Could anti-epileptic drugs be the next hope for treating and even preventing Alzheimer’s disease symptoms? People with AD suffer more epileptic seizures than do normal elderly, and there are hints the antiepileptic drug levetiracetam quells abnormal brain network activity associated with AD. In one small study, the drug was found to improve cognition in people with mild cognitive impairment (MCI). But preclinical data, a prerequisite for therapeutic development, have been lacking—until now. In the August 6 Proceedings of the National Academy of Sciences, researchers led by Lennart Mucke at the Gladstone Institute of Neurological Disease, San Francisco, California, report that levetiracetam not only quiets epileptiform activity in a mouse model of AD, but also reverses behavioral problems, including cognitive decline.

“This work is illuminating, and supports the idea that overactivity of brain networks is driving the disease, rather than the other way around,” Michela Gallagher told Alzforum (see full comment below). Gallagher, who works at Johns Hopkins University, Baltimore, Maryland, led the group that conducted the recent study of levetiracetam in MCI (see ARF related news story).

The finding comes after years of work in the Mucke lab studying overactive brain networks in mouse models of AD. Mucke told Alzforum that, while the field focused on synaptic suppression driven by amyloid-β (Aβ), his group found something more complex, specifically, that there were indications of hyperactivity in mice that produce abundant brain Aβ. First, the researchers found loss of the calcium-binding protein calbindin in the dentate gyrus of the hippocampus, a sign of excessive calcium influx into neurons (see ARF related news story). Four years later, they reported that subclinical, non-convulsive epileptic activity hampers the J20 mouse model (see ARF related news story). The hippocampus is hyperactive in people at risk for Alzheimer’s, including those with MCI (see Dickerson et al., 2005) and those carrying ApoE4 (see ARF related news story), the strongest genetic risk factor for sporadic AD. One interpretation of the human findings posits that this uptick in activation compensates for cognitive deficits, but the recent work from Gallagher’s group, and now this finding from Mucke’s lab, suggest that hyperactivation may be part of the problem, not a response to it.

First author Pascal Sanchez and colleagues tested a variety of approved anti-epileptic drugs in four- to six-month-old J20 mice. They found that only levetiracetam, both acute and chronic dosing, reduced aberrant firing seen in electroencephalograms. The longer, 28-day treatment also reversed behavioral problems, including hyperactivity in the open field and in the open arms of an elevated maze. Chronic levetiracetam corrected deficits in spatial- and context-dependent tests of learning and memory compared to a saline placebo. Interestingly, 14 days after chronic treatment ended, abnormal spike activity returned, and 21 days after that, the mice were back to behaving abnormally.

How does the anti-epileptic restore learning and memory? In the J20 mice, the observed hyperactivity is, in fact, a highly synchronous activity; it sets off a cascade of compensatory changes that leads to synaptic remodeling in the hippocampus, explained Mucke. This remodeling includes loss of calbindin, increased expression of neuropeptide Y—which has antiepileptic activity—and suppression of the transcription factor Fos. The changes culminate in loss of synaptic function. Levetiracetam normalized these molecular players, and restored long-term potentiation, a measure of synaptic plasticity. The drug reversed these molecular and behavioral changes without affecting APP processing or Aβ deposition.

Gallagher pointed out that a strength of the study is that it contains detailed physiological experiments that cannot be done in humans. “In both cases, preclinical and clinical, targeting overactivity was an effective treatment,” wrote Gallagher, adding “such translation across a preclinical AD model and clinical patients is rare, if not unprecedented, in the AD field.” However, she cautioned that the condition of hyperactivity in amnestic MCI might be much more subtle than in the J20 mice. Their abnormal physiology may represent one end of a spectrum of changes that happen between normal aging and the emergence of AD pathology (see comment below). In fact, Gallagher suggested that aging itself may be conducive to overactivity, and that this activity may permit pathology to take root. Mucke sees the relationship slightly differently. “We picture levetiracetam acting downstream of Aβ because, at least in our model, everything starts with elevated Aβ and the drug blocks the overexcitation of the network that ensues,” he said. His group plans to examine the effect of levetiracetam in J20 mice of different ages to determine if the beneficial effects depend on disease stage.

Why the other anti-epileptic drugs did not have the same effect is unclear. Levetiracetam is often described as an atypical anti-epileptic because of its broad effects. It is the only one of the drugs tested that binds to the synaptic vesicle protein SV2A. “Either through SV2A, or other actions on glutamate transporters, it probably downregulates the amount of glutamate around the synaptic cleft,” suggested Mucke.

Brad Hyman, Massachusetts General Hospital, Charlestown, told Alzforum that the potential differential efficacy of levetiracetam and other anti-epileptic agents generates the intriguing possibility that anti-seizure properties per se are not the critical issue for the effect. He suggested that subtle changes in excitatory/inhibitory balance might contribute to some neural system dysfunction in the J20 model. Other models, including APP23 and APP/PS1 mice, and certain strains of Tg2576 APP mice, also exhibit epileptic-like seizures (see Lalonde et al., 2005; Minkeviciene et al., 2009).

What do these new findings mean for the clinic? “These results, in conjunction with a series of other findings in the field (most importantly, Bakker et al., 2012), make a strong case for the need for further studies of physiologic neural network-level abnormalities in AD and related disorders, and also for the targeting of these abnormalities for novel approaches to treatment,” wrote Brad Dickerson, who also works at MGH (see full comment below). Mucke and Gallagher are pursuing the idea of using levetiracetam in clinical trials.—Tom Fagan

Comments

Make a Comment

To make a comment you must login or register.

Comments on News and Primary Papers

  1. This new study by Dr. Mucke and colleagues provides a great deal of important data to support the hypotheses that 1) abnormalities of neural network function are important contributors to behavioral deficits in mouse models of AD; 2) targeting these physiologic/functional abnormalities with specific treatments can ameliorate behavioral deficits; 3) such treatment does not appear to affect traditional histological markers of AD. These results, in conjunction with a series of other findings in the field (most importantly, Bakker et al., 2012), make a strong case for the need for further studies of physiologic neural network-level abnormalities in AD and related disorders, and also for the targeting of these abnormalities for novel approaches to treatment.

    References:

    . Reduction of hippocampal hyperactivity improves cognition in amnestic mild cognitive impairment. Neuron. 2012 May 10;74(3):467-74. PubMed.

    View all comments by Brad Dickerson
  2. This report by Sanchez et al. demonstrating improvements in molecular markers, synaptic function, and memory performance in hAPP mice points to overactivity as a therapeutic target. These data from a preclinical model of AD also match recent results from a study of amnestic mild cognitive impairment (aMCI) patients treated with the same therapeutic. In both cases, preclinical and clinical, targeting overactivity was an effective treatment. Such translation across a preclinical AD model and clinical patients is rare, if not unprecedented, in the AD field. The clinical results reported recently by us (Bakker et al., 2012) showed that neural network overactivity detected by functional magnetic resonance imaging (fMRI) was reduced in aMCI by a subclinical dose of levetiracetam, and concomitantly improved performance on a memory task. In the case of the physiological experiments in AD mice, which would not be possible in humans, multiple sites of synaptic dysfunction in the affected network were resolved, including reduced effective synaptic connections (e.g., in CA1 of hippocampus), consistent with the authors' view that synaptic remodeling is driven, at least in part, by overactivity rather than the other way around.

    The condition of overactivity treated in aMCI patients may be considerably more subtle than the aberrant activity occurring in J20 hAPP mice, and other mouse models based on familial AD. Indeed, the window of therapeutic efficacy, at subclinical doses of levetiracetam, is unlikely to be attributable to or aligned with anti-epileptic efficacy across the spectrum beginning with age-related memory loss and continuing to MCI and AD. A model of memory loss in aged outbred rats originally led to the discovery that neurons in the CA3 subregion of the hippocampus have elevated firing rates (Wilson et al., 2005). The CA3/CA1 synaptic connections in those animals also exhibit a compensatory profile (Nicholson et al., 2004) showing some additional similarity with the network remodeling occurring in AD mice. In contrast with the AD mice, however, no evidence of aberrant activity occurs in that condition of aging, based on extensive recordings from neurons and EEG monitoring. The fMRI studies of aMCI patients, guided by the studies of network dysfunction in aging, also closely resemble the findings from aged rats with memory loss, showing a localization of excess activation confined to the CA3/dentate gyrus (Yassa et al., 2010; Bakker et al., 2012). The more widespread and pathological aberrant activity in certain AD models may represent one end of a continuum in understanding how aging and the emergence of AD pathology affect the brain in the majority of late-onset cases.

    References:

    . Reduction of hippocampal hyperactivity improves cognition in amnestic mild cognitive impairment. Neuron. 2012 May 10;74(3):467-74. PubMed.

    . Reduction in size of perforated postsynaptic densities in hippocampal axospinous synapses and age-related spatial learning impairments. J Neurosci. 2004 Sep 1;24(35):7648-53. PubMed.

    . Age-associated alterations of hippocampal place cells are subregion specific. J Neurosci. 2005 Jul 20;25(29):6877-86. PubMed.

    . High-resolution structural and functional MRI of hippocampal CA3 and dentate gyrus in patients with amnestic Mild Cognitive Impairment. Neuroimage. 2010 Jul 1;51(3):1242-52. PubMed.

    View all comments by Michela Gallagher

References

News Citations

  1. Epilepsy Drug Calms the Hippocampus, Aids Memory
  2. Calbindin Study: Is Calcium the Molecular Handle on Dysfunction in AD?
  3. Do "Silent" Seizures Cause Network Dysfunction in AD?
  4. ApoE(ε)4 Brains Have to Work Harder

Paper Citations

  1. . Increased hippocampal activation in mild cognitive impairment compared to normal aging and AD. Neurology. 2005 Aug 9;65(3):404-11. PubMed.
  2. . Neurobehavioral characterization of APP23 transgenic mice with the SHIRPA primary screen. Behav Brain Res. 2005 Feb 10;157(1):91-8. PubMed.
  3. . Amyloid beta-induced neuronal hyperexcitability triggers progressive epilepsy. J Neurosci. 2009 Mar 18;29(11):3453-62. PubMed.

Further Reading

Primary Papers

  1. . Levetiracetam suppresses neuronal network dysfunction and reverses synaptic and cognitive deficits in an Alzheimer's disease model. Proc Natl Acad Sci U S A. 2012 Oct 16;109(42):E2895-903. PubMed.