Human stem cell derived neurones and astrocytes as an in vitro model of human prion infection and replication

Prion diseases have a profound impact on animal and human health. This is especially true in the UK where the effects of bovine spongiform encephalopathy (BSE) and variant Creutzfeldt-Jakob disease (vCJD) remain a live public health issue, and in North America where chronic wasting disease in deer is considered a potential threat to human health, but it is also true worldwide where one in a million people every year die from sporadic or familial forms of the disease. Prion diseases resemble more common causes of dementia, such as Alzheimer's disease, in that they are characterised by the deposition of abnormally folded protein in the brain, but unlike Alzheimer's disease, prion diseases are demonstratively transmissible from one person or one animal to another. This relative ease of transmission between individuals has meant that animal experimentation has been the main stay of prion disease research for decades, first using rodents and monkeys and then using increasingly sophisticated genetically modified mice to model animal and human prion diseases. Transgenic mouse models of human prion disease are among the most useful of any of the currently available transgenic mouse model of human neurodegenerative disease, however they carry both an economic and ethical cost.

The ability to infect human cells grown in culture with human prions would be a major advance and would lead to an acceleration in our understanding of prion diseases at the molecular and cellular level and consequently our ability to develop treatments for them. Previous attempts to infect human cells with human prion agent have been largely unsuccessful and to date there is no widely used cell culture model of any human prion disease. We have hypothesised that previous failures result from differences between the kinds of human cells that grow and multiply well in culture and the natural targets of prion in human brains, specialised brain cells called astrocytes and neurones. The ability to make functional human cells of many types (including astrocytes and neurones) from human embryonic and induced pluripotent stem cells offers much to biology and medicine, including cellular transplantation therapies, but it may also offer something extremely useful to prion disease research.

We have characterised a series of human stem cell lines for the sequence and expression of the normal prion protein and selected the ones that we predict would be susceptible to vCJD prion infection based on what we learned from patients and from experimental animal studies. We then made these stem cells into astrocytes and exposed them to small quantities of human brain tissues from patients who died from vCJD. Our results show that human prions can indeed replicate in human cells in culture, provided that care is taken to use cells that closely resemble the cells affected in the brains of patients.

We seek to capitalise on this observation by finding out (i) whether the neurones (in addition to astrocytes) can be infected with the vCJD prions, (ii) whether the prions associated with other forms of Creutzfeldt-Jakob disease can infect stem cell derived astrocytes and neurones, (iii) whether the cell culture system reproduces known genetic barriers to infection, (iv) if prion infection harms neurones and glia and (v) whether it affects the mutually supportive relationship between astrocytes and neurones. Lastly, we will develop a definitive method for measuring prion replication in cells. The results of these experiments will provide a sound basis on which to advocate the use of stem cell derived neurones and astrocytes as a replacement for experimental animals in specific kinds of basic and applied human prion research.

Prion diseases are fatal neurodegenerative conditions associated with misfolding of the normal form of the prion protein. The mechanism of neuronal dysfunction and death are not well understood and effective therapies are lacking. Some human prion diseases are acquired, most forms are transmissible, and all forms are thought to proceed by the seeded spread of prion protein misfolding in the brain. Factors known to affect susceptibility to prion infection include the strain of prion agent and host prion protein genotype. Prion disease research benefits from excellent animal models and more recently the molecular pathology has been modelled using cell-free assays, but there is a manifest gap in human prion disease research in the area of cell biology. At present there are no normal human cells that have been shown to be susceptible to infection with human prions in vitro. This places constraints on studies of human prion detection, replication, cellular susceptibility, neurotoxic mechanism and candidate therapeutic agents. We propose that neurones and astrocytes derived from human stem cells offer a promising route to a cell culture model of human prion infection and replication. Our preliminary data shows that astrocytes derived from human induced pluripotent stem cells can be infected with variant Creutzfeldt-Jakob disease prions from human brain tissue samples. Further, that susceptibility is genotype dependent. We wish to capitalise on this observation by extending these studies to induced pluripotent stem cell-derived neurones, examine effects of prion agent strain and host cell genotype on susceptibility and determine what the effects of prion infection are on neuronal and astrocytic function. These studies will demonstrate the potential for human stem cell-derived neurones and astrocytes to replace animal models in numerous aspects of human prion disease research.

The main driver of this research project is partial replacement of the use of animals (primarily transgenic mouse models) in basic and applied human prion disease research.

Literature searching the PubMed database [<> on 7th January 2014] using the search "prion" OR "transmissible spongiform encephalopathy" AND "mouse" identifies 4902 publications. Substitution of the term "transgenic" for "mouse" produces 1183 results. Introducing the filter species "human" results in 541 publications associated with transgenics, 182 of which were published in the last five years. All of these metrics show a rising trend over time. In contrast, only 137 publications are identified in the last five years using the search terms "prion" OR "transmissible spongiform encephalopathy" AND "cell culture" and this number is reduced to 57 if the "human" filter is applied. Searching for comparable studies using the two main cell-free prion model systems ("PMCA" AND "QUIC") produce 125 and 43 (with the "human" filter term added) in the last five years. These data confirm the informed impression that:

1. Prion disease research continues to rely heavily on animal based studies. This could be attributed to conservativism on the part of the research community, but probably also reflects the increasing availability and ease of construction of transgenic models and the historical difficulties associated with prion propagation in vitro.
2. Transgenic mouse studies modelling prion disease (including human prion disease) are on a rising trend. To illustrate this point, the numbers of publications identified in the PubMed database using the search terms "prion" OR "transmissible spongiform encephalopathy" AND "transgenic" in 1993 were 14, in 2003 were 45 and in 2013 were 87 [search date 13th January 2014].
3. Developments in cell culture have not kept pace with developments in the field of transgenesis, nor with the more recently developed cell-free assays.


An accurate estimation of the number of animal procedures rendered unnecessary by the development of the proposed system is difficult to make, however we can consider typical reported experiments. The definitive study of human susceptibility to the BSE/variant CJD agent reported by three of us involved 129 humanised and bovinised transgenic mice. A related study of the zoonotic potential of other animal prion disease strains reported by one of us used 918 humanised transgenic mice. A more recently published study used 47 non-human primates to address the zoonotic potential of just one animal prion disease. The inherent variability of animal transmission studies mean that large experimental group sizes are the norm in prion research, usually a minimum of one cage per experimental condition or time point (or ~6 mice) leading to typical experimental animal number of 100 animals per experiment described above. Experiments involving multiple time points, other interventions, or ranges of drug dosage levels multiply up this number correspondingly. If we make a very conservative estimate that leading European counties (France, Germany, Italy and Spain), Canada, the USA and Japan have a minimum of two centres each conducting such work and each running five such experiments each year we reach an annual estimated total of 1000 mice that could be wholly or partly replaced by an appropriate cell culture model.

Whilst these metrics should be considered as estimates it can be claimed with a considerable degree of certainty that the cell culture model system described here would be highly relevant to the following areas of research:

1. Studies of the basic mechanisms of prion replication and its toxicity to the cells of the nervous system
2. High-throughput early stage testing of candidate therapeutic agents for prion diseases
3. Studies of genotypic or species barriers to prion transmission
4. Analogous studies of other protein misfolding diseases.