Genetically engineered stem cell‐derived neurons can be rendered resistant to alpha‐synuclein aggregate pathology

Abstract

We discuss the implications of a recent study by Chen and collaborators which demonstrated that dopamine neurons derived from human pluripotent stem cells which had been genetically engineered to delete the alpha-synuclein gene are resistant to the experimental induction of Lewy pathology (Chen et al., 2018). Neural transplants in Parkinson's disease patients can develop Lewy pathology over a decade after the graft surgery, and this pathology might compromise the long-term survival and function of the grafted neurons. Therefore, the study by Chen et al. has potential clinical implications where alpha-synuclein null neurons might eventually be considered as possible donor cells in future intracerebral transplantation trials in Parkinson's disease. Alpha-synuclein aggregates are a salient feature of Parkinson's disease (PD) pathology and it has been suggested they participate in some way as key players in the pathogenic process (Goedert, Jakes, & Spillantini, 2017). A decade ago, alpha-synuclein aggregates, highly reminiscent of the Lewy pathology seen in the PD brain, were found inside a small proportion of dopamine neurons that had been grafted to the striatum of four PD patients between 10 and 16 years earlier (Kordower, Chu, Hauser, Freeman, Olanow, 2008; Kordower, Chu, Hauser, Olanow, Freeman, 2008; Li et al., 2008, 2010). They exhibited multiple markers that indicated they were bona fide Lewy bodies and Lewy neurites. Subsequent reports describing post-mortem findings from the brains of an additional four patients indicate that the phenomenon is robust and reproducible (Ahn, Langston, Aachi, & Dickson, 2012; Kordower et al., 2017; Kurowska et al., 2011; Li et al., 2016), and there are anecdotal reports of further cases with similar findings that are not yet reported in the literature. While Lewy pathology is consistently found in grafted neurons if the surgery was done a decade or more prior to death, there were no marked Lewy bodies in transplanted neurons in patients who died 18 months – 4 years after surgery (Brundin & Kordower, 2012; Chu & Kordower, 2010), which suggests that there is a lag time in Lewy pathology development. Whereas it is difficult to definitely prove how and why Lewy bodies developed in grafted neurons that were fetal at the time of implantation and as young as 10 years after surgery at the time of analysis, it was hypothesized that it was due to prion-like propagation of aggregated alpha-synuclein from the host brain into the grafted neurons (Brundin, Li, Holton, Lindvall, & Revesz, 2008). Over the decade that has followed after the initial observations, numerous cell culture and animal models of prion-like transfer of alpha-synuclein aggregates have been created and several independent laboratories have provided support for the hypothesis that this mechanism is likely to have played a key role in the generation of Lewy pathology in the PD patients, who received fetal neuron implants (Brundin, Ma, & Kordower, 2016; Lee, Bae, & Lee, 2014; Luk & Lee, 2014; Melki, 2018; Steiner, Quansah, & Brundin, 2018; Volpicelli-Daley & Brundin, 2018). Notably, animal models involving intracerebral grafts of neurons also have shown that host-derived alpha-synuclein can invade the transplanted neurons (Angot et al., 2012; Desplats et al., 2009; Hansen et al., 2011; Kordower et al., 2011). Whereas the phenomenon of alpha-synuclein aggregates in grafted neurons in PD patients is a robust and reproducible finding, the consequences of these aggregates on graft functionality are not yet clear. The first alpha-synuclein aggregates appear to take about a decade to develop, and at that time they can be only found in as few as 1%–2% of grafted dopamine neurons (Li et al., 2008). Obviously, it is impossible to define the relationship between time elapsed after clinical transplant surgery and the proportion of neurons exhibiting Lewy pathology in detail. However, the few patients described in the literature suggest that the proportion increases slightly over the years, reaching between 12% and 27% at 16–24 years after surgery (Kordower et al., 2017; Li et al., 2016). Notably, these observations cannot take into account if some neurons that have developed Lewy pathology die and therefore the reported proportions could represent underestimates. One report has described sustained clinical benefit for 15–18 years following transplantation of fetal dopamine neurons in two PD patients (Kefalopoulou et al., 2014), but we do not know if this benefit will be permanent. In the case where 12% of the grafted dopamine neurons contained Lewy pathology at the time of death, 24 years after surgery, there was no longer any detectable clinical benefit from the graft. This benefit was marked 10 years after surgery, and started to decline around 14 years after implantation (Li et al., 2016). A study on the PD substantia nigra has suggested that nigral dopamine neurons only live for around 6 months after forming Lewy bodies (Greffard et al., 2010). If this is correct, and also applies to immature dopamine neurons implanted into the PD striatum, it would imply that over the years following surgery the implanted dopamine neurons will gradually die. Therefore, it is desirable to develop techniques to prevent the formation of Lewy pathology in the transplanted dopamine neurons. The future of neural transplantation therapy for PD does not lie in the use of fetal neurons as donor material. Instead, dopamine neurons differentiated from pluripotent stem cells will be the way forward. There are already several ambitious programs in this regard at different stages of clinical development (Barker, Parmar, Studer, & Takahashi, 2017; Barker et al., 2016; Kirkeby, Parmar, & Barker, 2017). Notably, 1 month ago, a team from Kyoto, Japan, reported that they had performed the first dopamine neuron transplantation surgery in one PD patient using induced pluripotent stem cells (iPSCs) as starting material (Cyranoski, 2018). By using pluripotent stem cells as starting material, it is possible to standardize the cells that are used and produce them in bulk using Good Manufacturing Practice (GMP) protocols (Kirkeby et al., 2017). The approach also opens up the possibility to genetically manipulate the stem cells so that they are more resistant to, for example, alpha-synuclein pathology. It has been speculated that one can reduce the uptake of extracellular alpha-synuclein aggregates in transplanted neurons, and this could be achieved by knocking down the proteins potentially involved in the uptake process (Holmes et al., 2013; Mao et al., 2016). However, the uptake mechanisms which appear to involve endocytosis, are not fully understood and it is not clear if interfering with these can affect other vital cellular processes in the grafted neurons. It has also been speculated that it might be possible to genetically engineer cells to boost the lysosomal-autophagy system of the grafted neurons, so that they more efficiently degrade any alpha-synuclein aggregates that they take up. However, while this concept has been demonstrated to be valid both in cell cultures and an in vivo model of PD (Xilouri et al., 2013), it has still not been tested in neural transplants. In this issue, Chen and coworkers took another approach (Chen et al., 2018). They used CRISPR/Cas9 technology to generate human embryonic stem cells (hESC) and iPSCs with deletion of one or both alleles of the alpha-synuclein gene. The rationale is that it is known from animal studies that even if alpha-synuclein aggregates might be taken up by alpha-synuclein null neurons, they are obviously do not capable of seeding further aggregation in a prion-like manner in such a setting. Despite alpha-synuclein having one functional role in synaptic transmission (Burré, 2015), it is also known for many years that alpha-synuclein knock out mice are remarkably normal, with only subtle changes in synaptic transmission and altered responses to neurotoxins described (Abeliovich et al., 2000; Dauer et al., 2002). When injected intracerebrally with alpha-synuclein fibrils, such mice do not develop Lewy pathology (Luk et al., 2012). Similarly, cultured neurons from alpha-synuclein null mice do not develop aggregates when exposed to preformed alpha-synuclein fibrils (Volpicelli-Daley et al., 2011). This suggests that partial or complete knockdown of alpha-synuclein is a viable approach to generating transplantable neurons that are resistant to the development of Lewy pathology. If generated following GMP protocols, such neurons could then potentially be used in clinical trials. Genome editing using CRISPR technology has been catapulted into prominence due to the amazing precision by which this can be done and the mind-boggling potential benefits that can be achieved both in vitro (experimental systems) and in vivo (correcting genetic defects). This powerful technology may be employed to understand gene function by precision knock-out of target genes. The CRISPR strategy used in this study provides a template (no pun) of how gene knock-outs should be done and verified. The authors provide the following thorough strategy: (a) Different pairs of gRNAs were used to delete parts of the coding exon 2 of the SNCA gene to ensure that subsequent results are not due to off-target effects. (b) A nickase mutant of Cas9 was used to negate further off-target double-strand breaks in the DNA that may trigger unwanted DNA repair mechanisms. (c) Mutant cloned were screened by PCR and verified by sequencing. (d) Alpha-synuclein expression was monitored by western analyses to verify lack of expression, and finally, (e) in subsequent experiments two SNCA−/− bi-allelic and two SNCA+/− mono-allelic deletion clones were used – such replication lends credence to the reported results. With all the controls in place, the results of subsequent experiments can be properly interpreted. When Chen and coworkers differentiated the genetically engineered hESCs and iPSCs into dopamine neurons and exposed them to preformed alpha-synuclein fibrils – using a protocol that triggers formation of alpha-synuclein aggregates in non-engineered neurons (Volpicelli-Daley, Luk, & Lee, 2014; Volpicelli-Daley et al., 2011) – as expected they observed no aggregates forming in the alpha-synuclein null neurons. Taken together, the study by Chen and collaborators is an elegant and state-of-the-art demonstration of CRISPR/Cas9-mediated deletion of the alpha-synuclein gene in human stem cells (Chen et al., 2018). These cells will definitely be welcome laboratory tools for those who are studying mechanisms and consequences of propagation of alpha-synuclein aggregates, and can be viewed as a “control” paradigm because the cells seem incapable to generating alpha-synuclein aggregates. As mentioned by the authors, future animal transplantation studies will define whether these neurons are also resistant to propagation of alpha-synuclein aggregates in an in vivo transplantation setting. More detailed functional studies on these neurons will also be necessary to clarify if they are as efficient at producing and releasing dopamine, and can elicit the expected behavioral responses after grafting, despite partial or complete lack of alpha-synuclein. If that proves to be the case, it is possible that these types of genetically modified stem cell-derived neurons will be used in future clinical transplantation trials in PD. Both authors are grateful for the support of the Van Andel Institute. P.B. has received commercial support as a consultant from Renovo Neural Inc., Fujifilm-Cellular Dynamics, Axial Biotherapeutics, CuraSen, Living Cell Technologies Limited, Roche, Teva Inc, Lundbeck A/S, NeuroDerm, AbbVie, IOS Press Partners. He is conducting sponsored research on behalf of Roche and Lundbeck A/S. He has ownership interests in Acousort AB. G.A.C. reports no conflicts of interest.