Introduction
Parkinson’s disease (PD) is a neurodegenerative disease that usually develops late in life and is characterised by the loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc). Patients with PD manifest mainly hypokinetic motor features like bradykinesia, muscular rigidity, postural and gait impairment. Levodopa, which is a precursor to dopamine, is taken up by the dopaminergic neurons and converted to dopamine. In levodopa-induced dyskinesia (LID), patients usually manifest hyperkinetic abnormal spasm of body parts, which most commonly affect foot or leg and rarely on the arm or trunk. Levodopa is the gold standard drug for treating PD, but its long-term use is complicated by motor fluctuations and dyskinesias. Reference Albin and Leventhal1,Reference Eusebi, Romoli, Paoletti, Tambasco, Calabresi and Parnetti2 Hence, LID could be considered as an adverse effect of levodopa therapy. The disorder is poorly understood, and no peripheral biomarkers has been established that may shed light into its mechanisms. Increased dopamine metabolism due to prolonged levodopa treatment might amplify the pre-existing oxidative damage and neuroinflammatory pathology in PD. Lymphocyte activation gene 3 (LAG3) and toll-like receptor (TLR) proteins that were shown to be involved in PD could also play key roles in the pathophysiology of LID. Alternative splicing of LAG3 generates LAG3 that is available in the serum. The toll-interacting protein (TOLLIP) is an inhibitory adaptor protein of the TLR pathway, participating in the endo-lysosomal degradation of the interleukin (IL)-1R. TOLLIP overexpression impairs IL-1β-induced activation of nuclear factor kappa B (NF-κB) and inhibits inflammatory signalling. Reference Zheng, Liu, Shen, Hu and Zhao3
In the context of alpha-synuclein-induced inflammation, cytosolic NLRP3 inflammasome activates inflammatory cytokines like IL-1β. Reference Schroder and Tschopp4 Our earlier work suggested that alpha synuclein-NLRP3-mediated inflammation may underlie PD pathophysiology and serves as a novel therapeutic target in PD. Reference Chatterjee, Roy and Banerjee5–Reference Roy, Banerjee and Choudhury7 Further, we identified a significant decrease in reduced glutathione (GSH), IL-10 and brain-derived neurotrophic factor (BDNF) levels with PD severity which was reverse for pro-inflammatory cytokines and superoxide dismutase-1 (SOD1) enzyme activity. Reference Roy, Mondal and Banerjee8
Levodopa, the chemical precursor of dopamine, contributes to free radical formation and might exacerbate antioxidant dysfunction in PD. To our knowledge, there is no report available on peripheral inflammatory and oxidative stress markers in LID patients. Here, we present a snapshot of the oxidative-inflammatory status in PD patients with LID. We hypothesise that there should be a distinct inflammatory-oxidative profile in PD patients with LID compared to general PD population which is clinically less discriminable.
Materials and Methods
Subject Recruitment
Ten PD patients without LID and ten PD with LID were recruited as per the UK Brain Bank Criteria Reference Hughes, Daniel, Kilford and Lees9 at the Department of Neurology, Institute of Neurosciences Kolkata (I-NK), India. Prior signed informed consent was obtained from the subjects for recruitment in the study as per the approval of the Institutional Ethics Committee. Age- and gender-matched 15 healthy controls (HC) were recruited. Severely disabled patients (H&Y 4 and 5) and those under anti-inflammatory medications were excluded. PD and LID patients were both disease duration H&Y stage matched.
Patients with any other neurological signs other than PD, secondary parkinsonism, PD dementia were excluded from the study. To minimise intrinsic and extrinsic confounding factors on serum biomarkers of interest, we excluded participants with history/complaint of any acute or chronic inflammatory or infective conditions. C-reactive protein (CRP) being an acute-phase reactant but a non-specific marker for inflammation, we excluded PD patients with a CRP level >5 mg/mL avoiding ambiguity to establish chronic inflammatory pathophysiology of PD.
Medical conditions/co-morbidities other than PD may also cause chronic inflammation which could confound our study findings. To avoid this, during subject recruitment, we recorded the subject’s last one-month history of chronic inflammatory conditions like osteoarthritis, osteoporosis, fever and allergic condition, which could affect the level of inflammatory markers.
Motor Assessment
MDS-UPDRS Part III, IV including H&Y scale and AIM scale were used for motor and dyskinesia assessment. Reference Guy10 Patients were assessed only in the “OFF” phase of dopaminergic medication.
Cognitive and Behavioural Assessment
Montreal cognitive assessment (MOCA) was used for cognitive assessments. Reference Lihala, Mitra and Neogy11 Becks depression inventory (II) was used on LID patients to determine behavioural abnormalities. Reference Steer, Rissmiller and Beck12
Blood Sampling and Protein Estimation
Peripheral venous blood (5 ml) from PD and controls was collected into EDTA-free tubes for serum isolation. The sera were stored at −20°C in aliquots. Standard Lowry method was used to detect the protein content. Reference Roy, Mondal and Banerjee8
IL-1β, NLRP3, sLAG3, TOLLIP Estimation
ELISA kits from MyBiosource, USA specific for sLAG-3 (Cat no. MBS2515716), TOLLIP (Cat No. MBS1607132) and NLRP3 (Cat No. MBS917009) were used. IL-1β was also measured by ELISA method (Cat No. KHC0011; Invitrogen, USA). iMark Microplate Reader, BIORAD, USA, was employed to measure the absorbance of the samples.
Serum GSH, SOD1, Catalase Estimation
Serum samples were treated with 0.1 ml of 25% TCA and the resulting precipitate was pelleted by centrifugation at 3900 × g for 10 min. The free endogenous sulfhydryl was assayed in a mixture of 3 ml volume (2 ml of 0.5 mM DTNB prepared in 0.2 M phosphate buffer, with 1 ml of cell supernatant). The thiol group of GSH reacting with DTNB formed a yellow complex, whose absorbance was read at 412 nm. Reference Moron, Depierre and Mannervik13
SOD1 specific activity was determined by pyrogallol autoxidation method Reference Marklund and Marklund14 with slight modification. Briefly, the serum sample was added to tris-cacodylic acid buffer (62.5 mM), followed by the addition of pyrogallol (4 mM). The autoxidation of pyrogallol was monitored at 420 nm. Specific activity was defined by dividing the activity with protein concentration. Finally, standard spectrophotometric assay UV range (240 nm) was performed to detect serum catalase level. Reference Yumoto, Sawabe and Ueno15
Statistical Analysis
Descriptive statistics was presented using standard measures for central tendency, dispersion for numerical variables and frequency for categorical variables. Normality of the data was assessed by Shapiro–Wilk test. The values are represented as mean ± SD. A p value < 0.05 was considered insignificant to reject the null hypothesis. Pearson correlation test and Spearman’s correlation test were performed to determine association between the two parametric and non-parametric data. Mann–Whitney U test was performed to check for difference between two non-parametric unpaired data. One-way ANOVA was performed to check for difference in three different group means and Tukey’s test used as post hoc analysis. Analysis of covariance (ANCOVA) was done to adjust for covariates like age, gender, disease duration and LEDD. The statistical analysis was performed using SPSS software, Version 20.
Results
Clinical and Demographic Data
Clinical and demographic characteristics of PD, LID patients and HCs are presented in Table 1.
Table 1: Demographic and clinical characteristics for PD with and without LID and demography of healthy participants
# Association checked only within PD and LID for UPDRS III, H&Y, Disease duration, LEDD, and PD variants, Fisher’s Exact test used. UPDRS = Unified Parkinson’s disease rating scale; H &Y = Hoehn and Yahr scale; LEDD = L-dopa equivalent daily dose; MMSE = Mini-Mental State Examination; PIGD/TD = postural instability and gait disorder / tremor-dominant Parkinson’s disease.
Serum GSH Content and SOD1, Catalase Activities
SOD1 activity and GSH content (nmoles/µl) were significantly lower in LID compared to PD and HC (p = 0.0001 and p = 0.003, respectively), (Figure 1A, B). Catalase activity was significantly higher in LID than in HC (p = 0.0001) but not comparable to PD (Figure 1C).
Figure 1: Serum levels of Oxidative and Inflammatory markers in LID. GSH, SOD1 and catalase activities (A-C); sLAG3, TOLLIP, NLRP3 and IL-1β concentrations (D-G) in age-matched healthy controls (HC), Parkinson’s disease (PD) and Levodopa-induced dyskinesia (LID), Mann–Whitney U test used for inter-group comparison; data presented as Mean ± SD. P < 0.05 considered significant * & P < 0.01 as **.
Serum sLAG3, TOLLIP, NLRP3 and IL-1β Levels
sLAG3 concentration (pg/ml) was significantly lower in LID compared to PD (p = 0.0001) (Figure 1D); TOLLIP concentration (pg/ml) was significantly higher in LID compared to PD (p = 0.0001) (Figure 1E). LAG3 and TOLLIP concentrations were not significantly different between LID and HC. NLRP3 (ng/ml) levels were significantly lower in LID compared to PD (p = 0.0001) (Figure 1F), but the level in LID was not different from HC. IL-1β concentration (pg/ml) in LID was significantly higher compared to HC (p = 0.0001) and significantly lower when compared to PD patients without LID (p = 0.002) (Figure 1G). The significant difference in oxidative and inflammatory markers retained after post hoc correction for multiple comparison and analysis of covariance (ANCOVA) considering all clinical and demographic variables (p < 0.01).
Association of Motor, Dyskinesia and Cognitive Severity with Peripheral Inflammatory and Oxidative Markers
We failed to show any significant association of motor severity with peripheral inflammatory and oxidative markers (p > 0.05). There was no significant association between dyskinesia severity (AIMS scores) and any of the inflammatory or oxidative markers (p > 0.05) (Supplementary Figure A-G). Significant positive association was found between MMSE scores and GSH conc., SOD1 activity (p = 0.000, R 2 = 0.403; p = 0.006, R 2 = 0.286, respectively) (Figure 2 A-B). No association was found between inflammatory markers and MMSE scores (Supplementary figure H-K). Dose of levodopa and disease duration were also comparable between PD with or without LID patients (p = 0.190; p = 0.853). Moreover, gender also had no effect in any of these markers (p = 0.187) among the groups.
Figure 2: Association of peripheral markers with disease severity. Scatter plot between serum levels of GSH conc., SOD1 activity and MMSE score (A-B) show a strong linear relationship (Spearman’s rho 0.403 and 0.286 respectively, p 0.01). Positive correlation in serum levels of GSH conc., SOD1 activity (A-B) and MMSE scores. P < 0.05 considered significant, Dotted line indicates line of best fit.
Discussion
Inflammatory-oxidative pathophysiology in PD is gaining ground due to emerging clinical evidences and intensive research in this field. On the other hand, the role of dysregulated inflammatory redox system in LID, which occurs in at least 50% of PD patients after 5 to 10 years of levodopa treatment, has not been explored and warrants further investigation.
Classically, levodopa-induced toxicity was considered to be due to the production of free radicals. The proponents of this hypothesis believed that despite using peripheral dopa-decarboxylase inhibitors with levodopa, some amount is still transformed to dopamine in the periphery and that could be one of the sources of free radicals and hydrogen peroxide-induced neuronal damage. Reference Simuni and Stern16 Nevertheless, multiple clinical studies failed to provide sufficient evidence regarding the toxic effect of levodopa in PD patients.
We could not find any significant difference between the groups (p = 0.190; p = 0.853) in the dose and duration of levodopa between PD with or without LID. LID risk is influenced by both patient and drug factors. Moreover, genetic polymorphism is reported to play an important role on dyskinesia in PD. Reference Magistrelli, Ferrari and Furgiuele22 From the current pilot data, it may suggest that occurrence of LID in a PD patient possibly depends on characteristics and the inflammatory-oxidative marker profile of the patient.
Neuroinflammation is associated with levodopa treatment in PD, suggesting a role in LID. Reference Yan, Song, Zhang, Wang and Liu28 In this context, studies have highlighted that in the pathophysiology of LID, an adverse effect of levodopa therapy could involve activation of free radical-induced oxidative stress that can facilitate activation of this hyperkinetic movements. Reference Mulas, Espa and Fenu23
Only a few experimental works showed that synthetic antioxidants could ameliorate abnormal hyperkinetic movements in LID animal models. Reference Zhang, Xie, Lin, Wang, Wang and Liu24 To our understanding we present here for the first time direct clinical evidence for the involvement of a distinct pattern of low-grade oxidative profile related to inflammation in only patients with LID. Thus, the general PD and PD with LID harboured different inflammatory oxidative pathomechanisms.
The novel role of neuroimmune response in LID has received attention lately. The neuroinflammatory dysfunction in PD alone and its contribution towards LID pathophysiology and symptoms is equally intriguing. Involvement of NLRP3 inflammasome and its downstream inflammatory cascade have been investigated in pre-clinical models of PD. Our group also explored the involvement of NLRP3 mediated inflammatory activation in PD. Reference Chatterjee, Roy and Banerjee5 Our recent work on PD patients provided clear evidence of involvement of TLR-mediated inflammation which was also associated with disease progression. Even in PD, we showed the involvement of sLAG-3 and TOLLIP in the context of NLRP3-induced inflammatory activation in PD. Reference Roy, Choudhury, Banerjee, Basu and Kumar6 In line with this, we observed an upregulation of sLAG-3, NLRP3, IL-1β and downregulation of TOLLIP in PD patients. Unexpectedly, we found no difference of sLAG-3, NLRP3 and TOLLIP between LID patients and healthy subjects. Neuroinflammation, a central component of PD pathology, is also thought to contribute actively to LID onset and perpetuation. Reference Del-Bel, Bortolanza, Dos-Santos-Pereira, Bariotto and Raisman-Vozari25 Experimental data collected over recent years have demonstrated that a reactive gliosis process takes place in the dorsal striatum of dyskinetic rats. Reference Bortolanza, Cavalcanti-Kiwiatkoski and Padovan-Neto26
Preclinical studies also support distinct inflammatory pathophysiology in LID. Reference Bido, Marti and Morari27 To our knowledge, this is the first study exploring a distinct inflammatory profile in LID subgroup of PD patients. Interleukin 1-β, the major downstream proinflammatory cytokine in this pathway, remained high in LID patients compared to HC, as a reflection of low-grade inflammation in LID, which was further validated by upregulated TOLLIP levels (inhibitor of TLR-mediated inflammation) in LID compared to PD. This could be a novel finding for inflammatory burden in PD patients having LID. Whether this is a consequence of normal inflammatory condition or a result of compensatory activation of anti-inflammatory agents (TOLLIP) beyond a threshold of inflammatory burden is impossible to comment from this study. Post hoc power calculation with the significant different outcomes revealed a > 90% power of the study with an effect size of 1.462 (Cohen’s D).
The acute-phase protein, CRP which is regulated by pro-inflammatory cytokines, is the most studied biomarker of systemic inflammation. Reference Gabay and Kushner17 Previously studies have shown a link between CRP and chronic inflammatory and neurodegenerative diseases, such as cardiovascular disease, diabetes, stroke, and Alzheimer’s disease, as well as PD. Reference Luan and Yao18 Few epidemiological studies have also explored CRP levels and PD risk. However, results in the literature regarding CRP levels in PD patients are contradictory. While some studies found a significant increase of CRP levels in subjects suffering from PD compared to HCs, Reference Song, Cho, Kim, Park and Lee19 others did not report a similar tendency. Reference Hall, Janelidze, Surova, Widner, Zetterberg and Hansson20 Our study did not find any significant difference in CRP level between groups which validates independent inflammatory pathophysiology in LID.
Interestingly, we have observed a positive association of antioxidant markers with cognitive profile of our study patients. Indeed, higher levels of antioxidant enzymes have the potential to cleave oxidative radicals, which may have beneficial role in cognitive improvement. But to keep the homogeneous population we have excluded patients with cognitive impairment.
Recently, it has been shown that systemic inflammation exacerbated pre-existing neuroinflammation and thereby may play a crucial role in the development of LID. Reference Yan, Song, Zhang, Wang and Liu28 Neuroinflammation in the development and expression of LID is proved in the 6-OHDA rat model. Reference Bortolanza, Cavalcanti-Kiwiatkoski and Padovan-Neto26 In the current study, we could show involvement of only one inflammatory marker (IL-1β) in LID but failed to show any significant association between disease severity (UPDRS III scores) and any of the inflammatory or oxidative markers. A previous study showed occurrence of resting tremor as an initial manifestation of PD, which could lead to LID, Reference Nicoletti, Mostile and Nicoletti29 whereas another study suggested postural instability and gait dominant (PIGD) could be a useful indicator of severity and prognosis in PD by itself. Reference van der Heeden, Marinus, Martinez-Martin, Rodriguez-Blazquez, Geraedts and van Hilten30
Overall, this pilot study reveals a distinct peripheral molecular profile in PD patients with LID and provides a ray of light into probable activation of inflammatory and oxidative stress pathomechanisms in LID. Further research investigating upstream activation of NF-κB by oxidative stress or downstream receptor-mediated activation of the inflammatory-redox axis in LID might unlock windows for therapeutic intervention.
Conclusion
The relationship between inflammation and LID is still obscure. Our current study observations strengthened our hypothesis for the involvement of inflammatory pathomechanism in LID. To our understanding of the available literature on inflammation in LID, we may be the first to establish a unique low-grade inflammatory profile among LID patients that is distinct from PD in general. The findings from this pilot clinical study may be useful for conducting large-scale studies and for identifying novel therapeutic targets in LID.
Limitations
Despite the potential implication of this study, it lacks in terms of sample size and possibly the reason why we failed to find significant correlations between disease severity and the serum markers. Phenotypic differences among PD variants and inclusion of only selective blood molecular markers could also impede our better understanding of LID.
Supplementary Material
To view supplementary material for this article, please visit https://doi.org/10.1017/cjn.2023.8.
Acknowledgements
We are thankful to the participants for their consent and to those who provided direct and indirect support to the success of the study.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author [H.K]. The data are not publicly available due to the information content that could compromise the privacy of research participants.
Funding
Institutional (I-NK) research fund of HK was used for project consumables. Personal research fellowship to AR for this project was supported by DST INSPIRE Fellowship (IF-170628) supervised by Prof. Sanjit Dey and Dr Hrishikesh Kumar.
Conflict of Interest
All the authors have nothing to declare.
Statement of Authorship
Swagata Sarkar: Study concept and design, acquisition of data, analysis and interpretation, first draft of the manuscript, critical revision of the manuscript for important intellectual content.
Akash Roy: Study concept and design, acquisition of data, analysis and interpretation, first draft of the manuscript, statistical analysis, critical revision of the manuscript for important intellectual content.
Supriyo Choudhury: Study concept and design, analysis and interpretation, statistical analysis, critical revision of the manuscript for important intellectual content.
Rebecca Banerjee: Study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Sanjit Dey: Study concept and design, critical revision of the manuscript for important intellectual content, study supervision.
Hrishikesh Kumar: Study concept and design, critical revision of the manuscript for important intellectual content, study supervision.
