Alzheimer’s poisons the mind into silence. Millions of neurons wither and die; networks stop working; whole regions of gray matter shrink. But at the outset of this process a curious thing happens. Key memory regions become overactive, for years before dementia sets in. Some have thought that this overactivity is beneficial—that it represents an automatic compensation for failing circuitry, without which memory functions would decline more swiftly. But scientists have been finding evidence to the contrary. Recent studies in rodents and humans have shown that quieting this overactivity, for example with an anti-epilepsy drug, partly reverses memory impairments. Further clinical trials of this strategy might lead to new Alzheimer’s treatments as well as a better understanding of the disease.
“We think that this excess brain activity is driving disease progression,” says Michela Gallagher, a neuroscientist at Johns Hopkins University.
Hyperactivity as a disease driver
Evidence that brain overactivity occurs in early or pre-symptomatic Alzheimer’s has been building for over a decade. A UCLA brain-imaging study in 2000, for example, noted that middle aged and elderly people with the high-Alzheimer’s-risk apoE4 gene showed abnormally high activation, during memory tasks, of the hippocampus and other memory-related brain regions that are typically affected in the disease. Similarly, a Harvard Medical School study published late last year found hippocampal overactivity in patients with mild cognitive impairment (MCI), which often leads to Alzheimer’s. This hippocampal overactivity was greater in those patients who showed more thinning of the cortex.
Gallagher was part of a team that found, in 2005, that aged, memory-impaired rats showed overactivation in the CA3 region of the hippocampus. To see if this overactivation was a cause rather than a result of memory-impairment, she and her colleagues gave the animals levetiracetam, an FDA-approved anti-epilepsy drug, which has the effect of quieting CA3 activity. As they reported in 2010, the treated animals improved on memory tasks. She and her team have now reported, in the journal Neuron this May, that just two weeks of low-dose levetiracetam restored CA3 activity to normal, and improved performance on a CA3-relevant memory task, in a small group of people with mild cognitive impairment.
These results have been confirmed to some extent by the laboratory of neurologist Lennart Mucke, who directs the University of California at San Francisco’s Gladstone Institute of Neurological Disease and is a member of the Dana Alliance for Brain Initiatives. Mucke has reported in recent scientific meetings that his group has been able to use levetiracetam to reverse memory impairments in mice that are genetically engineered to develop a condition resembling Alzheimer’s.
The meaning of overactivity
In the most popular model of Alzheimer’s these days, toxic small aggregates of amyloid beta protein, called oligomers, accumulate in key brain regions, harm synapses, and eventually trigger a death-spiral of other neuron-killing processes. Where do the new hippocampal-overactivation results fit in to this model?
That is so far unclear, but there are two main hypotheses. In one, hippocampal overactivity causes the accumulation of toxic amyloid beta oligomers. In the other, hippocampal overactivity results from oligomer accumulation.
According to the first hypothesis, overactivity within the hippocampus occurs—as shown in Gallagher’s aged rats—perhaps only as a natural consequence of getting old. But this overactivity of hippocampal neurons causes them to secrete amyloid beta faster than they would otherwise. “As the neurons are firing [more often], they’re more likely to secrete amyloid beta, and as amyloid beta is being released by the neurons it’s aggregating and forming toxic subspecies,” says William Jagust, a neuroscientist at the University of California at Berkeley who co-authored a related review paper on synaptic activity and amyloid beta in 2011.
Gallagher suspects that this is what is going on. “Our work is consistent with others’ findings that increased neural activity can increase the production of amyloid beta,” she says.
Mucke and his colleagues have focused more on the second hypothesis, according to which hippocampal overactivation is a consequence of amyloid beta toxicity. “We think that, even though at individual synapses amyloid beta can depress synaptic transmission, overall it leads to aberrant excitatory network activity,” Mucke says. In other words, amyloid beta impairs cognition not just by shutting down synapses but, especially in early stages of disease, by a “network effect” in which it disrupts the normal rhythms of memory-related circuits.
Scientists have long noted links between Alzheimer’s and epileptic seizures, particularly in familial Alzheimer’s and in some Alzheimer’s mouse models. Mucke suspects that seizures in Alzheimer’s-affected brains are merely an intense form—“the tip of the iceberg,” he says—of the harmful network disruptions that amyloid beta can induce. Three recent studies add weight to this view. In one, published in April, a team led by Arthur Konnert at the Technical University of Munich found that soluble amyloid beta—which presumably contained toxic oligomers—caused hippocampal neurons to become overactive in the brains of live mice. In another study, also published in April, a team including Mucke and led by his Gladstone Institute colleague Jorge Palop traced the hippocampal network hyperactivity in “Alzheimer’s mice” to impairments in “inhibitory interneurons”—cells that normally keep brain regions from becoming dangerously overactive. The result hints that amyloid beta oligomers, which such mice overproduce, exert their memory-disrupting effect at least partly by harming this population of interneurons.
The third study, published in 2010 by Yadong Huang, another Gladstone Institute scientist, found evidence in mice that hippocampal interneurons appear to be harmed also by Alzheimer’s-risk-boosting apoE4. To Mucke, these separate findings hint that hippocampal interneurons are crucial targets of the Alzheimer’s disease process, so that when they are impaired past a certain point, hippocampal networks become dysfunctionally overactive, and memory begins to fail.
Whether amyloid-beta is principally a cause or an effect of hippocampal overactivity remains to be proven, but Jagust points out that “it’s not necessarily the case that one explanation excludes the other.” If both are true, then a vicious spiral may be at work, so that toxic oligomers of amyloid beta cause neuronal overactivity, which in turn creates more toxic amyloid beta oligomers.
In any case, Gallagher and Mucke are continuing to test levetiracetam in rodents and in people. The drug seems to work better, in the context of MCI or Alzheimer’s, than other anti-epilepsy drugs. “We’ve looked at many different anti-epilepsy drugs that are FDA-approved,” Mucke says, “and most of them either have no effect on hyperactivity in our mouse models, or make it worse, so there seems to be something very specific about the benefits that come from levetiracetam.”
Gallagher, whose team recently began a clinical study to determine the best dose of the drug in MCI patients, looks forward to longer-term, larger-scale trials, and hopes to show that the quieting of hippocampal overactivity stops or at least greatly slows the underlying disease process. “We’re looking for more than a symptomatic improvement,” she says.