Native α-Synuclein, 3-Nitrotyrosine Proteins, and Patterns of Nitro-α-Synuclein-Immunoreactive Inclusions in Saliva and Submandibulary Gland in Parkinson’s Disease

Background. Salivary α-synuclein (aSyn) and its nitrated form, or 3-nitrotyrosine-α-synuclein (3-NT-αSyn), hold promise as biomarkers for idiopathic Parkinson’s disease (IPD). Nitrative stress that is characterized by an excess of 3-nitrotyrosine proteins (3-NT-proteins) has been proposed as a pathogenic mechanism in IPD. The objective is to study the pathological role of native αSyn, 3-NT-αSyn, and 3-NT-proteins in the saliva and submandibulary glands of patients with IPD. Methods. The salivary and serum αSyn and 3-NT-proteins concentration is evaluated with ELISA in patients and controls. Correlations of αSyn and 3-NT-proteins content with clinical features of the disease are examined. Immunohistochemical 3-NT-αSyn expression in submandibulary gland sections is analyzed. Results. (a) Salivary concentration and saliva/serum ratios of native αSyn and 3-NT-proteins are similar in patients and controls; (b) salivary αSyn and 3-NT-proteins do not correlate with any clinical feature; and (c) three patterns of 3-NT-αSyn-positive inclusions are observed on histological sections: rounded “Lewy-type” aggregates of 10–25 µm in diameter, coarse deposits with varied morphology, and spheroid inclusions or bodies of 3–5 µm in diameter. “Lewy-type” and coarse inclusions are observed in the interlobular connective tissue of the gland, and small-sized bodies are located within the cytoplasm of duct cells. “Lewy-type” inclusions are only observed in patients, and the remaining patterns of inclusions are observed in both the patients and controls. Conclusions. The patients’ saliva presents a similar concentration of native αSyn and 3-nitrotyrosine-proteins than that of the controls, and no correlations with clinical features are found. These findings preclude the utility of native αSyn in the saliva as a biomarker, and they indicate the absence of nitrative stress in the saliva and serum of patients. As regards nitrated αSyn, “Lewy-type” inclusions expressing 3-NT-αSyn are observed in the patients, not the controls—a novel finding that suggests that a biopsy of the submandibulary gland, if proven safe, could be a useful technique for diagnosing IPD. Finally, to our knowledge, this is also the first description of 3-NT-αSyn-immunoreactive intracytoplasmic bodies in cells that are located outside the nervous system. These intracytoplasmic bodies are present in duct cells of submandibulary gland sections from all subjects regardless of their pathology, and they can represent an aging or involutional change. Further immunostaining studies with different antibodies and larger samples are needed to validate the data.

It is important to understand the role of nitrative stress and αSyn modifications in the saliva and salivary glands in PD pathogenesis. The objectives are as follows: (a) to study the expression of native and nitrated αSyn in the saliva and submandibulary gland in patients with idiopathic PD (IPD) and control subjects, (b) to detect the presence of nitrative stress in the saliva of patients and controls through the quantification of 3-NT-proteins, and (c) to explore the relationship of salivary αSyn and 3-NT-proteins with specific motor and non-motor features of the disease. Among salivary glands, the human submandibulary gland is the most active, producing~65% of the total salivary volume [26].

Participants
For this cross-sectional and observational study, 50 patients with idiopathic PD and 30 control participants were enrolled at Hospital Valme, Sevilla, Spain. Five saliva samples were discarded due to high hemoglobin concentration or technical problems, and hence the final number of patients samples was 45. Patients were diagnosed with PD if they presented all three classic motor signs of Parkinsonism (bradykinesia, rigidity, and resting tremor) [7,27], as well as a reliable loss of dopamine-transporter signal on basal ganglia, as measured with 123 I-Ioflupane DAT-SPECT [28]. All SPECT scans were performed, quantitatively analyzed, and visually assessed by expert physicians at the Service of Nuclear Medicine, following established criteria [28,29]. All patients had a disease duration longer than 3 years, and the age at PD onset was from 45 to 75 years. To exclude hereditary forms of Parkinsonism, those patients with atypical deficits, family members with PD, or younger than 45 years old were discarded. Control participants were recruited from volunteers, and they were group-matched by age and sex to PD subjects. Controls were excluded if they had a first-degree family member with PD or a neurological disorder.

Clinical Information
Standard demographic information was obtained from patients. Clinical data included the International Parkinson and Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS), the modified Hoehn-Yahr staging, and the modified Schwab-England activities of daily living scale. Age at PD onset and disease duration in years were also evaluated according to the year of motor symptoms onset, as reported by the patient. Individuals presenting with any liver, renal, cardiovascular, and hematological dysfunctions, as well as cancer, autoimmune disorders, or AIDS were not included because markers of oxidative and nitrative stress could be affected [30,31]. In addition, all participants were non-alcohol drinkers, non-smokers, and non-coffee drinkers [31][32][33][34]. Alcohol abuse was defined as drinking >210 g of alcohol per week. Smoking was defined as current smokers who consume cigarettes on a daily basis, or occasional smokers who consume cigarettes less than on a daily basis. Coffee drinking was defined as a person who intakes coffee drinks containing more than 300 mg of caffeine on a daily basis (e.g., more than 3 standard 8 oz cups of brewed coffee) [33].

Serum and Saliva Collection and ELISA Analysis
Blood was collected by cephalic vein puncture. A total of 5 mL of blood was collected in gel-coated tubes to induce blood coagulation and to obtain serum (BD Vacutainer, Madrid, Spain). The blood serum was centrifuged at 2500 rpm for 10 min to separate clots and trapped cells, and then serum was immediately frozen at −80 • C in 0.5 mL aliquots. Three milliliters of saliva were collected in 5 mL polypropylene tubes (Eurotube DeltaLab, Barcelona, Spain). The saliva was centrifuged at 2500 rpm for 10 min to precipitate cells, and then the liquid portion was immediately frozen at −80 • C in 0.5 mL aliquots. Hemoglobin concentration in a fresh 0.5 mL saliva aliquot was quantified as recommended [7], and those samples with a hemoglobin concentration higher than 1200 mg/mL were discarded. The serum and saliva aliquots were unfrozen and sonicated with homogenizing solution (150 mM NaCl, 50 mM HEPES, 1 mM phenylmethylsulfonil fluoride, 0.6 µm leupeptin, 1% Triton X-100, pH 7.4).

Histological Examination
All sections were reviewed by researchers blinded to the clinical information (E.F.E. and J.S.). Sections with positive immunoreactivity were visualized using a standard optical microscope (Nikon Instruments Europe B.V., Amstelveen, the Netherlands), coupled to the NIS-Elements Imaging Software 3.00 (Nikon). We screened 4-6 serial sections per subject with the anti-nitroα/β-synuclein antibody and the anti-IYD polyclonal antibody. The degree of immunoreactive inclusions within different regions of the submandibulary gland was assessed in contiguous tissue sections according to a five-point rating scale: not detectable (0), mild (1), moderate (2), frequent (3), and very frequent (4). These regions encompassed acini, ducts, and interlobular connective tissue (that contain blood vessels and autonomic nerves supplying the gland). The presence of immunoreactive inclusions was also assessed in other intraglandular cells such as adipocytes, endothelial, and mononuclear blood cells [24].

Statistics and Ethics
Comparisons of dichotomous variables were carried out with the χ 2 test. As for quantitative variables, two groups were compared with Student's t test, or the Mann-Whitney U test if distribution was a non-parametric one. Correlations between two dependent variables were carried out with the Pearson's test, or the Spearman's test in non-parametric distribution. Normalization of data was verified with the Shapiro-Wilk test. All the protocols were approved by the Internal Ethics and Scientific Board of Hospital Universitario Valme (ref. 10

Participant Characteristics
Basic demographic features are not found to be different between patients and controls, as shown in Table 1. Clinical parameters of patients are also shown in Table 1.

Salivary and Serum Native αSyn
Total αSyn concentration in the saliva and serum, as measured through ELISA, does not differ between patients and controls (Table 1). Individual αSyn concentration values are shown in Figure 1. The saliva/serum ratio of native αSyn is similar in both patients and controls (~1:700, Table 1). No correlations are found between the native αSyn concentration in saliva and demographic and clinical parameters, as shown in Table 2.

Salivary and Serum 3-Nitrotyrosine Proteins, and Nitrative Stress
The concentration of 3-NT-proteins is measured in saliva and serum, since it is a biochemical parameter that could be indicative of nitrative stress. The total 3-NT-proteins concentration in saliva and serum, as measured through ELISA, does not differ between patients and controls (Table 1). Individual 3-NT-proteins concentration values are shown in Figure 2. The saliva/serum ratio of 3-nitrotyrosine proteins is not found to be different in patients relative to controls (Table 1). No significant correlations are found between the 3-NT-proteins concentration in saliva and demographic and clinical parameters, as shown in Table 2.

3-Nitrotyrosine-αSyn in the Submandibulary Gland
Immunohistochemical study of the submandibulary gland reveals three patterns of inclusions expressing 3-NT-αSyn: (1) rounded "Lewy-type" aggregates, (2) coarse aggregates of varied morphology, and (3) small-sized bodies of spheroid shape (Figure 3). First, rounded "Lewy-type" inclusions are located within the interlobular connective tissue (Figure 3a-c). They resemble Lewy bodies because they are 10-25 µm in diameter, and show a peripheral halo or radiating fibrils, as described elsewhere [9,38,39]. Importantly, these "Lewy-type" inclusions are observed in patients with IPD, not in controls. Second, the interlobular connective tissue also contains coarse inclusions of varied morphology, including ribbon-like, spindle-like, and ovoid shapes (Figure 3d-i). It seems that the different shapes might result from a different viewing angle. Coarse inclusions are found in 5/6 IPD patients and 4/6 controls. Third, rounded bodies with a diameter of 3-5 µm are seen within the cytoplasm of IYD-positive duct cells (Figure 3c,i-l). All sections from the patients and control subjects showed 3-NT-αSyn-positive bodies. IYD-negative acinar cells, adipocytes, endothelial or blood cells are devoid of immunoreactive inclusions. A role of β-synuclein in immunoreaction signals cannot be discarded, since an anti-α/βsynuclein antibody was used. Density of the different patterns of inclusions and their location in the submandibulary glands in patients and controls are shown in Table 3. Finally, there were no significant differences in age (patients with IPD, 65.3 ± 8 years; control participants, 61.8 ± 10 years), and gender between the patients and controls.

Discussion
In this cross-sectional study, the concentration of native αSyn in the saliva is found to be similar in patients with IPD and control participants, which is consistent with many other studies [1][2][3][6][7][8]. No correlations are detected between the native αSyn concentration and motor and non-motor features of IPD, in accordance with most studies [1,5,7,8]. Blood serum contains αSyn, without difference in protein levels between patients and controls [6,7]. The results of the present study would preclude using native αSyn in the saliva as a biomarker for IPD. Regarding 3-nitrotyrosine proteins, these molecules are quantified in saliva and serum to detect the presence of nitrative stress, a type of oxidative stress that is characterized by elevated levels of 3-NT-proteins. No differences are observed between patients with IPD and controls, indicating the absence of nitrative stress in the saliva or serum of IPD patients.
The saliva/serum ratio of native αSyn is~1:700 in both patients and controls, a result that indicates a normal or "physiological" secretion of native α-synuclein in forming saliva in patients with PD. This result is of interest because the source of salivary native αSyn is unknown. Given that αSyn can be secreted by neurons [40], it is possible the salivary nerves release αSyn into saliva. However, it is also possible that αSyn derives from the cellular secretion of saliva or blood. Our findings support the hypothesis that salivary native αSyn comes from blood, because the saliva/serum ratio was similar in all subjects regardless of the pathology. The saliva/serum ratio of nitrated proteins is similar in patients and controls, a finding that confirms the absence of nitrative stress in these biofluids in patients with IPD.
As regards the histological study, the submandibulary gland is an exocrine gland with multiple lobules that contain secretory acinar cells. These lobules are separated by connective tissue that contain excretory ducts, blood vessels, and autonomic nerves supplying the gland. The immunohistochemical study reveals three different patterns of 3-NT-αSyn immunoreactivity, including rounded "Lewy-type" aggregates, coarse inclusions of varied morphology, and small-sized intracytoplasmic bodies.
Rounded "Lewy-type" inclusions are seen in the interlobular connective tissue, and they are considered as specific to PD because they are observed in patients, not in controls. These inclusions resemble Lewy bodies because they are 10-25 µm in diameter, and they show a peripheral halo or radiating fibrils [9,19,38,39]. The detection of "Lewy-type" aggregates expressing nitrated αSyn in the submandibulary glands of patients is, to the best of our knowledge, a novel finding. All previous studies on Lewy pathology in human submandibulary glands are based on oligomeric and phosphorylated αSyn, other oxidationrelated forms of the molecule [13,14]. This is an important issue because formation of Lewy aggregates is linked to the etiology of Parkinson's disease and other neurodegenerative disorders [9,41].
The interlobular connective tissue also contains coarse inclusions of varied morphology. They are observed in histological sections from patients and controls. These deposits have ribbon-like, spindle-like, and ovoid shapes; morphologies that are the likely consequence of different viewing angles. Of note is that these morphologies resemble previous descriptions of phosphorylated-αSyn-positive inclusions within the interlobular connective tissue of salivary glands [14,16,24,[42][43][44]. Therefore, these morphological shapes seem to be a common feature of αSyn-positive inclusions within the interlobular tissue of salivary glands.
Previous double-staining studies with antibodies against neural markers such as neurofilaments or protein-G product indicate the neuronal identity of "Lewy-type" and coarse deposits within the interlobular tissue [42][43][44]. Although we are unable to confirm the neuronal identity of these inclusions, their location within the interlobular connective tissue (where salivary neural fibers supplying the submandibulary gland are located) would suggest that these 3-NT-αSyn-positive aggregates derive from neuronal secretion [40]. However, it cannot be disregarded that aggregated nitrated αSyn can derive from cells other than neurons, and more studies are required.
Another novel finding is that the cytoplasmic staining of IYD-positive duct cells is condensed into defined small-sized spheroid inclusions or bodies that express nitrated αSyn. Intracytoplasmic inclusions of nitrated αSyn are well documented in the brain [45][46][47] but, to our knowledge, this is the first description of 3-NT-αSyn-immunoreactive bodies within the cytoplasm of non-neural cells, specifically within duct cells of the submandibulary gland. These bodies have a diameter of 3-5 µm, and they are frequently observed in all subjects regardless of their pathology. These observations suggest that these small-sized inclusions are not of pathological significance, and it is possible that they represent an aging or involutional change in duct cells. Further studies are needed to analyze these intracytoplasmic bodies, and their physiological or pathological significance.
There are some limitations to our study. Thus, it is important to note that several authors have proposed that some immunoreactions in salivary glands may be explained as cross-reactions due to endogenous enzyme activity [24], and that patterns of α-synucleinimmunoreactive inclusions are highly dependent on the specificity of primary antibodies [24,[42][43][44]. In addition, a role for nitro-β-synuclein in immunoreactions cannot be discarded because an anti-nitro-α/β-synuclein antibody was used. Further validation of our immunostaining findings with additional antibodies is needed to validate the usefulness of nitrated αSyn expression in the submandibulary gland as a diagnostic tool for PD. Finally, the sample size in the histological study is small, and the results must be confirmed by means of a larger sample in future studies.

Conclusions
The saliva of patients with idiopathic PD presents a similar concentration of native αSyn and 3-nitrotyrosine-proteins than that of controls, and no correlations with clinical features are found. These findings preclude the utility of native αSyn as a biomarker, and they indicate the absence of nitrative stress in the saliva and serum of patients. As regards nitrated αSyn, "Lewy-type" inclusions expressing 3-NT-αSyn are observed in patients, not in controls. It is a novel finding which suggests that a biopsy of the submandibulary gland, if proven safe, could be a useful technique for diagnosing IPD. It is also described for the first time the presence of small-sized bodies within the cytoplasm of cells that are located outside the nervous system. These intracytoplasmic bodies are present in duct cells of submandibulary gland sections from all subjects regardless of their pathology, and they can represent an aging or involutional change. Finally, it is important to further validate the immunostaining findings with additional antibodies and larger samples.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.