Parkinson’s disease (PD) has become the second most common neurodegenerative disease in the elderly. The incidence and prevalence of PD rise with age in a steadily increasing manner (de Rijk et al., 1995; Pringsheim et al., 2014). As the global population ages, the prevalence of PD is projected to double over the next two decades (Dorsey et al., 2018a,b). It occurs rapidly worldwide without significant epidemiological differences between regions (Dorsey et al., 2018a; Bloem et al., 2021), but the number of patients is increasing rapidly in China, which accounted for more than a quarter of all patients in 2020 (Dorsey et al., 2018a,b; Qi et al., 2021).

PD is a chronic and incurable disease that is characterized by the presence of bradykinesia, muscular rigidity, resting tremor, and postural instability. It is also associated with various non-motor symptoms such as cognitive impairment, hyposmia, constipation, and mood disorders (Jankovic and Tan, 2020; Bloem et al., 2021). As a result, PD leads to a significant reduction in the quality of life of patients and their families with symptoms worsening over time (Hoogland et al., 2017). The main pathological changes of PD, the loss of dopaminergic neurons in the substantia nigra (SN) and the presence of Lewy bodies (LBs) and Lewy neurites (LNs; Dickson, 2018; Foffani and Obeso, 2018; Chen et al., 2020; Bloem et al., 2021), which consist of α-Synuclein (de Rijk et al., 1995; Hoogland et al., 2017), are universally acknowledged. However, the precise cause of this pathological change still remains unclear. Genes, lifestyle, aging, oxidative stress, and other factors are responsible for the degeneration and death of dopamine neurons (Martignoni et al., 2007; Chen et al., 2020; Jankovic and Tan, 2020; Bloem et al., 2021). Thus, the identification of treatable conditions that contribute to PD may reduce the burden on public health and the economy in a rapidly aging society (Fredericks et al., 2017).

Methionine, an essential amino acid derived from proteins in food, is metabolized in the erythrocytes to produce homocysteine (Hcy), a sulfur-containing amino acid. Homocysteine is primarily metabolized in the liver. The conversion of methionine to active methyl donor S-adenosylmethionine (SAM) occurs through the catalysis of S-adenosylmethionine synthetase, in which methionine combines with ATP. Subsequently, SAM is converted to S-adenosine homocysteine (SAH) by methyltransferase, resulting in the hydrolysis of Hcy and adenosine (Martignoni et al., 2007; Tinelli et al., 2019). The metabolism of Hcy involves two foremost pathways (de Rijk et al., 1995; Postuma and Lang, 2004; Martignoni et al., 2007; Sharma et al., 2015; Murray and Jadavji, 2019; Tinelli et al., 2019): The remethylation pathway, which converts Hcy to methionine, requires methionine synthetase (MS) and vitamin B12 as coenzymes. 5-Methyltetrahydrofolate (5-MTHF) supplies the methyl group for this pathway (Pringsheim et al., 2014); The sulfidation pathway is an irreversible reaction that requires cystathionine-β-synthetase (CBS) and vitamin B6 as coenzymes. It converts Hcy to cystathionine, which is then converted to cysteine (Figure 1). Based on the metabolism of Hcy, the enzyme MTHFR plays an important role in converting Hcy to methionine with vitamin B12 and folate as cofactors (Martignoni et al., 2007; Hu et al., 2013). Hence, it is hypothesized that the plasma levels of Hcy are negatively correlated with plasma folate and vitamin B12 levels.

Studies have shown that hyperhomocysteinemia (Hhcy), characterized by plasma Hcy levels exceeding 15 μmol/L, may be an independent risk factor for Parkinson’s disease (Martignoni et al., 2007; Licking et al., 2017; Xie et al., 2017; Li et al., 2020). PD patients have been found to have significantly higher Hcy levels compared to healthy individuals worldwide (Dong and Wu, 2020; Rahnemayan et al., 2022). Furthermore, Hcy levels in PD are associated with folate, vitamin B12, age, genetics, sex, L-dopa treatment, cognitive impairment, and so on (Hu et al., 2013; Collier et al., 2017). Both in vitro and in vivo studies have demonstrated that Hcy plays a role in the pathogenesis of mesencephalic dopaminergic neuronal death in PD (Wei et al., 2016).

However, the association between Hcy levels and the severity of PD has not been previously reported (Deng et al., 2022). In fact, previous studies have suggested that Hcy does not play a significant role in the pathogenesis of PD. For instance, studies by Rodriguez-Oroz and Camicioli found an unclear relationship between plasma Hcy levels and cognitive impairment in PD patients (Camicioli et al., 2009; Rodriguez-Oroz et al., 2009), contradicting the results of other studies (Song et al., 2013; Shen, 2015; Rahnemayan et al., 2022).

To sum up, the association between Hcy levels and PD remains controversial. Consequently, our study aimed to provide valuable insights and useful references for the clinical practice of PD, by systematically analyzing the association between plasma Hcy, folate, vitamin B12 levels, and PD.

The meta-analysis was a quantitative systematic analysis with 25 articles included. Drawing on this, we systematically analyzed the association between Hcy, folate, and vitamin B12 levels and the risk of PD. Our results confirmed that PD patients had significantly higher Hcy levels compared to the controls, as well as lower folic acid and vitamin B12 levels. These results of our meta-analysis were in accordance with most studies (Xie et al., 2017; Christine et al., 2018; Dong and Wu, 2020; Boelens Keun et al., 2021; Periñán et al., 2023). On the contrary, Wei et al.’s (2016) found no correlation between PD pathogenesis and high Hcy levels, nor did it find any association with vitamin B12 and folate deficiency. We think there are two reasons for this discrepancy. First, the study only included cases from Luliang City, a mountainous region in China, making the results non-representative of the wider population in China or globally. Second, Wei employed qualitative analysis instead of quantitative analysis to analyze plasma Hcy levels. Furthermore, the cases were divided into two subgroups using a cut-off of 22.175 𝜇mmol/L instead of the recommended 14 𝜇mmol/L serum Hcy, which could have significantly impacted the result.

Previous studies have shown that plasma folate and vitamin B12 levels are decreased in patients with Parkinson’s disease when compared to healthy individuals (Postuma and Lang, 2004; dos Santos et al., 2009; Jankovic and Tan, 2020). PD patients may experience reduced absorption of vitamin B12 and folate due to gastrointestinal dysfunction, a common nonmotor symptom of PD, accompanied by changes in gut microbiota (Chiang and Lin, 2019; Rosario et al., 2021). However, the treatment of PD patients with levodopa and catechol-O-methyltransferase inhibitors (COMT-I) may lower vitamin B12 and folate levels (Triantafyllou et al., 2007; Anamnart and Kitjarak, 2021), although the reasons for this effect remain unclear. Vitamin B12 provides protection against neuronal apoptosis by preventing DNA damage and is also a cofactor for two essential enzymes, methionine synthase, and methylmalonyl Coenzyme A mutase, involved in myelination. A deficiency in vitamin B12 also contributes to myelination of the peripheral and central nervous system (Moore et al., 2012; Christine et al., 2018). Similarly, folate acts as an antioxidant, reducing plasma Hcy levels and protecting DNA from damage (Martignoni et al., 2007; Murray and Jadavji, 2019). Additionally, studies have shown that older individuals with low blood concentrations of folate and vitamin B12 perform poorly on cognitive tests, such as tests of memory and nonverbal abstract thinking (Ajibawo-Aganbi et al., 2020). Supplementing with vitamin B12 and folate has been proven to significantly reduce vascular events and dementia (Ansari et al., 2014; Craenen et al., 2020; Anamnart and Kitjarak, 2021; Boelens Keun et al., 2021). However, a prospective, population-based cohort study suggested that dietary vitamin B6 may decrease the risk of PD, rather than folate and vitamin B12 (de Lau et al., 2006). Nonetheless, we still recommend the supplementation of vitamin B12 and folate in all PD patients as an adjuvant.

Furthermore, our study confirmed that hyperhomocysteine had a significant relationship with PD since a deficiency in vitamin B12 and folate leads to an increase in Hcy as they take part in the remethylation pathway of Hcy metabolism (Figure 1). Hhcy may play a crucial role in neurodegeneration in several ways. (1) High levels of Hcy induce neuronal apoptosis and worsen dopaminergic degeneration in PD patients through oxidative stress (Postuma and Lang, 2004; Martignoni et al., 2007), a clear cause and early feature of PD. Moreover, Hcy increases the vulnerability of dopaminergic neurons to oxidative stress damage (Postuma and Lang, 2004; Martignoni et al., 2007; Obeid et al., 2009). Hcy results in the generation of vast amounts of oxygen free radicals through two main mechanisms. First, the accumulation of SAH, a consequence of high Hcy, inhibits methyltransferases and increases catecholamine levels (Coppola et al., 2000), while decreasing the activity of antioxidant enzymes in cells such as SOD and catalase (Bhattacharjee and Borah, 2016). Second, high Hcy increases the activity of NADPH oxidase, further promoting oxidative stress (Takeno et al., 2016). (2) The dysfunction of ATP synthesis in mitochondria is related to Hcy, which leads to the death of neurons (Mattson and Shea, 2003; Jankovic & Tan, 2020). One possible mechanism for this dysfunction is the aggregation of α-synuclein (Zhou et al., 2023), a protein that regulates dopamine biosynthesis and homeostasis (Foffani and Obeso, 2018; Rocha et al., 2018), participates in endoplasmic reticulum/Golgi trafficking (Obeid et al., 2009), and is also influenced by Hcy (Zhou et al., 2023). (3) Hcy can directly damage neurons through nerve immune inflammation via glial cells (Postuma and Lang, 2004; Chen et al., 2015; Yan et al., 2018; Lee et al., 2019) and through a neurotoxic effect (Postuma and Lang, 2004; Martignoni et al., 2007; Murakami et al., 2010; Tinelli et al., 2019) that leads to cell death by increasing intracellular calcium levels (Foffani and Obeso, 2018). (4) It is commonly recognized that Hcy is a risk factor for cardiovascular disease as it damages vascular endothelial cells (Martignoni et al., 2007; Foffani and Obeso, 2018). Thus, Hcy may indirectly damage neurons by impacting the nutrition and blood supply to the nervous system, hastening the onset and progression of PD.

Subgroup analysis was conducted to investigate the factors contributing to the high heterogeneity in Hcy levels. While the results of subgroup analyses align with the overall analysis regarding the relationship between plasma Hcy level and PD, high heterogeneity remains apparent in certain subgroups.

Levodopa, the gold standard for the treatment of PD (Bloem et al., 2021), is acknowledged as an important factor contributing to an increase in plasma Hcy and SAM levels (Postuma and Lang, 2004; Lamberti et al., 2005; Hu et al., 2013). The methylation process of L-dopa to 3-O-methyldopa is catalyzed by catechol-O-methyltransferase (COMT), which consumes SAM as the methyl-group donor (Brosnan et al., 2004; Postuma and Lang, 2004; Lamberti et al., 2005; Zoccolella et al., 2009). Consequently, the levels of SAH, which can be readily hydrolyzed to Hcy, increase, ultimately leading to Hcy accumulation. Recognizing the association between elevated plasma Hcy and reduced methylation capacity, we conducted a subgroup analysis to evaluate the effect of levodopa therapy. The result showed that both levodopa-treated and non-levodopa-treated PD patients had higher Hcy levels compared to healthy individuals. Moreover, Hcy levels were higher in levodopa-treated PD patients compared to non-levodopa-treated PD patients. This indicates that both levodopa treatment and PD itself contribute to an increase in Hcy levels. Given that hyperhomocysteinemia is a well-established risk factor for multiple health complications, including cardiovascular disease and Alzheimer’s disease, we strongly recommend strict monitoring of the metabolic pathway components of levodopa in PD patients undergoing levodopa treatment.

Aging is confirmed as another main risk factor for nigrostriatal degeneration in several studies (Martignoni et al., 2007; Foffani and Obeso, 2018). No difference was found between Hcy and PD patients under the age of 60 in our research. The vulnerability of the brain region to oxidative insult in PD is attributed to the declining capacity of nigral dopamine neurons to offset ROS production with aging (Trist et al., 2019). Furthermore, a study discovered that α-syn within the cytoplasm of SN neuron cell bodies increases with aging (Collier et al., 2017), partly resulting in mitochondrial dysfunction and PD. Moreover, subgroup analysis also revealed varying concentrations of Hcy in PD patients of different ages. It is difficult to conclude that aging is a significant risk for high plasma Hcy, considering that older patients may have been receiving increasing doses of levodopa for many years and could have already developed PD. In summary, measuring plasma Hcy may be an alternative approach for predicting the onset and progression of PD in older individuals, particularly those aged 60 and above. Nonetheless, further research is necessary to establish critical values of Hcy for different age groups.

Cognitive impairment in PD is inevitable within 20 years of diagnosis, and up to 80% of PD patients develop dementia over the course of the disease (Hoogland et al., 2017). In addition to apoptosis, excitotoxicity, and oxidative stress, atherosclerosis resulting from high Hcy levels can also contribute to cognitive impairment (Postuma and Lang, 2004). The severity of motor symptoms in PD is measured using the Hoehn and Yahr Scale. Stages 1 and 2 represent a mild motor disorder that may be easily ignored by most patients. Stages 2 and 3 represent a moderate motor disorder that may affect the patient’s daily life. Stages 4 and 5 represent a severe motor disorder that prevents patients from living independently. We found that Hcy was also positively correlated with the severity of symptoms in PD, including cognitive and mobility impairment. Our results showed a significant increase in Hcy levels only in mild and advanced stages of PD. However, individual studies on the association between plasma Hcy level and the risk of dementia occurrence did not reach a consensus. There are four possible explanations for the conflicting results of Rodriguez-Oroz et al. (2009) and Camicioli et al. (2009), who found that high Hcy levels were not a risk factor for cognitive decline. First, their sample sizes were small, with only 89 and 51 PD patients respectively, which may not be representative of all PD patients. Second, they did not apply MMSE or PD-CRS to identify PDD, instead using Petersen’s criteria for Rodriguez-Oroz and the Dementia Rating Scale for Camicioli. Additionally, Camicioli focused on patients aged 71.5 whose Hcy levels may increase physiologically. Moreover, detailed data on Hcy in Rodriguez-Oroz’s study suggested a ceiling effect, indicating that the concentration of plasma Hcy might be lower than the actual levels. Hence, we still think that reducing plasma Hcy levels may be an effective approach to delaying disease progression. Although we observed significant differences in PD patients with cognitive impairment or mild/severe mobility impairment, we cannot solely attribute high plasma Hcy to the severity of PD due to the progressive nature of the disease and potential confounding factors such as levodopa usage. Nevertheless, Hcy could be a useful supplementary tool rather than a definitive method for assessing the severity of PD.

The risk of developing PD is twice as common in men than women (Dorsey et al., 2018a; Cerri et al., 2019), but we failed to find a significant difference in Hcy between genders. However, the SMD for Hcy in males was slightly higher than in females. It is well known that estradiol increases the synthesis, release, reuptake, and turnover of DA (Kompoliti, 2003). Additionally, estradiol has anti-inflammatory properties that protect neurons by attenuating microglia activation and modulating microglia polarization toward a cytoprotective phenotype (Cerri et al., 2019; Bustamante-Barrientos et al., 2021). However, estrogen and its derivatives decline in women after menopause, which typically occurs around 50 years old. Since the average age of patients included in our study was mostly over 60, the age at which the protective effect of estradiol appears to diminish in women, our study found no apparent difference between males and females.

A dramatic variation in Hcy levels among PD patients was observed across different regions. One possible explanation for this difference is the influence of lifestyle. Previous studies have indicated that an increased consumption of dairy products and alcohol is associated with high Hcy levels in PD patients (Ishihara and Brayne, 2005; Park et al., 2005; Jankovic and Tan, 2020). Conversely, some studies have suggested that the intake of tea or coffee and smoking may potentially decrease the risk of PD. (Ishihara and Brayne, 2005; Li et al., 2015; Ascherio and Schwarzschild, 2016) Furthermore, there is evidence to suggest that exercise is related to a reduction in Hcy levels in PD patients (Nascimento et al., 2011; Yang et al., 2015).

This study had several limitations. First, only studies published in English were included in the analysis. Second, our meta-analysis did not include any prospective research that proves Hcy is a risk factor for PD. Moreover, the studies included in our analysis were predominantly conducted in East-Asian and European countries. The reason for the lack of American studies is that the research from America we found primarily focused on PD patients and lacked data on healthy individuals (Licking et al., 2017). In addition, it is worth noting that although some studies indicated a relationship between Hcy and other factors such as genetics (Postuma and Lang, 2004; Todorović et al., 2006; Zhu et al., 2015), types of PD (Christine et al., 2018; Shen et al., 2019), depression (Zhang et al., 2023), and minor hallucinations (Zhong et al., 2022), we did not analyze these connections in our study.

Our investigation included all studies that provided detailed raw data on Hcy, vitamin B12, and folate levels in PD and healthy individuals. To ensure comprehensiveness, we conducted thorough searches of five databases. To mitigate the effect of different units and measurement methods, we employed SMD to estimate the difference between PD patients and controls. Subgroup, sensitivity, and bias analyses were performed to identify potential sources of high heterogeneity. As a result, our results can be considered reliable and precise.

In summary, PD patients had lower folate and vitamin B12 levels, but significantly higher Hcy levels. These findings have implications for the treatment of PD in patients. Although hyperhomocysteinemia is not specific to PD, the measured Hcy can be seen as a surrogate marker of vitamin B12 and folate deficiency. Nonetheless, further in-depth research is warranted to confirm if Hcy can predict the onset and estimate the progression of PD, as well as to determine if it can be targeted for treatment.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.

YQ: Conceptualization, Data curation, Formal analysis, Methodology, Validation, Visualization, Writing–original–draft, Writing – review & editing. ZG: Data curation, Formal analysis, Methodology, Visualization, Writing – review & editing. QX: Data curation, Formal analysis, Visualization, Writing – review & editing. JX: Data curation, Formal analysis, Visualization, Writing – review & editing. RO: Formal analysis, Methodology, Visualization, Writing – review & editing. HS: Formal analysis, Methodology, Project administration, Supervision, Writing – review & editing. QW: Conceptualization, Data curation, Formal analysis, Methodology, Writing–original–draft, Writing – review & editing.