Introduction

Parkinson’s disease (PD) is the most common neurodegenerative movement disorder currently affecting around 5–6 million predominantly elderly individuals worldwide. PD is expected to be even more commonplace in the near future as the world’s population rapidly ages. In 2030, about 10 million or more are expected to be afflicted with PD (Dorsey et al. 2007). Clinically, the disease is characterized by a constellation of motoric deficits including resting tremor, bradykinesia (slowness in movements), postural instability and rigidity that arises from the depletion of striatal dopamine—a result of the progressive loss of midbrain dopaminergic neurons in the substantia nigra pars compacta (SNpc) that innervate the striatum. This is accompanied by the characteristic neuropathological pattern of eosinophilic intracytoplasmic inclusions known as Lewy bodies (LBs) in surviving neurons in the SN. Notably, α-synuclein, a presynaptic protein whose mutations are causative of familial PD, is a major component of LBs (Polymeropoulos et al. 1997). Although a small percentage of PD cases are inheritable as a result of mutations in genes including α-synuclein, Parkin, LRRK2, PINK1, and DJ-1, exposure to environmental toxins and pesticides, such as paraquat and rotenone, and synthetic toxins, such as 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), can also lead to PD (Goldman 2014). To date, the exact disease mechanisms underlying PD pathogenesis are not fully understood, but studies have consistently implicated aberrant mitochondrial and protein homoeostasis as key contributors to the development of PD, with oxidative stress likely acting as an important nexus (Lim and Zhang 2013).

At present, therapeutic strategies for the PD patient remain largely symptomatic and more often than not, existing pharmacological treatments come with undesirable side effects. Indeed, pharmacological replacement of dopamine with l-Dopa remains as the gold standard for PD treatment despite its association with diminishing effects and problematic drug-induced dyskinesia after prolonged intake. These inadequacies of the gold standard treatment highlight an urgent need to develop more effective disease-modifying drugs for PD. In recent years, several alternative approaches to delay the progression of the disease have been considered and coming under the spotlight recently is nutraceuticals. Functional foods and nutritional supplements, which are common sources of nutraceuticals, are beginning to gain international recognition due to the potential health benefits they harbour when consumed as part of a varied diet on a regular basis and at optimal levels. Consequently, both the scientific community and food industry are motivated to exploit these benefits for the prevention and even treatment of chronic ageing diseases, including PD.

Nutraceuticals and PD

As the word suggests, “nutraceuticals” refers to compounds that are derived from natural sources, and these food or derivatives therein have been clinically substantiated with reasonable scientific evidence to support their beneficial role in the prevention and/or treatment of a particular disease (Chao et al. 2012). It is this very reason that nutraceuticals are more readily accepted by the general populace as a form of treatment. There is a greater tendency to believe that there are fewer side effects associated with compounds derived from natural sources as compared to the many known side effects of synthetic drug compounds.

In the context of PD, several nutraceuticals have been shown to provide neuroprotection in experimental models and may serve as alternatives to synthetic drug compounds like l-Dopa that is known to cause many undesirable side effects. The mechanisms by which they work can be broadly classified into the following themes: (1) reactive oxygen species (ROS)/free radical scavenging; (2) anti-inflammation; (3) iron chelation; (4) modulation of cell signalling pathways; (5) anti-apoptosis; and (6) mitochondrial homoeostasis, although several nutraceuticals essentially function via a myriad of mechanistic pathways rather than adhere to a single mechanism (Fig. 1). Table 1 summarizes existing nutraceuticals that have been found to confer neuroprotection in PD. We will discuss a few examples of these below.

Fig. 1
figure 1

Nutraceuticals as therapeutics for PD. Nutraceuticals for PD can be grouped broadly into six themes based on their neuroprotective properties: (1) iron chelation; (2) cell signalling modulation; (3) anti-inflammation; (4) anti-oxidation; (5) anti-apoptosis; and (6) mitochondrial homoeostasis. However, several nutraceuticals hold multiple properties and function via a myriad of mechanistic pathways rather than adhere to a single mechanism

Full size image
Table 1 Summary of nutraceuticals in PD
Full size table

The seeds of the Mucuna plant, also affectionately known as dopa bean, are well known for containing l-Dopa, the go-to drug for treating PD. Although some species of Mucuna contain more l-Dopa than others, the Mucuna plant is generally favoured for the exploitation of l-Dopa due to its relative abundance of which compared to other plant families that have been studied (Patil et al. 2015). Other microbial and chemical means of synthesizing l-Dopa have also been explored (Surwase et al. 2012; Krishnaveni et al. 2009; Ali et al. 2007; Sikander and Ikram ul 2006), but the Mucuna plant has been preferred as it is a natural and inexpensive source, and it provides additional benefits as an antioxidant (Manyam et al. 2004). In fact, a species of Mucuna plant, Mucuna pruriens, has been shown in both the PD mice model and patients to be more effective than l-Dopa without the accompanying increase in dyskinesia (Hussian and Manyam 1997; Katzenschlager et al. 2004).

Apart from the Mucuna plants, there are many other nutraceuticals that appear to be neuroprotective due to their anti-oxidative properties. Such properties are particularly important in the context of PD as several studies have pointed to oxidative stress, which results in ROS generation and inflammation, as a pivotal contributor to age-related neuronal loss in PD (Jenner 1998). An example of a nutraceutical that possesses both anti-oxidative and anti-inflammatory properties is ginsenoside, a phytoestrogen that is extracted from several species of ginseng (Chen et al. 2005). It executes its anti-oxidative properties by maintaining glutathione levels, and its anti-inflammatory properties are a result of the regulation of several inflammatory pathways including the ROS-NFκB, JNK, P13K/AKT, ERK, IGF-1 receptor signalling pathways and oestrogen receptor pathway. In addition, ginsenoside also reduces the levels of nigral iron of MPTP-treated mice by regulating the expression of iron transport proteins (Wang et al. 2009b). This is of importance as the build-up of iron in conjunction with ROS at the site of neurodegeneration is thought to constitute a major trigger in neurotoxicity and neuronal demise in PD (Zecca et al. 2004). As such, nutraceuticals like ginsenoside that can inhibit pro-inflammatory and oxidative processes should, in theory, be able to attenuate dopaminergic neuronal damage. Indeed, it has been demonstrated that ginsenoside protects against toxicities and dopaminergic neuronal loss induced by PD toxins including 6-hydroxydopamine (6-OHDA) and MPTP (Chen et al. 2005; Xu et al. 2009). Due to its role in the regulation of JNK signalling, ginsenoside also possesses anti-apoptotic properties. Hence, another postulated mechanism through which the neuroprotective effect of ginsenoside is facilitated is its reduction of c-Jun phosphorylation, which prevents pro-apoptotic JNK signalling and dopaminergic neuronal loss during MPTP-induced neurotoxicity (Leppa and Bohmann 1999).

Besides ginseng, dietary soy and peanut products have also been reported to have similar anti-apoptotic effects. Soy and peanut are rich sources of genistein, a phytoestrogen-like ginsenoside. Genistein acts as a tyrosine kinase inhibitor that attenuates protein kinase C (PKC) activation and thereby downstream apoptotic effects (Kaul et al. 2005; Baluchnejadmojarad et al. 2009). Another potent anti-apoptotic nutraceutical that has been shown to protect against PD toxin-induced neurotoxicity is Ginkobiloba extract EGb 761. EGb 761 prevents the formation of apoptosome and the apoptotic cascade by blocking cytochrome-c release (Liu et al. 2008; Yeh et al. 2009; Nevado et al. 2010). Like ginsenoside, EGb 761 also attenuates the phosphorylation of c-Jun (Shi et al. 2009) and furthermore inhibits the cleavage of caspase-3 (Liu et al. 2008; Shi et al. 2009), thereby preventing DNA fragmentation, a hallmark of apoptosis. By blocking apoptosis through various mechanistic pathways, genistein and EGb 761 were found to attenuate dopaminergic neuronal loss and reduce associated locomotion impairment in 6-OHDA and MPTP mice models (Ahmad et al. 2005; Baluchnejadmojarad et al. 2009; Rojas et al. 2012; Yang et al. 2001).

Nutraceuticals and Mitochondrial Homoeostasis

As mentioned earlier, aberrant mitochondrial homoeostasis is commonly implicated in PD pathogenesis. Intuitively, one would propose that nutraceuticals that have a role in mitochondrial regulation can potentially mitigate PD pathology. Coenzyme Q10 (CoQ10) is a component of the mitochondrial electron transport chain and participates actively in ATP generation. It is noteworthy to mention that in PD animal models, CoQ10 attenuates MPTP-induced neurotoxicity, possibly due to its unique electron-accepting property, rendering it critical to the electron transfer between mitochondrial complex 1 and other complexes of the electron transport chain (Beal et al. 1998; Cleren et al. 2008). Although it has been proposed as a therapeutic strategy for PD (Shults 2005), clinical trials involving CoQ10 have been conflicting. While one study by Shults et al. reported a dose-dependent reduction in functional decline, another study by Muller et al. observed only mild symptomatic benefit (Shults et al. 2002; Muller et al. 2003). Other nutraceuticals that are reported to be neuroprotective due to their role in preserving mitochondrial complex 1 activity include curcuminoids from turmeric (Jagatha et al. 2008) and the earlier-mentioned Mucuna plant (Manyam et al. 2004). Yet another key player in the maintenance of ATP levels is phosphocreatine, an energy reserve in skeletal muscles and brain. Notably, treatment with creatine appears to rescue parkinsonian phenotypes in both human subjects and animal models. Specifically, diet supplement of creatine was found to improve the mood and reduce the dosages required for dopamine (DA) replacement therapy in PD patients (Bender et al. 2006), as well as reduce dopaminergic neuronal loss in SNpc of MPTP-treated mice (Matthews et al. 1999). It certainly seems that mitochondrial homoeostasis is a common targeted pathway for nutraceutical therapy notwithstanding the controversy surrounding CoQ10.

Epigallocatechin-3-gallate (EGCG), a main green tea-derived catechin, is a nutraceutical that is frequently featured in PD, perhaps due to its numerous putative neuroprotective mechanisms that is not limited to mitochondrial homoeostasis (Pan et al. 2003). These include anti-oxidation, iron chelation, ROS scavenging and anti-apoptotic properties. Moreover, EGCG crosses the blood–brain barrier easily, making it an attractive compound for therapy. Besides the myriad of properties, EGCG has been reported to be an AMPK activator (Spasic et al. 2009; Hwang et al. 2009). It increases cytosolic Ca2+ levels, thereby influencing the activity of Ca2+-/calmodulin-dependent protein kinase kinase (CaMKKβ), an upstream kinase of AMPK (Kim et al. 2014). The activation of AMPK in the presence of EGCG is therefore likely to be mediated by CaMKKβ. As discussed below, AMPK activation by EGCG has been demonstrated to be neuroprotective.

AMPK Activation and Neuroprotection

AMPK is a central energy sensor and regulator that is normally activated in response to diminishing energy supply, e.g. ATP depletion or glucose starvation (Li et al. 2012). Given the critical role of AMPK in energy homoeostasis, it is perhaps not surprising to note that AMPK has profound influence on mitochondrial homoeostasis amidst a plethora of metabolic events that it governs. It is well documented that AMPK works through peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) to promote mitochondrial biogenesis (Lee et al. 2012). Although the mechanism by which AMPK upregulates PGC-1α activity remains unclear, studies have suggested that it could directly phosphorylate PGC-1α (Ng et al. 2012b) or indirectly activate the transcriptional coactivator by promoting its deacetylation through the NAD+-dependent deacetylase SIRT1 (Ng et al. 2012b). Interestingly, a recently identified AMPK target is UNC-51-like kinase 1 (ULK1), a mammalian ortholog of the yeast Atg1 kinase that acts as a key initiator of the autophagy cascade (Carroll et al. 2014; Martin et al. 2014). The activation of ULK1 by AMPK promotes autophagy, including mitophagy. Accordingly, when ULK1 function is impaired, it results in the accumulation of abnormal mitochondria with reduced potential (Martin et al. 2014). Similarly, when the known AMPK-mediated phosphorylation sites on ULK1 are abolished, autophagy is also impaired (Martin et al. 2014), suggesting that the clearance of damaged mitochondria is dependent on the AMPK–ULK1–autophagy pathway. More recently, a study by Toyama et al. reported that AMPK is required for rotenone-induced mitochondrial fission (Toyama et al. 2016), an essential process to isolate damaged mitochondria and promote mitophagy (Twig et al. 2008). Importantly, they identified a novel substrate of AMPK, mitochondrial fission factor (MFF), and found that the presence of non-phosphorylatable MFF resulted in defective mitochondrial fission, further emphasizing the importance of AMPK in regulating mitophagy.

Notwithstanding earlier discussion that AMPK activation helps to maintain mitochondrial quality control and should theoretically promote cellular survival, the role of AMPK activation in neuroprotection remains controversial. Similarly, AMPK activation is also a double-edge sword in the case of PD, promoting neurodegeneration under some circumstances yet aggravating in others. For instance, a study by Kim et al. contradicts a neuroprotective role of AMPK in PD and found that AMPK mediates dopaminergic neuronal atrophy in 6-OHDA-lesioned mice. Moreover, metformin-induced AMPK activation accelerates rather than retards 6-OHDA-induced neuronal loss in these mice (Kim et al. 2013). Subsequently, Xu et al. also observed in a related study similar detrimental effects of AMPK activation in primary neurons treated with 6-OHDA, MPTP or rotenone (Xu et al. 2014). Nonetheless, there are also several reports supporting a neuroprotective role of AMPK in PD. In a recent study by Patil et al., MPTP-treated mice on chronic metformin regimen demonstrated enhanced antioxidant activity and brain-derived neurotrophic factor (BDNF) levels, thereby rendering protection against dopaminergic neuronal loss induced by MPP+ (Patil et al. 2015). In a related study, AMPK is activated upon MPTP treatment, and when AMPK activity is downregulated by compound C, neurotoxicity is enhanced (Choi et al. 2010). Supporting this, a recent study demonstrated similar findings albeit in cultured cells exposed to rotenone (Wu et al. 2011). Given that both MPP + and rotenone are complex I inhibitors, the rescue of the in vivo and in vitro PD models by AMPK activation is consistent with its role in the maintenance of mitochondrial homoeostasis. Importantly, it is also noteworthy to mention that a recent cohort-based study involving 800,000 subjects in a Taiwanese population revealed that metformin-inclusive sulfonylurea therapy significantly reduces PD risk in individuals with type 2 diabetes (Wahlqvist et al. 2012), suggesting the neuroprotective effects of AMPK activation. Collectively, these findings suggest that AMPK activation may be beneficial for the disease.

On a related note, we have recently found that EGCG facilitates neuroprotection in PD by mediating mitochondrial regulation via AMPK activation. Using Drosophila as a model, we found that EGCG administration ameliorates the pathological phenotypes of parkin null PD flies, including prominent mitochondrial abnormalities and progressive loss of selected dopaminergic neuronal clusters that are accompanied by an age-dependent decline in locomotor ability (Ng et al. 2012a). These are disease phenotypes that bear resemblance to that of human PD (Green and Kroemer 2004; Whitworth et al. 2005). Importantly, the EGCG-mediated protective effects require AMPK as genetic inactivation of AMPK abolishes the neuroprotective effects while subsequent genetic restoration of AMPK and pharmacological activation of AMPK with potent AMPK activators (metformin or AICAR) reproduce these beneficial effects (Ng et al. 2012a). In contrast, treatment of parkin null flies with another compound, i.e. Baicalein—an established antioxidant (Shieh et al. 2000), failed to ameliorate the observed parkinsonian phenotypes (unpublished observation). Accordingly, we speculated that AMPK activation rather than anti-oxidation may be involved in EGCG-mediated protective effects. In a similar fashion, Drosophila LRRK2 mutants could be rid of its pathological phenotypes via pharmacological treatment with EGCG, metformin or AICAR or the co-expression of a constitutively active AMPK mutant (Ng et al. 2012a), suggesting that this approach may be relevant to different forms of PD. Consistent with our results, Ferretta et al. demonstrated similar benefits using resveratrol (Ferretta et al. 2014), another nutraceutical found in the skin of grapes and berries that is known to be a relatively strong AMPK activator (Dasgupta and Milbrandt 2007). In patient’s fibroblasts harbouring parkin mutations, they found that resveratrol increased mitochondrial biogenesis and improved oxidative phosphorylation (Ferretta et al. 2014).While AMPK activation brings about a virtually complete rescue of PD pathological phenotypes in flies, we are currently on the endeavour to dissect precisely how it happens. Nevertheless, existing knowledge of AMPK has shed light on the possible mechanistic pathways via which AMPK-mediated neuroprotection may occur. As mentioned earlier, AMPK is capable of positively regulating mitochondrial biogenesis via PGC-1α and also mitophagy. This is further supported by the resveratrol-based study by Ferretta et al. that also revealed the ability of AMPK to enhance autophagy flux in parkin-mutant fibroblasts (Ferretta et al. 2014). Hence, it is likely that enhanced mitochondrial biogenesis and/or mitophagy could help in the maintenance of a viable pool of bioenergetically competent mitochondria necessary for dopaminergic neuronal survival. Accordingly, approaches towards promoting these processes may be of therapeutic value for PD.

Conclusion

As with many other neurodegenerative diseases, PD is a debilitating disorder that gradually robs an individual of his/her fundamental bodily functions. Although much effort has been put into advancing therapeutic strategies for this disease, many conventional and existing treatment options are unfortunately accompanied by several undesirable effects despite their ability to provide symptomatic relief. Fortunately, with the advent of nutraceuticals, an alternative avenue to tackle this seemingly evasive biological problem has been provided. Nutraceuticals, by virtue of their origin from naturally available food or food products, appear to be a favourable treatment option since harnessing therapeutic strategies from natural resources can potentially avoid side effects. As a matter of fact, many of the nutraceuticals discussed in this review have been shown to be not only preventive but also therapeutic for PD. Significantly, we have highlighted in this paper how we and others have demonstrated the ability of green tea-derived catechin EGCG to rescue PD pathological outcomes, possibly through enhancing mitochondrial homoeostasis. However, despite many promising reports about the role of nutraceuticals in neuroprotection for PD, we acknowledged that it is early days yet as several mechanistic gaps remain unanswered. Notwithstanding this, the recognition that nutraceuticals might be of therapeutic benefits offers countless opportunities to explore other natural compounds that have not been looked at in terms of their potential neuroprotective roles in PD. For example, ergothioneine (EGT) is a naturally occurring amino acid found in mushrooms that protects mitochondria from oxidative stress (Cheah and Halliwell 2012) and was found to accumulate at significantly lower levels in PD patients compared to healthy controls (Hatano et al. 2015), thus suggesting a therapeutic potential of EGT for the disease. In addition, future research could also direct efforts towards better understanding the effects of nutraceuticals in combination with existing drug therapies for PD patients, in order to derive improved outcomes for the PD patient. The delivery of nutraceuticals could also be optimized in order to maximize their neuroprotective effects. As a parting note, it is hopeful to envisage that one day we would be able to simply modify our diet to prevent or mitigate the progression of this debilitating disease.