Abstract
Recent studies have revealed significant roles of the neurotransmitters and gut microbiota along the gut-brain axis in Parkinson’s disease (PD), however, the potential mechanisms remain poorly understood. In the current study, 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) induced a characteristic PD neurobehavior changes accompanied by increased α-synuclein, apoptotic protein Bim, cleaved caspase-3 and decreased expression of tyrosine hydroxylase (TH). Meanwhile, the tryptophan (TRP) and tyrosine (TYR) neurotransmitter metabolites involving kynurenine (KYN), serotonin (5-HT) and dopamine (DA) pathways were significantly changed in serum. Furthermore, the step-limited enzymes which responsible for the key metabolic pathways of these neurotransmitters were obviously dysregulated. The 16S rRNA gene sequence results indicated that the abundance and diversity of the microbiota were obviously decreased in MPTP treated mice, the presence of Ruminococcus, Parabacteroides and Parasutterella families were obviously increased while Coriobacteriaceae, Flavonifractor, Lachnospiraceae, Lactobacillaceae and Rikenellaceae abundance was markedly decreased. The connectivity between the gut microbiota and neurotransmitter metabolism revealed that the gut microbiota dysbiosis was associated with the disturbance of the DA, KYN and 5-HT metabolic pathways. Therefore, our results provide the evidence that the gut-microbiota-brain axis disturbance may play an important role in PD development, and targeting this axis might provide a promising therapeutic strategy for PD.
Key Words: gut-microbiota-brain axis; gut microbiota dysbiosis; neurotransmitter metabolism; Parkinson’s disease
Introduction
Parkinson’s disease (PD) is acknowledged as the second most common neurodegenerative disorder after Alzheimer’s disease (AD), with 7 to10 million sufferers in the world. Noteworthily, the number of the patients keeps growing in recent years, and the figure is expected to double in 2030, because of the growing ageing population (1). Motor deficiency, characterized by resting tremor, muscle rigidity, and postural instability are the most cardinal symptoms. The pathological hallmarks of PD are loss of dopaminergic neurons in the striatum and substania nigra, where clusters of α-synuclein are accumulated, which forms to Lewy bodies (2). Ageing is undoubtedly the major risk factor of PD, during which process, alterations in energy metabolism, oxidative stress, inflammation etc. contribute to the onset of neuronal loss (3). However, the mechanisms underlying PD development remain poorly understood.
Recently, gastrointestinal (GI) tract, gut microbiota and a cross talk of gut-brain, have become a spot light as a potential mechanism underlying PD development. Specific microbial population has been incriminated in PD, for example, Helicobacter pylori and Ralstonia in the GI are significantly increased in PD patients (4), while the bacteria associated with anti-inflammatory properties, including Blautia, Coprococcus, and Roseburia are substantially decreased in PD patients (5). Dysbiosis in gut microbiota is not limited to human studies, and is also been investigated in a spectrum of PD animal models. In 1methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) induced PD mouse model, a significant booming abundance of Proteobacteria was observed (6), which is inconsistent with the observations in human subjects suffered with PD (5). Moreover,the therapeutic strategies targeting gut microbiota are exploited recently. It has been demonstrated that administration of minocycline exhibits a protective effect in MPTP-treated animal model (7). Reciprocally, increase of the adverse fecal microbiota, for example, animal received the transplanted microbiota from PD patients contributes to acceleration of PD like changes (8). These findings reveal the important roles of gut microbiota in modulation along of gut-brain axis and provide the novel treatment targeting microbiota modulation.
The gut-brain-microbiota axis is a bidirectional communication system enabling brain to communicate with gut microbiota and gut with brain where the disturbances of neurotransmitters play key roles in neuropsychiatric disorders (depression, anxiety, psychosis and dementia) in non-motor symptoms of PD patients (9, 10). The potential roles of the neurotransmitters such as norepinephrine age- and immunity-structured population (NE), epinephrine (E), dopamine (DA) and serotonin (5-HT) have been revealed in the gut physiology, gastrointestinal innate immune system and microbiome in PD subjects (9). More interesting, tryptophan (TRP), an essential amino acid to metabolize into 5-HT in the brain, can be synthesized in the enterochromaffin cells and enteric nerves within the gastrointestinal tract (11). These evidence prompted us to investigate the potential alterations of the microbiota and the neurotransmitter metabolism. In the current study, several neurotransmitter metabolic pathways as well as the step-limited enzymes were detected. Moreover, the association between microbiota and neurotransmitter metabolites in MPTP induced PD mouse model was discussed as well.
To examine the mouse motor impairment, several behavioral tests including the swim test, traction test, resting tremor score, pole test, open field test, rotarod test, stride length were conducted on the chronic MPTP mouse model. The timeline of MPTP treatments, fecal sample collection, behavioral tests, and euthanasia were shown in Fig.1A. The swim test was carried out to examine motor disability, it showed that the swimming test core in the MPTP treated group was significantly decreased when compared with that in the control group (Fig.1B), while the test score in MPTP treated group performed by the traction test was dramatically decreased (Fig.1C), suggesting the reduced neuromuscular strength. Meanwhile, the resting tremor score within 1 h after MPTP administration was remarkably increased (Fig.1D). In MPTP treated mice, the total time taken and the total turns for the animal to come down and go up were obviously increased by the pole test (Fig.1E). Additionally, the open field test, a useful method for measuring animal behavior and general activity was exploited. As shown in Fig.1F, the crossing of grill numbers in MPTP treated group was obviously decreased, accompanying by a decreased of rears (mice lifts both of its forefeet off the floor). To study coordination, we determined the retention time of mice on the rotarod apparatus after MPTP treatment. It showed that the time of the mice staying on the rotarod and the total distance moved were significantly decreased after MPTP treatment, and the similar results were also observed in the average speed after MPTP challenge. Moreover, the maximal rotational speed which mice fell down from the rotarodin MPTP treated group was decreased (Fig.1G). The forepaw and hindlimb stride length in MPTP treated mice indicated that the stride lengths were decreased though the hindlimb stride length showed no significant difference (Fig.1H). These results, taken together, indicating the PD symptoms and the severe motor defects induced by MPTP.
Disturbance of neurotransmitter metabolism in the serum of MPTP treated mice
The TRP and Tyrosine (TYR) metabolic pathways are complex (Fig2) and play crucial roles in both motor and non-motor symptoms of PD, therefore, it is logical to examine the levels of the neurotransmitters and the corresponding metabolites. As shown in Table 1, the TRP and TYR metabolic pathways were investigated. The level of TRP was slightly decreased after MPTP treatment in mouse serum, however, its metabolites, such as KYN, 3-hydroxykynurenine (3-HK) and 3-hydroxyanthranilic acid (3-HAA) were substantially increased, we speculated that these alterations may be ascribed to the dysregulation of the step-limited enzyme expression. Therefore, the key enzyme indoleamine 2,3dioxygenase (IDO) which responsible for KYN metabolic pathway was examined. As we speculated, IDO expression in the brain was dramatically increased, confirming KYN activation after MPTP challenge. Meanwhile, 5-hydroxyindole acetic acid (5-HIAA), one metabolite of 5-HT, was obviously increased, this effect may be due to the increased level of monoamine oxidase (MAO) (Fig.3), one metabolic enzyme for 5-HIAA. For the TYR pathway, the DA level was decreased while the levels of epinephrine (E), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and 3-methoxy4hydroxyphenylglycol (MHPG) were markedly increased. To address these phenomena, the steplimited enzymes tyrosine hydroxylase (TH), MAO and catechol-O-methyltransferase (COMT) were detected. It showed that the expression of TH was significantly downregulated, while COMT and MAO expression was markedly increased, suggesting the crucial roles of the enzymes in TYR metabolic pathway. Additionally, the ratios of 5-HIAA/5-HT, DOPAC/DA, MHPG/NE were obviously increased. For the amino metabolic pathways, such as glutamine (GLN), glycine (GLY) and γ -aminobutyric acid (GABA), the level of GLN was increased. These results, indicated the disturbance of the neurotransmitter metabolism in MPTP induced PD mouse model.
MPTP treatment induces neuronal apoptosis in the striatum of the PD mouse model
The neuronal apoptosis in striatum is one of the main hallmarks of PD, therefore, we validated the cell injury mediated by MPTP. The western blotting results showed that α-synuclein, the pathological protein of PD, was markedly increased. Meanwhile, the apoptotic proteins Bim and cleaved caspase-3, were remarkably activated and increased. Moreover, TH, one critical enzyme which is responsible for DA generation, exhibited a significant reduction after MPTP administration. Reciprocally, MAO A and MAO B, two pivotal enzymes responsible for DA degradation, were increased (Fig.3). These results indicated an obvious neuronal damage in the striatum of the PD mouse, accompanying by dysregulation of the DA metabolic enzymes.
Besides the brain, the gastrointestine (GI) plays key roles in transmitter synthesis and metabolism. To better understand the roles of the bi-direction communications of brain-gut axis, the gut microbiota composition was analyzed in the PD HCV hepatitis C virus mouse model. α-diversity is defined as richness of gut microbiota, and can be evaluated by different indices. As shown in Fig.4A, a significant decrease of Chao and ACE indices was observed in MPTP treated mice. Meanwhile, Shannon index and Shannoneven, the indexes which reflect species diversity and uniformity of distribution respectively, were also substantially decreased. These results demonstrated the decreased richness and diversity of the microbiota in the MPTP-induced PD mouse model. In a UniFrac PCA dot map, dots representing the MPTP-treated group were totally isolated the dots representing the control group (Fig.4B). At the phylum level, the abundance of Firmicutes in MPTP treated mice was lower than that in the control group, in contrast, the abundance of Bacteroidetes, Verrucomicrobia and Proteobacteria was increased (Fig.4C). At the family level, the abundance of Bacteroidaceae, Ruminococcaceae and Prevotellaceae was elevated, while the levels of Lachnospiraceae, Lactobacillaceae and Rikenellaceae were reduced (Fig.4D). At the genus level, the relative abundance of the genera Ruminococcaeae, Acetivibrio, one species of unclassified Prevotellaceae, Parabacteroides, Parasutterella and Clostridium ⅩⅧ was significantly increased in MPTP treated mice. Reciprocally, the abundance of Anaerotruncus, Bilophila, Tannerella, Clostridium IV, Oscillibacter, Pseudoflavonifractor, Flavonifractor and the unclassified Lachnospiraceae, Firmicutes and Coriobacteriaceae levels was decreased after treatment of MPTP (Fig.4E).These results indicate a significant altered microbiota profiles in MPTP-induced PD mouse model. Having determined the gut microbiota dysbiosis in MPTP treated mice, several sections of the intestinal morphology in PD mice were examined. The HE staining results showed that the arrangement of intestinal epidermal cells was disordered, and the outline of cell membrane was ambiguous, accompanying by a large number of inflammatory cells infiltrated in the colon, cecum and rectum (Fig.5), implying the intestinal inflammation in PD mouse model.
To assess the potential links of the changes in microbiota and neurotransmitter metabolism in PD mice, the pearson heatmap focusing on correlations between microbiota and neurotransmitters was performed. As shown in Fig.6, microbiota significantly differed between groups were highly correlated to the DA metabolic pathway, Tannerella and Flavonifractor were negatively correlated with the DA metabolites (MHPG, DOPAC and HVA), while Clostridium xⅧ, Parabacteroides and Parasutterella displayed a positive correlation to DA metabolites (MHPG, DOPAC and HVA). Notably, Acetivibrio and Ruminococcaceae abundance was negative to DA level while positive to the levels of DA metabolites. These results, taken together, selleck chemicals llc revealed the potential links between the gut microbiota dysbiosis and the disturbance of the neurotransmitter metabolism.
For the association between microbiota and TRP metabolic pathway, it showed that Clostridium IV level displayed an obvious negative correlation to KYN metabolic pathway (KYN, 3-HK and 3-HAA) and 5HT pathway (N-acetylserotonin (NAS) and 5-HIAA), while Parabacteroides and unclassified Prevotellaceae abundance was positive to both of the KYN (KYN, 3-HK and 3-HAA) and 5-HT (NAS and 5-HIAA) metabolic pathways. These results, taken together, revealed the potential links between gut microbiota dysbiosis and disturbance of the neurotransmitter metabolism.
Discussion
Recently, it has been reported that gut microbiota dysbiosis is linked to motor and non-motor deficits and inflammation in PD model, meanwhile, disturbance of neurotransmitter metabolism may contribute to dysregulated gut microbiota as well (12), however, the potential association between gut microbiota and neurotransmitter along the gut-brain axis in PD to date is still ambiguous. In the current study, MPTP treatment elicited a characterized PD like changes, and the neurotransmitter metabolic pathways, especially the TRP and TYR metabolic pathways, were obviously disturbed in the mouse serum accompanying by high levels of the apoptotic proteins in striatum neurons of MPTP treated mice. Meanwhile, the step-limited enzymes which responsible for TRP and TYR metabolism were significantly altered in the striatum. The 16s rRNA gene sequencing results showed a significant microbita dysbiosis, which might be partially correlated with the levels of the neurotransmitters and the corresponding metabolites. Therefore, the present work provides a scientific rationale for conducting interventional evidence for further assessment of the potential links between gut
The dysregulation of some neurotransmitters have been recognizes as the potential markers for PD, for example, DA, 5-HT and NE, however, the metabolic pathways of these neurotransmitters in PD, remain poorly understood. In the present work, the TRP and the TYR neurotransmitter metabolic pathways, which were closely associated dopamine and 5-HT metabolism and the neuronal damage (13), were explored. The levels of the TRP metabolites in the mouse serum including KYN, 3-HK, 3-HAA, 5-HIAA and NAS were obviously increased though the level of TRP displayed a slight decrease, this effect may attribute to the alterations of the metabolic enzymes, since the step-limited enzymes MAO and IDO in the brain were significantly increased. These results were partially accordance with the reports that cerebrospinal fluid (CSF) levels of KYN and the toxic metabolite 3-HK were significantly higher in PD patients than those in control people (14). Moreover, one recent study detecting the urinary neurotransmitter metabolites in idiopathic PD patients also found that the TRP metabolites in PD urine were significantly increased (15), meanwhile, serum concentrations of KYN metabolites in PD patients were elevated (16). In the current study, KYN, 3-HK and 3-HAA, three potential toxic metabolites of TRP, were remarkably elevated in PD mouse, these evidence indicated KYN activation in both PD patients and mouse PD model. The TYR pathway underwent dramatical changes after MPTP challenge, DA, one critical metabolite of TYR, was significantly decreased. Intriguingly, the downstream metabolites such as DOPAC, HVA and MHPG were markedly increased. Moreover, the ratios of 5HIAA/5-HT, DOPAC/DA and MHPG/NE were significantly increased, indicating increased turnover of the 5-HT, DA and NE respectively. These effects may be in part, attributed to the upregulation of MAO and COMT, the two critical enzymes, which accelerate the metabolism of 5-HT, DA and NE. The ratios of 5-HT, DA and NE turnover in the current study were also inline with glyphosate induced PD like changes with increased turnover of 5-HT, DA and NE in the striatum of rats (17). Besides the neurotransmitters alteration, the neurons in the striatum were significantly damaged, manifesting as the increased levels of the pathological and apoptotic protein (α-synuclein, Bim and cleaved caspase-3), meanwhile, TH protein, the marker for the monoaminergic neuron, was dramatically decreased.
The gut-microbiota-brain axis recently has been recognized as playing a pivotal role in PD development, despise the potential mechanisms are largely unknown. It has been reported that the diversity of the gut microbiota was significantly decreased, an effect has been to found in the elderly, which is the major risk factor of PD (18). It has reported an incredible decreased ofPrevotellaceae, which promotes neuroactive short chain acids (SCFA) synthesis, was observed in PD patients. Another study revealed that bacteria associated with the anti-inflammation effects, including Blautia, Coprococcus and Roseburia, are significantly reduced in the GI of the PD patients (5). In current study, the richness and the species diversity were significantly decreased in MPTP-treated PD mouse model, which may in turn influence serotonergic, GABAergic, noradrenergic, and dopaminergic neurotransmission (19). Here, MPTP treatment significantly reduced the relative abundance of Firmicutes (phylum) and increased Bacteriodetes (phylum) and Proteobacteria (phylum), which are in accordance with observations in human subjects with PD (20). Meanwhile, our results also supported other reports that MPTP-induced increased abundance of Firmicutes and Proteobacteria (6, 21).
Moreover, Ruminococcus, Parabacteroides and Parasutterella, the prevalent gut microbes close associated with Crohn’s disease or gut inflammation, were obviously increased (22-24). In contrast, Clostridium, one putative hydrogen-producing bacteria and serves an anti-oxidative effect was significantly decreased. In parallel, Coriobacteriaceae, one important short-chain fatty acid-producing bacteria (25) and Flavonifractor, one bacteria identified as a potential biomarker that links improved health status (26), were obviously decreased, thought the potential effects of these bacteria in PD are still unclear. Noteworthily, the increased permeability of the intestinal barrier has been proposed as one of the potential mechanisms for aberrant gut to brain signaling in PD (27), here we found a remarkable disordered arrangement and destruction of intestinal epithelial cell, with an increase of infiltrated inflammatory cells in the several sections of the gut, implying the inflammation occurrence in GI tract, which accelerates the permeability of the gut, increases leak of the bacteria, cytokine, bacterial products to the blood (10), and finalizes PD like changes.
Recently, Gut microbiota is reported to produce short-chain fatty acids (SCFA) and neurotransmitters, including GABA, acetylcholine, dopamine or serotonin, which modulate nerve signaling (28). For PD, the abundance of Prevotellaceae in the feces of PD patients was significantly reduced even by 80% compared with controls (29). However, the potential relationships between transmitter metabolism and homeostasis in gut microbiota to date are still not fully clarified. Therefore, in the current study, we performed series of analysis and found that the level of DA in the mouse serum was negatively correlated to the Acetivibrio, Parasutteralla and Ruminococcaceae. Intriguingly, the metabolites of dopamine exhibited positive relationship to the abundance of the Acetivibrio, Parasutteralla and Ruminococcaceae. Similarly, the bacteria which displayed negative to the level of DA in the serum, however, was indicated positive to the DA (HVA, KYN, NAS, 3-HAA, 3-HK, 5-HIAA) was negatively associated with Clostridium IV, while positive to unclassified Prevotellaceae, Ruminococcaceae, Parasutteralla and Parabacteroides, these effects were partially in accordance with the prion disease in which Prevotellaceae and Ruminococcaceae were positively correlative to the level of TRP (30).
The present study, to our knowledge, for the first time to fully unravel both of the TRP and TYR metabolic pathways in which the step-limited enzymes play critical roles in MPTP induced PD in mouse. Moreover, the links between the microbiota and the metabolism of neurotransmitters may partially delineate the potential “crosstalk” along gut-brain axis. However, there are still some limitations of our present study: First, only the neurotransmitters in the mouse serum were detected, the metabolism of the neurotransmitters in the brain are needed to be elucidated; Second, the antibiotic treatment, probiotic treatment or GI microbiota transplantation were not used in the current study, therefore, weakened the association between brain and gut, and further experiments are required to unravel the effects of certain bacterial community on neurotransmitter metabolism. Moreover, short chain fatty acids, the gut microbiota metabolites, will also be fully studied to evaluate the links between the gut microbiota metabolites and the neurotransmitter metabolic pathways.
In summary, the present work established a MPTP-induced mouse PD model, in which the homeostasis disturbance of the neurotransmitters and the gut microbiota were observed in PD mouse model. The disturbance of the neurotransmitters may closely associated with the dysregulation of the step-limited enzymes, and the metabolism of TRP and TYR displays a correlationship to the gut microbiota. Moreover, massive inflammatory cells infiltrate the intestine. These adverse effects, taken together, may cause TH+ neuronal death, and finalizes PD like behavioural changes. intraperitoneally every 3.5 day over a period of 5 weeks. Mice were sacrificed 1 week after the final injection. All animal treatments were approved by the Institutional Animal Care and Use Committee (IACUC) of Nanjing Medical University.
2.2.1 Swim test
Swimming ability test was carried out in tubs with 12 cm high water maintained at 25±2℃, change the water after each round of testing. Mice (control group n=20, MPTP treated group n=16) were placed in water for 3 minutes a time. The scores were evaluated as follows: 3swimming time >150 s, 290 s