Pre-clinical models of alpha-synuclein neuropathology as tools to explore novel neuroprotective strategies

Parkinson’s disease (PD) is a common neurodegenerative disorder characterized by an array of motor symptoms, including bradykinesia, resting tremor, and postural instability, as well as non-motor symptoms such as olfactory and gastrointestinal deficits, REM sleep disorder, and cognitive impairment. Pathological features of PD include the degeneration of dopaminergic neurons that project from the substantia nigra (SN) to the striatum, referred to as nigrostriatal degeneration, as well as the presence of inclusions referred to as Lewy bodies (LB) and Lewy neurites (LN) largely comprised of fibrillar forms of the presynaptic protein alpha-synuclein (aSyn).  Although the motor symptoms of PD can be managed satisfactorily for several years, there are as of yet no disease-modifying therapies for PD patients, in large part because the root causes of the disease are incompletely understood.

In PD and other synucleinopathies, aSyn undergoes accelerated aggregation in the presence of phospholipid membranes by adopting an exposed alpha-helical structure, a state that favors membrane-induced self-assembly. In turn, aSyn aggregation at membrane surfaces can cause a disruption of dopaminergic vesicles, a potential cause of the nigral dopaminergic neuronal death observed in the brains of PD patients. Endosulfine-alpha (ENSA), a cAMP-regulated phosphoprotein expressed throughout the body including the CNS and is known to be downregulated in the brains of individuals with Alzheimer’s disease, Down syndrome, dementia with Lewy bodies, or PD interacts with aSyn at the surface of phospholipid membranes and interferes with aSyn aggregation, aSyn-mediated vesicle disruption, and aSyn neurotoxicity although these inhibitory effects are abrogated by the S109E substitution mimicking protein kinase A-mediated ENSA phosphorylation of residue S109.  In the studies outlined in this thesis, we have characterized multiple pre-clinical models of aSyn neuropathology and used a subset of these models to examine potential neuroprotective effects of ENSA in synucleinopathy disorders.  

In one set of studies (outlined in Chapter 3), we examined effects of ENSA expression in an in vivo rAAV-aSyn model. As a starting point, we established an rAAV-aSyn model in our lab by carrying out a study in which we characterized rats expressing human WT aSyn and two familial aSyn mutants (A53T and A53E) in the substantia nigra (SN) using behavioral and histopathological endpoints. We observed that rats expressing all three –aSyn variants showed evidence of motor deficits, nigrostriatal degeneration, and aSyn neuropathology. These results set the stage for delineating potential neuroprotective effects of ENSA expression in this model. Following a pilot study that was conducted to optimize the viral load for adequate protein expression without non-specific toxicity, rats were injected in the SN with rAAV-stuffer (vector control virus), rAAV-aSyn + rAAV-stuffer, or rAAV-aSyn + rAAV-ENSA-WT, -S109A, or -S109E. Unexpectedly, nigral aSyn expression levels in midbrain sections prepared 19 weeks after the injections were found to be markedly lower in rats co-transduced with rAAV-aSyn and rAAV-stuffer compared to animals co-expressing aSyn and each of the ENSA variants, apparently because of a non-specific effect of the high viral load on aSyn expression or clearance. Our observation that animals co-expressing aSyn and ENSA had higher aSyn levels implied that ENSA had a stabilizing effect on aSyn, perhaps by interacting with the protein on phospholipid membranes. Rats co-expressing aSyn and the ENSA variants showed evidence of striatal DA terminal depletion and the formation of aggregates enriched with aSyn phosphorylated on residue S129 (pSer129-aSyn), a known pathological form of the protein, presumably because of the elevated aSyn levels in the brains of these animals. Stereological analysis revealed a trend towards a decrease in the number of neurons staining positive for the dopaminergic marker tyrosine hydroxylase (TH) in the SN of rats co-transduced with rAAV-aSyn and rAAV-stuffer or – animals co-expressing aSyn and each of the ENSA variants, although additional immunohistochemical data obtained by staining nigral sections for the general neuronal marker HuC suggested that the results reflected a downregulation of TH rather than an overt loss of neurons. Data from behavioral tests showed evidence of motor deficits in animals expressing human ENSA-WT, -S109A but not –S109E, potentially because of modulatory effects of ENSA on dopaminergic neurotransmission in the striatum that were ablated by the pS109-mimicking glutamate substitution. Overall, these results revealed for the first time a stabilizing effect of ENSA on aSyn, as well as the ability of ENSA to modulate motor function linked to striatal neurotransmission, in an in vivo model.

In the studies outlined in Chapter 4, we investigated the effects of ENSA over-expression or down-regulation on aSyn neurotoxicity and the propagation of aSyn aggregates in rat primary neuronal cultures and human iPSC-derived neuronal cultures. We hypothesized that ENSA over-expression should alleviate aSyn neurotoxicity and interfere with aSyn propagation, whereas ENSA knockdown or gene disruption should have opposite effects. We observed that ENSA over-expression interfered with the formation of insoluble cytosolic pSer129-aSyn inclusions in rat cortical neurons treated with preformed fibrils (PFFs) prepared from the recombinant mouse protein. Knockdown of ENSA sensitized cultured rat neurons to toxicity associated with virally-mediated aSyn expression and to PFF-mediated aSyn propagation. The propagation of aSyn aggregates also occurred with enhanced efficiency in PFF-treated cortical neurons derived from human iPSCs with a CRISPR/Cas9-mediated ENSA gene disruption compared to the corresponding parental iPSCs. This phase of my PhD research revealed that ENSA up-regulation could be a viable therapeutic strategy to alleviate aSyn neurotoxicity and the propagation of aSyn aggregates in the brains of PD patients. Moreover, our evidence from these cell culture studies that ENSA interferes with PFF-mediated aSyn propagation provides a strong premise for investigating this inhibitory activity in in vivo models of PFF-induced aSyn pathology.

In the studies outlined in Chapter 5, we explored molecular mechanisms that could account for the increased risk of PD associated with traumatic brain injury (TBI). Data from previous studies suggest that some TBI patients develop PD symptoms and have PD-related pathological features (including nigrostriatal degeneration and the presence of aggregated forms of aSyn and the microtubule-binding protein tau) that are evident upon post-mortem examination of the brain. We hypothesized that aSyn expression, leading to aSyn aggregation, plays a central role in nigrostriatal degeneration and comorbid tau pathology associated with TBI.  To test this hypothesis, we characterized a rat in vivo TBI model and a TBI-on-a-chip cell culture model in terms of PD-related neuropathology, and we examined the effects of aSyn downregulation on tau accumulation in cultured neurons subjected to an impact injury. Rats subjected to a single- or triple-blast injury showed evidence of (i) an apparently reversible decrease in TH+ terminal density in the striatum; and (ii) aSyn pathology in the SN and cingulate gyrus, as well tau pathology in the striatum. By analogy with the co-morbid aSyn and tau pathology observed in the brains of blast-injured rats, we found that aSyn and tau were both up-regulated in neurons subjected to an impact injury in the TBI-on-a-chip model. Neurons depleted of aSyn via shRNA-mediated knockdown prior to the impact showed reduced tau up-regulation, suggesting that aSyn expression was necessary for the observed tau up-regulation in this model. Overall, the studies outlined in this chapter led to the development of two pre-clinical TBI models with defined neuropathological features, setting the stage for future studies of candidate therapeutic strategies including targeting aSyn aggregation via ENSA expression.

In summary, the results presented here have been instrumental in validating multiple models of aSyn neuropathology, and in shedding light on the neuroprotective potential of ENSA in PD and other synucleinopathy disorders. Data from additional studies involving in vivo models of aSyn aggregate propagation would further advance our understanding of the neuroprotective role of ENSA, which has so far only been explored in cellular models. Ultimately, evidence of ENSA-mediated neuroprotection in vivo would set the stage for designing therapies to slow nigrostriatal degeneration via ENSA up-regulation or activation in the brains of PD and TBI patients.