Neuroprotective Effects of Bifidobacterium breve CCFM1067 in MPTP-Induced Mouse Models of Parkinson’s Disease

There is mounting evidence that the microbiota–gut–brain axis (MGBA) is critical in the pathogenesis and progression of Parkinson’s disease (PD), suggesting that probiotic therapy restoring gut microecology may slow down disease progression. In this study, we examined the disease-alleviating effects of Bifidobacterium breve CCFM1067, orally administered for 5 weeks in a PD mouse model. Our study shows that supplementation with the probiotic B. breve CCFM1067 protected dopaminergic neurons and suppressed glial cell hyperactivation and neuroinflammation in PD mice. In addition, the antioxidant capacity of the central nervous system was enhanced and oxidative stress was alleviated. Moreover, B. breve CCFM1067 protected the blood–brain and intestinal barriers from damage in the MPTP-induced mouse model. The results of fecal microbiota analysis showed that B. breve CCFM1067 intervention could act on the MPTP-induced microecological imbalance in the intestinal microbiota, suppressing the number of pathogenic bacteria (Escherichia-Shigella) while increasing the number of beneficial bacteria (Bifidobacterium and Akkermansia) in PD mice. In addition, the increase in short chain fatty acids (acetic and butyric acids) may explain the anti-inflammatory action of B. breve CCFM1067 in the gut or brain of the MPTP-induced PD mouse model. In conclusion, we demonstrated that the probiotic B. breve CCFM1067, which can prevent or treat PD by modulating the gut–brain axis, can be utilized as a possible new oral supplement for PD therapy.


Introduction
By 2040, more than 17.5 million people will have Parkinson's disease (PD), a frequent neurodegenerative disorder that affects the central nervous system (CNS) [1]. Despite this, there is presently no treatment that may prevent or cure PD. Two important pathological markers of PD are the degenerative death of midbrain nigrostriatal dopamine (DA) neurons and a significant reduction in striatal DA content, which in turn leads to motor deficits, such as stiffness, resting tremors, gait impairment, and bradykinesia in Parkinson's disease [2][3][4]. People with PD usually have gastrointestinal problems and constipation years before they start having motor symptoms [5,6]. This suggests that gut disorders may be linked to the start of PD.
The gut microbiota may influence brain function and behavior through neurological, immunological, and endocrine pathways, establishing the microbiota-gut-brain axis (MGBA) [7,8]. Clinical investigations have shown that both the composition and abundance of gut bacteria and the produced metabolites are drastically altered in patients with PD compared to those in healthy patients [9]. One clinical study showed that the bacterial family Prevotellaceae is reduced, and beyond that, Faecalibacterium prausnitzii (a bacterial

Bifidobacterium breve CCFM1067 Preparation
The biological genesis of Bifidobacterium breve CCFM1067 is descri

Bifidobacterium breve CCFM1067 Preparation
The biological genesis of Bifidobacterium breve CCFM1067 is described in Supplementary Materials S1.1. B. breve CCFM1067 was incubated at 37 • C for 48 h after scribing onto MRS solid medium (containing 0.05% cysteine) to obtain a single colony. This single colony was then selected and inoculated in MRS liquid medium and cultured at 37 • C for 24 h for activation. Next, to obtain a B. breve CCFM1067 bacterial solution, the activation solution was inoculated at a concentration of 2% (v/v) with MRS liquid medium and incubated at 37 • C for 24 h. Then, the bacterial solution was centrifuged at 8000× g for 10 min to obtain B. breve CCFM1067. Suspensions were maintained at −80 • C.

Behavioral Tests for Motor Functions
Each mouse's motor function was evaluated using four behavioral tests: pole test (PT), narrow-beam test (NBT), rotarod test (RTR), and open field test (OFT). Refer to Supplementary Materials S1.2 for specific behavioral experimental protocols. On days 29-31, all of the mice got behavioral training once a day. Tests of behavior were conducted 24 h following the final injection of MPTP on day 37.

Neurochemical and Biochemicall Analyses
Sample collection: Immediately after dissection, the striatum and midbrain were frozen at −80 • C for chemical analyses and mRNA expression experiments. For immunohistochemical staining, whole brains were removed and placed in a solution of 4% paraformaldehyde. Colon tissues were rinsed with saline and subjected to fast freezing in liquid nitrogen to preserve them at −80 • C for later use in chemical analysis and mRNA expression experiments. Mouse feces were collected from the colon for 16S rDNA gut microbiota analysis and SCFA extraction, and stored at −80 • C.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)
Following the protocol described using total RNA extraction reagents (Vazyme, R401-01), RNA was extracted from the brain and intestinal tissue. Using the qPCR kit (Vazyme #R333, Nanjing, China), the extracted RNA was reverse-transcribed into cDNA. SYBR Green Supermix was used for quantitative PCR on a BioRad-CFX384 machine (Bio-Rad, California, USA). We used the cycle threshold (Ct) and normalized using the 2 −∆∆Ct technique relative to Gapdh to measure the relative expression of genes. In the Supplementary Materials S1.3, the RT-qPCR primers were detailed.

Immunohistochemistry
The paraformaldehyde-fixed and dehydrated brain tissue was embedded in paraffin and then processed according to standard paraffin section immunohistochemistry (IHC) experimental procedures. Specific procedures included the dewaxing of paraffin sections in water, antigen repair, endogenous peroxidase blocking, serum closure, and with  TH (1:500, EP1536Y, ab51253, Abcam, Cambridge, MA, USA) and secondary  antibodies (goat anti-rabbit IgG, 1:200, GB23303, Servicebio, Wuhan, China) incubation.  3,3 -diaminobenzidine (DAB) staining was then used to observe SNpc TH-immunoreactive  (TH-IR) neurons, next with re-staining of the nuclei, and then the sections were dehydrated and sealed. We used the Confocal laser scanning microscope equipment (ZEISS, Oberkochen, Germany) to digitize the stained slides, and then Image-J 1.53k was used for analysis.

Gut Microbial and Bioinformatics Analysis
The process for sequencing fecal 16S rRNA is discussed in full in Supplementary Materials S1.4. To analyze the raw sequencing data, we used the platform Quantitative Insights into Microbial Ecology (QIIME). Selected OTUs from the Greengenes 13.5 database was based on a nucleotide identity criteria of 97%. Species richness and distributional evenness among very uncommon OTUs were used to calculate the Observed, Shannon, and Simpson indices, which were used to evaluate the α-Diversity. β-Diversity was calculated using Bray-Curtis distances, displayed using principal component analysis (PCA) and principal coordinate analysis (PCoA), and differences were determined using permutational multivariate analysis of variance (perMANOVA). Microbiological biomarkers were separated using linear discriminant analysis (LDA) effect size (LEfSe) (Wilcoxon rank-sum test, p < 0.05 and log LDA > 4.0 was chosen as the criteria), which was shown as a taxonomic cladogram tree. Predicted functional pathways were annotated using the KEGG orthology, and the metagenomes of gut microbiota were computed from 16S rRNA sequences using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) method. Using the Wekemo Bioincloud (https://www.bioincloud.tech (accessed on 20 June 2022)), the aforementioned studies were conducted. The network of associations was constructed using Spearman's correlation coefficients and the OmicStudio tools (https://www.omicstudio.cn/tool (accessed on 29 June 2022)). Unless otherwise stated, suggested settings were always utilized.

Statistical Analysis
The statistical analyses were conducted using the GraphPad Prism 9.1 software. Data are presented as the mean ± SEM. The experimental data were checked for conformance to a normal distribution using the D'Agostino-Pearson test. If the datasets were normally distributed, Student's unpaired t-test or one-way analysis of variance (ANOVA) with Dunnett's test was used to compare two or multiple datasets, respectively. Conversely, if the datasets were not normally distributed, the Mann-Whitney test or Kruskal-Wallis with Dunn's test was conducted to compare two or multiple datasets, respectively.

B. breve CCFM1067 Improves MPTP-Induced Motor Impairments
The experimental design is shown in Figure 1A. Mice in the MPTP group started to lose weight on day 35, during the development of the PD model ( Figure 1B). As shown in Figure 1C, L-DOPA therapy significantly lowered weight loss in MPTP-induced mice, whereas B. breve CCFM1067 treatment reduced weight loss marginally, but not significantly (p > 0.05). Motor function tests, including PT, NBT, RTR, and OFT, were used to assess the effectiveness of B. breve CCFM1067 on the locomotor coordination of PD mice. Compared to the control group, mice in the MPTP group displayed significant motor impairments, including decreased motor agility ( Figure 1D Figure 1G, T (12) = 12.88, p < 0.0001), where they had a shorter residence time ( Figure 1H, T (14) = 12.53, p < 0.0001) and fewer entries ( Figure 1J, T (14) = 18.44, p < 0.0001) than those of normal control mice, which could be significantly ameliorated by L-DOPA and B. breve CCFM1067 treatment (all p < 0.0001). Collectively, these results suggest that therapy with B. breve CCFM1067 mitigates MPTP-induced motor impairment.

B. breve CCFM1067 Ameliorated the Dysbiosis of the Mouse Gut Microbiota of MPTP-Induced
To determine how B. breve CCFM1067 protects the MPTP-induced PD mouse model by modifying the microbial community structure, we sequenced 16S rRNA from feces collected six days after MPTP treatment. Nine bacterial phyla dominated the gut microbiota in all four groups ( Figure 4A). As shown in Figure 4B, the phyla Patescibacteria, Proteobacteria, and Tenericutes showed a considerably increased abundance in PD mice, while the phyla Verrucomicrobia, Cyanobacteria, and Deferribacteres had a significantly decreased abundance (vs. control). Mice with PD showed significant microbial changes, which B. breve CCFM1067 was shown to correct.
According to the α-diversity analysis, the microbiota of PD mice exhibited a lower Observed index ( Figure 4C, T (10) = 5.147, p = 0.0004), which is closely linked to microbial richness, and lower Shannon and Simpson index values ( Figure 4D, T (10) = 3.468, p = 0.006; Figure 4E, T (10) = 4.738, p = 0.0008), which are closely correlated with microbial diversity (vs. control). Treatment with B. breve CCFM1067 considerably increased the Shannon ( Figure 4E, F (2,15) = 5.858, p = 0.0122) and Simpson ( Figure 4E, F (2,15) = 9.485, p = 0.0081) values (vs. MPTP), whereas L-DOPA therapy had no effect on these (all p > 0.05, vs. MPTP). These findings show that B. breve CCFM1067 treatment may reverse the reduction in microbial community richness and diversity in PD mice. Moreover, the β-diversity analysis indicated that the gut microbial community of the MPTP group was considerably different from that of the control group, whereas that of the B. breve CCFM1067 and L-DOPA groups was similar to that of the control ( Figure 4F, PERMANOVA: F = 5.382, p = 0.001). The results of the principal component analysis (PCA) were the same as that above ( Figure 4G,H). The findings on intestinal microbiota diversity and structure demonstrate that therapy with B. breve CCFM1067 effectively controlled alterations in the gut microbiota of MPTPtreated mice.
We evaluated the relative abundance of microorganisms at the taxonomic genus level in various groups ( Figure 5A) and screened seven differential genera in the control, MPTP, L-DOPA, and B. breve CCFM1067 groups using LEfSe analysis ( Figure 5B). In addition, the results of LEfSe analysis between the control and MPTP groups, the MPTP and B. breve CCFM1067 groups, and the MPTP and L-DOPA groups for each two groups were shown in Figure S4. MPTP-induced mice exhibited a lower relative abundance of Akkermansia ( Figure 5F, T (10) = 3.082, p = 0.0116) and a higher relative abundance of Bacteroides ( Figure 5C  Twenty-one genera exhibited significant changes in bacterial abundance when examined using the univariate approach for the factorial Kruskal-Wallis test (p < 0.05, FDR < 0.05). These 21 genera were used to create a heatmap of gut microbial marker correlations ( Figure 5I). Because the gut microbiota interacts to maintain a dynamic equilibrium, examining the interactions between various genera may help us understand how these species affect the onset of PD. To further show the impact of microbiota on mice treated with MPTP, correlations between the abundance of bacterial genera and other experimental outcomes were analyzed ( Figure 5I,J). These findings suggest that B. breve CCFM1067 alleviates microbiota dysbiosis in mice with PD, highlighting the importance of microbial dysbiosis in PD progression.

Functional Predictions Suggested That B. breve CCFM1067 May Modify Functional Modules of the Gut Microbiota
We conducted a Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis to better understand the functions of these significantly altered microorganisms. Twenty-four KEGG (level 3) pathways were differentially enriched between the control and MPTP groups (p < 0.01, with Q value < 0.01) ( Figure 6A), and MPTP increased the pathways involved in benzoate degradation, human papillomavirus infection, viral carcinogenesis, drug-metabolizing enzymes, butanoate metabolism, pentose phosphate, renal cell carcinoma, Cushing syndrome, carbon metabolism, and ribosome biogenesis in eukaryotes ( Figure 6A). Thirteen KEGG (level 3) pathways were differentially enriched between B. breve CCFM1067 and MPTP (p < 0.05, Q value < 0.05) ( Figure 6B), and B. breve CCFM1067 increased pathways involved in the biosynthesis of carotenoids, terpenoids and steroids, sesquiterpenoids and triterpenoids, steroids, and insulin resistance ( Figure 6B). Differences in predicted functions between L-DOPA and MPTP groups at the KEGG pathways at level 3 were shown in Figure S5 (p < 0.05, Q value < 0.05). In addition, the Spearman correlation analysis between 21 different genera and 32 different KEGG pathways showed that many different genera were significantly correlated with different KEGG pathways ( Figure S6). inversely proportional to the amount of time required to complete the pole descent test. Spearman's rho correlation coefficient and the regression line are shown. Data are means with SEM; # p < 0.05, ### p < 0.001, #### p < 0.0001, compared with the control group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 vs. MPTP group, "ns" means there was no noticeable difference between the groups.

Functional Predictions Suggested That B. breve CCFM1067 May Modify Functional Modules of the Gut Microbiota
We conducted a Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis to better understand the functions of these significantly altered microorganisms. Twenty-four KEGG (level 3) pathways were differentially enriched between the control and MPTP groups (p < 0.01, with Q value < 0.01) ( Figure 6A), and MPTP increased the pathways involved in benzoate degradation, human papillomavirus infection, viral carcinogenesis, drug-metabolizing enzymes, butanoate metabolism, pentose phosphate, renal cell carcinoma, Cushing syndrome, carbon metabolism, and ribosome biogenesis in eukaryotes ( Figure 6A). Thirteen KEGG (level 3) pathways were differentially enriched between B. breve CCFM1067 and MPTP (p < 0.05, Q value < 0.05) ( Figure 6B), and B. breve CCFM1067 increased pathways involved in the biosynthesis of carotenoids, terpenoids and steroids, sesquiterpenoids and triterpenoids, steroids, and insulin resistance ( Figure 6B). Differences in predicted functions between L-DOPA and MPTP groups at the KEGG pathways at level 3 were shown in Figure S5 (p < 0.05, Q value < 0.05). In addition, the Spearman correlation analysis between 21 different genera and 32 different KEGG pathways showed that many different genera were significantly correlated with different KEGG pathways ( Figure S6). (C) Concentration (µ mol/g feces) of SCFAs. (D) Total SCFAs: the amount of acetic, propionic, isobutyric, butyric, isovaleric, and valeric acids add up to the total SCFAs (n = 5/group). (E) A network of correlations between different experimental findings in the microbiota-gut-brain axis. The different node colors show different classifications, the red, pink, purple, and blue nodes correspond to the experimental results regarding the gut microbiota, brain, colon, and motor function of the Figure 6. B. breve CCFM1067 treatment affects gut microbial functions in MPTP-induced PD mice, which may be mediated by microbial metabolites. (A) Differences in predicted functions between control and MPTP groups in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways at level 3 (Welch's t test, two-sided, Storey FDR q < 0.01). (B) Differences in predicted functions between B. breve CCFM1067 and MPTP groups in the KEGG pathways at level 3 (Storey FDR q < 0.05). (C) Concentration (µmol/g feces) of SCFAs. (D) Total SCFAs: the amount of acetic, propionic, isobutyric, butyric, isovaleric, and valeric acids add up to the total SCFAs (n = 5/group). (E) A network of correlations between different experimental findings in the microbiota-gut-brain axis. The different node colors show different classifications, the red, pink, purple, and blue nodes correspond to the experimental results regarding the gut microbiota, brain, colon, and motor function of the mice. Negative Spearman correlation coefficients are represented by gray, dashed lines (R < −0.5), whereas positive correlation coefficients are shown with solid orange lines (R > 0.5). Spearman correlation coefficient values below −0.5 (negative correlation) are indicated with gray dotted lines, and coefficient values above 0.5 (positive correlation) are indicated with solid orange lines. The strength of the correlation is indicated by the thickness of the line. Data are means with SEM; unpaired Student's t-test #### p < 0.0001, compared with the control group; one-way ANOVA with Dunnett's test for multiple comparisons between MPTP, L-DOPA, and B. breve CCFM1067 groups, * p < 0.05, *** p < 0.001, **** p < 0.0001 vs. MPTP group, "ns" means there was no noticeable difference between the groups.

Correlations Support the Involvement of the MGBA in the MPTP-Treated Mouse Model
We measured the levels of SCFAs in intestinal contents ( Figure 6C,D). The levels of acetic acid, butyric acid, valeric acid, and isovaleric acid (all p < 0.001) were significantly lower in the MPTP group compared to those in the control group. In contrast, B. breve CCFM1067 greatly increased the concentration of SCFAs ( Figure 6C: acetic acid, F (2, 21) = 87.78, p < 0.0001; butyric acid, F (2, 24) = 11.08, p = 0.0002; isobutyric acid, F (2, 24) = 26.18, p < 0.0001; isovaleric acid, F (2, 24) = 159.0, p < 0.0001) with the exception of propionic acid, the level of which was not affected by any of the treatments ( Figure 6C). Therefore, B. breve CCFM1067 affected the MPTP-induced SCFA decline in the mouse colon.
We conducted a correlation analysis between key experimental data to gain a deeper understanding of the processes involved in the MGBA in the pathogenesis of PD mice and the preventive benefits of B. breve CCFM1067 therapy. The relative abundances of Bifidobacterium, Bacteroides, Ruminococcus-1, Escherichia-Shigella, and Akkermansia represented the microbiota changes in the different groups. The total time in the pole test and total retention time on the rotarod test were selected as indicators for changes in motor ability. To illustrate the changes in the brain, the number of TH-positive cells in the SN; the expression of Iba-1, IL-6, and ZO-1 in the striatum; the levels of DA in the striatum; and the level of GSH in the midbrain were assessed. We also chose the level of butyric acid in colon feces and the expression of IL-6 and ZO-1 in the colon as important features of the gut.
Surprisingly, we discovered that the majority of associations between these markers were statistically significant. These correlation data jointly suggest the participation of the MGBA in the development of the MPTP-induced PD mouse model, as well as the protective effects of B. breve CCFM1067 therapy in this model.

Discussion
To test whether B. breve CCFM1067 helps regulate the intestinal microbiota and assess its neuroprotective potential, we used a subacute PD mouse model obtained through intraperitoneal MPTP injection. Our results show that B. breve CCFM1067 intake exhibited neuroprotective effects, significantly alleviating motor deficits in PD mice ( Figure 1D-J), protecting dopaminergic neurons in the midbrain (Figure 2A,B), and increasing the levels of striatal neurotransmitters ( Figure 2H-M). Ingestion of B. breve CCFM1067 also enhanced endogenous neuroprotective factors in the brains of mice ( Figure 2F,G) and alleviated MPTPinduced neuroinflammation (Figure 3C-E) and oxidative stress ( Figure 3A,B). Ingestion of B. breve CCFM1067 restored the BBB and the impaired intestinal barrier ( Figure 3K-P), alleviated dysbiosis of the intestinal flora (Figures 4 and 5), and increased the SCFA content in the feces of PD mice ( Figure 6). Therefore, we herein demonstrate that a particular strain of Bifidobacterium breve CCFM1067, may ameliorate PD-related symptoms through the MGBA.
DA in the brain works in tandem with the glutamate system to control motor and cognitive functions of the body [27]. DA synthesis is limited by the rate-limiting enzyme TH, and motor symptoms of PD emerge when TH levels drop below a particular threshold [28,29]. According to histopathological staining findings, ingestion of B. breve CCFM1067 reduced the death of dopaminergic neurons in the SN. Liao et al. demonstrated that probiotics protect neurons by preventing a decline in TH in dopaminergic neurons and boosting DA expression [7]. The neurotransmitters DA and 5-HT play essential roles in the brain [30]. In agreement with the findings of other studies, we found that B. breve CCFM1067 significantly enhanced the levels of neurotransmitters, such as DA, DOPAC, and 5-HT, in the striatum of PD mice. Neurotrophic factors, such as BDNF and GDNF, have been shown to protect dopaminergic neurons in PD mice by preventing neuronal death [31,32]. BDNF may modulate the neuroinflammatory response through the m-BDNF-tropomyosin receptor kinase B (TrkB) pathway, in addition to its role in maintaining neuronal survival [33][34][35]. In our study, after B. breve CCFM1067 intervention, PD mice exhibited a significant improvement in motor dysfunction, a significant reduction in dopaminergic neuron loss, increased neurotransmitter levels in the striatum, and restored BDNF and GDNF expression, suggesting B. breve CCFM1067 has a neuroprotective effect against MPTP-induced PD.
ROS are regarded as a major threat to bodily health, and their activation can perturb the redox equilibrium [36]. An increasing body of research indicates that ROS generation exacerbates α-synuclein misfolding in the brain, resulting in PD and other synaptic proteinopathies [37,38]. In clinical studies, probiotics have been shown to lower the development of α-synuclein aggregates by lowering ROS levels, thereby improving the condition of patients with PD [37,39]. In our study, MPTP-treated mice showed considerably lower CAT and GSH levels in the midbrain. Restoration of CAT and GSH antioxidant capabilities under MPTP toxicity after the intake of B. breve CCFM1067 is shown in Figure 3, indicating that this supplement partially attenuated MPTP-induced superoxide toxicity by increasing antioxidant levels. GSH levels in health and disease can provide important information regarding neuronal health [40,41]. Although GSH is not the sole antioxidant involved in PD, it seems to relate with disease severity because GSH depletion starts before other PD symptoms, including diminished complex I activity in the mitochondria and the development of Lewy bodies [42,43]. Increasing GSH levels in the brain is a possible therapy for PD, and multiple experimental investigations have shown this strategy works [44,45]. We discovered that B. breve CCFM1067 intake enhances the antioxidant capacity of the central nervous system, which helped PD mice reduce superoxide toxicity and alleviate oxidative stress. However, additional research is required to determine the precise mechanism by which B. breve CCFM1067 improves antioxidant levels in the brain.
Neuroinflammation is the primary cause of PD [46]. It accelerates the progression of PD by activating the brain microglia, which in turn generates pro-inflammatory substances through TLR4 receptors under the stimulation of MPP+ [47]. In our study, the brain tissue samples from MPTP-induced animals exhibited increased expression of both microglia (Iba-1) and astrocytes (GFAP). In contrast, mice administered B. breve CCFM1067 showed reduced MPTP-induced glial reactivity (Figure 2). Upregulated NLRP3 expression may enhance caspase-1-mediated inflammatory cascades and further stimulate the production of downstream inflammatory molecules, including IL-1β, IL-6, and TNF-α, all of which lead to an inflammatory response in a variety of neurological diseases [48,49]. In addition, in the postmortem brain, serum, and cerebrospinal fluid of patients with PD, inflammatory cytokines such as IL-1β, TNF-α, and IL-6 have been identified at excessively high concentrations [50,51]. This suggests that neuroinflammation plays a crucial role in the progression of neurodegenerative diseases [52]. According to our findings, the amount of IL-10 produced in mice with PD was drastically reduced, but the expression levels of IL-6, IL-1β, and TNF-α were markedly elevated ( Figure 3). Moreover, our findings also showed that treatment with the B. breve CCFM1067 strain was able to reduce these inflammatory reactions and, of note, was better than L-DOPA in suppressing the expression of pro-inflammatory cytokines (Figure 3).
Intestinal inflammation is a crucial contributor to PD development [53][54][55][56]. Systemic inflammation induced by intestinal inflammation and barrier disruption is key in gutbrain communication [57][58][59]. Therefore, we examined the levels of three crucial cytokines (TNF-α, IL-1β, and IL-6) in the colon. We found that MPTP-induced PD animals exhibited intestinal inflammation with changes in inflammatory factor gene expression, consistent with the results in brain tissues (Figure 3). When B. breve CCFM1067 was administered orally, the level of colonic inflammation was much lower, indicating that the probiotic activity of B. breve CCFM1067 may assist in decreasing intestinal inflammation. According to previous studies, systemic inflammation may disrupt the BBB, allowing pro-inflammatory molecules (such as cytokines and LPS) to enter the CNS [60][61][62]. Tight junction proteins, such as ZO-1 and occludin, contribute significantly to the maintenance of the intestinal barrier, which aids in protecting immune cells underneath the intestinal mucosa from invading infections [54,63,64]. Our findings further demonstrate that the intestinal epidermal mucosa in the colon of MPTP-induced PD animals had an enhanced permeability, and that intervention with B. breve CCFM1067 significantly alleviated this change ( Figure 3). Therefore, our data imply that intestinal barrier disruption caused by dysbiosis of the gut microbiota resulted in a pro-inflammatory cytokine leak, causing systemic inflammation. However, treatment with B. breve CCFM1067 reduced intestinal inflammation and prevented the development of PD. This was accomplished by improving the health of the intestinal barrier, preventing leakage of inflammatory molecules, and reducing inflammation in the brains of mice with PD.
A subsequent microbiota analysis revealed that the protective benefits of B. breve CCFM1067 could be mediated by the restoration of the normal microbiome. The αand β-diversity data indicated that the microbial community of B. breve CCFM1067 was comparable to that of the control group ( Figure 4). According to previous studies, MPTP induction may increase the occurrence of certain pathogenic bacteria and significantly impair the antioxidant and anti-inflammatory capabilities of the body [7,65], corroborating our findings. According to our research, MPTP intervention drastically changed the intestinal microbiota of PD mice ( Figure 5), increasing Bacteroides, Escherichia-Shigella, and Dubosiella, and decreasing Akkermansia ( Figure 5), which is consistent with the findings of previous studies [47,66,67]. In addition, treatment with B. breve CCFM1067 restored the altered levels of these bacteria; nevertheless, the MPTP-induced alteration of several taxa, most of which were members of the Bacteroidetes phylum, was irreversible. Interestingly, we discovered that the abundance of the Bifidobacterium genus increased after B. breve CCFM1067 administration, but there was no discernible variation in the genus after MPTP therapy. In the current study, pro-inflammatory cytokine levels were increased in the striatum and colon of MPTP-challenged mice but decreased after B. breve CCFM1067 administration. This indicates that B. breve CCFM1067 treatment may reduce inflammation and restore BBB integrity damaged by MPTP intoxication. This was further confirmed by the expression of three major tight junction proteins detected in the striatum and colon. It is worth noting that the correlation analysis revealed that the relative abundance of Bifidobacterium was substantially positively linked with relative ZO-1 expression in the striatum and colon. Meanwhile, our predicted metagenomic study showed MPTP-induced modifications in intestinal microbiota functions, specifically lowered "Biosynthesis of terpenoids and steroids," "Sesquiterpenoid and triterpenoid biosynthesis," "Steroid biosynthesis," and "Carotenoid biosynthesis" (Figure 6), but these alterations were reversed by B. breve CCFM1067. In conclusion, we found that the ingestion of B. breve CCFM1067 after MPTP treatment led to variations in microbial function and restored the MPTP-induced alterations on multiple metabolic levels. Further research is required to ascertain the changes in the Bifidobacterium genus in PD after MPTP treatment, as this probiotic species is widely recognized for its anti-inflammatory benefits. Furthermore, we found that MPTP-treated mice presented elevated levels of Enterobacteriaceae ( Figure 5), which is thought to be implicated in both PD and AD [68]. Intriguingly, our findings revealed a greater number of Escherichia-Shigella in PD mouse feces, which significantly decreased after oral administration of B. breve CCFM1067. In our study, the Escherichia-Shigella genus was negatively correlated with DA, 5-HT, and DOPAC levels in the striatum, and positively correlated with movement disorder-related indicators. The Escherichia-Shigella genus is inversely correlated with illness duration, as reported by Qian et al. [69]. Chen et al. demonstrated that fisetin ameliorated the disorder by decreasing the number of Escherichia-Shigella in the intestines of PD mice [70]. However, our results demonstrate that B. breve CCFM1067 alleviated the MPTP-induced increase in Enterobacteriaceae. Both the data presented by Petrov et al. [71] and our own analysis show that the abundance of Akkermansia in PD substantially decreased. Additionally, we discovered that the abundance of the genus Akkermansia increased with B. breve CCFM1067 treatment. It has been widely assessed that Akkermansia muciniphila, which is one of the most numerous bacteria in the gut microbiota, has a role in metabolic illnesses, including obesity [72] and diabetes [73]. This bacterium has been heralded as the "next generation" of probiotics [74]. Although its mode of action has not been completely elucidated, mounting data suggest that, through the gut-brain axis, A. muciniphila is crucial for proper brain function and has therapeutic potential in several neuropsychiatric diseases [75,76]. In our study, we found that the abundance of Dobusiella in MPTP mice treated with B. brev CCFM1067 was comparable to that in the control group. Overall, our findings show that supplementing mice with B. breve CCFM1067 may reduce the symptoms of MPTP-induced PD by restoring a healthy gut microbial population.
According to reports, SCFAs generated by bacterial fermentation in the colon serve as essential mediators for the gut microbiota to regulate gut-brain communication [77]. SCFAs play an important role in human neurological illnesses because of their anti-inflammatory properties and their capacity to strengthen the BBB [78,79]. In addition, SCFAs can act as neuromodulators by regulating the synthesis of neurotransmitters (especially 5-HT) [80] as well as the expression of their receptors, thus providing neuroprotection [81]. The present results also show that PD mice treated with B. breve CCFM1067 exhibited substantial alterations in the gut microbial composition and a considerable increase in fecal SCFA content. Sampson et al. showed that microbiota-derived SCFAs are crucial for α-synucleinmediated neuroinflammation, glial cell hyperactivation, and motor symptoms [10]. In our study, the increase in SCFAs may be one of the explanations for the anti-inflammatory action of B. breve CCFM1067 on the gut or brain of PD mice.

Conclusions
In summary, our study demonstrates that the probiotic B. breve CCFM1067 can exert neuroprotective effects on dopaminergic neurons in a subacute PD mouse model obtained via intraperitoneal injection of MPTP. We also showed that B. breve CCFM1067 improved the dyskinesia, death of dopaminergic neurons, and decrease in neurotransmitters caused by MPTP in mice. The possible mechanism by which B. breve CCFM1067 exerts neuroprotective effects in PD mice is through the increase in neurotrophic factor and SCFA levels, and the decrease in glial hyperactivation, resistance to oxidative stress injury, inflammatory response, and gut microbiota dysregulation, thus preventing dopaminergic neuron loss in the SN. Therefore, this study has preliminarily identified a new probiotic strain of B. breve CCFM1067, with promising applications as an effective oral supplement for PD prevention and therapy by influencing the gut-brain axis.