Certain brain regions falter early in neurodegenerative disorders, and scientists have pinpointed the entorhinal cortex (EC) as the site where some of the first abnormalities appear in Alzheimer’s disease. In the December 22 Nature Neuroscience, researchers led by Scott Small and Karen Duff at Columbia University, New York, narrow that focus to a specific subregion of the EC. Using a high-resolution variant of functional MRI, the researchers found that healthy older adults who went on to develop dementia exhibited weak metabolism in the lateral EC. The team found the same vulnerability in transgenic mouse models that express Aβ and tau, the major pathological proteins of AD. Using these mice, they found that Aβ likely accelerates the tau pathology that drives lateral EC dysfunction, which can spread to connected brain areas in a pattern similar to that observed in humans.

Scientists know that AD begins years or even decades before overt cognitive decline. Older evidence has established that neuronal loss starts in the EC, which comprises distinct populations of cells. Lateral EC (LEC) and medial EC (MEC) neurons connect to different parts of the brain and have distinct physiology and morphology. Small wondered if one of these subregions succumbs earlier in AD than the other. “To answer that, you need to be able to look at the earliest stages of AD, and you need a functional imaging technique that has very high spatial resolution,” he told Alzforum.

To address this latter requirement, first author Usman Khan and colleagues used high-resolution cerebral blood volume (CBV) functional MRI to assess brain metabolism and, by proxy, neuronal activity. They developed an automated system for marking regions of interest within the entorhinal cortex that allowed them to distinguish the MEC and the LEC—regions researchers had been unable to distinguish by manually examining fMRI images. “The automated analysis allowed us to target the whole brain and pinpoint exactly where in the EC things were happening,” Small told Alzforum.

Khan and colleagues imaged 96 healthy older adults. Over a period of 3.5 years on average, 12 of the volunteers developed mild AD. Compared with those who remained cognitively healthy, these 12 had low metabolism in the EC. More specifically, this shortfall occurred in the LEC but not the MEC. In addition, the impairment correlated with reduced metabolism in two other brain regions, the parahippocampal gyrus and the precuneus of the parietal cortex, suggesting that the EC impairment has a domino effect in some connected brain regions. Both the parahippocampal gyrus and the precuneus tend to show up in other structural, functional, or amyloid PET imaging studies looking for early AD-like changes in cognitively normal older people. 

To investigate EC dysfunction experimentally, the researchers used the same CBV-fMRI imaging technique with mouse models. Taking advantage of transgenic mice that overexpress human APP and/or tau predominantly in the EC, they tested if and how either protein drove LEC dysfunction. “Evidence indicates that tau accumulates in the EC with normal aging," noted Small. He wondered whether tau alone caused problems or both tau and APP were required for pathology, and if the latter, how they interacted.

The researchers discovered that tau was required to slow LEC metabolism. This depended on the age of the animal and whether it also expressed human APP. In young mice, both APP and tau were needed to impair the LEC. In older mice, tau alone was sufficient, although APP co-expression made things worse. “When you have both Aβ and tau, the LEC becomes particularly sensitive,” Small told Alzforum. “Our data suggest Aβ causes tau to be neurotoxic somehow.” This finding is consistent with previous mouse studies suggesting that Aβ and tau work synergistically in driving pathology (see Aug 2001 news storyMay 2007 news storyAug 2013 news story). In the authors’ view, recapitulating in the tau-expressing mice the LEC impairment seen previously in preclinical AD and finding that APP exacerbated the effect, together suggest that a tau/Aβ partnership may similarly contribute to the human LEC dysfunction.

In addition, the researchers observed weaker metabolism in two other mouse-brain regions that connect to the EC but expressed neither transgene: the perirhinal and posterior parietal cortices. This parallels the human data, since these regions are homologous to the parahippocampal gyrus and precuneus. The results suggest that disease spreads from the LEC to connected brain areas. “We saw a pattern in humans—a spread from the EC out to the parietal cortex—and we confirmed the same pattern in mice,” said Small.

How does disease-related dysfunction spread from the EC to other cortical areas? Duff and others had previously reported that mutant human tau expressed primarily in the EC can transfer across synapses to connected regions (e.g., the hippocampus) and seed aggregation (see Feb 2012 news storyde Calignon et al., 2012). Aβ may do the same (see Nov 2010 news story). Although Khan and colleagues detected human tau in the hippocampus of the tau-expressing mice, the metabolism there was unaffected. On the other hand, they observed no notable tau histopathology in the parietal cortex, despite the impaired metabolism there. Consequently, the authors noted that rather than supporting the idea of pathology spreading directly by transfer of pathological proteins across synapses, their findings are “more consistent with the idea of functional spread,” where abnormal signaling from the EC leads to secondary dysfunction in connected regions.  

These new data leave open the fundamental question of why the LEC seems especially vulnerable in AD. Previous studies have reported higher basal metabolism in this region compared with other parts of the brain. The researchers confirmed this in both healthy, young people and in young, wild-type mice. “Basal metabolism typically reflects dendritic complexity,” said Small. The LEC contains mostly fan neurons that form distinct, dense dendritic arbors. Small hypothesized that their unique dendritic morphology may make these neurons particularly vulnerable in AD.

Since tau normally accumulates in LEC dendrites throughout life, the high levels that exist at older ages could conspire with APP-related abnormalities to initiate AD, Small suggested. Why so much tau accumulates in the dendrites of the LEC, and whether it relates to the high metabolism remain to be investigated, he said.—Linda Lee


  1. This is a fantastic study from the Small and Duff labs, looking at the molecular origins of AD and the spread of pathology. It has great potential for identifying new biomarkers and therapeutic strategies to contain the pathology at a confined space, if this is possible.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Finally United? Aβ Found to Influence Tangle Formation
  2. APP Mice: Losing Tau Solves Their Memory Problems
  3. In Adult Mice, Reduced Tau Quiets Agitated Neurons
  4. Mice Tell Tale of Tau Transmission, Alzheimer’s Progression
  5. Insidious Spread of Aβ: More Support for Synaptic Transmission

Paper Citations

  1. . Propagation of tau pathology in a model of early Alzheimer's disease. Neuron. 2012 Feb 23;73(4):685-97. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Molecular drivers and cortical spread of lateral entorhinal cortex dysfunction in preclinical Alzheimer's disease. Nat Neurosci. 2014 Feb;17(2):304-11. Epub 2013 Dec 22 PubMed.