Alpha‐Synuclein as a Biomarker for Parkinson's Disease

Introduction

Parkinson's disease (PD) is a prevalent, chronic and progressive neurodegenerative disorder affecting approximately 1% and 4% of the general population over the ages of 60 and 80 years, respectively 25. It is the second most common serious neurodegenerative disease after Alzheimer's disease (AD) 25. Clinically, PD is characterized by motor symptoms, for example, bradykinesia, resting tremor and rigidity, as well as non‐motor symptoms such as anosmia, constipation, autonomic failure, depression and cognitive dysfunction 52, 100. Diagnosis is largely based on clinical symptoms, but definitive confirmation of the disease requires pathological examination at autopsy, where progressive degeneration of dopaminergic neurons in the substantia nigra, along with Lewy bodies in surviving neurons, is observed 76, 100, 103. However, PD shares common neuropathological, cognitive and clinical profiles with several other neurodegenerative diseases (Table 1), complicating its diagnosis. As a result, at least 15% of patients clinically diagnosed with PD do not meet strict clinical criteria for PD 94 and post‐mortem pathological examination of the brains of clinically diagnosed PD patients shows a different diagnosis in up to 35% of cases 46, 47, 67, 92. Additionally, dementia is present in 10% to 80% of PD patients depending on the stage of the disease 1, 3, such that the cognitive profile of PD with dementia (PDD) overlaps substantially with dementia with Lewy bodies (DLB), AD, and other diseases including multiple system atrophy (MSA).

Moreover, PDD and DLB, in addition to similar cognitive profiles, share a common pathological hallmark: ubiquitin‐positive neuronal Lewy bodies. Typical Lewy bodies are 5–25 microns in diameter, contain a dense eosinophilic core with a surrounding halo 53 and are found in the substantia nigra, as well as in regions such as the dorsal motor nucleus of the vagus, nucleus basalis of Meynert, the locus coerulus and more diffusely in the brain during the later stages of PD 35. Cortical Lewy bodies are usually less prominent on routine pathology examination and do not contain an eosinophilic core 53. Lewy neurites contain filaments similar to those of Lewy bodies, and the protein alpha synuclein (α‐syn) is one of the most prominent components of both 2, 10.

Earlier diagnosis and differentiation from similar diseases presents a serious problem in the care of PD patients. A significant proportion of nigral neurons are lost prior to the onset of motor symptoms 58, 59, meaning that clinical diagnosis is likely to occur too late for the administration of disease‐modifying therapies, such as putative treatments that might preserve the surviving population of dopaminergic neurons. Earlier diagnosis would also aid in future research by reliably identifying the correct population of subjects (early or preclinical PD) and subsequently running clinical trials on potential neuroprotective or neurorestorative treatments. Thus, determining the best biomarker for disease detection and progression could aid in accurate diagnosis, monitoring disease progression and response to treatment.

Clearly, clinical assessment of motor symptoms, despite being the current mode of diagnosis, does not overcome a number of obstacles in early diagnosis. Although non‐motor symptoms may arise many years prior to the onset of motor symptoms, and affect a significant portion of PD patients, they are not sufficiently specific to PD to aid in early diagnosis 110. Thus, other diagnostic tools, including structural and functional imaging, and biochemical measurements in body fluids, are under investigation as strategies to improve diagnosis and tracking of progression. Whether changes in brain structure and function that are detectable by current imaging strategies are sufficiently distinctive for PD and clinically useful for diagnosis is still unknown 34, 40. Furthermore, the changes detected could reflect either the disease process or compensatory responses, or might arise from treatment related complications. Imaging techniques also suffer from challenges related to radiotracer ligand exposure, as well as costs that make imaging‐based screening tests impractical. Therefore, detection of biochemical/molecular markers in readily available body fluids could be beneficial clinical tools.

A few widely tested biochemical markers of PD include α‐syn, protein deglycase (DJ‐1), tau, amyloid beta (Aβ), uric acid and glutathione 114. α‐Syn in particular is implicated in the pathogenesis of PD, as it is a major component of Lewy bodies. Additionally, point mutations in the SNCA gene that codes for α‐syn lead to familial PD 5, 88, as do relatively rare duplications or triplications of the SNCA locus, which result in elevated α‐syn in both brain and blood 73. Moreover, variants in the gene have been identified as a risk factor in Genome‐Wide Association Studies (GWAS) 48. Thus, α‐syn in blood, plasma or cerebrospinal fluid (CSF) is a promising candidate as a biomarker for diagnosing PD.

Structure and Physiological Function of α‐Syn

Synucleins are a family of small soluble proteins, which include alpha‐, beta‐ and gamma‐synuclein. α‐Syn, a relatively small (14 kDa) protein, is largely expressed in the brain at presynaptic terminals 49, 79, and thus plays a role in modulating the stability of the neuronal membrane, influencing presynaptic signaling, and membrane trafficking through vesicular transport 19. Furthermore, α‐syn accounts for up to 1% of total protein in soluble cytosolic brain fractions, suggesting that it may play a significant role in neuronal function 49. Although α‐syn is primarily an intracellular protein, it is also found in biological fluids such as CSF, blood and plasma 18, 32, 63, 74, 107.

At least four α‐syn isoforms, with varying numbers of amino acids and aggregation potential, are encoded by the same SCNA gene in humans and produced via alternative splicing 71. The major variant, containing 140 amino acids, retains all sites that undergo post‐translational modification 64, while the shorter isoforms (with 112, 126 or 98 amino acids) 17 do not, and may be at greater risk of abnormal aggregation 64. The whole transcript (α‐syn 140) can be divided into two functionally distinct regions, an N‐terminal and C‐terminal region. The N‐terminal region (residues 1 to 103), responsible for lipid binding, contains amphipathic apolipoprotein helical motifs that can bind to phospholipids and adopt a helical conformation 33. This N‐terminal region includes the non‐amyloid‐β component (NAC), a hydrophobic region responsible for protein–protein interactions, which allows α‐syn to undergo conformational change from a random coil to an aggregation prone β‐sheet structure, subsequently leading to amyloid fibril formation 33. The unfolded C‐terminal region (103–140) is believed to be responsible for mediating interactions of α‐syn with other cytosolic or membrane bound proteins 33.

Factors such as oxidative stress 43, proteolysis 30, 65, fatty acid concentration 55, 87, phospholipids and metal ions 43, 82 can modulate α‐syn structure, leading to alternative formations of the protein, which include oligomers and fibrils, the latter of which can develop into cytoplasmic inclusions 112. Additionally, post‐translational modifications, such as phosphorylation, ubiquitination, nitration and truncation can also result in altered protein size, structure or charge 16. Because approximately 90% of insoluble α‐syn in Lewy bodies is phosphorylated 22, 38, phosphorylation has been the most widely studied post‐translational modification. Although phosphorylation at Ser‐129 (a major phosphorylation site) is characteristic of PD and related synucleinopathies, whether phosphorylation affects fibril formation, a necessary step in the formation of cytoplasmic inclusions and subsequently Lewy bodies 101, is still unclear 83. A majority of in vitro studies suggest that authentic phosphorylation at Ser‐129 8, 38, 101, 106, 117, but not phosphorylation‐mimicking mutations 83, 95, leads to an increase in fibril formation. In vivo, phosphorylation at Ser‐129 was not correlated with alterations in fibril formation 9, 22, 70, 93, though outcomes are somewhat complicated by differences in model, and the use of phosphomimetic mutation versus genuinely phosphorylated α‐syn. Despite differences in findings between studies assessing fibril formation, the general consensus from animal models is that phosphorylation at Ser‐129 produced neurotoxic effects 22, 70, 93.

α‐Syn in Parkinson's Disease Versus Healthy Aging

α‐Syn, and subsequently Lewy bodies, are markers of other neurodegenerative diseases collectively termed α‐synucleinopathies, a group which includes PD (with or without dementia, ie PDD), DLB, Lewy body variant of AD and MSA. In PD, PDD and DLB α‐syn inclusions are mainly seen in cell bodies and in axonal processes of neurons at different stages 111. However, PDD and DLB cannot be distinguished pathologically 53. Comparatively, in MSA the inclusions are mainly seen in the cytoplasm of glia and neurons, with the majority of α‐syn inclusions observed in oligodendrocytes 53. Lewy bodies are also present in autopsy brain specimens of 10% to 12% of healthy elderly, aged 60 years and above, with no clinical signs of PD 57, but it remains controversial whether these cases represent pre‐clinical PD patients. Notably, although α‐syn aggregation is present in other neurodegenerative diseases, for the purpose of this review, only studies focusing on patients diagnosed with PD will be discussed.

Measurement of α‐Syn in Peripheral Tissue and Body Fluid

Pathological α‐syn in solid tissues, such as within Lewy bodies and Lewy neurites, has been visualized in multiple peripheral tissues using immunohistochemistry. α‐Syn has also been identified in CSF, plasma and saliva, and thus is readily secreted into extracellular spaces 62, 68, 78; a fraction of this secreted protein has been linked to exosomes, membrane vesicles of endocytic origin 7, 23, 24, 98. Total α‐syn, as well as its oligomeric and phosphorylated forms, can be measured in body fluids using western blot, enzyme‐linked immunosorbent assay (ELISA), Luminex assay or mass spectrometry, with each technique having its advantages and disadvantages (Table 2).

Discussion

Finding an adequate biomarker to aid in the diagnosis of PD as well as monitoring progression would be beneficial for the implementation and development of neuroprotective therapies, particularly at early stages, when diagnosis based on clinical characteristics alone is difficult. Although relatively invasive, tissue biomarker(s) could work in conjunction with clinical assessment and imaging to accurately diagnose and differentiate PD from other neurodegenerative diseases. Current studies suggest that using tissue α‐syn as a biomarker may be beneficial in PD diagnosis, although its role in monitoring progression needs to be further validated 105, 107, 113. Of the tissues tested, the gut mucosa 60, 61, 90, 91 and salivary glands 14, 21, 26 may be the most promising candidates for α‐syn detection and subsequent PD diagnosis. Nevertheless, different methods of α‐syn analysis (particularly when analyzing positive α‐syn staining), severity of PD between studies, site of biopsy as well as sample collection could have possibly led to variations in study outcomes. Thus, until sample collection protocols and methods of α‐syn analysis are standardized, the use of tissue biopsies in the diagnosis of PD may be challenging. Specificity of antibodies, methods of detection and technologies used, as well as cohort size 99 also contribute to variability, and must be addressed in future studies. Additionally, longitudinal studies are required to definitively address PD progression.

Assessment of CSF α‐syn in PD patients and controls has produced fairly consistent results, with a majority of studies (80%) demonstrating a reduction in total CSF α‐syn in PD patients compared with controls 42, 45, 54, 74, 75, 85, 86, 107, 113. Studies assessing the different variants of α‐syn are even more consistent. It appears that while monomeric α‐syn is not affected in PD 18, 51, both oligomeric 84, 86, 108 and phosphorylated 105, 113 α‐syn are increased in PD patients compared with controls. The general consensus is that CSF α‐syn may serve as a valuable marker of PD, but whether it can serve as a marker of disease progression/severity is still unknown as studies have shown inconsistent results 45, 105, 107, 113. While mechanisms underlying changes of various forms of α‐syn (total, monomeric, oligomeric and phosphorylated α‐syn) are largely unclear currently, a more direct comparison could be improved when all variants of α‐syn are measured simultaneously to minimize confounding factors of methods of detection. Finally, if future longitudinal studies take into consideration factors such as disease severity, methods of analysis, sample collection as well as extent of blood contamination it may be possible to accurately assess the use of CSF α‐syn as a marker of disease progression.

CSF α‐syn may also be beneficial in differential diagnosis of Parkinsonism caused by various neurodegenerative disorders. For example, Mollenhauer and colleagues 74 found that measurement of a combination of CSF α‐syn together with tau protein was valuable in distinguishing synucleinopathies (PD, DLB, MSA) from other neurological disorders. Similarly, an earlier study suggests that the mismatch in CSF α‐syn and tau, that is, higher tau accompanied with lower α‐syn, is associated with AD 109. Therefore, the measurement of CSF α‐syn along with the calculation of α‐syn to tau ratio could possibly distinguish PD from AD. Similarly, analysis of a combination of other CSF biomarkers along with α‐syn, such as Aβ₄₂, Flt3 ligand and fractalkine has been shown to be valuable not only in distinguishing PD from controls but also from patients with AD and MSA 97. As it appears that α‐syn improves the diagnostic sensitivity and specificity of other markers, measurement of additional biomarkers along with α‐syn may be essential in separating PD from other neurodegenerative diseases, diagnosing PD and possibly determining disease severity.

Despite many advantages of CSF biomarkers, CSF cannot be readily obtained in most clinical settings, especially in developing countries or remote areas of developed countries, and it is not the best specimen for routine monitoring due to the invasive nature by which it is obtained 99. To this end, plasma and serum, which are more accessible, could be valuable biofluids for use in biomarker detection. The current inconsistencies in studies assessing plasma α‐syn are most likely a result of hemolysis, contamination of platelets in plasma, inadequate age matched controls, or differences in detection methods and sensitivity or accuracy of antibodies used 99. If these factors are taken into consideration, plasma α‐syn levels may serve as a valuable tool in PD diagnosis. A way to overcome some of these concerns may be to assess exosomal α‐syn, which showed diagnostic sensitivity and specificity comparable to those determined by CSF α‐syn 98, making it potentially valuable in PD diagnosis and determining severity. Although measuring α‐syn in whole blood may represent another way of eliminating confounding factors, the feasibility of this strategy will depend on careful implementation to minimize variability. Thus, further studies assessing the effects of PD on blood α‐syn are required, while taking into consideration method of α‐syn detection as well as RBC count, which could lead to alterations in α‐syn levels. Additionally, while saliva is also readily accessible, further studies are required to assess saliva α‐syn while taking into consideration saliva collection methods and methods of analysis.

Finally, assessing the levels of different variations of α‐syn, particularly phosphorylated α‐syn, is important as studies suggest that the level of phosphorylated α‐syn (PS‐129) appears to be more effective than total α‐syn in differentiating PD from other neurodegenerative diseases and better correlates with disease severity 105, 113. However, further analysis is required to fully understand its changes in PD and related diseases, and whether it may be used in conjunction with other modifications remains largely uninvestigated.

Conclusion

Due to the current difficulty in diagnosing early‐stage PD, there is urgent need to identify biomarkers in readily obtainable biological specimens that can reliably diagnose PD or even objectively trace disease severity. The results reported on α‐syn concentrations in solid tissue samples, biopsies and biofluids and its correlation with PD suggest that an early diagnostic and severity biomarker is possible, with CSF and plasma exosomal α‐syn as the most promising candidates. The exploration of a combination of biomarkers and how they can work together to improve diagnostic accuracy is required, as there may not be one unique biomarker for PD. Additionally, in the future, further longitudinal studies with adequate controls are required to determine if the current leading candidate biomarkers of PD can adequately diagnose PD as well as if they correlate with PD severity.