The effects of Saccharomyces boulardii on rat colonic hypermotility induced by repeated water avoidance stress and the potential mechanism

Animals

The male Wistar rats (200–220 g) with a certified report were acquired from Vital River Laboratories (Beijing, China) and bred in a constant temperature (20 °C∼25 °C) and humidity (50%∼65%) environmentally controlled chamber with a 12-hour light/dark cycle. Rats received sterile food and water freely. Padding was changed twice a week. Rats were euthanized by inhalation of CO₂ at the end of the experiment. Animal ethics approval was granted by the Laboratory Animal Welfare and Ethics Committee of Renmin Hospital of Wuhan University (approval ID: NO.20210308) and all procedures were in accordance with the Experimental Animal Protection Committee of the Ministry of Health of the People’s Republic of China.

Repeated WAS protocol

We placed the rats on a circular platform (9 × 9 × 15 cm) fastened to the center of a tank (40 × 40 × 50 cm) containing water for one hour per day for ten days, according to prior test techniques in our laboratory. The amount of water in the tank was kept at a maximum height of one centimeter below the platform. The controls were put on the same platform as the experimental animals, but were not given any water. That is, the control rats received sham water avoidance stress (SWAS). The above experiment was implemented at 10:00 am each day. During the whole experiment, the defecation of each rat was recorded for subsequent statistical analysis.

Experimental design

The rats were randomly divided into three groups after three days of adaptive feeding (n = 8 per group): (i) control; (ii) NS + WAS; and (iii) Sb + WAS. Sb was purchased from Biocodex (Gentilly, France) and contained ≥1.3 × 10⁹ CFU/g viable Sb. Rats in the Sb + WAS group were gavaged for 14 consecutive days with Sb saline suspension (0.5 g/kg/d). The dose of Sb was defined based on the dose administered to rodents in previous studies. The NS+ WAS rats were given the same volume of NS via gavage. Only distilled water was given to the control rats. From the 15th day, the rats in the NS+WAS group and the Sb + WAS group were treated with WAS for 1 h after the conventional gavage for 2 h for 10 consecutive days, and the rats in the control group received SWAS. All rats were euthanized on day 25, and their blood was collected to obtain serum and stored at −20 °C for future analyses. The colon was removed aseptically to measure motility. Stool samples were also collected and set aside at −80 °C.

H&E staining

Rats were euthanized, and the excised proximal colon tissue was immersed in 4% paraformaldehyde and fixed at 4 °C for 24 h. Following that, samples were dehydrated in various concentrations of ethanol before being paraffin embedded. After that, the embedded paraffin samples were cut to a thickness of 4 µm. Subsequently, the sections were stained at room temperature with hematoxylin and eosin (H&E). The stained sections were scanned using NDP.view 2 software, and the structure, length, and integrity of colonic villi were observed.

Colonic motility tests

The proximal colons of rats were carefully removed, the intestinal lumen was opened along the mesenteric border, and the intestinal contents were then cleaned with cold Ca2 +-free saline solution perfused with 95% O₂ and 5% CO₂, sheared along the vertical and horizontal directions of the long axis of the colon to prepare CM and LM strips (8 mm long and 2 mm wide), respectively. We immersed the prepared smooth muscle (SM) strips in a thermostatic bath with 5 mL of Tyrode buffer containing a mixture of 95% O2 and 5% CO2. The ends of the colonic SM strips were fastened to the thermostatic bath containing the Tyrode’s buffer and to the isometric force transducer (JZJOIH, Chengdu, China). SM strips were incubated under initial resting tension (1.0 g for CM strips and 1.5 g for LM strips) until the amplitude was at equilibrium. The contraction of the CM/LM strips in each group of rat colon was measured using the RM6240 multichannel physiological signal system (Chengdu, China). In addition, SM strips were prepared from the isolated proximal colons of rats in the model group. Similarly, changes in colonic SM tone were measured with a tension transducer. After a period of regular contraction signals, incubation of colonic SM strips by adding a concentration gradient of Sb solutions to the thermostatic bath, we determined the in vitro administration concentration of Sb by pre-experiment (final incubation concentrations of CM strips were 1, 2, and 3 mg/mL, respectively; final incubation concentrations of LM strips were 0.5, 1, and 1.5 mg/mL, respectively). The changes in spontaneous contractile activity were observed and recorded.

Immunohistochemistry

The paraffin sections of the colon were dewaxed using an environmentally friendly dewaxing agent and then dehydrated in graded alcohol (100%, 95%, and 75%) for 5 min each. Antigen repair, blocking, and other steps were carried out in strict accordance with the kit instructions. Sections were incubated with rabbit anti-TLR4 antibody (1:200, bs-23927R, Bioss) overnight at 4 °C and with secondary antibody for 45 min at 37 °C. Finally, freshly prepared DAB working solution was added to the tissue sections and observed under the microscope, with the color development terminated by tap water. The sections were then re-stained with hematoxylin, dehydrated and dried in gradient alcohol, and sealed with an environmentally friendly sealant. The sections were scanned using NDP.view 2 software and a specialist pathologist evaluated the images and calculated HSCORES without knowing the grouping, where HSCORE = ΣPi(i +1), (Pi: The number of positive cells as a percentage of the number of all cells in the section; i: Coloring intensity).

q RT-PCR

Total RNA was extracted from rat colon using Trizol reagent (Takara, Japan), and cDNA was obtained with a reverse transcription reagent kit (Takara, Japan). The LightCycler 480 SYBR Green I Master (Roche, Basel, Switzerland) and the LightCycler 480 qPCR device were used for the qPCR. The relative expression level of TLR4 was calculated using 2^−ΔΔCT method. The primer sequences used for amplification were as follows: the upstream primer for TLR4 is 5′-TCCAGAGCCGTTGGTGTATC-3′, the downstream primer is 5′-GAGCATTGTTCCTCCCACTCG-3′; the upstream primer for GAPDH is 5′-GACATGCCGCCTGGAGAAAC-3′, the downstream primer is 5′-AGCCCAGGATGCCCTTTAGT-3′.

ELISA

ELISA assay kits (MULTI SCIENCES, Hangzhou, China) were used to measure IL-6, IL-1β, IL-10, and IFN-γ in serum and colonic tissue homogenates. The experimental steps were in strict accordance with the manufacturer’s instructions.

16S rRNA Gene Sequencing

Extraction of DNA from stool samples via the Mag-Bind® Stool DNA Kit (Omega, Guangzhou, China) as per the kit’s instructions. Electrophoretic separation was performed with a 1.2% agarose gel, and the purity and concentration of the isolated DNA were determined using a NanoDrop NC2000 spectrophotometer (Thermo Science, Waltham, USA). The primers 338F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) were utilized to amplify the hypervariable V3 and V4 regions of the bacterial 16S rRNA gene. PCR amplification was performed with Q5® High-Fidelity DNA Polymerase from New England Biolabs. The amplification cycle was controlled to minimize the number of amplification cycles and to ensure identical amplification conditions for the same batch of samples. PCR amplification and purification of amplification products were used for library preparation and pyrosequencing, which were performed at paired-end 250 bp on the Illumina NovaSeq platform (Illumina, San Diego, USA) by Personal Biotechnology, Co., Ltd. (Shanghai, China).

Statistical analysis

The data from this experiment was statistically analyzed using SPSS Statistics 28.0 (IBM SPSS, Chicago, IL). All results were expressed as mean ± SEM. Significant differences between all groups were assessed with the use of the independent sample t-test, one-way ANOVA test with subsequent Tukey’s multiple comparisons test, or Kruskal-Wallis test, Bonferroni correction was applied to multiple testing. P < 0.05 indicated the presence of statistical significance.

Histopathological observation

As shown in Fig. 1, no clear pathological changes were observed in the H&E staining of the colonic tissues of rats in each group. The colon structure of all three groups was clear, no inflammatory cells infiltrated or tissue damage was evident in the colonic mucosa of any of the groups. The mucosal epithelium appeared intact, the arrangement of intestinal glands was in good condition, and goblet cells were abundant. However, repeated WAS caused mild villi damage, and the microstructure of the colonic mucosa was similar in the WAS + Sb group and the control rats. Fig. 1 shows longer villi length in the WAS + Sb rat and shorter villi length in the WAS+NS rat. The observations outlined above suggest that oral administration of Sb could decrease the damage of colonic villi in IBS-D rats.

The fecal pellets expulsion

Chronic stress enhanced the motility of the colon significantly. Here, we assessed the colonic motor function of rats using the quantity of fecal pellets defecated during a 1-h period. Figure 2A represents the fecal pellet excreted by the rats from the control, NS + WAS, and Sb + WAS groups during the session. Figure 2B shows the average number of fecal pellets excreted by the three groups of rats during the 10-day WAS experiment. Repeated WAS accelerated fecal pellet expulsion. Rats in the NS+WAS group excreted more fecal pellets than controls (NS + WAS 10.01 ± 0.39 vs Control 2.82 ± 0.17, P < 0.0001). Rats in the Sb +WAS group defecated less than rats in the NS+ WAS group (Sb + WAS 6.40 ± 0.38 vs NS + WAS 10.01 ± 0.39, P < 0.0001), although it was still greater than controls (Sb + WAS 6.40 ± 0.38 vs Control 2.82 ± 0.17, P < 0.0001).

The effects of repeated WAS on colonic strip contractile function

In comparison to the control rats, repeated WAS dramatically enhanced spontaneous contraction of the isolated proximal colon. The mean contraction amplitude of CM strips in NS+WAS rats was statistically upregulated compared to controls (1.82 ± 0.23 g vs 0.53 ± 0.03 g, P < 0.0001). A similar phenomenon has been observed in Fig. 3D. The LM strips contraction amplitude was obviously higher in NS + WAS rats than in controls (1.00 ± 0.01 g vs 0.30 ± 0.04 g, P < 0.0001). However, oral administration of Sb reversed the enhanced contraction of the proximal colon in rats caused by repeated WAS in vivo.

Sb alleviate colonic hypermotility in vitro

Sb dose-dependently suppressed the spontaneous contraction of the CM/LM strips containing all layers. The average contractile amplitude in the CM strips (Fig. 4C) was 0.92 ± 0.22 g before Sb was added, but we found it reduced to 0.57 ± 0.10 g (p = 0.2517 VS Control), 0.45 ± 0.06 g (p = 0.0840 VS Control), and 0.32 ± 0.06 g (p < 0.05 VS Control) in the presence of Sb 5, 10, and 15 mg, respectively. Likewise, prior to supplementation of Sb, the average amplitude of LM strips (Fig. 4D) was 0.51 ± 0.05 g. After the addition of Sb 2.5, 5, and 7.5 mg, it was dropped to 0.42 ± 0.04 g (p = 0.3242 VS Control), 0.26 ± 0.03 g (p < 0.001 VS Control), and 0.17 ± 0.03 g (p < 0.0001 VS Control), respectively.

Immunohistochemical staining for TLR4

As shown in Fig. 5, repeated WAS caused a significant upregulation of TLR4 expression levels in rat colonic tissues compared to the control group (p < 0.0001). Compared with NS+WAS, TLR4 expression levels in rats in the Sb + WAS group were visibly decreased (p < 0.01), but still higher than rats in the control group (p < 0.001).

Colon TLR4 mRNA expression

Figure 6A illustrates that the TLR4 mRNA expression level in the NS + WAS group was higher than that in the control group (P < 0.01). However, oral intake of Sb reversed the higher expression level of TLR4 mRNA due to repeated WAS (P < 0.05). The control and Sb + WAS groups did not have a significant difference.

The concentrations of cytokines

The serum IL-6, IL-1β, and IFN-γ concentrations were significantly higher in the NS+WAS group than in the control group (P < 0.05), as shown in Figs. 6B–6D. Notably, oral Sb treatment significantly reversed the elevation of serum IL-6, IL-1β, and IFN-γ caused by repeated WAS (P < 0.05). But IL-10 was not detected in the serum. Meanwhile, the changes in IL-6, IL-1β, and IFN-γ levels in colonic tissue were consistent with those in serum (Figs. 6E–6G). In addition, oral administration of Sb upregulated IL-10 levels in the colon tissue of chronic stress-loaded rats (P < 0.05, Fig. 6H).

The gut microbiota diversity of rats

A total of 1,904,405 raw reads were produced from 18 rat fecal samples using high-throughput sequencing. On the basis of 97% sequence similarity, the V3 region and V4 region sequences were identified as 13,893 bacterial OTUs in the rat gut, with 1,176 common bacterial OTUs (Fig. 7A).

As shown in Figs. 7B–7C, repeated WAS decreased the Faith_pd index (p < 0.05) in comparison to controls. Furthermore, the results of the Faith_pd index and Goods_converage index indicated a trend towards higher alpha diversity in rats in the Sb + WAS group compared to the NS+WAS group, although statistical significance was not reached (Faith_pd, P = 0.34; Goods_converage, P = 0.42). Meanwhile, NMDS analysis was performed to visualize the difference between individual treatment groups. Three sets of samples were separated and divided into different groups, indicating that beta diversity was significantly different among the three groups (Figs. 7D–7E). These findings indicated dysbiosis in the gut microbiome of rats in the NS+WAS group.

The gut microbiota composition of rats

At several levels, we compared the differences in microbiota between the three groups. Four phyla were found to have relatively high abundance at the phylum level, with Firmicutes, Bacteroidetes, Proteobacteria, and Actinobacteria being the dominant phyla, which accounted for 99.6% of all bacteria (Fig. 8A).

The NS +WAS group had a greater relative abundance of Acidobacteria than the Sb + WAS group (P < 0.05, Fig. 8G). When compared to the NS + WAS group, the relative abundances of Patescibacteria (P < 0.01, Fig. 8D) and Cyanobacteria (P < 0.05, Fig. 8F) were higher in the control group. The control group had a substantially higher relative abundance of Epsilonbacteraeota than the NS + WAS and Sb + WAS groups (P < 0.01, Fig. 8E). Similarly, the control group had a higher relative abundance of Actinobacteria than the Sb + WAS group (P < 0.01, Fig. 8C).

Figure 8B shows the top 20 microbial genera. The relative abundances of Allobaculum (P < 0.01, Fig. 8I) and Subdoligranulum (P < 0.01, Fig. 8J) were lower in the control group, as well as Anaerostipes (P < 0.05, Fig. 8K) in the Sb +WAS group, when compared to the NS +WAS group. Furthermore, when compared to the controls, repeated WAS reduced Turicibacter relative abundances (P < 0.0001, Fig. 8L). Yet, Sb supplementation had a tendency to increase the relative abundance of Blautia when compared to controls (P = 0.060, Fig. 8H).

To gain insight into the differences in the gut microbiota communities of the three groups, we performed linear discriminant analysis effect size (LEfSe) analysis at the microbial genus level with LDA scores greater than or equal to 2.0. The control group contained large amounts of Turicibacter, Clostridium_sensu_stricto_1, and Bifidobacterium. The NS +WAS group had higher levels of Allobaculum, Subdoligranulum, and Faecalibacteriumin. A significant enrichment of Blautia was noticed in the Sb +WAS group (Fig. 9).