By its very nature, neuroscience tends to be interdisciplinary. To study the nervous system, especially the brain, takes data from imaging and the environment, anatomy and bioinformatics, as well as traditional genetic and behavioral animal models. But neuroscience is also the study of diseases — of how the otherwise-elegant nervous system fails. Today, many researchers are beginning to focus on gene expression to better understand diseases like Parkinson’s, which remain so out of reach that for many of them, the only way to truly diagnose the presence of a particular illness is by studying brain samples collected after a patient has died.
“When it comes right down to it, they’re just hard diseases to figure out,” says Michael Hauser, from Duke University’s Center for Human Genetics. “They are complicated because there’s genetic influence, there’s environmental influence, and unraveling the mechanisms of these is very difficult.”
The heavy hitters of neuroscience — Alzheimer’s, Huntington’s, Parkinson’s, and schizophrenia — affect millions of people. Even though the diseases are vastly different, neuroscientists in each of these fields are asking the same basic question: is there a difference in gene expression in the brain of someone with a neurological disease compared to someone without it? The approaches to answer this question vary as much as the diseases do.
This variety of methods and technologies is, in part, based on the disease under scrutiny and how global or localized a glimpse the researcher wants into the disease. Studying wider expression patterns can narrow the search for a disease gene or help pinpoint when a harmful mutant protein first crops up. More locally, a disease might preferentially prey on specific neurons while leaving others alone; understanding why certain cells survive could be a boon for treatment.
Searching for a Parkinson’s Gene
Michael Hauser, an associate research professor, takes a global approach to his search for the Parkinson’s disease gene. The disease is also a movement disorder, so often the first symptom of the disease is a tremor or another problem moving. Currently, the only way to definitively diagnose someone with Parkinson’s disease, like many other diseases of the brain, is a pathological diagnosis of a post-mortem brain. Hauser, though, is on the lookout for the Parkinson’s gene that could potentially be used as a diagnostic in living people.
Studies using family-based linkage analysis and case-controlled studies have given researchers an idea of regions on several chromosomes that might be involved in Parkinson’s disease. Using gene expression analysis, Hauser says, will narrow those hundred of possible genes down to a more manageable number.
To whittle away the possibilities, Hauser looks at gene expression in parts of the brain where the effects of Parkinson’s disease are most notable, especially the substantia nigra. Then he matches the genes expressed in those areas back to the suspected chromosomal region to check for a Parkinson’s disease susceptibility gene. “That was our logic. We could greatly reduce the number of genes we wanted to explore [and] evaluate by doing this kind of expression analysis,” he says. At the same time, gene expression analysis would systematize the diagnosis of Parkinson’s disease instead of relying upon histological methods, Hauser adds.
The Origins of Huntington’s
Ruth Lüthi-Carter, an assistant professor at the Federal Institute of Technology in Lausanne, Switzerland, also uses animal and post-mortem human brains in her work on huntingtin. Mutant huntingtin, the protein that leads to the development of Huntington’s disease, affects the brain globally. If Lüthi-Carter can understand how to trace the protein back to its first mutant roots, then the progression of Huntington’s disease to abnormal movements and eventually death could conceivably be understood and targeted with drugs. “If we can identify the earliest changes, hopefully that will be informative about the cascade through which the mutant protein is causing large and massive brain dysfunction,” Lüthi-Carter says. That brain dysfunction is what leads to neuronal death.
Though Lüthi-Carter uses both animal models and human brain samples in her research, post-mortem samples, she says, only give a slim and late-stage view into the Huntington’s disease. But by combining these late-stage genomic assays with animal disease models, she can cross-compare which genes are expressed. “We’re then able to trace back the effective mutant huntingtin to the very earliest changes,” Lüthi-Carter says.
Though the gene for Huntington’s disease is known, most Huntington’s disease drugs only treat the symptoms. If those early protein changes are found, then “you could hope to intervene in the disease at an early stage and prevent further damage,” Lüthi-Carter says.
Symptoms of Schizophrenia
Károly Mirnics also takes a region-specific approach to studying schizophrenia. As a neurological and psychiatric disease, schizophrenia is multi-faceted and its symptoms range from hallucinations to disordered thinking to delusions. Mirnics, an associate professor at Vanderbilt University’s Brain Institute, says that schizophrenia differs just as much at the molecular level as at the clinical level. Still, he wants to get to the root of schizophrenia’s symptoms.
Mirnics, who was one of the pioneers in applying microarray technology to human brain studies, studies gene expression patterns in post-mortem human brains as well as in animal models of schizophrenia.
Mirnics takes samples from parts of the brain known to be involved in schizophrenia, such as the prefrontal cortical area. Then, using DNA microarrays, he analyzes the gene expression patterns. “We try to find what is consistent and what is going on,” he says.
After looking at the genes common to human and mouse samples, Mirnics then correlates the overlapping genes. One gene involved in schizophrenia is BDNF, or brain-derived neurotrophic factor. By looking at BDNF knockout mice, Mirnics found more genes common to both the animal model and people with schizophrenia. Those new genes, Mirnics says, may be part of the cascade of genes in the human brain that are involved in the development of schizophrenia. “I think we are in very exciting times, from gene expression to environmental insult to transgenic animals. We are getting slowly there,” Mirnics says.
Singling out Alzheimer’s
Stephen Ginsberg takes a different approach in his study of Alzheimer’s disease. Instead of the global analysis of Lüthi-Carter or the regional assays of Hauser or Mirnics, Ginsberg looks at single cells.
In brains with Alzheimer’s disease, neurons can develop neurofibrillary tangles or just die, leading to memory loss. But some neurons are more susceptible to these changes than others. CA1 pyramidal neurons in the hippocampus are wracked by tangles while the related CA3 neurons are relatively unscathed. “I just really want to know what makes those two cell types different,” says Ginsberg, an assistant professor at New York University’s School of Medicine. To do that, he looks at the cell types and assays their gene expression individually.
In his lab, Ginsberg compares the post-mortem expression profiles of people with Alzheimer’s disease, people without it, and people with mild cognitive impairment. Through single-cell analysis, Ginsberg and his lab members are finding changes in many different genes involved with the basic cell biology of the disease, including signaling markers as well as markers for endosomes and lysosomes. They have found that certain nerve growth receptors are down-regulated in Alzheimer’s disease and that their down-regulation correlates to cognitive data from when the person was alive. “We’re seeing very vivid shifts in the expression profile,” Ginsberg says. Once scientists know what makes one set of cells vulnerable and the other safe, Ginsberg says, that lead will be the basis for pharmacogenomics research aimed at diagnosing and perhaps treating the disease.