-
Amino acid profiling in Parkinson’s Disease (PD) shows unique changes for each amino acid.
-
Elevated proline levels may indicate disturbed collagen/glutamate pathways.
-
Changes in ornithine, citrulline, and aspartate levels may be indicators of mitochondrial dysfunction, and increased ornithine, a precursor to glutamate, urea, and polyamines, may influence α-synuclein-associated PD pathology.
Introduction
Parkinson’s disease (PD) is one of the most prevalent neurodegenerative disorders with a worldwide prevalence of over six million individuals. This number is expected to double to over 12 million by 2040. Over the past generation, its prevalence has increased by 2.5 times, making it a leading contributor to neurological disability. Reference Feigin, Vos and Alahdab1,Reference Dorsey, Sherer, Okun and Bloem2
PD is characterized by the presence of involuntary or dysregulated motor symptoms, including tremors, rigidity and impaired coordination. The clinical presentation typically unfolds gradually, displaying a progressive nature. As the disease advances, patients often encounter challenges in ambulation and speech production. Additionally, various non-motor manifestations emerge, encompassing cognitive and affective alterations, sleep disturbances, depressive symptoms, cognitive impairments and persistent fatigue. 3
The cardinal features of PD result from the degeneration of dopaminergic neurons in the basal ganglia, a critical hub for motor function regulation. These neurons are responsible for synthesizing and releasing dopamine, a neurotransmitter essential for smooth motor control. The loss of dopaminergic neurons leads to insufficient dopamine levels, causing the characteristic motor impairments in PD. Reference Surmeier4
PD is also characterized by the presence of Lewy bodies and Lewy neurites, which are neural inclusions, along with cell loss in the substantia nigra and other areas of the brain. The primary components of Lewy bodies are aggregated and misfolded α-synuclein species, indicating that PD falls under the category of synucleinopathies. Reference Tolosa, Garrido, Scholz and Poewe5
Misfolded α-synuclein species include polyamines, leading to the accumulation and fibril formation of putrescine, spermine and spermidine. Its accumulation in neurons is playing a key role in PD development. Reference Chang, Cheng, Tang, Huang, Wu and Chen6,Reference Ghosh, Mehra, Sahay, Singh and Maji7 Polyamines are molecules including two to four amino groups and play a role in several mechanisms in living organisms including apoptosis, cell division and differentiation, cell proliferation, DNA and protein synthesis, gene expression, homeostasis and signal transduction. Dietary amino acid intake and serum amino acid profile contribute to polyamine biosynthesis, and the accumulation of polyamines has been reported to be related to several diseases including neurological diseases. Reference Handa, Fatima and Mattoo8,Reference Zahedi, Barone and Soleimani9
Furthermore, amino acids are the precursors of the neurotransmitters regulating all the neurological activities of the brain. A deficiency or excess of a specific amino acid can significantly impact neurotransmitter synthesis. Reference Dalangin, Kim and Campbell10
Therefore, it is important to indicate the serum amino acid profile of PD patients to understand the underlying mechanism and dietary factors for the development and treatment procedures of the disease.
Methods
Eligibility criteria
To investigate the serum amino acid profile and metabolism in PD, a comprehensive review of human studies was conducted, including those involving the serum amino acid level of PD patients. Data on mean and standard deviation were collected and used in the meta-analysis, without imposing any restrictions on age, gender, race or body mass index.
To be included in the meta-analysis review, studies had to meet the following inclusion criteria: (i) Measurement of serum amino acid levels with reported mean and standard deviation values; (ii) Provision of sample size information; (iii) Use of accepted diagnostic criteria for the disease; (iv) Adherence to appropriate sample collection conditions. Studies that did not meet these criteria were excluded (see Table 1). Detailed information about the included studies is provided in the “Characteristics of Studies” section and Table 2.
Systematic search
In this study, a systematic research was conducted in the PubMed, Scopus and Web of Science databases with keywords “serum amino acid” or “serum amino acid level” and “Parkinson” or “Parkinson’s disease” without any date restriction. The systematic screening was conducted until February 15, 2023. The PRISMA procedures were followed for searching and evaluating the data Reference Page, McKenzie and Bossuyt11 The articles that were identified as duplicates among the articles found commonly in the databases have been eliminated.
Statistical analysis
Pooled data were analyzed to examine the relationship between serum amino acid levels and PD. Heterogeneity among the studies was assessed using the I 2 statistic, which quantifies the percentage of total variation across studies that is due to heterogeneity rather than random chance. An I 2 value of 0% indicates no observed heterogeneity, while higher values suggest increasing degrees of heterogeneity.
Based on the I 2 statistic, either fixed-effect or random-effect models were chosen for combining the study results. Fixed-effect models assume that the effect size is the same across all studies, while random-effect models account for variability in effect sizes between studies. Sensitivity and specificity analyses were performed following established guidelines Reference Higgins12–Reference Borenstein, Higgins and Rothstein14 to assess the diagnostic accuracy and reliability of the findings.
Meta-analysis was carried out using RevMan 5.3, a specialized software for systematic reviews. GraphPad Prism 6 was used to generate figures and perform correlation analyses, including matrix visualizations of the data.
Risk of bias assessment
The risk of bias for each study was assessed either as low, unclear or high risk for each of the following criteria: selection bias, performance bias, detection bias, attrition bias and reporting bias and other as described in the Cochrane Handbook. The bias detections were given in Table 2 as relevant.
Results
A systematic search was conducted on three databases until February 15, 2023, yielding a total of 733 records. Using the exclusion criteria, 368 non-related studies; 145 reviews and meta-analyses; 136 animal studies; 25 non-English articles; 4 studies that did not include control groups; 3 articles unavailable in full text due to subscription barriers, institutional access restrictions or publication issues; and 2 studies that did not provide mean and SD values were excluded (Figure 1) regarding the criteria that are given in PICO table (Table 1). After combining common studies included in more than one database, 18 studies were available for meta-analysis. Included studies were published between 1992 and 2022 (as shown in Table 2). Reference Chang, Cheng, Tang, Huang, Wu and Chen6,Reference Zhang, He and Qian15–Reference Iwasaki, Ikeda, Shiojima and Kinoshita31
According to the meta-analysis results, it was observed that valine level was significantly increased in the PD group (p: 0.02, mean difference: 37.47, 95% CI: 6.86, 68.08) (as shown in Figure 2). The levels of isoleucine (p: 0.25) and leucine (p: 0.65) were not significantly different between groups.
Aliphatic non-essential amino acids were also analyzed, and it was seen that proline levels were significantly higher in the PD group than in the controls (p: 0.002, mean difference: 36.87 μmol/L, 95% CI: 13.54, 60.20). However, alanine and glycine levels weren’t significantly different between groups (p: 0.09 and p: 0.43) (Figure 2).
There were no significant differences seen in the serum levels of the aromatic amino acids, namely, phenylalanine, tryptophan and tyrosine (previously p-values of 0.95, 0.78 and 0.60, respectively, shown in Figure 3).
The acidic amino acids, aspartate and glutamate were analyzed and presented in Figure 3. It was found that the level of aspartate was significantly lower in the Parkinson’s group than in the controls (p: 0.01, mean difference: -10.45 μmol/L, 95% -18.63, -2.27). However, the level of glutamate did not show any difference between the groups (p: 0.73). The histidine level wasn’t significantly different between groups (p: 0.08, mean difference: 10.58, 95% CI: -1.44, 22.60); however, further sensitivity analysis showed that the study of Mally et al. in 1997 is responsible for these results. When this study was excluded from the meta-analysis, histidine became a significantly increased amino acid in PD patients. Therefore, it might be mentioned that histidine levels tend to be increased in PD development.
It was found that the arginine level wasn’t significantly different between groups (p: 0.17); however, the ornithine level was significantly higher in the disease group than in the control group (p: 0.002, mean difference: 11.78 μmol/L, 95% 4.21, 19.34). The citrulline level was significantly decreased in the PD group (p: 0.02, mean difference: -3.53 μmol/L, 95% CI: -6.50, -0.56). Similarly, the lysine level was also decreased in the patients with PD compared to controls (p: 0.01, mean difference: -8.58 μmol/L, 95% CI: -15.22, -1.94). Serine level was significantly decreased in the PD group compared to controls (p: 0.05, mean difference: -17.03 μmol/L, 95% CI: -33.81, -0.25). Threonine level wasn’t different between groups (p: 0.26) (Figure 4).
The levels of sulfur-containing amino acids, cysteine and methionine, were not found significantly different between groups (p: 0.70 and p: 0.14). However, homocysteine level was found significantly increased (p < 0.00001), as shown in Figure 5. Glutamine, taurine and gamma-aminobutyric acid (GABA) levels were not changed with PD (p: 0.39, p: 0.51 and p: 0.99 (Figure 5).
Based on our analysis of cumulative correlations between amino acids, we observed homocysteine showed positive correlations with glutamate (p < 0.001; r: 0.99) and ornithine (p: 0.02; r: 0.87). In our analysis of disease stage parameters and their relationship with amino acid levels, we also observed several significant correlations. The Unified Parkinson’s Disease Rating Scale III (UPDRS III) score, Hoehn and Yahr stage score, disease duration and levodopa equivalent daily dose (LEDD) were examined for their association with amino acids.
There was a significant negative correlation between LEDD and arginine levels (p: 0.02, r: -0.99), indicating that higher medication doses might be associated with lower levels of arginine. Additionally, the UPDRS III score, which reflects motor symptom severity, showed a significant negative correlation with phenylalanine levels (p: 0.01, r: -0.95). This suggests that lower phenylalanine levels may be associated with more severe motor symptoms. There was also a tendency for tyrosine levels to negatively correlate with UPDRS III scores (p: 0.059, r: -0.99), hinting at a possible link between tyrosine metabolism and motor symptom severity, though this correlation did not reach statistical significance.
Furthermore, disease duration showed a tendency toward a negative correlation with citrulline levels (p: 0.05, r: -0.99), suggesting that prolonged disease duration might be associated with reduced citrulline levels.
Three studies were found with a high risk of bias. In the studies held by Klatt et al. Reference Klatt, Doecke and Roberts17 and Celik et al., Reference Çelik, Çiğdem, Kapancik, Kiliçgün and Bolayir20 there were no defined diagnosis criteria, and/or age information was not given, which is accepted as a risk of selection bias and reporting bias. The assessment of amino acid levels was not clear in the studies of Celik et al. Reference Çelik, Çiğdem, Kapancik, Kiliçgün and Bolayir20 and Gunaydin et al., Reference Günaydın, Özer and Karagöz23 which is accepted as performance bias and reporting bias. Furthermore, subgroup analysis depending on patients’ characteristics could not performed due to the limitation of the data for meta-analysis.
Discussion
This study provides strong evidence that neurodegenerative diseases, particularly PD, are closely related to the amino acid profiles of individuals. Our results suggest that valine, proline, ornithine and homocysteine levels were increased in PD patients, while aspartate, citrulline, lysine and serine levels were significantly decreased. Furthermore, alanine and histidine levels had a tendency to change with PD progression.
A meta-analysis study found decreased aspartate, serine, tryptophan and lysine and increased proline and homocysteine levels in PD patients. Reference Jiménez-Jiménez, Alonso-Navarro, García-Martín and Agúndez32 However, according to our literature records and analysis, we observed that there are important differences between our results and this meta-analysis results.
Our systematic search resulted in more available and up-to-date data about PD patients’ amino acid profiles. While there are some similarities between our results and this meta-analysis such as increased proline and homocysteine levels and decreased aspartate, serine and lysine levels, our study also found significantly increased levels of ornithine and valine and a significant decreased in citrulline levels.
We believe that our findings significantly contribute to this meta-analysis Reference Jiménez-Jiménez, Alonso-Navarro, García-Martín and Agúndez32 through the inclusion of more recent and different studies evaluating blood amino acid levels. Reference Zhang, He and Qian15–Reference Calvani, Picca and Landi18,Reference Çelik, Çiğdem, Kapancik, Kiliçgün and Bolayir20,Reference Günaydın, Özer and Karagöz23,Reference Lee, Kim and Lee26–Reference Levin, Bötzel, Giese, Vogeser and Lorenzl28 In the discussion, we explored the pathways that may be associated with the different results we found and their potential relevance to PD development mechanisms. From these perspectives, we found that citrulline, valine, ornithine and proline metabolism may also be related to PD by using a meta-analytic approach for the first time in the literature.
Amino acid metabolism in Parkinson’s disease
Amino acid metabolism is a highly intricate and dynamic process that plays a vital role in the synthesis and catabolism of biomolecules in living organisms. The interconversion of amino acids through various metabolic pathways allows for a highly regulated and interconnected network of biochemical reactions that maintain the delicate balance of amino acid levels in the body.
The interaction of amino acids with transport systems underscores the complex balance of amino acid levels in the brain. For example, specific groups of amino acids share the same transporters across the blood-brain barrier, influencing their dynamic interplay. Branched-chain amino acids (BCAAs), methionine, phenylalanine, tryptophan, tyrosine, histidine, threonine and glycine are transported bidirectionally between the blood and brain via the Large Neutral Amino Acid Transporter 1. Similarly, arginine, lysine and ornithine are transported bidirectionally through the Cationic Amino Acid Transporter 1. Glutamate is transported into the brain either via its own System N transporter or through the Anionic Amino Acid Transporter System X/EAA T 1–3, which it shares with aspartate. Reference Smith, Mandula and Parepally33,Reference Zaragozá34
Changes in BCAA levels in Parkinson’s disease
BCAAs are essential amino acids that are required for growth, development, nutrient signaling, providing nitrogen to neurotransmitter synthesis and glutamate/glutamine cycling. These are crucial roles of BCAAs to maintain body and brain health. However, in recent years, there is an important debate about BCAA’s effects on neurodegenerative diseases that has arisen.
While some research groups suggest that BCAAs may benefit brain function and cognitive aging, there are differing hypotheses about their role in neurodegenerative diseases. Reference Coppola, Wenner and Ilkayeva35–Reference Yan, Yang and Sun41 One hypothesis is that excessive dietary intake of BCAAs may interfere with the uptake of neutral amino acids such as tryptophan and tyrosine, which are essential for serotonin and dopamine production. BCAAs and these neutral amino acids share the same transporters across the blood-brain barrier. Therefore, high BCAA levels could compete with tryptophan and tyrosine for transport into the brain, potentially impairing the synthesis of 5-hydroxytryptophan (5-HT) and dopamine. Our cumulative results showed a significant increase in valine levels in the PD group, while levels of isoleucine and leucine did not change significantly. This suggests that the altered BCAA profile, particularly the increase in valine, may influence the balance of neurotransmitter synthesis, potentially contributing to the neurochemical disruptions observed in PD.
A study reported that high plasma BCAA levels resulted in low central 5-HT levels. Anxiety-like behavioral changes were observed in rats fed a high-fat, high-carbohydrate diet enriched with BCAAs. Reference Coppola, Wenner and Ilkayeva35 With BCAA supplementation, a decrease was observed in tryptophan and its metabolite kynurenic acid. Reference Coppola, Wenner and Ilkayeva35 Oral BCAA supplementation during exercise has been shown to reduce brain catecholamine levels. Reference Choi, DiSilvio, Fernstrom and Fernstrom36 In another study, it was determined that tyrosine and dopamine levels in the brain were decreased with BCAA supplementation. Reference Le Masurier, Oldenzeil, Lehman, Cowen and Sharp37 In a study conducted on healthy people, when they evaluated the effect of BCAA supplementation on dopamine levels by plasma prolactin levels, it was shown that BCAA supplementation could cause an increase in prolactin secretion by suppressing dopamine levels. Reference Neuhaus, Goldberg and Hassoun38 In addition, the BCAA supplements used in the studies typically show short-term effects, and it remains unclear how neurotransmitter levels might be affected by long-term BCAA consumption over extended periods. Reference Yoo, Shanmugalingam and Smith39
The study by Zhang et al. (2022), which is also included in this meta-analysis, found that patients with PD have lower plasma BCAA and aromatic amino acid levels. Reference Zhang, He and Qian15 Another study investigating the effects of whey protein supplementation, which is a good source of BCAAs, on PD showed favorable results. Reference Yoo, Shanmugalingam and Smith39 With the whey protein intake, the serum homocysteine level was decreased, while oxidized glutathione level was increased in PD patients. However, they also reported that there were no changes in the clinical outcomes. Reference Tosukhowong, Boonla and Dissayabutra40 Also, in a recent animal study, researchers revealed that rotenone-induced PD may lead to alteration in the gut microbiota, which may be the possible cause of peripheral BCAA deficiency. When they fed the animals with a high BCAAs diet, they found improvements in the inflammatory levels, motor and non-motor dysfunctions and dopaminergic neuron impairment in PD mice. Reference Yan, Yang and Sun41
A very recent study in 2023, which analyzed data from 21,982 Alzheimer’s disease (AD) cases and 41,944 controls, found an association between decreased levels of BCAAs and the presence of AD in patients. Reference Qian, Liu and Zhang42
A study with app/ps1 double transgenic mice published in the Cells journal in 2022 reports that higher plasma BCAA levels were displayed in this AD model, and reducing BCAA daily intake alleviates AD-related pathology and cognitive decline. The same study highlighted a potential link between BCAAs and AD progression. Reference Siddik, Mullins and Kramer43
Furthermore, the consumption of BCAAs in a high-fat diet may be an important factor for neurological impairment as shown in an animal study of Alzheimer’s disease model. Mice fed with a high BCAA-high-fat diet showed higher tau neuropathology, and only 4 out of 13 animals survived. Mice fed with a low-BCAA diet showed higher threonine and tryptophan cortical levels. Reference Tournissac, Vandal and Tremblay44
In a recent study in 2023, leucine-rich α2 glycoprotein was investigated to be a biomarker of systemic inflammation in PD in 66 patients and 31 age-matched controls. It was significantly higher in the patient group than in the control group and also correlated with Charlson comorbidity index, C-reactive protein levels and dementia. Reference Ohmichi, Kasai and Shinomoto45
There are conflicting results in the literature; however, our study may help to clarify the relationship between BCAA intake and PD and neurological functions. By examining the serum levels of BCAAs in PD patients, as well as the mechanistic relationships of transporters, the effects of BCAA supplements and their potential impact on disease progression in the literature, our findings provide a more comprehensive understanding of how BCAAs influence PD and may inform future research and therapeutic strategies. It might be better to separate BCAAs as each of these amino acids might be changed in patients or affect the disease progression in a different way. To understand the underlying mechanism, more studies on BCAA’s effects on brain and neurodegenerative disease development are needed. Also, it is important to examine the BCAAs in the other dietary and lifestyle factors that can change the effects of amino acids and disease progression.
Altered proline metabolism
Besides BCAAs, there were several amino acids found to be significantly changed in the patients compared to control.
Our results showed that proline levels were significantly higher in the PD group than in the controls; however, alanine and glycine levels weren’t changed.
Proline constitutes one-third of amino acids in the collagen proteins, comprising nearly 30% of body proteins. To understand the increase in proline levels it is beneficial to examine the stress response of the body. L-proline has the role of a chemical chaperone, preventing protein unfolding or misfolding for endoplasmic reticulum stress; Reference Liang, Dickman and Becker46 however, at high levels, it can be harmful to neural activity.
L-proline metabolism might be related to the development of neurodegenerative diseases associated with the formation of protein aggregates, such as Parkinson’s, Alzheimer’s, etc. These diseases have a number of common features in their processes. Energy metabolism dysfunction, glutamate excitotoxicity and oxidative stress seem to be underlying in the pathophysiology of these disorders. Reference Nagaoka, Kunii and Hino47
The process of proline catabolism involves the conversion of proline to pyrroline-5-carboxylate (P5C) by proline dehydrogenase (PRODH) in the mitochondrial matrix. This generates glutamate through NAD-dependent P5C dehydrogenase (P5CDH), which is an essential excitatory neurotransmitter in neurons and a precursor to glutamine, GABA and mitochondrial TCA cycle intermediates. On the other hand, glutamate can be converted into a P5C intermediate through P5C synthase (P5 CS) and further reduced to proline by P5C reductases (PYCRs). All the enzymes involved in the proline metabolism pathway have been reported to be associated with neurological or psychiatric disorders in human and animal models. Reference Guo, Tang, Yang and Li48–Reference Liu, Borchert and Donald50 Also, hypovitaminosis B complex especially vitamin B6 and B12 are important for proline metabolism and hyperhomocysteinemia through impaired remethylation.
As a part of neural metabolism, proline has the capability to function as a metabolic precursor of L-glutamate. PRODH knockout mice have shown that the cytosolic accumulation of L-proline disrupts GABAergic transmission through glutamate decarboxylase blockade. Reference Cohen and Nadler51 Moreover, normally, proline induces L-glutamate synthesis and acts as a GABA-mimetic inhibitor of the GAD enzyme. Reference Crabtree, Park, Gordon and Gogos52 Therefore, proline can reduce the synthesis of the GABA neurotransmitter and accumulation of L-glutamate, thereby leading to synaptic dysfunction in some forms of disease. Reference Patriarca, Cermola and D’Aniello53 On the other hand, if proline cannot contribute glutamate or collagen synthesis due to the lack of these metabolic paths, it may accumulate. Dysfunctional proline metabolism could be one of the underlying mechanisms of muscle weakness in PD. Further studies are needed to elucidate this hypothesis.
Related to this excitotoxicity, particularly in neural glutamate metabolism, patients with genetic defects in PRODH or in P5CDH, called hyperprolinemia I/hyperprolinemia II, have higher proline levels up to 10–15-fold than in normal people and may have developed schizophrenia-related phenotypes (learning, memory and sensorimotor gating) and schizophrenia. Reference Nagaoka, Kunii and Hino47,Reference Clelland, Read and Baraldi54
Studies have shown that stress and anxiety can influence the concentrations of hydroxyproline and proline in urine. Reference Lee, Kim, Park and Lee55 Long-term exposure to proline has also been linked to behavioral changes in zebrafish, which were reversed by antipsychotics, indicating a relationship between proline and psychiatric diseases. Reference Savio, Vuaden, Piato, Bonan and Wyse56 Given these findings, it is important to explore how elevated proline levels observed in PD patients might be related to similar stress or behavioral mechanisms.
In vitro studies have revealed that differentiated neurons depleted of PYCR2 showed thinner neuronal fibers and significantly increased axonal beading, an early morphological hallmark of neuronal injury. Reference Escande-Beillard, Loh and Saleem57 Furthermore, there is a hypothesis that PRODH deficiency could lead to hyperactivation of the dopaminergic system due to dysregulated astroglial control of dopamine homeostasis. Reference de Oliveira Figueiredo, Bondiolotti, Laugeray and Bezzi58 Other studies suggest that in the hypothalamus, proline is taken up by astrocytes and converted into lactate, which is then released from astrocytes and taken up by neurons. In neurons, lactate is oxidized for energy production. Reference Arrieta-Cruz and Gutiérrez-Juárez59
There is significant evidence that PRODH polymorphisms are associated with susceptibility to schizophrenia and defects in PRODH have been linked to glutamatergic and GABAergic neuron dysfunctions. Reference Guo, Tang, Yang and Li48,Reference Crabtree, Park, Gordon and Gogos52
In the literature, it was also stated that proline-rich protein 14, which is involved in the alteration and activation of the mTOR signaling pathway, was upregulated in PD patients. Reference Jin, Tan and Shi60 Its expression was found higher in whole blood, substantia nigra and medial substantia in the PD group than in healthy controls. They mentioned that the detection of proline-rich protein 14 levels in serum/plasma can be a biomarker for PD. Reference Jin, Tan and Shi60
Another reason for proline accumulation might be also related to the arginine-ornithine cycle. Our results showed that while arginine levels in PD patients were decreased, there was a significant increase in ornithine levels. In the urea cycle, ornithine-arginine transferase converts L-ornithine to proline-5 carboxylase, which is a precursor of L-proline via proline-5 carboxylase reductase. Reference Morris61
Examining proline metabolism could provide valuable insights into the mechanisms underlying neurodegenerative diseases. By understanding how proline levels and their metabolic pathways interact with disease processes, we can gain a better understanding of the causes of PD and potentially identify new therapeutic targets.
Excessive ornithine levels in Parkinson’s disease
Ornithine levels were found to be elevated in PD patients and were also associated with the severity of the disease. Advanced-stage PD patients had higher ornithine levels than early-stage PD patients. Ornithine levels were found correlated with α-synuclein protein, which includes polyamines leading to the accumulation and fibril formation of putrescine, spermine and spermidine. Its accumulation in neurons is playing a key role in PD development. Reference Chang, Cheng, Tang, Huang, Wu and Chen6,Reference Ghosh, Mehra, Sahay, Singh and Maji7
Polyamines are molecules including two to four amino groups. They play a critical role in several mechanisms in living organisms including apoptosis, cell division and differentiation, cell proliferation, DNA and protein synthesis, gene expression, homeostasis and signal transduction. Polyamine biosynthesis initiates from arginine and ornithine. Arginine is first converted to ornithine, and ornithine decarboxylase leads to putrescine synthesis. Then putrescine can be converted to spermidine and spermine. The accumulation of polyamines has been reported to be related to several diseases. Reference Handa, Fatima and Mattoo8
The major metabolic ways that can be activated when ornithine levels are increased include the synthasis of L-proline, L-glutamate, GABA and polyamines and the activation of urea cycle. Reference Majumdar, Barchi and Turlapati62 Also, the activation of the urea cycle may contribute to the excessive urea production, which can increase of osmolarity in the cerebellum, cerebral cortex and brain stem. Reference Çelik, Erşan, Kılıçgün, Kapancık and Erşan63 Ornithine-related osmotic pressure also affects plasticity in the hippocampal region with the tonic inhibition of the GABA receptors. Reference Glykys, Mann and Mody64 An in vitro study showed that osmotic stress is also a causative factor for α-synuclein accumulation. Reference Fragniere, Stott and Fazal65
Homocysteine as a biomarker of pathogenesis
The non-proteinogenic α-amino acid homocysteine is directly linked to diseases such as dementia, heart disease and stroke, with elevated blood levels indicating underlying physiological issues. Reference Smith, Refsum and Bottiglieri66,Reference Kaplan, Tatarkova, Sivonova, Racay and Lehotsky67
As it is expected and well-studied and approved with meta-analysis studies, homocysteine level was found significantly increased in PD, as high homocysteine leads to nerve cell apoptosis, oxidative stress and DNA damage of neuron cells. Reference Fan, Zhang and Li16,Reference Dong and Wu68 Increased homocysteine levels in PD patients showed an association with frontal cortical thinning and microstructural damage in frontal and posterior-cortical regions. Reference Sampedro, Martínez-Horta and Horta-Barba69
Therefore, our results suggest that its positive correlation with glutamate and ornithine may need to be considered in the potential relationships that could be established with PD.
Histidine tendency to increase in Parkinson’s disease
We found histidine levels tend to be increased in PD patients, and further sensitivity analysis showed that this was due to one study that gave contrary results to the rest of the studies. Reference Babu, Gupta and Paliwal21 With the elimination of that study, histidine levels significantly increased in the PD group (p: 0.04). The dimerization of α-synuclein protein requires histidine, which is an important aggregation for PD development. Reference Abeyawardhane, Fernández and Heitger70 Histidine is converted into histamine by the enzyme histidine decarboxylase. Regarding the effects of histamine on PD, animal experiments showed that increased histamine levels may lead to the degeneration of dopaminergic neurons in the substantia nigra. Reference Liu, Chen, Liu, Hu and Luo71 The enzyme responsible for the clearance of histamine, histamine N-methyltransferase, has been found to play an important role in the pathogenesis of PD. Reference Shan, Bossers and Luchetti72 A negative correlation was observed between the mRNA expression of histamine N-methyltransferase in the substantia nigra and disease progression inPD patients. Reference Shan, Bossers and Luchetti72 Evidence indicated that excessive histamine production may be related to PD and other neurodegenerative diseases, which are also related to increased histidine amino acid levels. Reference Shan, Bao and Swaab73
Also, histidine is a precursor of carnosine (β-alanyl-L-histidine), which is a dipeptide synthesized from b-alanine and histidine. Blockage of the carnosine pathway or over-hydrolysis of carnosine may result in histidine accumulation. Carnosine is a neuroprotective agent and a potential additive for Parkinson’s patients. Reference Schön, Mousa and Berk74
Decrease in specific amino acids in Parkinson’s disease
Aspartate
We found that aspartate levels were lower in PD patients. Aspartate is an acidic amino acid that plays a role in the citric acid cycle. Aspartate is also involved in the production of other amino acids and in the synthesis of nucleotides, which are the building blocks of DNA and RNA.
There is significant evidence suggesting that L-aspartate can be transported into nerve cells through excitatory amino acid transporters, stored in vesicles and released through a process called Ca2+-dependent vesicular exocytosis in various parts of the brain. Based on these properties, L-aspartate has the potential to be considered a classical neurotransmitter in the central nervous system (CNS) from a presynaptic standpoint. Reference Cavallero, Marte and Fedele75 In the literature, it was stated that N-acetyl-aspartate is an important metabolite for energy metabolism of the brain especially in stress-induced brain injury, cognitive impairment and PD. Reference Moffett, Arun, Ariyannur and Namboodiri76–Reference Nie, Zhang and Huang79
Citrulline
Furthermore, we found citrulline levels decreased in PD patients, and disease duration was negatively correlated with citrulline levels in PD patients. Citrulline is a non-proteinogenic amino acid that is synthesized in the urea cycle, which is responsible for removing toxic ammonia from the body. Citrulline is also involved in the production of nitric oxide, which is important for blood vessel dilation and cardiovascular health. Reference Yabuki, Shioda and Yamamoto80 Citrulline has not been extensively studied in the context of PD, but it has been shown to have a neuroprotective effect in other neurological disorders, such as stroke and traumatic brain injury. This may be due to its ability to increase the production of nitric oxide, which can improve blood flow and reduce inflammation in the brain.
There is a research on the potential neuroprotective benefits of L-citrulline for CNS disorders, such as brain ischemia. Previous studies have indicated that L-citrulline could not only prevent neuronal cell death but also prevent capillary loss in the hippocampal region caused by cerebral ischemia. The protective effect of L-citrulline on cerebrovascular function is thought to be linked to the restoration of endothelial nitric oxide synthase expression in the hippocampus. Reference Yabuki, Shioda and Yamamoto80 Furthermore, as an important relation, a study showed that citrullinated proteins in control substantia nigra only occur in astrocytes. However, in PD substantia nigra, citrullinated proteins are also found in the cytoplasm of neurons, including dopamine neurons. Abnormal protein citrullination might be related to PD, including Lewy bodies and prion diseases. Reference Nicholas81 Together with our results, the decrease in blood citrulline levels in patients might depend on the accumulation of citrulline in neurons.
Lysine
Lysine is a basic amino acid that is involved in protein synthesis and is important for the growth and maintenance of tissues. Lysine is also involved in the production of carnitine, which is important for energy metabolism, and in the regulation of gene expression.
There are several studies that show the impaired acetylation/deacetylation of lysine in histones and alpha-synuclein-Lewy bodies in PD patients. Impairment in acetylation/deacetylation mechanisms is also closely related to aging. According to these studies, lysine acetylation is aberrant in PD patients, while deacetylation mechanisms were decreased. Reference Park, Tan, Garcia, Kang, Salvesen and Zhang82–Reference Guedes-Dias and Oliveira84 Therefore, a lower level of serum lysine level might be a marker for this mechanism in PD patients.
Serine
We found that serine levels were decreased in PD patients. Serine is a neutral amino acid that is involved in the synthesis of proteins, nucleotides and phospholipids, which are important components of cell membranes. Serine is also involved in the regulation of enzyme activity and in the synthesis of neurotransmitters, which are chemical messengers in the brain. Reference Jiang, Li, He and Huang85
According to recent research, parkin, an E3 ubiquitin ligase involved in PD, regulates the synthesis of serine through its interaction with PHGDH, an enzyme involved in the serine biosynthesis pathway. Parkin ubiquitinates and degrades PHGDH, thereby suppressing serine synthesis. Reference Liu, Zhang and Wu86
Studies have demonstrated that the L-serine pathway is related to both Parkinson’s and Alzheimer‘s diseases. Beta-N-methylamino-L-alanine (L-BMAA) leads to protein misfolding in neurons and symptoms of Alzheimer’s-like dementia and Parkinsonism. It was shown that L-serine suppresses the erroneous incorporation of L-BMAA into proteins in the nervous system and improves cognition and electrophysiological dysfunction. Reference Cai, Tian, Zhang, Jiang and Han87
L-serine serves as a precursor for D-serine, which acts as a co-agonist of synaptic NMDA receptors that are necessary for synaptic plasticity. It was observed that L-serine-supplemented diet effectively prevents both synaptic and behavioral deficits in AD mice. Reference Le Douce, Maugard and Veran88,Reference Maffioli, Murtas and Rabattoni89 The effects of serine in AD are a topic of ongoing debate, and its role in PD remains poorly understood. Therefore, further studies need to clarify the relation between serine and neurodegenerative diseases.
Motor and non-motor symptoms
Our cumulative analysis revealed correlations between amino acid levels and disease severity indicators such as the UPDRS III score, Hoehn and Yahr stage and disease duration. Notably, arginine levels negatively correlated with the LEDD, suggesting that alterations in arginine metabolism might be associated with the progression of motor symptoms and medication requirements. Arginine serves as a precursor for nitric oxide (NO) synthesis, which is crucial for neurotransmission and neurovascular function. Reference Yabuki, Shioda and Yamamoto80 Elevated levels of neuronal and inducible NO synthase (NOS) have been observed in the substantia nigra of PD patients and animal models. Reference Aquilano, Baldelli, Rotilio and Ciriolo90 The negative correlation between LEDD and decreasing arginine levels may indicate the depletion of arginine for NO synthesis. However, arginine levels were not significantly different between groups, highlighting the need for further studies to clarify its role in PD symptoms.
Similarly, phenylalanine levels negatively correlated with UPDRS III scores, suggesting a potential link between this amino acid and motor symptom severity, although phenylalanine levels were not significantly different between PD and healthy groups. As a precursor to tyrosine, which is then converted to dopamine, lower phenylalanine levels may reduce the availability of tyrosine and subsequently dopamine, Reference Shnitko, Taylor and Stringfield91 potentially leading to more severe motor symptoms as reflected by the UPDRS III score.
Additionally, citrulline levels tended to decrease with disease duration, highlighting its potential role in disease progression. Our findings also showed that citrulline levels were significantly decreased in the PD group compared to controls, indicating a strong mechanism underlying PD. Citrulline, like arginine, is crucial in the NO production pathway. Although NO has neuroprotective effects, Reference Yabuki, Shioda and Yamamoto80 elevated levels of neuronal and inducible NOS observed in the substantia nigra of PD patients Reference Aquilano, Baldelli, Rotilio and Ciriolo90 suggest that disruptions in these pathways could negatively affect neuronal health and function, potentially influencing disease progression.
We acknowledge several limitations in our study that arise from the quality of the studies and the data available in the literature. A primary limitation is the insufficient reporting of amino acid levels by disease stage, gender and ethnicity, which restricted our ability to perform analyses across these variables. Consequently, we were unable to conduct subgroup analyses based on disease stage or demographic characteristics, limiting our exploration of potential variations in amino acid profiles. Additionally, the studies included in our meta-analysis did not provide information on amino acid levels at various stages of PD or before disease onset, precluding an assessment of when these changes first occurred.
Our meta-analysis also encountered challenges due to high statistical heterogeneity, as reflected by the I2 values in each amino acid analysis. This heterogeneity is indicative of variations in study design, diagnostic criteria and measurement methods, which can affect the consistency of findings across studies. While clinical heterogeneity is expected in meta-analyses, it underscores the necessity for future research to address these issues by providing more detailed and consistent reporting. Future studies should aim to include data stratified by disease stage and patient characteristics.
Conclusion
In conclusion, amino acid metabolism appears to play a critical role in PD pathogenesis. Our study found that BCAAs, particularly valine, might have a negative impact on PD. The effect of BCAA intake on neurotransmitter levels needs to be further investigated in long-term studies. Proline levels were also found to be significantly higher in PD patients, which may contribute to the formation of protein aggregates associated with neurodegenerative diseases or may accumulate due to the metabolic disturbs in the production of collagen and glutamate. Ornithine levels may also contribute to the disease progression as a precursor for glutamate, urea and polyamines, which are closely related to the α-synuclein protein. Furthermore, there were close relationship between decreased citrulline and serine levels in PD patients. This cumulative analysis provides evidence that understanding the mechanisms underlying amino acid metabolism in PD could lead to new therapeutic strategies.
Author contributions
SA: design, data collection and processing, analyses, literature review, writing and editing; DY: data collection; NY: design, supervision, analyses, literature review, writing and critical review.
Funding statement
No funding was received for conducting this study. Sevginur Akdas was supported by TUBITAK-BIDEB 2211 and Turkish Republic Higher Education Institution (YÖK) 100/2000 programs for her PhD degree at Ankara University Institute of Health Sciences under the supervision of Prof. Dr. Nuray Yazihan.
Competing interests
The authors have no relevant financial or nonfinancial interests to disclose.