By conceptualizing the tryptophan-melatonin pathway as an integral aspect of core cellular function via its interactions with mitochondrial metabolism, and therefore in the regulation of oxidant/antioxidant ratio, ROS-driven microRNAs (miRNAs) and therefore patterned gene expression as well as ATP production, mitophagy regulation and α-synuclein levels/fibrillization, the tryptophan-melatonin pathway is intimately linked to wider aspects of PD pathophysiology.
4.1. Platelets and Erythrocytes
Importantly, wider systemic cells may be similarly dysregulated in PD, with relevance to α-synuclein production, including erythrocytes [
28] and platelets [
29]. Both erythrocytes and platelets are significant producers of α-synuclein, with erythrocyte α-synuclein proposed to seed the increase in gut α-synuclein in PD [
28], thereby initiating alterations in the gut microbiome and gut permeability, whilst also providing α-synuclein to be retrogradely transported via the vagal nerve to the CNS [
1]. There is a growing appreciation of the role of platelets in neurodegenerative disorders and other medical conditions, including cancer [
30]. Importantly, like SNpc dopamine neurons, platelets are regulated by systemic processes, including circadian melatonin and gut microbiome-derived butyrate and lipopolysaccharide (LPS), with platelet-derived serotonin being an important serotonin source for the initiation of the melatonergic pathway across body and CNS microenvironments [
30]. Platelet function may therefore be synchronized with wider changes in PD via regulation by systemic processes as well as genetic factors and epigenetic processes. Platelets are also a significant source of sphingosine-1-phosphate (S1P), which regulates enteric glial cell function and lymphocyte chemotaxis to the gut [
30], as well as attenuating SNpc dopamine neuron loss [
31]. The enteric nervous system [
32], and especially enteric glial cells may be crucial regulators of the interface of the gut, immunity and vagal nerve in PD, as in other gut-linked medical conditions [
9].
4.2. Stress, HPA axis and morning cortisol awakening response
As with many other medical conditions, stress exacerbates PD pathophysiology, including via glucocorticoid receptor (GR) effects on motor and non-motor PD symptoms, as shown in preclinical PD models [
33]. Stress induced hypothalamus-pituitary-adrenal (HPA) axis activation is classically associated with the emergence of psychiatric conditions, especially depression, in PD [
34,
35] and many other medical conditions [
36,
37]. However, there is a growing realization that such psychiatric presentations do not simply represent comorbidities. Rather these conditions emerge in association with pathophysiological changes relevant to classical PD pathophysiology, including from cortisol/GR effects [
33]. Importantly, cortisol/GR effects do not arise solely from stress-linked HPA axis activation but also from the morning cortisol awakening response (CAR). The morning CAR surge starts just before awakening and lasts for 30 minutes. The morning CAR is an accentuation of the rise in plasma cortisol seen typically over sleep at night [
38,
39], see
Figure 2. Although cortisol/GR activation regulates thousands of genes across all body cells, including immune cells, there is a surprising lack of investigation as to the role of CAR. Generally, CAR, along with the corelease of adrenal aldosterone, is proposed to increase respiration and blood pressure in ‘preparation for the coming day’ [
40]. However, recent work indicates that CAR may be an important regulator of pathophysiology across diverse medical conditions, including neurodegenerative conditions and cancer, as well as aging [
41,
42].
The GR is typically held in a cytoplasmic complex with heat shock protein (hsp)90 and p23. Following GR activation by cortisol, the GR is translocated to the nucleus where it binds to the glucocorticoid response element (GRE) in the promotor of thousands of genes. The GR can also interact with other transcription factors in the nucleus to significantly regulate their transcription efficacy. Two of the factors shown to be decreased in PD, namely melatonin and butyrate, prevent the GR nuclear translocation, thereby suppressing the consequences of both stress and CAR driven GR activation [
43,
44]. Butyrate effects are mediated via its capacity of a HDACi, thereby acetylating the GR and/or hso90 to prevent GR nuclear translocation [
45]. It has been proposed that many medical conditions may arise at night, involving aging-linked suppressed melatonin and therefore a decrease in the immune- and antioxidant-dampening effects of melatonin at night [
46,
47]. The suppression of night-time pineal and local melatonin as well as gut microbiome-derived butyrate will therefore have consequences for how night-time processes shape the nature of how morning CAR primes body and CNS cells/systems ‘for the coming day’. This may be of some importance given the dramatic impact of cortisol/GR on thousands of genes in all cells.
Another factor intimately associated with the regulation of the GR as well as providing protection in PD is bcl2-associated athanogene (BAG)-1. BAG-1 restores the function of the PARK7 gene protein, DJ-1, in an in vitro model of hereditary PD, whilst also attenuating α-synuclein toxicity in PD models [
48]. Two common medications, lithium and sodium valproate, show some efficacy in PD clinical trials and models, with effects proposed to be mediated via increased antioxidants [
49,
50]. However, both lithium and valproate increase BAG-1 [
51], suggesting a role of BAG-1 in their efficacy. Like melatonin and butyrate, BAG-1 also prevents GR nuclear translocation [
51], with recent data indicating that BAG-1 may also translocate the GR to mitochondria, where it can cross into the mitochondrial matrix to bind with IL-6 and PDC and thereby impacting on mitochondrial ATP production [
52]. As melatonin, via epigenetic processes and possibly directly, can increase BAG-1 [
42], BAG-1 may be another factor to regulate stress-linked HPA axis activation and morning CAR, with relevance to PD pathoetiology and pathophysiology.
In the absence of an attenuation of GR nuclear translocation by melatonin, butyrate and BAG-1, no increase in cortisol production during stress or CAR would be necessary for an elevation in GR nuclear translocation. Such relative enhancement of HPA axis and CAR driven GR activation of thousands of genes with the GRE in their promotor will have significant wide-ranging consequences, including for the tryptophan-melatonin pathway. GR activation of the GRE increases tryptophan 2,3-dioxygenase (TDO), which is especially highly expressed in the liver and astrocytes, to drive the conversion of tryptophan to kynurenine and other kynurenine pathway products, thereby depleting the availability of tryptophan for the tryptophan-melatonin pathway. Circulating kynurenine is readily taken up over the blood-brain barrier into the CNS, like tryptophan, via LAT-1 (figure 1). 60% of CNS kynurenine is derived from the circulation, and therefore from cells out-with the CNS [
53]. When taken up into the CNS, kynurenine can be converted to a number of powerfully neuroregulatory factors, including the excitatory picolinic acid and the excitotoxic quinolinic acid, as well as to kynurenic acid, which inhibits the N-methyl-d-aspartate (NMDA) receptor and activates the aryl hydrocarbon receptor (AhR). All of these factors are relevant regulators of PD pathophysiology [
1]. Cortisol/GR/GRE effects therefore parallel the effects of the raised pro-inflammatory cytokines evident in PD, which increase indoleamine 2,3-dioxygenase (IDO), with similar consequence on the tryptophan/kynurenine ratio and kynurenine pathway products effects. There is considerable interest in the role of the kynurenine pathway in PD pathophysiology [
54,
55] and treatment [
56]. Overall, the suppression of pineal melatonin, gut microbiome butyrate and BAG-1 in PD not only enhances stress-linked HPA activation of the GR, but also enhances the GR activation following the morning CAR, thereby differentially regulating the how the morning CAR primes body cells and systems (including immune and glial cells) ‘for the coming day’. Such differential priming includes increased kynurenine pathway products regulating patterned neuronal activity as well as kynurenine and kynurenic acid activation of the AhR, thereby regulating patterned immune responses [
57].
4.3. Stress, α-synuclein, transcription factors and glutamatergic regulation
As well as suppressing pineal melatonin [
58], stress, via CRH induction [
59] and GR activation [
60], coupled to raised pro-inflammatory cytokines [
1], increases gut permeability/dysbiosis, thereby lowering butyrate levels and increasing circulating LPS, leading to enhanced activation of the transcription factors, nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB) and yin yang 1 (YY1) [
61]. Numerous studies link the TLR4/NF-kB pathway to PD pathophysiology [
62], including via exogenous α-synuclein activation of astrocyte TLR4/NF-kB leading to released proinflammatory cytokines and oxidants [
63]. TLR4 can also induce YY1, with astrocyte YY1 a significant driver of PD pathophysiology [
64], partly mediated via the suppression of the excitatory amino acid transporter (EAAT)2, thereby increasing glutamate and excitatory neuronal damage [
65]. Interestingly, erythrocyte extracellular vesicles containing α-synuclein can cross the blood-brain barrier, with α-synuclein accumulating in astrocyte end-feet, in association with suppressed EAAT2 function [
66], being another route whereby systemic factors and processes modulate PD pathophysiology. Interestingly, both NF-kB and YY1 can increase the tryptophan-melatonin pathway and melatonin release thereby allowing melatonin to have autocrine and paracrine effects that dampen inflammation and oxidant induction, whilst optimizing mitochondrial function [
67,
68].
Astrocytes constitutively express and release melatonin, as first shown in 2007 [
69], suggesting that the suppressed capacity of astrocytes to upregulate the tryptophan-melatonin pathway and melatonin efflux, may drive a maintained pro-inflammatory milieu, including from maintained YY1 suppression of EAAT2 [
3]. As well as dysregulating glutamatergic activity in the SNpc, glutamatergic dysregulation will also significantly modulate cognition in PD via changes in other regions, including the hippocampal CA2 region [
70]. Overall, the suppressed capacity of astrocytes to efflux melatonin will dramatically alter the nature of the interactions of SNpc dopamine neurons with their immediate microenvironment as well as in other CNS regions, coupled to the attenuation of melatonin’s suppression of α-synuclein, thereby heightening α-synuclein toxicity presynaptically and in mitochondria [
71,
72]. A number of studies suggest that α-synuclein spread in astrocytes and CNS cells may be dependent upon the heightened levels of hyperlipidemia, perhaps especially raised 27-hydroxycholesterol levels, and the 27-hydroxycholesterol modification of mitochondrial α-synuclein [
72]. Notably, melatonin decreases intestinal lipid absorption, cholesterol synthesis and hyperlipidemia [
73,
74]. This may be of some importance to systemic alterations in PD, given that α-synuclein accumulates in all body organs in PD, including in the liver in association fatty liver disease [
75]. Butyrate is also associated with the suppression of hyperlipidemia [
76]. Overall, data indicate α-synuclein has effects in the liver, as well as in the gut, in the modulation of CNS α-synuclein spread, with the effects of melatonin and butyrate in other organs attenuating α-synuclein spread in the brain.
4.4. Astrocytes, α-synuclein, and neuronal mitochondrial metabolism
The suppression of astrocyte melatonin induction not only prolongs the pro-inflammatory milieu in the SNpc but also prevents melatonin from inducing lactate dehydrogenase, as shown in other cell types [
77], thereby not only decreasing lactate as an energy substrate for SNpc dopamine neurons but also increasing the requirement for astrocytes to upregulate their antioxidant protection under challenge. The enhancement of astrocyte antioxidant protections seems to be mediated via cystine-glutamate antiporter (System X
c-) upregulation, whereby cystine is taken up in exchange for glutamate efflux in the course of glutathione (GSH) synthesis. System X
c- upregulation increases extracellular glutamate that contributes to excessive excitatory activity, especially when coupled to the α-synuclein/LPS/HMGB1-TLR4-YY1 suppression of EAAT2. The loss of astrocyte melatonin, coupled to suppressed EAAT2 levels, enhances extracellular glutamate and suppresses lactate for neuronal mitochondrial energy production making SNpc neurons vulnerable to a variety of challenge, especially the mitochondrial changes posed by various PD susceptibility genes. As the suppression of astrocyte melatonin prevents melatonin from suppressing α-synuclein levels and toxicity, SNpc neurons are significantly challenged by α-synuclein, irrespective of its source, as well as by the PD susceptibility genes that challenge mitochondrial function, including parkin, PINK1, and DJ-1, as well as α-synuclein [
21,
22]. Astrocytes, and the astrocyte tryptophan-melatonin pathway may therefore form an important hub for integrating CNS and systemic processes that ultimately drive PD pathophysiology, including from the melatonin regulation of astrocyte BMAL1 and BAG-3, which protects against both α-synuclein and hyperphosphorylated tau pathology [
78].
The importance of the tryptophan-melatonin pathway, perhaps especially in astrocytes, in the pathophysiology of PD is given some support by wider data. A number of factors indicate suppressed melatonin production in the SNpc, including 1) decreased melatonin receptors in SNpc dopamine neurons, which may be indicative of suppressed melatonin, given that melatonin induces its own receptors [
12]; 2) suppressed tryptophan levels and increased kynurenine pathway products in PD [
1]; and 3) the efficacy of monoamine oxidase inhibitors on PD motor symptom treatment [
79]. Importantly, decreased levels of different 14-3-3 isoforms significantly potentiate SNpc neuron loss with 14-3-3 isoforms attenuating α-synuclein aggregation and toxicity [
80]. As noted, different 14-3-3 isoforms are important determinants of the tryptophan-melatonin pathway function (see
Figure 1), with 14-3-3 isoforms proposed to modulate an array of processes linked to PD pathophysiology, including: 1) LRRK2 mutant effects [
81]; 2) 14-3-3e stabilization decreasing α-synuclein [
82]; 3) 14-3-3γ haploinsufficiency decreases SNpc dopamine production and induces motor deficits [
83]; 4) 14-3-3ζ interacts with α-synuclein to maintain it in a monomeric form, thereby limiting α-synuclein toxicity [
84]; 5) 14-3-3 interacts with, and regulates, parkin [
85], thereby having relevance to the suppressed mitophagy, which is proposed to underpin the rise in MHC-1 and the chemoattraction of CD8
+ T cells that drive SNpc dopamine neuron destruction, paralleling other ‘autoimmune’/’immune-mediated’ disorders [
9].
The regulation of 14-3-3 isoforms may be intimately linked to the alterations in miRNAs evident in PD. A number of miRNAs, including miR-7, miR-375 and miR-451 can repress 14-3-3 isoforms, which is proposed to drive the suppression of the melatonergic pathway in intestinal epithelial cells, platelets and pinealocytes in autism [
86]. As many miRNAs are induced by ROS, it will be interesting to determine whether the suboptimal mitochondrial function in PD drives ROS-regulated miRNAs that act to suppress 14-3-3 isoforms that are crucial in driving the enzymatic processes of the tryptophan-melatonin pathway (see
Figure 1). The suppressed melatonergic pathway in astrocytes may, via decreased melatonin and lactate, raise mitochondrial ROS levels to drive miRNAs that act to suppress the melatonergic pathway in neurons. This is parsimonious with recent conceptualizations of cell elimination in a given microenvironment, where other cells act to suppress the melatonergic pathway in a cell to make it vulnerable to immune-mediated cell elimination [
9,
11].
The AhR and its activation by kynurenine is another factor that suppresses melatonin availability. The AhR induction of cytochrome P450 (CYP)1A2, CYP1B1 and CYP1A1 can suppress melatonin availability by two processes, namely: 1) the O-demethylation of melatonin to its immediate precursor NAS, which seems primarily mediated by CYP1A2 and CYP1B1; 2) the hydroxylation of melatonin to 6-hydroxymelatonin, which may be mediated predominantly by CYP1A1 and CYP1B1. The variable effects of these CYP1 isoforms arise from factors that bias the protein-protein interactions of melatonin with these CYP1 isoforms [
87,
88]. The plethora of endogenous and exogenous factors activating the AhR, such as air pollutants and kynurenine, may therefore contribute to the suppression of melatonin availability, The stress/HPA axis and morning CAR induction of TDO/kynurenine/AhR activation pathway allows heightened GR nuclear translocation and TDO induction to suppress melatonin availability by a number of means, including by decreasing the availability of tryptophan and by increasing melatonin hydroxylation and/or O-demethylation.
The role of brain-derived neurotrophic factor (BDNF) activation of the tyrosine receptor kinase (TrkB)-full length (FL) receptor is generally beneficial to neuronal survival and frequently recommended as a treatment target in neurodegenerative conditions, including PD [
89]. However, TrkB also has a number of truncated forms, predominantly TrkB-T1, which generally has pro-apoptotic effects, and is linked to PD pathophysiology [
90]. TrkB effects are further complicated by the expression of both TrkB-FL and TrkB-T1 on the plasma membrane and/or mitochondrial membrane [
91]. This has consequences for AhR activation. As NAS is a BDNF mimic via TrkB activation [
92], the AhR/CYP1A2/CYP1B1 O-demethylation of melatonin to NAS will activate TrkB, with differential consequences at the TrkB-FL vs TrkB-T1 and whether either form of TrkB is expressed on the mitochondrial and/or plasma membrane. NAS therefore has potentially negative consequences given the raised levels of the AhR evident in SNpc dopamine neurons and astrocytes as well as in the striatum, as shown in PD models [
93], coupled to the raised levels of circulating and CNS kynurenine to activate the AhR [
1]. As TrkB-T1 is induced under conditions of heightened ROS-driven miRNAs [
94] the raised levels of AhR induced NAS may contribute to SNpc dopamine neuron loss in PD. Importantly, the heightened glutamatergic NMDAr activation evident in PD, including as induced YY1 suppression of astrocyte EAAT2 as well as by amyloid-β, upregulates TrkB-T1 levels [
95].
The roles of TrkB and BDNF in neurodegenerative disorders may therefore also be intimately intertwined with the wider regulation of the melatonergic pathway, including by the AhR.
4.5. Melatonergic pathway regulation of amyloid-β and tau interactions with α-synuclein
There is a growing appreciation of the interactions of α-synuclein, with pathophysiological factors classically associated with Alzheimer’s disease, namely amyloid-β and hyperphosphorylated tau. Heightened amyloid-β levels that are evident in PD and Lewy Body diseases where they can increase α-synuclein aggregation [
96,
97], indicating a role for excessive amyloid-β and tau production in the pathophysiology of PD, especially when associated with dementia, and other tauopathies. Interestingly, melatonin affords powerful protection in dementia models where is it decreases the TRL4/NF-kB/YY1 induction of β-site amyloid precursor protein-cleaving enzyme (BACE)1 [
98], whilst also preventing directly, and indirectly via decreased amyloid-β, tau hyperphosphorylation in neurons [
99]. Melatonin also increases the non-amyloidogenic α-secretase activities of ‘A disintegrin and metalloproteinase domain-containing protein 10’ (ADAM10) and ADAM17, as well as suppressing the amyloidogenic β- and γ-secretases [
100,
101]. Such data would indicate that the attenuation of the melatonergic pathway in astrocytes and neurons may be an important commonality in the pathophysiological processes underpinning diverse neurodegenerative conditions [
3]. See
Figure 3. Notably, the suppression of melatonin and butyrate in PD allows plasma membrane GR activation to increase BACE1 and amyloid-β [
102] whilst the GR also upregulates presenilin (PSEN)1 assembly on the endoplasmic reticulum (ER), which allows amyloid-β to accumulate on the ER mitochondrial associated membrane (MAM) [
103]. It requires experimental investigation as to the nature of melatonin and butyrate’s modulation of such GR induced BACE1 and amyloid-β induction in the modulation of α-synuclein.