Share this article on:

AT THE BENCH: Erik D. Roberson, MD, PhD On Learning and Memory Formation, and the Role of Tau in Alzheimer's and FTD

Hiscott, Rebecca

doi: 10.1097/01.NT.0000473459.52706.1d
Back to Top | Article Outline




Erik D. Roberson, MD, PhD, the recipient of the American Neurological Association's 2015 Derek Denny-Brown Young Neurological Scholar Award in Basic Science, discusses the personal and professional influences and events that shaped his career.

In September, Erik D. Roberson, MD, PhD, an associate professor of neurology and neurobiology and the Virginia B. Spencer professor of neuroscience at the University of Alabama at Birmingham (UAB) and co-director of UAB's Center for Neurodegeneration and Experimental Therapeutics, received the 2015 Derek Denny-Brown Young Neurological Scholar Award in Basic Science, along with Annapurna Poduri, MD, MPH, at the annual meeting of the American Neurological Association in Chicago. The award recognizes neurologists and neuroscientists in the first 10 years of their career who have made outstanding clinical and scientific advances toward the prevention, diagnosis, treatment, and cure of neurologic diseases. Dr. Roberson spoke with Neurology Today about the mentors and events that shaped his career.

Dr. Roberson nearly abandoned his dream of becoming a scientist in high school after a summer spent working in a research lab — not because the work was too demanding, but because it was too easy. The project, which involved determining why glial cells were beneficial for growing cultured neurons, was simply handed to him. “I had a kit to do an assay, I did it, and it worked,” he told Neurology Today. “It was too clean, too ‘cookbook.’”

A few years later, as an undergraduate student majoring in molecular biology at Princeton, Roberson found himself without a summer job. So he called up Robert E. Fellows, MD, PhD, then chair of the University of Iowa's department of physiology and biophysics, where he'd spent that unfulfilling summer in high school. Dr. Fellows invited him back to the research program. “That summer, nothing worked,” Dr. Roberson recalled. “We got no data. It was a new project and we didn't know how to do it. We made a lot of progress, though, and that whole process of trying to figure things out really captured me.”

Roberson had no particular interest in neuroscience or neurologic disease when he began his medical training at Baylor College of Medicine in Houston. “I was going there to do virology, oncogenes, growth control, cellular senescence,” he said. But then he attended a lecture by J. David Sweatt, PhD, then an assistant professor in the department of neuroscience. The professor discussed his work inducing “learning and memory in a brain slice, then grinding it up and figuring out the biochemical changes underlying learning and memory,” Roberson said. “I had no idea you could even do that.” The young scientist was intrigued.

He applied to do his thesis in Dr. Sweatt's lab, investigating the biochemistry of long-term potentiation in the hippocampus and the role of protein kinases in memory formation. (Dr. Sweatt, who is now professor and chairman of the department of neurobiology in the McKnight Brain Institute at UAB, recommended Dr. Roberson for the ANA award.)

Back to Top | Article Outline


Dr. Roberson's early work in Dr. Sweatt's lab sought to unravel the molecular mechanisms underpinning learning and memory formation in the hippocampus. “One hypothesis was that [an event or stimulus] activated a protein kinase, which can then lead to a self-sustaining signal if the kinase phosphorylates itself and leads to its own activation,” he said. “The other main hypothesis was that the activation of the kinase would lead to changes in gene expression, and that you'd get a sustained change in gene expression in the nucleus in the cell.”

He and his fellow student in the lab, Joey English, MD, PhD, began investigating the MAP kinase cascade because it was linked to both of these processes, Dr. Roberson said. They showed that several other kinases that are activated during learning and memory cross-talk with the MAP kinase cascade to change gene expression through activation of cAMP response element-binding (CREB) protein. Dr. Roberson and Dr. English shared first author credit on the study that reported these findings in the Journal of Neuroscience in 1999. The paper has been cited nearly 400 times since.

“We now know that changes in gene expression are critical for long-term memory,” Dr. Roberson said. “If you don't have new protein synthesis and changes in gene expression, memory can't be sustained for more than a few hours.”

Back to Top | Article Outline


In 2003, after completing a neurology residency at the University of California, San Francisco (UCSF), Dr. Roberson began a fellowship in behavioral neurology at the university, working jointly with Bruce L. Miller, MD, FAAN, and Lennart Mucke, MD, of the Gladstone Institute.

Dr. Roberson admired Dr. Mucke's work in amyloid beta (Abeta) and amyloid precursor protein (APP) mouse models of Alzheimer's disease (AD). “I wanted to learn from him and apply that in FTD [frontotemporal dementia] models,” he said. “And of course that meant working with tau, because at that point the only FTD gene we had was tau.”

While he waited for the FTD mice to be imported into Gladstone and bred to an appropriate age, Dr. Roberson undertook what he initially thought of as a side project to pass the time: looking at the role of tau in mouse models of amyloid-beta toxicity. “Nobody had looked at the role of tau in these mice. There had been a paper suggesting that tau played an important role downstream of Abeta in cultured neuron models of Abeta toxicity, and we thought, ‘Let's see if that really happens in vivo.’”

So Dr. Roberson and his colleagues crossed Dr. Mucke's APP mice with tau knockout mice. And “lo and behold, [we found that] if you take out the tau, the mice do much better” in terms of memory function, even though amyloid plaque load remained unchanged.

This work, published in Science in 2007, has been credited with sparking the research community's interest in the role of tau in the pathogenesis of Alzheimer's disease. Importantly, it also suggested that reducing tau levels or function might be an effective therapeutic approach in AD, without needing to target tau phosphorylation or aggregation.

Back to Top | Article Outline


Dr. Roberson moved to UAB in 2008, where he continued to research tau's role in AD and FTD.

“We've finally gotten around to some of those FTD experiments I had wanted to do at the outset,” he said, explaining that his lab uses mouse models with the genetic mutations that cause tau overexpression in humans, leading to FTD.

Dr. Roberson and his colleagues continued to investigate why eliminating or lowering tau was beneficial in mice and how that might be captured therapeutically. They found that “tau knockout mice and tau heterozygous mice that have 50 percent of tau are relatively resistant to seizures induced by a variety of agents” — including genetic forms of epilepsy. This led to the understanding that tau plays an important role in regulating neuronal excitability, and that abnormal excitability in the hippocampus likely contributes to cognitive dysfunction in Alzheimer's.

Back to Top | Article Outline


Dr. Roberson's work with patients once a week at the UAB Memory Disorders Clinic continues to underscore the importance of doing translational work, he said. He has been focusing on the interaction of tau with FYN kinase, which he believes may be central to the pathogenesis of AD.

“There's a lot of evidence implicating FYN in Alzheimer's disease and in learning and memory, and also in excitability,” he said. “If you knock out FYN, you get an effect that's similar in some ways to the effect of knocking out tau in terms of resistance to Abeta.” He and his colleagues have started a drug discovery project to identify compounds that can inhibit the interaction between tau and FYN. So far, they have screened 100,000 compounds, identifying several that have attractive chemical properties for further development. The lab recently received funding to proceed to the hit-to-lead phase, Dr. Roberson said.

He and his colleagues have also identified a possible therapeutic target in the FTD mouse, based on the finding that an insufficiency of N-methyl-D-aspartate (NMDA) receptor signaling appears to mediate the repetitive behaviors and changes in social behavior that characterize the disease. “We found we could correct the deficits in the mice by boosting NMDA signaling with a drug called D-cycloserine. It's a co-agonist of the NMDA receptor that's already FDA-approved as an anti-tuberculosis agent, so there may be a drug repurposing opportunity.”

It's not enough to study these devastating diseases in mice and call it a day, he said. “Obviously we're very interested in understanding the mechanisms of disease because that's the basis for rational design of therapeutics, but our goal shouldn't be to understand or treat a mouse. It's eventually to treat people. We try to stay focused on that as much as possible.”

Back to Top | Article Outline


•. UAB faculty profile for Dr. Roberson:
    •. Published research by Dr. Roberson:
      © 2015 American Academy of Neurology