within this cluster. Sequencing of patient DNA samples at the HOXA1 locus uncovered single nucleotide changes in both alleles that are predicted to prematurely shorten the protein and render it nonfunctional.

HOX genes, whose genomic structure and patterns of expression have been painstakingly preserved within the animal kingdom, play a fundamental role in conveying spatial information to axial tissues like the brain during embryonic stages. Now the challenge is to understand how wrinkles in these coordinates can have such far-reaching effects on brain function. This study suggests that “the brainstem is really playing a much larger role, not just in breathing or controlling facial muscles, but perhaps also in guiding the development of the higher brain,” said lead author Max Tischfield, a neuroscience graduate student. Another mystery that remains to be solved is how mutations in HOXA1 lead to the cardiovascular defects seen in affected individuals, since such a mechanism has not yet been identified in mice or humans.

—Nicole Davis

First Rodent Model of Schizophrenia Mimics Human Brain Changes

Schizophrenia research has added a powerful tool to its toolbox. Scientists from HMS and McLean Hospital have created the first animal model of schizophrenia that closely mimics the brain changes seen in human patients. The study was led by Barbara Gisabella, a research fellow in the laboratory of Francine Benes, HMS professor of psychiatry (neuroscience) at McLean Hospital, and appears in the Sept. 13 Proceedings of the National Academy of Sciences.

Postmortem research in schizophrenia has been hampered by the lack of a model system for validating and extending research results. “There has never been a model that has grown out of postmortem studies because schizophrenic brains do not display the clear histopathological changes seen with diseases like Alzheimer’s or Huntington’s,” said Benes. To compensate for this gap, Benes and others have relied on complicated postmortem microscopic analyses of human brains.

Such analyses eventually revealed subtle changes in the schizophrenic brain, especially in the hippocampus, which receives input from the amygdala to produce emotion-related learning. Another group had reported in the 1980s that the uptake of GABA, which inhibits neuronal activity, is significantly decreased in the amygdala of schizophrenic patients. Consequently, the Benes group hypothesized that deficiencies in GABA-related activity in the amygdala may allow for an abnormally increased flow of excitatory signals to the hippocampus in schizophrenic patients. “We know that patients with schizophrenia are often frightened by daily events…. This overactivity of fear-based responses is consistent with our hypothesis,” said Benes.

Their hypothesis was supported by Benes’s previous observation that sectors CA3 and CA2 of the hippocampus, which receive abundant projections from the amygdala, show a selective decrease of neurons involved in GABA-related inhibitory activity in schizophrenics.

To test their idea, the researchers created a rodent model in which picrotoxin, an antagonist of the GABA receptor, was infused into the amygdala of living rats. After 96 hours, hippocampal slices were obtained. Patch clamp recordings of single neurons showed that inhibitory GABA currents were decreased in sectors CA3 and CA2, but not CA1, when compared with controls. This distinctive pattern closely mimicked the one that the Benes’s lab had observed in postmortem studies of human schizophrenic brains.

“Now we have a rat model that we can manipulate at will to help us interpret changes in postmortem schizophrenic brains. This will be a great help to us in linking subtle changes in neural circuitry to functional abnormalities at the cellular and molecular levels,” said Benes. “This is a big step forward for schizophrenia research.”

—Jillian Lokere

National Health Data Network Would Require Billions More in Federal Investment

To wire the nation’s health care system so that medical data is on computers instead of paper and the computers can talk to each other without error, the country would need to invest $156 billion in hardware, software, and training in the next five years, and then pay out about $48 billion a year in annual operating costs.

This capital investment, estimated by an expert panel, is roughly equivalent to the extra money the country will spend in the next two years as health care costs rise, report Rainu Kaushal, HMS instructor in medicine at Brigham and Women’s Hospital, and her colleagues in the Aug. 2 Annals of Internal Medicine. “Seamless interchange of medical information across the nation is extremely useful, as demonstrated by the victims of hurricane Katrina, whose medical information is now inaccessible,” Kaushal said, referring to media reports of doctors needing to guess at medication dosages or facing the prospect of re-immunizing children.

“ Although the price tag appears enormous, the national health information network (NHIN) would increase the nation’s health care budget by a small percentage annually,” writes Peter Basch, medical director of MedStar e-Health in Washington, D.C., in an accompanying editorial. “Many would argue that the NHIN is affordable at almost any cost since the aggregate cost of errors and unnecessary duplication possibly exceeds Kaushal and colleagues’ estimate.”

On its current trajectory, the country will spend a total of about $24 billion over the next five years, a fraction of the cost of an NHIN, Kaushal and her co-authors report.

Federal investment will be the key to achieve the goals mentioned by President George W. Bush in the 2004 State of the Union address: “By computerizing health records, we can avoid dangerous medical mistakes, reduce costs, and improve care.”

So far, despite its clinical potential, computerized information is used most effectively for submitting claims electronically and checking the eligibility of patients for care, Kaushal and her colleagues say in a related paper in the September/October 2005 Health Affairs.

Large hospitals are adopting the health information technology needed to link into an NHIN, but smaller stakeholders are lagging behind. Incentives and standards, would ensure widespread equitable adoption, the researchers conclude.

—Carol Cruzan Morton

top