Bifico Ameliorates Neurological Deficits After Ischemic Stroke in Mice: Transcriptome Profiling

Abstract

Background/Aim

Evidence suggests that gut microbiota can affect various neurological diseases, including stroke. Stroke patients have an increase in harmful gut bacteria and a decrease in beneficial bacteria. This increases intestinal permeability, increases the risk of infection, and even affects many inflammatory factors. While probiotics may affect stroke prognosis by improving the gut environment. This study aimed to investigate the effect of probiotic Bifico on the neural function in mice after focal cerebral ischemia and explore its mechanisms of action.

Materials and Methods

A focal cerebral ischemia model was established in mice. Four weeks before modeling, animals were divided into three groups: Stroke plus Vehicle group, Stroke plus Pre-Bifico group and Bifico group. The infarct volume and neurobehaviors were evaluated. Whole-gene expression profiling was performed at different days after treatment (D1, D7, D14, D28) by RNA-seq. Differentially expressed genes (DEGs) were the processed for Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG). Some inflammation and immune related genes were screened and their expression was analyzed.

Results

Compared to the Stroke plus Vehicle group and Bifico group, the infarct volume and neurological score were significantly reduced in the Pre-Bifico group. There were 2 DEGs at D1, 193 DEGs at D7, 70 DEGs at D28 between Stroke plus Pre-Bifico group and Stroke plus Vehicle group. For GO analysis, there were 139 significant terms at D7 and 195 at D28. For KEGG, there were 2 significant pathways at D7 and 9 at D28. Among 87 genes related to inflammation and immunity, 6 DEGs were identified. The expression of CCL9 was significantly elevated at most time points after stroke compared to the Stroke plus Vehicle group, while that of CCL6, CXCL10, CD48, CD72 and CLEC7A was highly expressed only in the recovery stage of stroke.

Conclusion

Oral pre-treatment with Bifico for 28 days can reduce cerebral infarction and promote recovery of neurological function in stroke mice, which may be ascribed to the regulation of immunity and inflammation in the brain.

Ischemic stroke is a major cause of disability and death and has been a critical public health problem which causes a heavy social and economic burden worldwide. This is expected to get worse as population is aging. However, due to the short thrombolytic time window and concerns of safety and medical cost, only a very small number of patients receive thrombolysis and endovascular treatment. Therefore, to explore the mechanisms of ischemic brain injury and repair and to identify potential therapeutic targets for stroke have been hot topics in neuroscience researches.

Recently, studies have shown that the intestinal microbiota plays a significant role in the interaction between brain and distant organs such as the intestine, which is commonly referred to as the intestinal brain axis or the microbiota intestinal brain (MGB) axis (1). MGB axis acts as a communication portal integrating nerve, hormone and immune signals in the host. Some studies have emphasized the complexity of this interaction, and intestinal microflora may become a therapeutic target in many nervous system diseases including stroke (2-4).

The intestinal barrier includes the mechanical barrier, chemical barrier, immunological barrier and biological barrier (5). There is evidence showing that the intestinal flora of stroke patients displayed a distinct imbalance (6). The 16S rRNA sequence analysis of the intestinal microbiology showed that the intestinal flora of stroke patients were enriched in Enterobacteriaceae, Oscillatobacterium (Oscillospira) and Parabacteroides, while PrevoElla, Roseburia and Faffaliens were dominant in the intestinal flora of healthy individuals (7). In addition, clinical studies have indicated that the intestinal microbiological area is a source of infection after stroke. Due to the activation of sympathetic nervous system during stroke and subsequent series of events, intestinal permeability increases rapidly and significantly. As a result, the intestinal barrier is destroyed, and ultimately the intestinal bacteria transmit to the surrounding tissues (8).

It has been confirmed that probiotics are beneficial to health (9). Two meta-analyses revealed that the incidence of gastrointestinal complications, infection and flora imbalance was substantially lower and the hospital stay was markedly shorter in the enteral nutrition and probiotics group than in the simple enteral nutrition group. The combination of early enteral nutrition and probiotics in stroke patients can effectively improve the nutritional status of patients, enhance immune function, regulate intestinal mucosal barrier function and intestinal flora, and decrease the incidence of complications and gastrointestinal dysmotility. In addition, probiotics can exert anti-inflammatory effect by reduce interleukin-6 (IL-6), interleukin-10 and tumor necrosis factor-α (TNF-α) (10,11).

A study has shown that oral administration of Bifidobacterium brevis, Lactobacillus casei, Lactobacillus bulgaricus and Lactobacillus acidophilus can reduce ischemic injury after experimental stroke in mice, but the change in microbial levels has not been confirmed, and no behavioral benefits have been observed (12). In a study, stroke mice were treated with a mixture of probiotics (including Clostridium butyricum and Bifidobacterium, Lactobacillus casei and L. delbroeckii subsp) for 2 weeks. Results showed Lactobacillus acidophilus reduced ischemic area and improved neurological deficits through anti-oxidation and anti-inflammation (13).

It has been reported that probiotics may improve intestinal epithelial barrier, balance intestinal flora, and regulate inflammatory immune response. In fact, inflammatory immunity is involved in the whole process of stroke. We hypothesized that pre-treatment with oral probiotics could improve the harmful consequence of stroke by regulating inflammatory immune response.

In this study, the RNA-seq technique was employed to determine the gene expression profile of different treatment groups in mice intracranial infarction tissue at different time points. The differentially expressed genes (DEGs) were identified, and subjected to Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, in order to explore the mechanism by which probiotics exert protective effects on the stroke.

Materials and Methods

Experimental animals. Specific pathogen-free adult male C57BL/6 mice (23.0-26.0 g; 8-12 weeks) were purchased from Beijing Vital River Laboratory Animal Technology Co. Ltd., (Beijing, PR China). All the mice were housed in a stable environment with the humidity at 60%±5%, temperature at 22˚C±3˚C and 12 h light/12 h dark cycle. Animals were given ad libitum to water and food. This study was approved by the Animal Research Committee of Capital Medical University.

Animal models and reagents. As previously described, the distal middle cerebral artery (MCA) was ligated and both common carotid arteries (CCA) were occluded for 7 min. In brief, mice were anesthetized with 0.3% pentobarbital (0.1-0.2 ml/10 g, i.p.), and a 7-mm incision was made on the right side between the orbit and the ear. Following separating the muscle, MCA was exposed by drilling a 3-mm hole. The distal branches of MCA were ligated with 10-0 sterile sutures while both common CCA were occluded for 7 min (14). Finally, the wound was closed. The body temperature of animals was maintained at 37±0.5˚C during surgery and for 2 h after surgery.

Bifico capsules with at least 1.0×10⁷ c.f.u. Entero-coccus faecalis, 1.0×10⁷ c.f.u. Lactobacillus acidophilus, and 1.0×10⁷ c.f.u. viable lyophilized Bifidobacterium longum per capsule (210 mg) were purchased from Shanghai Sinepharm (Shanghai, PR China).

Experimental procedure. Figure 1 illustrates the experimental timeline of this experiment. According to the sample quantity calculation method (15), there were 3 groups: Stroke plus Vehicle group, Stroke plus Pretreatment Bifico (Stroke plus Pre-Bifico) group and Stroke plus Bifico group (n=21 per group). In the Stroke plus pre-Bifico group, mice were intragastrically treated with Bifico daily; in the Stroke plus Vehicle group, mice were intragastrically treated with the same amount of 0.9% sodium chloride; in the Stroke plus Bifico group, mice were treated with blank for 28 days. The dose of Bifico was 105 mg.

Figure 1. Timeline of experiments in three groups. After surgery, probiotics treatment continued. Behavioral tests were performed at each time point.

After 28 days, the ischemic stroke model was established as described above. After operation, animals continued to receive corresponding treatments after recovery from anesthesia. At 24 h after operation, 12 animals in each group were sacrificed, 9 of them were used for 2,3,5-triphenyltetrazolium chloride (TTC) staining and the infarct area was determined, and in the remaining 3 mice, the ischemic lesions (including core and penumbra area) were collected from brain tissues. After deep anesthesia, animals were perfused with 25 ml of saline phosphate buffer (PBS: 0.1 M, pH 7.3) prior to sample collection.

At 7 days, 14 days and 28 days after operation, 3 animals were sacrificed and the cortical infarct tissues were collected for the following experiment.

The adhesive removal test was performed at 1 and 7 days after stroke. A mouse was allowed to stay in a cage for 1 min, and then a 50-mm² tape was applied to the distal end of the left forelimb as a tactile stimulus. The time to removal of the tape was recorded with a cut-off time of 120 s for each test. For each animal, test was performed three times and an average was calculated. Mice were trained 3 times before surgery to ensure that the mice could remove the tape. The investigators responsible for the neurological assessment were blind to grouping.

RNA-sequencing. RNA was isolated from brain tissues and then subjected to RNA-seq analysis (16). In each group, a total of 3 μg of RNA was processed for further analysis. Figure 2 shows the process of analysis. Data filtering criteria: reads contaminated by the connector, low-quality reads, and reads with a N content greater than 5% were removed.

Figure 2. RNA-seq analysis. RNA was extracted from brain tissue, and after the total RNA sample passed the detection, magnetic beads with Oligo (dT) were selected for enrichment and purification. Addition of fragmentation buffer to the purified mRNA to make its fragments short. Use of the fragmented RNA as a template and using random primers for reverse transcription, cDNA synthesis is achieved. After end repair, addition of base A, addition of sequencing connectors, and PCR amplification, the entire library preparation process was completed and sequenced on the machine. In order to ensure the quality of information analysis, the raw data obtained from sequencing is filtered to obtain high-quality sequences for comparison with the reference database.

The FPKM method was employed to calculate gene expression. By eliminating the effects of differences in gene length and sequencing volume on the calculated gene expression, the FPKM method can be used to directly compare the differences in gene expression between samples. The gene expression was calculated as follows: FPKM=(10×3*F)/(NL/10×6). FPKM(A) referred to the expression of gene A; L referred to the length of exonic region of gene A; N referred to the total number of fragments uniquely aligned to the reference gene; F referred to the number of fragments uniquely aligned to gene A.

The DESeq 2 package was used to analyze data. The differences of p<0.05 and |log2(Foldchange)|≥0 were identified as statistically significant. GO enrichment analysis of DEGs was performed with the GOseq R package. The KEGG pathway analysis was performed with KOBAS software. A value of p<0.05 was considered statistically significant.

Inflammatory immune related genes. The genes that play an important role in the immune and inflammatory processes after focal cerebral ischemia were screened, and the DEGs were identified and their expression was detected at different time points.

Statistics. Statistical analysis was performed with GraphPad Prism 8.0 software. The infarct volume and results from behavioral tests were compared with one-way analysis of variance (ANOVA). A value of two-tailed p<0.05 was considered statistically significant.

Ethics approval and consent to participate. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of Beijing Friendship Hospital, Capital Medical University.

Results

Bifico treatment reduces infarct volume and improves neurological deficits. As shown in Figure 1, at 24 h after surgery, the infarct volume was determined after TTC staining. Among the three groups, mice in the Stroke plus Pre-Bifico had the smallest infract volume (p<0.05) (Figure 3A and B).

Figure 3. Bifico treatment reduces infarct volume and improves neurological function. (A) Twenty-four h after surgery, 2,3,5-triphenyltetrazolium chloride (TTC) staining of brain slices in different groups. (B) Infract volume of three groups at 24 h after surgery was calculated and compared. Among three groups, the Stroke plus Pre-Bifico group had the lowest infract volume (n=9), *p<0.05. (C) At 1 d after ischemic stroke, the time to removal of the sticker was recorded (n=22), *p<0.05. (D) At 7 d after ischemic stroke, the time to removal of the sticker was recorded (n=9).

At 24h after operation, behavioral test showed that the adhesive removal time in the Stroke plus Pre-Bifico group was significantly shortened as compared to the other two groups (p<0.05) (Figure 3C). At 7 days after surgery, the adhesive removal time was comparable among three groups (Figure 3D).

Temporal genome-wide expression. Above results showed the neurological function in the Stroke plus Pre-Bifico group was better than in the remaining groups, especially the Stroke plus Vehicle group. Then, the gene expression profile in the ischemic lesion was further detected to explain the potential molecular mechanism. At D1, D7, D14 and D28 after stroke, temporal gene expression profile was compared with that in the Stroke plus Vehicle group. First, the DEGs in the Stroke plus Pre-Bifico group and Stroke plus Vehicle group were screened at different time points. The screening criteria were p<0.05 and |log2(Foldchange)| ≥0. At D1, there were 2 DEGs, of which 1 was upregulated and 1 was downregulated; at D7, there were 193 DEGs, of which 111 were upregulated and 82 were downregulated; at D14, DEGs were not identified; at D28, there were 70 DEGs, of which 62 were up-regulated and 18 were down-regulated. As shown in Figure 4, two genes at D1 and top 10 genes at D7 and D28 were displayed (Table I).

Figure 4. Cluster map and volcano map demonstrate the DEGs between the Stroke plus pre-Bifico group and Stroke plus Vehicle group at different time points. At D1, there were 2 DEGs, of which 1 was up-regulated and 1 was down-regulated; at D7, there were 193 DEGs, of which 111 were up-regulated and 82 were down-regulated; at D28, there were 70 DEGs, of which 62 were up-regulated and 18 were down-regulated.

Table I. Top 10 DEGs at different time points.

Gene ontology enrichment analysis of DEGs. The DEGs were subjected to GO analysis of biological processes, cellular components and molecular functions. NO significant term was enriched at D1 and D14. In respect of biological processes, there were 98 significant terms at D7 and 146 at D28. The most enriched terms included response to tumor necrosis factor, regulation of protein modification process, regulation of phosphorylation, positive regulation of response to stimulus, cellular response to interleukin-1, regulation of response to stimulus, regulation of inflammatory response, neutrophil aggregation, complement activation, cell migration.

In respect of cellular components, there were 14 significant terms at D7 and 12 at D28. The most enriched terms included side of membrane, receptor complex, plasma membrane protein complex, extracellular space, high−density lipoprotein particle, external side of plasma membrane.

In respect of molecular functions, there were 27 significant terms at D7 and 37 significant terms at D28. The most enriched terms included receptor regulator activity, passive transmembrane transporter activity, Toll−like receptor binding, immunoglobulin receptor binding, double−stranded RNA binding. Of them 10 GO items with the highest enrichment were represented by bubble plots as shown in Figure 5. The larger the bubbles, the higher the Rich Ratio was; the redder the color, the higher the enrichment was.

Figure 5. Top 10 GO items with the highest enrichment from biological processes, cellular components and molecular functions. The higher the Rich_Ratio, the redder the color is, and the higher the degree of enrichment is.

Pathway analysis of DEGs. To further investigate the function of DEGs, DEGs were subjected to KEGG pathway analysis. There were 2 significant pathways at D7 and 9 at D28 (Figure 6). Pathways that clustered significantly at different points were selected and divided into two groups: function-related and disease-related pathways. The function-related pathways included RIG−I−like receptor signaling pathway, complement and coagulation cascades, NOD−like receptor signaling pathway, cytokine−cytokine receptor interaction, IL−17 signaling pathway, and chemokine signaling pathway. The disease-related pathways included Influenza A, Measles, Hepatitis C and Epstein−Barr virus infection.

Figure 6. Scatter plot to display the results of KEGG pathway enrichment. Two significant pathways at D7 and 9 significant pathways at D28 in the bubble chart. The higher the Rich_Ratio, the redder the color is, and the higher the degree of enrichment is.

Expression of inflammatory immune related genes. A total of 87 genes that may play roles in the inflammatory and immune processes after focal cerebral ischemia were identified, including cytokines and receptors, complement and receptors, surface antigens, C-type lectin family and major histocompatibility complexes. Then, 6 DEGs were screened out of these 87 genes, including CCL6, CCL9, CXCL10, CD48, CD72 and CLEC7A. Figure 7 shows the expression of 6 DEGs at different time points.

Figure 7. Six DEGs were screened out of these 87 genes that may play roles in the inflammatory and immune processes after focal cerebral ischemia. The graph shows the expression levels of genes at different time points. Compared to the Stroke plus Vehicle group, the expression levels of CCL9 at D1, D14 and D28 increased in the Stroke plus Pre-Bifico group. The expression levels of CCL6, CXCL10, CD48, CD72 and CLEC7A increased at D28 which is the recovery stage of stroke.

Results showed chemokine CCL9 was highly expressed at most time points during stroke (D1, D14 and D28) as compared to the Stroke plus Vehicle group, while chemokine CCL6, CXCL10, CD48, CD72 and CLEC7A were highly expressed only during the recovery stage of stroke (D14 and D28). We speculated that CCL9 exerted protective effect in the acute stage of stroke. CCL6, CXCL10, CD48, CD72 and CLEC7A conferred protective effect in the recovery stage of stroke.

Discussion

In the present study, an animal model of focal ischemic stroke was established by permanent MCA occlusion coupled with transient CCA ligation, which is similar to the partial reperfusion injury in stroke patients. As an intermediate model for permanent ischemia and complete reperfusion after transient ischemia, this model is well matched. Moreover, this surgical method reduces the mortality rate of experimental animals, which greatly reduces the number of experimental animals used. This animal model is widely used for stroke research (17,18).

It is estimated that the number of bacteria in the gastrointestinal tract is about 3.8×10¹³, similar to the total number of human cells. However, the total mass of intestinal microflora in healthy people accounts for only 0.3% of total body weight (19). Interestingly, there are 9 million unique genes in the intestine observed in the metagenome study, which is 450 times that of the whole human genome. The huge genome of intestinal flora also creates an opportunity for us to find new targets for the treatment of diseases.

The criteria for the selection of probiotics include 1) selective fermentation of potentially beneficial microbiota in the colon, 2) resistance to upper digestive tract digestion, 3) selective stimulation of probiotic growth, 4) stability under various food/feed processing conditions and 5) beneficial effects on host health (20). Based on the above criteria, Bifico was used for probiotic preparation in our study. Each Bifico capsule contains at least 1.0×10⁷ c.f.u. Entero-coccus faecalis, 1.0×10⁷ c.f.u. Lactobacillus acidophilus, and 1.0×10⁷ c.f.u. viable lyophilized Bifidobacterium longum (210 mg). In the pilot study, the dose of Bifico capsule was investigated, and finally, 105 mg of Bifico capsule was determined as the dose used in this study. At this dose, the animals can eat normally and have a continuous increase in body weight.

In our study, results showed mice in the Stroke plus Pre-Bifico group after infarction achieved better behavioral recovery as compared to the Stroke plus Vehicle group. Therefore, in the gene expression analysis, we focused on the Stroke plus Pre-Bifico group and Stroke plus Vehicle group. Results showed there were DEGs at 1, 7 and 28 days between two groups. At 7 days after infarction, the expression of end-regulated LAO1 significantly increased. Our results showed that LAO1 was associated with the adjustment of the intestinal flora and related to the amino acid metabolism. At 28 days after infarction, the expression of DEGs was down-regulated. The top 10 DEGs were mainly immunoglobulin-related genes. Thus, we speculate that immune related factors play an important role in the recovery of infarction. GO and KEGG analyses showed that CytoKine-CytoKine Receptor Interaction was an enriched signal pathway at both day 7 and day 28. The intestinal flora affects the central nervous system and immune system of the host to improve cerebral diseases by regulating the immune factor. Therefore, we focused on the analysis of inflammatory immune related genes.

Some studies have explored probiotics, inflammation and immunity. Probiotics may reduce the risk of cerebrovascular diseases by inhibiting inflammatory markers, such as high-sensitivity C-reactive protein (hs CRP), total cholesterol and low density lipoprotein (LDL) cholesterol and some cytokines involved in inflammatory response (21). Supplementation with probiotics can significantly improve mental health and metabolic status of biomarkers, such as hsCRP, nitric oxide, parameters related to inflammation and oxidative stress, as well as low density lipoprotein or total cholesterol in patients (22-24). Sánchez et al. proposed four different mechanisms by which probiotics exert protective effects on human health: 1) probiotics can improve epithelial barrier function; 2) probiotics can compete with pathogens through nutrients and adhesion sites; 3) probiotics can affect other tissues through the production of immune system and neurotransmitters; 4) probiotics can regulate immune system (25). Studies have shown that oral probiotics in the elderly can reduce circulating pro-inflammatory cytokines, such as IL- 6, IL-1β and TNF-α (26).

There is evidence showing that probiotics significantly reduce the malondialdehyde and TNF-α in ischemic brain (13). Another study investigated colitis related cancer. Results showed a set of genes were identified as potential targets for Bifico treatment, including CXCL1, CXCL2, CXCL3, CXCL5, and ligands for C-X-C kinesin receptor 2 (CXCR2) (27). These studies suggest that probiotics can regulate inflammation and immunity, which is consistent with our findings.

In our study, animals were pretreated with Bifico for 28 days, which may greatly balance the intestinal flora, increase beneficial bacteria and reduce harmful bacteria. After subsequent stroke, animals with Bifico pre-treatment showed better recovery ability, which may be ascribed to the regulation of inflammatory immune response. Our study provides evidence on the prophylactic use of probiotics in the future, such as in patients with high risk for perioperative stroke.

There were limitations in our study. First, only healthy animals were used. In future studies, diseased animals (such as hyperlipidemia, arteriosclerosis and hypertensive animals) can be investigated in stroke models. Second, this was a preliminary study, and the impact of probiotics on intestinal flora was not further investigated. In addition, the roles of other pathways in the pathogenesis and/or recovery of stroke should be further explored.

Funding

This study was supported by the National Natural Science Foundation of China (81671191 to YZ) and Capital Health Research and Development of Special (2018-4-2025 to YYZ).

Data Availability

The datasets used and analyzed during the present study are available from the corresponding author on reasonable request.

Conflicts of Interest

The Authors declare that there are no conflicts of interest regarding the publication of this paper.

Authors’ Contributions

YH: performed experiments and responsible for data analysis and manuscript writing. HX, ST, YZ, ZW: performed the experiments and responsible for data analysis. YYZ and YZ: responsible for conception and experimental design.