Illuminating synucleinopathies: Advances in α-synuclein PET tracer development for in vivo neuroimaging
Abnormal α-synuclein aggregation is a pathological hallmark of Parkinson's disease, multiple system atrophy, and dementia with Lewy bodies. A suitable radiotracer that can noninvasively map synucleinopathies through positron emission tomography (PET) will lead to breakthroughs in early diagnosis, monitoring disease progression, and evaluating treatment responses. However, the development of PET tracers for α-synuclein is lagging due to several challenges. In this perspective, we provide a brief review of the advancements in PET tracers targeting α-synuclein and summarize recent clinical studies aimed at mapping synucleinopathies in neurodegenerative patients using these PET tracers.
Introduction
Aggregation of α-synuclein is found in Lewy bodies (LBs) and Lewy neurites (LNs), which are the pathological hallmarks of Parkinson's disease (PD) and related disorders known as synucleinopathies (1, 2). However, the regional distribution, conformation, and seeding capacity of synucleinopathies are heterogeneous among the diseases. In PD, neuronal inclusions of LBs and LNs first emerge in the brainstem, spreading through the midbrain/substantia nigra to the medial temporal cortex, and through the mesocortex to the neocortex (3). In comparison, the cellular pathology of dementia with Lewy bodies (DLB) is characterized by glial cytoplasmic inclusions found in oligodendrocytes, which are particularly prominent in the white matter of the brainstem and cerebellum (4). SNCA gene located in chromosome region 4q21-4q23 is responsible for encoding the α-synuclein protein in humans. The protein consists of three distinct regions: (1) an amphipathic domain (residues 1–60) that contains apolipoprotein lipid-binding motifs, which are predicted to form amphiphilic helices, thereby conferring a propensity to adopt α-helical structures upon membrane binding, (2) a non-amyloid β-component (NAC) (residues 61–95), which is crucial for potential β-sheet aggregation, and (3) an acidic domain that is highly negatively charged and prone to being unstructured (Figure 1) (5–7). In 1997, Polymeropoulos et al. identified the first missense variant in SNCA, leading to an A53T amino acid change and, as a result, a prolonged β structure that is prone to aggregation (8). Since then, several other point mutations in SNCA have been reported, including E46K (9), G51D (10), E83Q (11), and V15A (12) depicted in Figure 1. Nevertheless, a common mechanism by which SNCA point mutations might result in synucleinopathies has yet to be discovered, indicating that the diverse genetic architecture may influence clinical and pathological presentations through distinct signaling pathways (13). In addition, a diverse set of variants contributing to mitochondrial and mitophagy function, lysosomal and trafficking pathways, and so on, are inversely or directly correlated with the level of synucleinopathies, such as LRRK2, PARK7 (also known as DJ-1), PRKN (also known as PARK2), and PINK1 (14).
Citation: Genomic Psychiatry 2025; 10.61373/gp025p.0032
Molecular probes with an appropriate affinity, high selectivity, and specificity in vivo for α-synuclein will be useful in the understanding and monitoring of synucleinopathy-related diseases using positron emission tomography (PET). This could be exemplified by radioligands targeting tauopathies, which provide PET-based Braak staging as an effective method to differentiate between phases of the AD continuum (15). Our recent studies have shown that synaptic loss in the brain measured by [18F]SynVesT-1, a 18F-radiolabeling radiotracer that targets synaptic vesicle glycoprotein 2A, correlates with clinical, fluid, and imaging biomarkers of neurodegeneration (16, 17). This finding supports the use of PET imaging as a valuable tool for assessing interconnected pathologic processes, including pathologic structures, neuroinflammation, and synaptic dysfunction. Furthermore, in vivo PET quantification can expedite drug discovery and development by providing valuable information on accessing target occupancy and monitoring treatment feedback. Aβ-targeting monoclonal antibodies such as aducanumab and lecanemab gained accelerated FDA approval based on the reductions of amyloid-β (Aβ) PET signals in clinical trials. Undoubtedly, there is much interest to develop radiotracer for mapping synucleinopathies in the brain regions. In this perspective, we aimed to briefly review the development of PET tracers targeting α-synuclein (Figures 2 and 3) and highlight recent advances that may illuminate the path for future development.
Citation: Genomic Psychiatry 2025; 10.61373/gp025p.0032
Citation: Genomic Psychiatry 2025; 10.61373/gp025p.0032
Thioflavin-T derivative [11C]Pittsburgh compound-B ([11C]PIB) and benzoxazole [18F]BF227 were first investigated as non-selective probes for β-sheet structures of α-synuclein aggregates. Using α-synuclein or Aβ1−42 fibrils, [18F]BF227 showed high binding affinity to β-sheet structures of both species with two classes of binding sites on Aβ1−42 fibrils (dissociation constants Kd1 = 1.31 and Kd2 = 80 nM, respectively) and one class of binding sites on α-synuclein fibrils (Kd = 9.63 nM) (18). Fluorescent BF227 staining of the substantia nigra from patients with PD showed also colocalization with immunohistochemistry staining using an α-synuclein–targeting antibody. However, no binding of [18F]BF227 was detected in pure DLBs homogenates in the absence of Aβ plaques. This observation suggests that the structure of recombinant fibrils may not accurately reflect the structure of fibrils existed in brain homogenates, particularly due to the potential impact of posttranslational modifications and protein interactions on the conformation of β-sheet structures and the accessibility of binding sites in vivo. Bagchi and Yu et al. subsequently demonstrated that the phenothiazine analog [125I]SIL23 bound to a site identified on recombinant α-synuclein fibrils, as well as to fibrillar α-synuclein in LBs and LNs found in the brains of patients with PD (19). The density of corresponding binding site was proven to be sufficiently high to be detected by PET imaging using high affinity radioligands. However, not yet a tricyclic structure based on SIL23 has been reported for imaging synucleinopathies in patients. [18F]46a reported by Chu et al. was optimized from selective fluorescent dyes and displayed good selectivity on α-synuclein fibrils (20). The high lipophilicity and potential reduction of the nitro group in the structure were believed to contribute to a high level of nonspecific binding in vivo, hindering its application as a suitable PET probe for neuroimaging. Inhibition pathological aggregation of prion protein and α-synuclein using diphenylpyrazole led to the discovery of [11C]anle253b, from which [11C]MODAG-001 was developed (21, 22). In dynamic PET imaging of normal mice, [11C]MODAG-001 demonstrated good brain penetration; however, several radiometabolites were identified in the brain homogenates at 5 min postinjection. Further in vitro autoradiography studies showed no significant binding on brain sections from patients with Lewy body dementia using [3H]MODAG-001. Like [18F]46a, its high logD value was proposed as the reason for the low signal-to-noise ratio in human brain tissue with synuclein pathology. Arylpyrazolethiazole derivative [11C]APT-13 was recently reported to have a Ki value of 27.8 ± 9.7 nM and a 3.3-fold selectivity over Aβ. In preliminary studies, it exhibited high initial brain penetration in healthy mouse brains and emerged as a lead for further development (23). Diarybisthiazole derivative [18F]FS3-1 showed promising sensitivity in a rat model overexpressing human E46K-mutated α-synuclein (24).
Meanwhile, numerous structures containing polyphenols such as gallic acid and flavonoids such as quercetin have been found to have good inhibitory effect on α-synuclein aggregation (25, 26). It's hypothesized that phenol or catechol group promotes the interaction with α-synuclein–containing species. Accordingly, the dimethylamino group on BF227 may be accountable for its affinity toward Aβ fibrils, sabotaging its specificity. A novel PET tracer targeting α-synuclein ([18F]F-0502B, Figure 2) has recently been reported (27). It is structurally similar to [18F]BF227 but features a free phenol group, with molar activities ranging from 18.5 to 37 GBq/μmol. Using saturation binding assays, [18F]F-0502B exhibited Kd values of 3.68, 107.8, and 151.2 nM in brain homogenates from PD, Alzheimer's disease (AD) with Aβ fibrils and AD with tau fibrils, respectively, demonstrating its specific interactions with α-synuclein fibrils. In healthy non-human primates, [18F]F-0502B showed an initial brain uptake with a standardized uptake value (SUV) below 1.5 and fast washout within 5 min postinjection. The tracer was further investigated in non-human primate models of PD with nigrostriatal degeneration induced by intracranial injection of adeno-associated virus encoding A53T mutant human α-synuclein (AAV-A53T-α-Syn) or preformed fibrils of α-synuclein. PET images averaged from 30 to 60 min postinjection of [18F]F-0502B showed a higher radioactivity accumulation in the striatal regions of PD models compared with the control group. Further investigations, including radiometabolite analysis, kinetic modeling, and human translation, are warranted to see whether [18F]F-0502B could serve as a useful tracer for imaging α-synuclein aggregates in living patients. Notably, Maurer et al. disclosed the development of a library of 2-styrylbenzothiazoles in 2023 (28). Using human recombinant α-synuclein fibrils and [3H]PIB, less lipophilic analog MFSB also exhibited enhanced affinity to α-synuclein aggregates (Ki = 10.3 ± 4.7 nM) compared with that of PFSB. PET imaging demonstrated significant brain penetration of [18F]MFSB, with a SUV of 1.79 in healthy wild-type mice. However, the slow washout of [18F]MFSB from the brain, along with increased radioactivity accumulation in white matter–rich areas such as the midbrain and brainstem, indicates a high degree of nonspecific binding in vivo.
[18F]SPAL-T06 with an (E)-hex-2-en-4-yne linker in the backbone structure was reported by Higuchi et al. (29). The affinity of [18F]SPAL-T06 was determined to be 2.49 nM using putamen homogenates of patients with multiple system atrophy (MSA) with predominant parkinsonism. It successfully visualized the synucleinopathies in patients with MSA without Aβ deposition. However, its rapid metabolism and, consequently, insufficient intact radioligand in the bloodstream preclude its ability to capture α-synucleinopathies in cases of PD and DLB with low target abundance. The same group further developed [18F]C05-05 to overcome the rapid clearance issue of [18F]SPAL-T06 (30). In in vitro autoradiography, [18F]C05-05 showed specific accumulation in the amygdala from a patient with DLB and substantia nigra from a patient with PD with dementia. The concentrations of this tracer to induce 50% homologous inhibition were determined to be 1.5 and 1.7 nM in DLB and MSA homogenates, respectively. Ten patients meeting clinical diagnostic criteria for PD or DLB, and eight healthy controls were included for an exploratory clinical PET study of [18F]C05-05. The study focused on the synucleinopathies in the midbrain, as the midbrain substantia nigra is a common area affected by Lewy pathologies in ‘‘body-first’’ and ‘‘brain-first’’ subtypes of PD and DLB at a clinical stage. Using the deep white matter as the reference region, subjects in PD/DLB group showed significantly higher ratios of SUVmidbrain/SUVreference region, which is well correlated with the degree of motor impairments assessed by Movement Disorder Society revised Unified Parkinson's Disease Rating Scale part III scores. Although patients with AD pathologies were pre-excluded, this study provided the first essential evidence on the capability of PET probe for imaging α-synuclein pathologies in humans. However, it is important to note that both [18F]SPAL-T06 and [18F]C05-05 interact with the groove-like binding pocket in the β-sheet structure of α-synuclein fibril cores. This binding characteristic complicates the achievement of high selectivity over tau pathology, due to the significant resemblance of cross-β structures. Consequently, the accumulation of radioactivity from these tracers in vivo will be confounded by the presence of tau aggregations. Meanwhile, the increased nonspecific binding of [18F]C05-05 to myelin components may potentially elevate background noise in other regions with high white matter fractions. Further studies in a relatively large cohort are warranted to investigate whether the aforementioned factors have a profound impact on its clinical value regarding disease progression and companion diagnosis.
[18F]ACI-12589, developed by the biotech company AC Immune, was recently published and demonstrated promising results in distinguishing MSA from other neurodegenerative diseases (31). By in vitro autoradiography, the Kd values of [3H]ACI-12589 was estimated to be 17 nM using brain tissues from a familiar PD and 28 nM using brain tissues from a MSA case. High-resolution autoradiography revealed radioactive accumulation on individual α-synuclein inclusions, aligning with the pathologies identified through immunohistochemical staining. Using AD tissues with Aβ and tau aggregates, [3H]ACI-12589 showed a Kd value of 300 nM and a low maximal binding capacity, indicating its excellent specificity towards α-synuclein pathologies. In the preliminary clinical studies, compared to patients with PD, DLB, and healthy controls, participants diagnosed with MSA exhibited a greater retention of [18F]ACI-12589 in the cerebellar white matter, particularly in the phenotype dominated by parkinsonism (MSA-P) rather than in the phenotype dominated by cerebellar ataxia (MSA-C). The low density of α-synuclein pathologies, along with the variations in conformation and posttranslational modifications across different synucleinopathies, may explain the lack of specific accumulation in PD and DLB. Impressively, the study further included participants with progressive supranuclear palsy (PSP, three cases), hereditary ataxias (two cases), and AD (five cases). The radioactivity accumulation of [18F]ACI-12589 exhibited overlaps with pathologies identified by the tau tracer [18F]RO948, but showed a weak correlation with the positive regions observed by the Aβ PET tracer [18F]flutemetamol. Similarly, the retention of [18F]ACI-12589 in PSP matched the expected tau pathology. This may be clarified by the co-pathologies of α-synuclein in AD and PSP. Further characterizations of the tracer's binding sites in various neurodegenerative diseases, along with an in-depth clinical investigation involving larger patient cohorts, would provide an answer.
Immense efforts have been devoted to developing α-synuclein PET tracers with enhanced selectivity. Mach et al. introduced heterocyclic moieties, such as diazaspirocyclic or bridged amino cores, to replace the piperazine in the nonselective lead compound (32). The key structure-affinity relationship is illustrated in Figure 3A. In comparison to compound 4o, linking the 4-methoxy-N-phenylbenzamide and pyridine with either 2,8-diazaspiro[4.5]decane (4f), 1,4-diazepane (4m), or 3,8-diazabicyclo[3.2.1]octane (4i and 4p) enhanced the selectivity of the structures toward α-synuclein. [11C]4i was subsequently obtained with high molar activity (106 ± 56 GBq/μmol) and peaked brain uptake in non-human primates with SUV values of 1.68 ± 0.54 at 4 min postinjection. In vitro binding assays using [3H]4i suggest an off-target to 4R tau, which may limit the application of [11C]4i in certain circumstances. [3H]asyn-44, featuring a pyridothiophene core structure, was reported by Neil et al. to have a potent Ki value of 1.85 nM using PD homogenates (Figure 3B, left) (33). In in vitro autoradiography studies, it generated a distinct radioactive signal in brain sections from MSA and PD, aligning with neuropathology visualized through anti-pS129 α-synuclein immunohistochemistry. The corresponding PET tracer [18F]asyn-44 was hindered by the penetrance of radiometabolite in the brain, preventing further evaluations. Imidazo[2,1-b][1,3,4]thiadiazole derivatives were proposed by Cui et al. as a novel scaffold, and [18F]FITA-2 was screened out with moderate affinity (IC50, α-synuclein = 245 nM, Figure 3B, right) (34). It possessed suitable brain uptake with sufficient clearance and good stability in healthy SD rats and is currently being evaluated in patients. Other chemical structures that may fulfill α-synuclein neuroimaging are discussed in recent reviews (35, 36).
In summary, extensive posttranslational modifications in vivo, including phosphorylation, truncation, and acetylation, may lead to changes in aggregation characteristics, such as structure and properties within the binding pocket. This results in varying binding potency of the structures to synthetic α-synuclein aggregates and human tissues. It seems to be an unavoidable trend to include brain sections or homogenates from donors with neurodegenerative conditions for compound screening. Binding assays of this kind include also the inherently low density of synucleinopathies. Molecular docking and photoaffinity labeling, alongside cryo-electron microscopy techniques, may assist in identifying potential binding sites and optimizing structure (32, 37). Second, α-synucleinopathies are often accompanied by Aβ and tau aggregations that share a similar β-sheet structure, making it important and somewhat challenging to achieve selectivity. Recently, immunomagnetic cell sorting following in vivo radiotracer injection dissected the cellular allocation of 18-kDa translocator protein (TSPO)–PET signals in human glioma samples (38). We speculate that this approach may serve as a valuable tool to untangle the sources of radioactive signals in vivo from newly established α-synuclein PET tracers. In addition to the points mentioned above, the development of α-synuclein PET tracers encounters the typical challenges faced by molecular probes for central nervous system. It is essential to ensure sufficient molar activity, adequate brain penetration and to avoid confounding signals from radiometabolites in the brain (39). Nonetheless, the first promising clinical results have been disclosed. We believe that continuous scientific contributions from multiple disciplines will eventually pave the way for the development of α-synuclein PET tracers to illuminate synucleinopathies during disease progression.
Acknowledgments
A special thanks to Dr. Xiaoqing Song (Shanghai United Imaging Healthcare Advanced Technology Research Institute) for constructive scientific interactions.
Author Contributions
Y.H. and L.Q. performed the literature review and wrote the original draft. Q.Y. and X.F. participated in reviewing and editing the manuscript. The manuscript has been read and approved by all authors. No related work is under consideration elsewhere.
Funding Sources
This publication was supported by STI2030-Major Projects (2022ZD0213800) and Shanghai Pujiang Program (23PJ1401500).
Author Disclosures
The authors have confirmed that no conflict of interest exists.
Schematic of α-synuclein domains.
Chemical structures of α-synuclein PET tracers.
(A) Structure-affinity relationship of 4i and its derivatives (32). aThe compound was labeled with carbon-11 and evaluated as a PET tracer targeting α-synuclein. bN.A stands for not active (no displacement at 1000 nM). (B) Chemical structures of [18F]asyn-44 and [18F]FITA-2.
Contributor Notes
Publisher's note: Genomic Press maintains a position of impartiality and neutrality regarding territorial assertions represented in published materials and affiliations of institutional nature. As such, we will use the affiliations provided by the authors, without editing them. Such use simply reflects what the authors submitted to us and it does not indicate that Genomic Press supports any type of territorial assertions.
