A Summary of Phenotypes Observed in the In Vivo Rodent Alpha-Synuclein Preformed Fibril Model.

The use of wildtype recombinant alpha-synuclein preformed fibrils (aSyn PFFs) to induce endogenous alpha-synuclein to form pathological phosphorylation and trigger neurodegeneration is a popular model for studying Parkinson's disease (PD) biology and testing therapeutic strategies. The strengths of this model lie in its ability to recapitulate the phosphorylation/aggregation of aSyn and nigrostriatal degeneration seen in PD, as well as its suitability for studying the progressive nature of PD and the spread of aSyn pathology. Although the model is commonly used and has been adopted by many labs, variability in observed phenotypes exists. Here we provide summaries of the study design and reported phenotypes from published reports characterizing the aSyn PFF in vivo model in rodents following injection into the brain, gut, muscle, vein, peritoneum, and eye. These summaries are designed to facilitate an introduction to the use of aSyn PFFs to generate a rodent model of PD-highlighting phenotypes observed in papers that set out to thoroughly characterize the model. This information will hopefully improve the understanding of this model and clarify when the aSyn PFF model may be an appropriate choice for one's research.

INTRODUCTION

Parkinson’s disease (PD) is a neurodegenerative disorder affecting approximately 1%of the population over the age of 60. Characterized by motor disturbances as well as non-motor symptoms, the pathology of PD involves deposits of aggregated, phosphorylated alpha-synuclein (aSyn) protein in affected tissues and brain structures and degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNpc). Given that PD is a human-specific condition, various models have been developed to enable research and therapeutic development for this disease. Common models include injection of neurotoxins to trigger degeneration of the dopaminergic neurons of the SNpc, transgenic rodent models carrying PD-related genetic mutations, and induction of aSyn pathology through viral vector-mediated overexpression of aSyn, among others [1-3]. All models present with advantages and disadvantages, so selection of the model should be based on the desired pathology for the intended research question.

In the last 10 years, a model has arisen that capitalizes on the observations made by Braak and colleagues that aSyn pathology progressively accumulates in different brain regions following a spatiotemporal pattern that suggests spreading [4-7]. This model, dubbed the aSyn preformed fibril (PFF) model, uses injection of recombinant aSyn protein that has been stimulated to form aggregates and sonicated to produce short fibrils [8-10]. These aSyn PFFs cause templating of endogenous aSyn into pathological species characterized by phosphorylation at S129 (pS129 aSyn), beta-sheet formation, and aggregation, followed by increases in autophagy and neuronal dysfunction [11]. The flexibility of this model allows injection of different forms of aSyn PFFs (e.g., mouse vs. human aSyn, mutated aSyn), unilateral or bilateral injection, targeting of different brain regions and administration through different peripheral routes to model distinct aspects of the disease. This flexibility is a strength of the model but also serves as a weakness, as the distinct protocols lead to different pathologies which has hampered cross-study comparisons. To better understand the various study designs employed for the aSyn PFF model and the resulting pathologies, a survey of the literature was performed and is summarized within this manuscript.

GUIDE TO READING AND INTERPRETING THE TABLES

As hundreds of studies using the aSyn PFF model have been published, Tables 1–9 herein contain information specifically from publications that sought to phenotype the effects of injection of recombinant wildtype aSyn PFFs into rodents to develop a PD model. As a result, the tables are not comprehensive in nature but do contain reports from a variety of studies across laboratories.

Studies focusing on the uptake of aSyn following injection have been excluded as the study is not designed to thoroughly assess resulting pathology. Studies using the aSyn PFF model to test the effect of an intervention have been excluded as the focus is on the therapeutic intervention tested rather than the characterization of the pathological process and timelines. Studies injecting aSyn PFFs to model another disease (e.g., Multiple System Atrophy) were excluded to focus specifically on PD. Studies injecting aSyn PFFs into non-human primates or using aSyn PFFs in cell culture were excluded for the sake of focus. Studies injecting rodent/patient brain-derived material were excluded due to concerns that the injectate is not homogenous and the concentration of aSyn and other protein components cannot be known or compared across studies. Although a number of studies have been published analyzing the differences in pathogenicity of fibrils of different conformations [12-18], different aSyn mutations [19-22], different aSyn truncations [23-25], and different aSyn post-translational modifications [26], these were excluded from the summary tables as the objective of these experiments is to compare pathogenicity relative to wildtype aSyn PFFs and therefore the nuanced information requires a different venue.

Tables 1–9 are organized by categories such as: injected species (mouse vs. rat), route of administration of aSyn PFFs, and species of aSyn PFF (human vs. mouse). To understand the variation in observed phenotypes within the model, readers should compare only within categories rather than across categories. Please note that there may be differences in study design within categories (e.g., unilateral vs. bilateral injection, wildtype vs. transgenic rodent) that should be taken into account when drawing conclusions on timelines and robustness of phenotypes.

Papers included within the tables are organized chronologically, with high-level information on study design, outcome measures, and notes that may provide additional context for the reader. Information on study design includes the rodent strain used, the injectate, the dose of aSyn PFFs with information on whether this dose was administered bilaterally or unilaterally (for bilateral injections, the total dose noted was for each hemisphere), and the days post-injection (DPI) at which time the model was analyzed. Reported phenotypes are separated by category to facilitate comparisons of common readouts across studies. The time post-injection at which the phenotype was observed is included, with a “+” indicating the phenotype was also observed at the later timepoints. If later timepoints were analyzed within the study but the “+” sign is absent, this indicates that either the phenotype was not analyzed at the later timepoints or was analyzed but not observed. If a phenotype was observed in a particular structure, the structure is included in parentheses. Readouts that were not included in the study are denoted as “N/A”. Please note, to fully understand all reported or absent phenotypes in the models, a separate literature review is required.

SUMMARY OF PHENOTYPES REPORTED IN THE ASYN PFF MODEL

The earliest aSyn PFF model studies were performed by injecting aSyn PFFs into the mouse striatum. Table 1 provides a summary of studies that used unilateral or bilateral intrastriatal injection of mouse aSyn PFFs in wildtype (WT) mice. Table 2 provides a summary of studies that used intrastriatal injection of human aSyn PFFs in WT mice. Table 3 provides a summary of studies that used intrastriatal injection in transgenic mice.

Table 1

Injection of mouse aSyn PFFs into the wildtype mouse striatum

aSyn, alpha-synuclein; PFFs, preformed fibrils; TH, tyrosine hydroxylase; DA, dopamine; N/A, not analyzed; SNpc, substantia nigra pars compacta; STR, striatum; AMY, amygdala; ROS, reactive oxygen species.

Table 2

Unilateral injection of human aSyn PFFs into the wildtype mouse striatum

aSyn, alpha-synuclein; PFFs, preformed fibrils; TH, tyrosine hydroxylase; DA, dopamine; N/A, not analyzed; SNpc, substantia nigra pars compact.

Table 3

Unilateral and bilateral injection of aSyn PFFs into transgenic mouse striatum

aSyn, alpha-synuclein; PFFs, preformed fibrils; Hu, human; TH, tyrosine hydroxylase; DA, dopamine; N/A, not analyzed; SNpc, substantia nigra pars compacta; CPu, caudate putamen; CTX, cortex.

Others have chosen to inject non-striatal brain regions to model prodromal or non-motor features of PD in mice. Table 4 provides a summary of studies injecting aSyn PFFs into the olfactory bulb (OB) or sublaterodorsal tegmental nucleus (SLD) to model olfactory dysfunction and sleep disturbances, respectively. Table 5 provides a summary of studies injecting aSyn PFFs into the hippocampus, cortex, and SNpc as alternate ways to induce pathology in the mouse.

Table 4

Unilateral and bilateral injection of aSyn PFFs into the wildtype and transgenic mouse olfactory bulb or sublaterodorsal tegmental nucleus

aSyn, alpha-synuclein; PFFs, preformed fibrils; Hu, human; N/A –not analyzed; OB, olfactory bulb; AON, accessory olfactory nucleus; HPC, hippocampus; SLD, subdorsolateral tegmental nucleus; SNpc, substantia nigra pars compacta; GI, gastrointestinal; RBD, REM sleep behavior disorder; LFP, local field potential; TH, tyrosine hydroxylase; DA, dopamine.

Table 5

Unilateral or bilateral injection of aSyn PFFs into the wildtype or transgenic mouse hippocampus, cortex, or substantia nigra

aSyn, alpha-synuclein; PFFs, preformed fibrils; Hu, human; N/A, not analyzed; KI, knockin; HPC, hippocampus; CTX, cortex; SN, substantia nigra.

Although most studies to date have focused on phenotyping mice injected with aSyn PFFs, rats have also been used for this model. Table 6 provides a summary of studies injecting aSyn PFFs into the rat striatum or SNpc.

Table 6

Unilateral or bilateral injection of aSyn PFFs into the wildtype or knockout rat striatum or substantia nigra

aSyn, alpha-synuclein; PFFs, preformed fibrils; SNpc, substantia nigra pars compacta; WT, wildtype; KO, knockout; TH, tyrosine hydroxylase; DA, dopamine; VMAT, vesicular monoamine transporter; DAT, dopamine transporter; STR, striatum.

In addition, both mice and rats have been used for peripheral administration of aSyn PFFs to study the seeding capabilities of aSyn PFFs and peripheral-to-central spread of synuclein pathology. Table 7 provides a summary of studies injecting aSyn PFFs into the gut of rodents to model GI dysfunction and gut-to-brain transmissibility of aSyn pathology. Finally, Table 8 provides a summary of studies performing intramuscular injections of aSyn PFFs into rodents and Table 9 provides a summary of studies performing intraperitoneal, intravenous, intraneural, and intravitreal injection of aSyn PFFs into rodents.

Table 7

Injection of aSyn PFFs into the wildtype or transgenic rodent gut

aSyn, alpha-synuclein; PFFs, preformed fibrils; Hu, human; DMV, dorsal motor nucleus of the vagus; MG, myenteric ganglia; SC, spinal cord; GI, gastrointestinal system; CNS, central nervous system; SNpc, substantia nigra pars compacta; DA, dopamine; KO, knockout.

Table 8

Unilateral or bilateral injection of aSyn PFFs into the transgenic mouse muscle

aSyn, alpha-synuclein; PFFs, preformed fibrils; Hu, human; KI, knockin; SC, spinal cord; DRG, dorsal root ganglia; CNS, central nervous system; dMRI, diffusion magnetic resonance imaging; fMRI, functional magnetic resonance imaging.

Table 9

Injection of aSyn PFFs into the wildtype or transgenic rodent peritoneum, vein, nerve, or eye

aSyn, alpha-synuclein; PFFs, preformed fibrils; Hu, human; SC, spinal cord; DRG, dorsal root ganglia; CNS, central nervous system; GI, gastrointestinal.

A visual representation of timelines of phenotypes reported in common iterations of the aSyn PFF model is provided in Fig. 1. Replicated phenotypes that have been reported in more than one study are provided along the timeline of the model. Phenotypes that were only investigated in one study are also included but denoted as “underexplored phenotypes”. An inset containing phenotypes that were reported as absent is also included.

Fig. 1

Visual representation of the various phenotypes reported in common iterations of the alpha-synuclein preformed fibril (aSyn PFF) model. Replicated phenotypes (reported in > 1 study) and underexplored phenotypes (observed in only 1 study) are mapped across the timeline of the model. Phenotypes that were investigated but found to be absent are also included in an inset to the right of the table. Italicized phenotypes are those that vary across studies by either their presence/absence (denoted by superscript A) or timing of appearance (denoted with superscript T). For all italicized phenotypes, the most common time at which the phenotype is observed is reported.

DISCUSSION

For all studies, one of the earliest phenotypes reported is the presence of pS129 aSyn within brain regions innervating the injected structure. As the model progresses, the density of pS129 aSyn pathology and regions displaying pS129 aSyn pathology increase. This pathology is at times accompanied by cell loss, inflammation, behavioral deficits, and/or other readouts of pathology. Importantly, the phenotypes observed in this model are not always reproducible and their presence/absence varies between studies (Fig. 1). This can be noted when analyzing the phenotypes listed in Tables 1–9 when comparing studies of similar designs with regard to injection site, unilateral vs bilateral injection, wildtype vs transgenic rodent, etc.

An example of this can be found in motor deficits observed following intrastriatal injection. Despite using the same dose of aSyn PFFs, some report motor deficits following unilateral intrastriatal injection of mouse aSyn PFFs as early as 90 DPI [28, 30, 31] while others do not observe motor impairments until 180 DPI [19, 27] (Table 1). Others still do not observe motor impairments even at 180 days following bilateral injection [35]. Some of these differences may be attributed to the behavioral assays employed. For instance, Henderson et al (2019) used two behavioral tests in the same cohort—grip strength and rotarod—and demonstrated differences in grip strength upon aSyn PFF treatment but no effect of aSyn PFF treatment on rotarod performance [30]. These differences in detecting an effect of aSyn PFF treatment on motor function or non-motor function could be due to the physiology probed within these assays, the sensitivity of the tests, or confounds that may impact the readouts [77].

Another phenotype that greatly varies between studies is pS129 aSyn pathology in the brain following injection of aSyn PFFs to the gut (Table 7). Roughly half of the studies observe pS129 aSyn pathology spread to the midbrain/forebrain [62, 64, 65, 68] whereas the other half observe pathology in the periphery/brainstem that never progresses to the midbrain/forebrain [61, 63, 64, 66, 68]. As mentioned in a recent review by Bindas et al (2021), the reason for this is unclear but could relate to gastrointestinal conditions, amount of pathology generated, site of pathology, and type of pathology induced by the aSyn PFFs [78].

When attempting to understand the variability within the aSyn PFF model, it is important to understand the various factors that can influence the pathogenicity of the aSyn PFFs. Some factors may be obvious and easily accounted for, such as dose or days post-injection. Other factors are not so clear. The source and method of preparing the aSyn PFFs can greatly influence their pathogenicity. Multiple studies have noted that endotoxin may impact the aSyn PFF protein [8, 14, 78]. Endotoxin should not only be accounted for due to its ability to generate an immune response that is independent of the aSyn [8, 79], but also for its ability to alter the structure and pathogenicity of the aSyn fibrils themselves [14]. The buffers, temperature, and sonication protocol used to generate aSyn PFFs from monomeric starting material can also lead to variations in the structure of the PFF aggregates that dramatically affect pathogenicity [12-18]. In addition, downstream steps such as storage (duration and temperature) can impact aSyn PFF performance while injection coordinates can impact the pathology observed in the various structures [8].

Taken together, the aSyn PFF model is a popular model due to its ability to recapitulate the pathological hallmarks of PD through the templating of pathology in the endogenous aSyn protein. The model has been used by many to study PD biology and therapeutic interventions targeting aSyn spread, inflammation, neurodegeneration, etc. Although many groups have adopted the model successfully, it is very important to acknowledge the variation in phenotypes between labs. The tables provided in this paper will hopefully assist groups who wish to learn more about the model and clarify which phenotypes are reproducible between labs to prevent issues in adopting the model for one’s studies.