Subscribe to eTOC

Two Labs, Same Conclusion: Alzheimer Disease Spreads from Neuron to Neuron


investigators Dr. Karen Duff, Dr. Scott Small, and Li Liu.

Two independent laboratories — one at Columbia University; the other, at Harvard Medical School — have devised a clever set of experiments to prove that the pathology that leads to tangle formation in Alzheimer disease (AD) spreads across the brain from neuron to neuron rather than selectively hitting vulnerable regions at different time points over the course of the disease.

The finding answers a pivotal question as to how AD progresses, and could open the door to novel treatments that could stop the disease from spreading and damaging key cognitive circuits.

Karen Duff, PhD, Scott Small, MD, and their colleagues at Columbia reported their finding on Feb. 1 in the open-access journal PLoS One and Bradley Hyman, MD, PhD, and his team at Harvard published similar results in the Feb. 23 issue of Neuron.

It's been known since the 1980s that the entorhinal cortex is the first brain region to develop the neurofibrillary tangle pathology of AD. To scientists studying Alzheimer's dementia, this made sense: the entorhinal cortex has major projections to the hippocampus and other memory circuits, and tangle formation is most closely associated with cognitive decline. As the disease progresses and symptoms worsen, the pathology appears to spread from the original site in the entorhinal cortex to many other brain regions.

Until these recent studies, however, no one was quite sure how the disease makes its way across the brain. There were two alternating hypotheses that generated a lot of theories over the decades. Were brain regions hard hit by AD connected through the entorhinal cortex? Or were the damaged areas affected later in the disease just less sensitive than the entorhinal cortex?

To answer this puzzle, the Harvard investigators turned to transgenic mouse models of tau pathology. Specifically, they engineered mice to express human tau predominantly in the entorhinal cortex and then followed the animals over time to see how the disease spreads. As expected by the model, early on the expression of pathological human tau led to tangle formation in the entorhinal cortex. Then, they noticed something surprising upstream of the entorhinal cortex in the dentate and the cingulate: tangle formation that included both mouse and human tau. How did it get there?

They did a series of experiments to determine whether the tau and tangles they were seeing outside of the entorhinal cortex were of human origin. Over and over again they concluded that they were. “We think that tau is transported down axons, released at synapses and taken up by neighboring cells,” explained Dr. Hyman, a professor of neurology at Harvard and Massachusetts General Hospital and director of the Massachusetts Alzheimer's Disease Research Center. “The human forms of pathological tau latch on to tau in connected neurons and trigger the formation of new pathological, misfolded tau.”

The Columbia researchers found the same phenomenon. Dr. Duff, a professor of pathology and cell biology and senior author on the PLoS One study, has been creating transgenic animal models of AD for almost two decades. Expressing human tau in the entorhinal cortex of the mouse, the Columbia investigators were able to map the path of the abnormal tau protein and connect the dots across the brain. “Because of the distinct route that the pathology followed, we could predict that tau was leaving one cell and being taken up by a neighboring cell,” said Dr. Duff. “This is rather unusual as tau is a protein involved in stabilizing the cell skeleton and it should not be found outside of the cell.”

Dr. Duff and colleagues identified human tau in some cell bodies of the entorhinal cortex and the subiculum by 10 months of age. Most of the human protein was in the axons within superficial layers of the medial and lateral entorhinal cortex, the subiculum, and at the terminal zones of the perforant pathway in the dentate gyrus. By 22 months and older, they found more mature human tau pathology in the neuronal cell bodies in the same areas; but also a substantial number of hippocampal pyramidal neurons especially in CA1; and in dentate gyrus granule cells that didn't contain human tau protein earlier.

Because granule cells in the dentate gyrus are one synapse away from cells in the entorhinal cortex that make the human tau protein, there was clear propagation of pathology out from the entorhinal cortex, added Dr. Duff. The data “support a trans-synaptic mechanism of spread along anatomically connected networks, between connected and vulnerable neurons,” she and co-authors wrote. “In general, the mouse recapitulates the tauopathy that defines the early stages of AD and provides a model for testing mechanisms and functional outcomes associated with disease progression.”


: “We think that tau is transported down axons, released at synapses and taken up by neighboring cells. The human forms of pathological tau latch on to tau in connected neurons and trigger the formation of new pathological, misfolded tau.”

“It all seemed to follow the same pattern,” added Dr. Small, a professor of neurology at Columbia. “It spreads across the brain like falling dominos.”


The next big question is how pathological forms of tau move in and out of cells and how it converts normal tau to the pathological form along the way. “We think tau leaks out of the dying cell and is taken up by a neighboring cell,” said Dr. Duff. “This finding opens up a whole area of pathobiology for us to explore.”

The hope is that scientists can now test drugs that stop the escalation of tau pathology through the whole brain, possibly even before it spreads beyond the entorhinal cortex. As the spread of tau from hippocampal areas to the neocortex maps with the worsening of cognitive symptoms, preventing spread to this region could be of benefit in preserving the cognitive abilities of patients.

Dr. Duff said that an immunotherapy approach could capture tau and block its spread. “If tau is spreading in the way we think it is, through spaces between cells, this approach could work,” she added. “But we are a long way from proving this.”

Columbia's Dr. Small said that similar spreading probably occurs in other neurodegenerative conditions.

In 2010, Lennart Mucke, MD, director and senior investigator at the Gladstone Institute of Neurological Disease, and professor of neurology and neuroscience at the University of California, San Francisco, published a paper in Neuron showing amyloid beta expression in the entorhinal cortex of a transgenic animal. The investigators observed local plaque formation but no spreading beyond the area. This makes sense given what scientists know about tau, the executioner and driver of amyloid-beta pathology, said Dr. Small.


“The unrecognized pioneer in this area is Markus Tolnay, MD, a professor of neuropathology at the University of Zurich who injected human neurofibrillary tangles into the brains of rodents in 2009 and reported what appeared to be transsynaptic spread of tangle pathology,” said Sam Gandy, MD, PhD, director of the Center for Cognitive Health at Mount Sinai Medical Center and professor of neurology and psychiatry, who was not involved with the study.

“Virginia Lee, PhD, professor of pathology at the University of Pennsylvania, demonstrated that synuclein could display similar phenomena, and she also confirmed the Tolnay report,” he said. “The most important implication is that if we can dissect the transport pathway, we might be able to arrest the spread (and symptoms) at very early stages. This wasn't totally surprising. We recognize more now that there are released vesicles (exosomes) and nanotube connectors between cells so that things move around and move from cell to cell in ways that we used to think were forbidden.”

Marc I. Diamond, MD, associate professor of neurology at Washington University in St. Louis, MO, first published evidence that aggregated tau in cell culture could be passed from one cell to another. Dr. Diamond said that he began thinking about this possibility in 2004. “I was struck by the similarities between prion disease and common neurodegenerative diseases. They begin with a small problem that spreads and becomes a bigger problem.”

In all of these conditions, from Alzheimer's to prions, protein misfolding is a common theme. “If we take a page from prion biology, these proteins can propagate pathology.” In 2009, Dr. Diamond and his colleagues published two papers in the Journal of Biological Chemistry showing that in cultured cells tau aggregates can act like the prion protein to “be taken into the cell and convert it to a fibril and released and transferred on to the next cell.”

He said that the new mouse papers are “a big advance. They showed that aggregates moved to neurons that shared synaptic connectivity. And the template conformational change triggers more aggregation.”

He added that the findings “support a propagation model of neurodegeneration and it means that there could be new ways to treat these diseases.” That pathological proteins are spending time outside of the cell means that antibodies can target the proteins and block its spread. “My prediction is that this finding will revolutionize how we diagnose and treat neurodegenerative diseases.”



A new finding that pathological tau spreads from neuron to neuron in Alzheimer disease could open the door to novel treatments that could stop the disease from spreading and damaging key cognitive circuits. Listen here as Dr. Karen Duff, of Columbia University, and Dr. Bradley T. Hyman, of Harvard, describe their findings and why they could be game-changers for the field:


• Liu L, Drouet V, Duff K, et al. Trans-synaptic spread of tau pathology in vivo. PLoS One 2012; 7(2): e3130.
    • de Calignon A, Polydoro M, Hyman BT, et al. Propagation of tau pathology in a model of early Alzheimer's disease. Neuron 2012; 73(4): 685-697.
      • Harris JA, Devidze N, Mucke L, et al. Transsynaptic progression of amyloidβ-induced neuronal dysfunction within the entorhinal-hippocampal network. Neuron 2010;68(3):428-441.
        • Frost B, Jacks RL, Diamond MI. Propagation of tau misfolding from the outside to the inside of the cell. J Biol Chem 2009; 284(19):12845-12852.
          • Frost B, Ollesch J, Diamond MI, et al. Conformational diversity of wild-type Tau fibrils specified by templated conformational change. J Biol Chem 2012; 284(6):3546-3551.
            • Clavaguera F, Bolmont T, Tolnay M, et al. Transmission and spreading of tauopathy in transgenic mouse brain. Nat Cell Biol 2009;11(7):909-913.