Infectious prions are known to spread from cell to cell in an insidious process that unfolds over weeks and months, ultimately leading to neurodegeneration. Now, researchers report that the misfolded prions inflict damage soon after they come into contact with neurons. Armed with a new cell culture system, researchers led by David Harris at Boston University found that neurons retract their dendritic spines within 24 hours of encountering PrP-Sc, a pathological form of the prion protein. Interaction between the N-terminus of PrP-Sc and normal prion proteins on the neuron’s surface mediated the spine loss. The findings, published May 26 in PLOS Pathogens, reinvigorate the debate over how the slow process of prion propagation relates to its neurotoxicity. The data may also help scientists understand other neurodegenerative diseases. This early synaptic hit mediated by misfolded PrPSc resembles what occurs in Alzheimer’s and other proteinopathies, Harris told Alzforum.

“This work is very well done and addresses the cellular mechanism of PrPsc neurotoxicity convincingly,” wrote Dennis Selkoe of Brigham and Women’s Hospital in Boston. He added that while other neurodegenerative diseases share the characteristic of early synapse loss, the mechanism described here is likely specific to prions.

Total Collapse.

Dendritic spines fall apart in wild type neurons (top panels) after exposure to PrP-Sc (right). Neurons lacking normal PrP (bottom panels) are spared. [Courtesy of Fang et al., PLOS Pathogens 2016.]

Infectious misfolded prions spread throughout the brain by corrupting normal prion proteins. While much research has focused on the mechanisms of this prion propagation, fewer studies have examined how the misfolded proteins destroy neurons. The neuroblastoma cell lines commonly used to study prion propagation are largely impervious to the toxic effects of prions. To study toxicity, researchers have used in vivo models and brain slice cultures; however, these studies have been restricted to events that occur after prion infection sets in, which takes weeks (see Jeffrey et al., 2000; Campeau et al., 2013). Limited studies that have examined prion infection in primary neurons focused on later events as well (see Cronier et al., 2004Hannaoui et al., 2013). What has been lacking in the field is an analysis of the earliest neurotoxic events inflicted by prions, which may even precede detectable prion propagation, Harris said. 

First author Cheng Feng and colleagues set out to create such a model, and to examine changes to dendritic spines, which they predicted would be the first structures to crumble. “We got a big clue from the AD field, because one readout for Aβ oligomer toxicity is exactly this effect—damage to spines,” said Harris. The researchers isolated hippocampal neurons from normal mice, and grew them in culture next to a feeder layer of astrocytes for three weeks. At this point, the neurons had formed mature axons and dendrites studded with mushroom spines, and they had formed synapses. The researchers then treated the cells with brain homogenates from mice infected with Rocky Mountain Laboratory (RML) scrapie, a prion strain that originally came from sheep. After 24 hours, the researchers analyzed the morphology of dendritic spines using fluorescently labeled phalloidin, which binds actin.

Compared to homogenates from normal mice, those from animals infected with prions ablated about three-quarters of the dendritic spines on the hippocampal neurons. The cytoskeleton bolstering each spine had collapsed, leaving behind a crumpled patch of actin. The researchers obtained similar results when they used purified PrP-Sc rather than crude homogenates, a result that strongly implicated the scrapie prion in the spine assault. They also found that spines succumbed equally to PrP-Sc that was sensitive to digestion by proteinase K (PK), which cleaves off the first 65 amino acids from the protein’s N-terminus, and to PrP-Sc that resisted the protease. Purified PrP-Sc consists of both. It has been proposed that PK-resistant PrP-Sc represents the more aggregated form, although the importance of PK-sensitivity is an unresolved issue in the field (see Saverioni et al., 2013). The fact that both PK-resistant and PK-sensitive forms of PrP-Sc triggered spine loss indicated that the first 65 amino acids of PrP-Sc were dispensable for toxicity, and hinted that different forms of PrP-Sc aggregate were equally toxic to spines.

On the other hand, no spine loss occurred when the researchers added PrP-Sc to neurons from PrP-C knockout mice. This led the researchers to hypothesize that an interaction between normal and misfolded prions facilitated the spine collapse. To further tease out this relationship, they added PrP-Sc to neurons from mice carrying truncated prion genes. Neurons from animals lacking prion residues 23-111 or 23-31 resisted PrP-Sc treatment and their dendritic spines remained completely intact.

The researchers concluded that an interaction between PrP-Sc and the 23-31 amino acid region in PrP-C triggered the spine defect, but how? Answering that question will be the focus of future studies, Harris said. It is unclear whether the conversion of PrP-C to PrP-Sc must occur to trigger spine loss, or whether the interaction mediates a signaling cascade independent of prion conversion, Harris said. Joel Watts of the University of Toronto, who praised the elegant simplicity of the work, commented that it takes much longer than 24 hours to detect conversion of prions in culture. “That they get a toxic effect so early hints that prion conversion is not required for this form of toxicity,” he said. To address this question directly, researchers could use neurons from mice expressing prion proteins resistant to this process, he suggested.

The results address whether prion propagation occurs independently of prion toxicity, and if the same or different strains take part in each process, Watts added. It also remains to be seen if these early changes in hippocampal cultures reflect the process going on in vivo, and if they are connected to progressive pathology, he said.

Harris said his lab plans to use the culture model to address many of these questions, and to understand if other misfolded proteins, including Aβ, trigger spine loss via similar mechanisms. His and other labs previously reported that an interaction between Aβ oligomers and PrP-C promoted synaptotoxicity (see Feb 2009 conference newsMay 2011 newsFluharty et al., 2013). Harris will investigate whether common signaling pathways downstream from PrP-C binding trigger spine loss in different neurodegenerative diseases.—Jessica Shugart


  1. This is an interesting study that reports on an experimental platform to explore the cellular mechanisms of prion toxicity. Prions are ordered aggregates of the cellular prion protein (PrPc) that provide a template on which PrPc monomer is thought to bind and undergo conformational rearrangement in autocatalytic manner. Thus, prions constitute the (protein only) infectious entity in prion disease. While much is known about prion infectivity, little is known about how, or if, prions actually cause neuronal compromise. As with most neurodegenerative disorders, synaptic loss occurs early in prion disease.

    Here, Feng et al. report on the use of primary mouse hippocampal neurons to monitor the effects that various extracts from prion-infected mouse brain have on structural post-synaptic elements, i.e., dendritic spines. They show that crude brain extracts from prion-infected mouse brain, and prions purified with and without the use of proteases, each reduce spine number and density. They also show that purified prions treated with proteinase K (PK) induce loss of dendrites. However, due to the facts that (i) the crude extract contained ~7.5 ug/ml of PrPc and the purified preps ~4.4 ug/ml PrPSc, and (ii) neurons in different cultures have different numbers of spines, and spines in different cultures have different areas, reliable quantitation of the relative toxicity of the preps is not possible. Nonetheless, the data suggest that classical PK-resistant PrPsc can induce loss of dendrites (Fig. 4) and that PK sensitive PrPsc may cause shrinkage of surviving dendrites (Figs. 1 and 2).

    Employing cultures prepared from Prnp-null mice and mice expressing PrPc with one of two N-terminal deletions, the authors nicely show that PrPSc toxicity requires expression of full-length PrPc with a specific need for residues 23-31. The requirement for this short, 9-amino-acid polybasic sequence is particularly important since previous work indicated that this stretch of amino acids is necessary for binding of PrPC to PrPSc and for toxicity of PrP. This dendrite toxicity assay should enable further experiments to differentiate if residues 23-31 are required solely for binding or ifr D23-31 allows binding, but not toxicity. This approach may also provide a springboard for testing other forms/preparations of PrPSc and the downstream signaling events required for toxicity. In terms of Alzheimer’s disease, it will be interesting to determine whether PrPSc and Aβ protofibrils, which are known to bind PrPc and induce dendritic toxicity, operate by common mechanisms. Thus this work provides an important tool on the long road to understanding prion toxicity and how it may relate to synaptotoxicity seen in more common neurodegenerative conditions.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Keystone: Partners in Crime—Do Aβ and Prion Protein Pummel Plasticity?
  2. Patient-Derived Aβ Needs Prion Protein to Harm Synapses

Paper Citations

  1. . Synapse loss associated with abnormal PrP precedes neuronal degeneration in the scrapie-infected murine hippocampus. Neuropathol Appl Neurobiol. 2000 Feb;26(1):41-54. PubMed.
  2. . Prions can infect primary cultured neurons and astrocytes and promote neuronal cell death. Proc Natl Acad Sci U S A. 2004 Aug 17;101(33):12271-6. PubMed.
  3. . Analyses of protease-resistance and aggregation state of abnormal prion protein across the spectrum of human prions. J Biol Chem. 2013 Sep 27;288(39):27972-85. PubMed.
  4. . An N-terminal fragment of the prion protein binds to amyloid-β oligomers and inhibits their neurotoxicity in vivo. J Biol Chem. 2013 Mar 15;288(11):7857-66. PubMed.

Further Reading

Primary Papers

  1. . A Neuronal Culture System to Detect Prion Synaptotoxicity. PLoS Pathog. 2016 May;12(5):e1005623. Epub 2016 May 26 PubMed.