Researchers have for the first time recorded the firing of hippocampal place cells in Alzheimer disease model mice. Published online in the early edition of the Proceedings of the National Academy of Sciences (PNAS), the results show an age-related degradation in the function of place cells, which are specialized neurons that help an animal identify its location. Place cells are crucial for spatial memory and navigation. Lead author John O'Keefe of University College London (UCL) said the paper demonstrates that in vivo study of place cell function can provide a sensitive assay of the rate and amount of spatial-memory deterioration in an AD surrogate. O'Keefe and other investigators cautioned, however, that the paper's accompanying observations correlating place cell function with amyloid plaques in the animals' hippocampus remain inconclusive until more experiments are done.

Using techniques and equipment modified from his pioneering studies of place cells in rats, O'Keefe, along with first author Francesca Cacucci and colleagues Ming Yi and Thomas Wills of UCL and Paul Chapman of the GlaxoSmithKline Centre for Cognitive and Neurodegenerative Disorders in Singapore, implanted microelectrodes in the hippocampal pyramidal cells of 42 animals. The investigators divided the mice into four groups: young and old controls, and young and old Tg2576 mice. The researchers then tested the spatial memory of each group over 12 days of trials while simultaneously recording from the behaving mice. Cacucci and colleagues used a T-maze rather than the traditional water maze because, as O'Keefe noted, live wires and water don't mix. The place cells of the aged Tg2576 mice showed functional differences that are known to coarsen the resolution of spatial mapping: the field size of the cells was larger, for example, and the amount of spatial information they processed was lower than in the other groups. Those differences correlated with deficits in the animals' spatial memory.

Because such physiological changes may actually occur before overt behavioral deficits, O'Keefe says, space-cell malfunction could become an early indicator of disease onset. "We suspect we'll begin to see signs of changes in the function of the cells before we see changes in behavioral tasks." Deficits in spatial memory are among the early signs of Alzheimer disease (see, e.g., deIpolyi et al., 2007).

O'Keefe and his colleagues also found that hippocampal plaque burden correlated with the level of place cell degradation. This observation taps into an ongoing debate in the field about which forms of Aβ peptide (and tau protein, for that matter) do the most damage to neuronal function during the pathogenesis of AD. Besides having developed the Tg2576 mouse, Karen Hsiao Ashe of the University of Minnesota Medical School in Minneapolis is a leading investigator on that issue. Ashe commented by e-mail that more than two time points would be needed to implicate amyloid plaques in causing the place cell deficits (see extended comment below). O'Keefe is quick to agree. He is planning further studies with more time points, but "it didn't make any sense to do intermediate time points until we knew there were differences in the first and last ones," O'Keefe says. In these further experiments, O'Keefe plans to collaborate with John Hardy, who last year moved from the U.S. National Institutes of Health back to his native U.K., and UCL.

Hongxin Dong of Washington University School of Medicine in St. Louis, Missouri, told this reporter that the place cell work opens the door to several other avenues of study. "It would be interesting to…inspect the morphology of the individual place cells as well as their location in reference to nearby plaques," she noted in an e-mail with colleague Nicole Hicklin (see comment below). Dong has carefully analyzed the relationship between synaptic loss and amyloid deposits in the Tg2576 mice (see Dong et al., 2007).

Another independent group that studies transgenic mouse models of AD is Lennart Mucke's at the Gladstone Institute of Neurological Diseases at the University of California, San Francisco. Erik Roberson and Jorge Palop in that group said the paper closes a gap between behavioral studies showing learning and memory deficits in aged Tg2576 mice and in-vitro cellular studies showing deficits in synaptic plasticity, particularly long-term potentiation, as demonstrated by the landmark paper Chapman et al., 1999. "This observation interposes between those two previous findings quite nicely," Roberson said. Palop added that the place cell deficits may be more relevant to disease symptoms than the LTP deficits observed at the cellular level.

But Palop cautioned that the place cell recording technique is time-consuming and elaborate. "It's going to be restricted to a subset of labs that have the capacity to do this electrophysiological tracking," Roberson agreed. At present, just a handful of other investigators are equipped to do place cell recordings in living, freely moving mice: Eric Kandel's lab at Columbia University, New York, for example, and Susumu Tonegawa's and Matt Wilson's at MIT. To his knowledge, neither group is looking at AD-model mice, O'Keefe said. He hopes to generate interest in AD among other place cell researchers, so that place cell recordings may eventually help to identify the pathology of AD and serve as a platform for testing therapeutic interventions.—Karen Wright

Karen Wright is a freelance writer in New Hampshire.


  1. The electrophysiological findings in this interesting paper appear solid. However, the suggestion that the place cell abnormalities are due to plaques appears premature to me. It is based on an incomplete set of data from mice at one time point, i.e., 16 months. Future studies should test a second and maybe third time point, for example, at 10 months or at 22 months, because in Tg2576 mice the behavioral impairment, and presumably the electrophysiological abnormalities in place cells, fail to change at rates that are commensurate with changes in plaque deposition. The conclusion that the place cell abnormalities are due to plaques can be drawn only if the correlations hold, when data from two widely separated time points are combined and analyzed together.

  2. This article represents a clear, logically sound experiment in which the function of hippocampal place cells was monitored relative to spatial memory performance in an AD animal model. The authors’ rationale in investigating the relationship among place cells, amyloid, and impaired spatial function is justifiable and worthy. Another notable discussion point lies in the finding that behavioral deteriorations correlated significantly with all amyloid concentrations, while amyloid levels did not, in turn, correlate well with place cell functional quality. I agree with the interpretation of this as being indicative of the importance of the recruitment of the neocortex in the T-maze task. This, too, as addressed, would likely be an advantageous avenue of further study.

    The results of this experiment provide another tool that may perhaps be helpful in the study of the pathology and the correlated behavioral deficits associated with AD. It would be interesting also to take this approach further and inspect the morphology of the individual place cells as well as their location in reference to nearby plaques. Also, it’s not clear that the authors checked to make sure the electrodes were actually implanted in place cells. I am surprised that the old wild-type mice did not do significantly poorer than the young groups.

  3. This article on hippocampal place cell firing and memory deficits and plaque burden provides a crucial piece of the amyloid-β puzzle: the relationship between neuropathology on the one hand and neuronal function on the other. While cognitive deficits have been well documented in APP mouse models, this is the first time that a direct neural correlate of such behavioral deficits has been shown. The central finding is the reduction of spatial information carried by the activity of hippocampal place cells in Tg2576 mice at ages at which significant plaque aggregation has occurred. This was directly observed by measuring changes in place field size and spatial information.

    The correlation of the place cell activity deficits with percentage of plaque coverage in both hippocampal and cortical neurons is especially interesting. It strongly suggests that this particular deficit arises from the amount of aggregated amyloid-β, providing evidence for a specific effect of plaques on neuronal activity underlying behavior. The evidence is further bolstered by the finding that mean firing rate is not affected in these neurons, suggesting that the electrical properties of the neurons are not the cause of the change in functional neuronal properties. It is also interesting that the deficits in place cell firing are highly variable. This would be expected if the deficits are a result of distortion of the connectivity of the neuronal network, since the spatial distribution of plaques is also highly variable.

    Although the findings of the current paper do not directly address the mechanism of the information deficit in the place cells, an in vivo intracellular recording study in similarly aged Tg2576 mice has shown similar losses of specific evoked neuronal response reliability in neocortical neurons without affecting spontaneous activity. Those results suggest that temporal disruption of synaptic activity (caused by the spatial distortion of the neuropil by plaques) results in reduction of the neurons’ ability to integrate the necessary afferent information (Stern et al., 2004).

    The current paper shows behaviorally relevant loss of information-processing capability by individual neurons in brain areas with significant plaque aggregation during behavior—and correlated with performance deficit. The information deficit is in a (if not the) major functional role of these neurons, i.e., that of spatial localization and mapping. To my knowledge, this is the first study directly correlating AD neuropathology with changes in neuronal activity directly related to behavior—and correlated with behavioral deficit.

    One further comment is warranted: these findings in no way detract from the numerous studies implicating the soluble, oligomeric forms of amyloid-β on neuronal activity; they simply provide evidence that the aggregated forms of amyloid-β play a separate, specific role in the disease process, and that role (in addition to that played by soluble amyloid-β) must be addressed in any therapeutic approach to the disease.


    . Cortical synaptic integration in vivo is disrupted by amyloid-beta plaques. J Neurosci. 2004 May 12;24(19):4535-40. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Spatial cognition and the human navigation network in AD and MCI. Neurology. 2007 Sep 4;69(10):986-97. PubMed.
  2. . Spatial relationship between synapse loss and beta-amyloid deposition in Tg2576 mice. J Comp Neurol. 2007 Jan 10;500(2):311-21. PubMed.
  3. . Impaired synaptic plasticity and learning in aged amyloid precursor protein transgenic mice. Nat Neurosci. 1999 Mar;2(3):271-6. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Place cell firing correlates with memory deficits and amyloid plaque burden in Tg2576 Alzheimer mouse model. Proc Natl Acad Sci U S A. 2008 Jun 3;105(22):7863-8. PubMed.