Scientists Use Neuroimaging, Genetic Studies to Probe Biology of Tourette Syndrome

Cleveland—In a darkened conference hall, Kirk Frey, MD, PhD, nodded toward the screen showing a slide listing all the published studies on neurochemical changes in the brain related to Tourette syndrome (TS). The list did not take long to read—it contained just 5 citations.

"This is the literature in its entirety," said Frey, director of nuclear medicine at the University of Michigan Medical Center, Ann Arbor.

Despite being named and discovered more than a century ago by French neurologist Georges Gilles de la Tourette, the disorder, characterized by verbal and motor tics, remains vastly understudied and underdiagnosed. Getting at the root cause of TS has been a tough proposition for researchers. They have not been able to glean much from autopsy and postmortem data because those studies have included small numbers of individuals, and the neurochemical changes reported have been subtle. Those data also are confounded by the fact that they often come from patients who had more than one neurobehavioral diagnosis and took medications that could cause changes in the brain.

On the living, breathing side of TS, research has been difficult because the disorder primarily affects children whose developmental changes and comorbid diagnoses—TS has been linked with obsessive-compulsive disorder, attention-deficit/hyperactivity disorder, anxiety, and depression—confound study results.

The variable nature of TS symptoms also makes it a complicated neuropsychiatric puzzle. Motor tics can range from simple twitches to rapid-fire jerks of the head or jumping and twirling about. Vocal tics might be as mild as throat clearing or sniffing, or as severe as coprolalia, the sudden outburst of obscene words.

Tourette syndrome differs from other movement disorders like chorea and dystonia in that tics are stereotyped—the same ones repeat over and over. Patients' tics also wax and wane repeatedly over time, with the intensity of symptoms typically peaking around age 11 years.

Despite the challenges, Frey and other researchers who met here in June for the Fourth International Scientific Symposium on TS are hammering out a strategy. They hope that new methods in magnetic resonance imaging (MRI) of the brain coupled with recent and emerging data from genetic linkage studies of affected families will produce an endophenotype for TS. (Endophenotypes are disease characteristics that cannot be seen by the naked eye but are more specific for a particular condition than the phenotype.)

Endophenotypes act as a bridge between genotype and phenotype. In complex conditions such as TS that most likely involve multiple genes, identification of an endophenotype can help researchers zero in on the actual genes responsible for the condition.

"What we're proposing is that perhaps subtle brain shape distinctions might actually be the endophenotype that we're looking for," said Frey.

The Tourette Syndrome Association (TSA), headquartered in Bayside, NY, which sponsored the symposium along with the Cleveland Clinic in Ohio, supports consortia of international researchers whose work is focused on these neuroimaging and genetics studies. Now in its third year, the TSA Neuroimaging Consortium is organizing an effort to build a data set of magnetic resonance brain scans from approximately 600 individuals with TS. Researchers will obtain the brain scans from some 300 sibling pairs who already have documented phenotypes and genotypes from their participation in the TSA International Consortium for Genetics' family studies.

Since the early 1990s, family studies have established that first-degree relatives of individuals with TS have approximately a 10% risk of also having the disorder. In addition, first-degree relatives have about a 15% risk of chronic tic disorder, which is believed to be a milder form of TS, and a 10% to 15% risk of obsessive-compulsive behavior. The studies have also shown that TS affects about 3 or 4 times as many males as females.

To date, family studies have provided details of TS inheritance patterns, and segregation analyses indicate that a number of genes—some that appear recessive and others dominant—control the expression of TS symptoms. But the search for candidate genes has been disappointing. "We haven't hit any home runs," said David Pauls, PhD, professor of psychiatry and director of the Psychiatric and Neurodevelopmental Genetics Unit at Harvard Medical School, in Boston.

Now further analyses are coming up to bat. Pauls said that in the past 5 years, several studies have begun to implicate some of the same chromosomes. His work with colleagues in the TSA genetics consortium suggests that chromosome 8 may be involved in TS (Am J Hum Genet. 1999;65:1428-1436). So does a recent study of 91 Afrikaner families in South Africa (Am J Med Genet. 2001;105:163-167). Additional studies have implicated chromosomes 5 and 11.

Pauls and colleagues currently are analyzing data from a second genome scan for 240 more sibling pairs in their ongoing family studies. One of the potential pitfalls, he said, is that if phenotypes have not been described accurately enough in the families, genetic analyses will not be able to pinpoint the exact genes involved in TS. "Probably the most critical component of any genetic study is to get our phenotypes right," he said.

That is where the neuroimaging consortium and its search for an endophenotype come in. Reanalysis of gene-specific data from the families, with neuroimaging data factored in, provides more power in the hunt for the gene or genes responsible for TS. Once the neuroimaging group has completed MR scanning of the sibling pairs—still in the early stages—it hopes to use innovative brain mapping techniques to determine whether differences in brain structures in individuals with TS have any link with a genetic influence for the condition, said Frey.

Frey said brain mapping techniques developed by the International Consortium on Brain Mapping (ICBM), based at the University of California at Los Angeles (UCLA), bring precision to the analysis of MR images of brain size and shape. Even meticulous analyses can be open to some interpretation; the highly skilled experts who draw the boundaries of the brain structures under study may vary in how they define those boundaries. The ICBM, said Frey, has developed a "toolbox" of approaches that produce an objective MR image of brain structures and also determine what type of tissue is contained within a structure—gray or white matter or cerebrospinal fluid.

"One can get an individually specific, high-resolution objective map without having subjective decisions about the boundaries of brain structure," said Frey.

Researchers at UCLA have used these 3-dimensional maps to show how genes and brain structure may be linked. One controlled study of twins with schizophrenia showed that monozygotic twins and affected dizygotic twins had smaller hippocampal volumes than unaffected dizygotic twins and controls, a finding that suggests a genetic link between hippocampal morphology and schizophrenia (Neurobiol Dis. 2002;11:83-95).

By using more precise brain mapping techniques to analyze MR scans of TS patients, the neuroimaging group hopes to answer lingering questions about the structure and function of the basal ganglia, which have been identified as the major cerebral players in TS.

In the past decade or two, clues from animal studies, basic human physiology and anatomy, and clinical observations have pointed to malfunctions in the basal ganglia as a starting point in the brain for TS.

"Normal basal ganglia function is complex, but one of those functions appears to include the inhibition of unwanted movements," said Jonathan Mink, MD, PhD, of the University of Rochester Medical Center. The basal ganglia also appear to play a role in emotional and cognitive functions.

Roger Albin, MD, professor of neurology at the University of Michigan, noted some additional signs that point to the involvement of certain brain regions in TS. He said tics respond to treatment with dopamine antagonists and that obsessive-compulsive behaviors, which have been linked with TS, respond to selective serotonin reuptake inhibitors. The fact that another TS comorbid condition, ADHD, responds to stimulants provides another dopaminergic clue.

"All of this clinical pharmacology suggests involvement of regions of the brain involving serotonin and dopamine, so we automatically think of the striatum," said Albin. The striatum is the part of the basal ganglia comprising the caudate, putamen, and nucleus accumbens. Research by Albin and others has indicated that available dopamine is greater in the striatum of people with TS, and that this may change how the striatal circuitry is regulated.

Interest also has focused on whether the volume of these brain structures relates to TS. Research in the early 1990s showed a reduction in basal ganglia volume in individuals with TS (Neurology. 1993;43:941-949). Attempts to replicate the finding have been inconsistent, but a more recent study has shown an average volume reduction of 5% to 8% in the caudate nucleus of individuals with TS (Arch Gen Psychiatry. 2003;60:415-424).

The more recent study, says Frey, "is the kernel of evidence for reduced striatal volume" as a possible explanation for at least some of the symptoms seen in patients with TS.

Another indicator not only of striatal involvement in TS but of its predictive value for future tic severity was described in a poster presentation at the symposium. In a study by researchers at Yale University School of Medicine in New Haven, Conn, and Columbia University College of Physicians and Surgeons, in New York City, 43 children with TS underwent MRI before they reached age 14 years. Clinical examinations after the children turned 16 years showed that reduced caudate nucleus volume on the original MRI accurately predicted more severe tics and obsessive-compulsive symptoms later.

As the neuroimaging consortium delves further into its work on MRI scans of sibling pairs, Frey hopes the group's results eventually will resolve questions about any relationship of the volume of the basal ganglia and striatum to TS. They will try to replicate the previous studies that showed such links as part of their search for an endophenotype for TS. Their results may go a long way toward unraveling this neuropsychiatric mystery.