Distinct Limosilactobacillus reuteri microcapsule models: construction and therapeutic evaluation in DSS-induced colitis mice

Abstract

The limited stability of L. reuteri in liquid formulations during storage, transport, and gastrointestinal transit presents a major challenge for its application as a probiotic. To address this, our study developed two distinct microcapsulation models for L. reuteri A-1, tailored for specific release profiles: slow-release and quick-release model. Utilizing single-factor experiments and response surface methodology, we optimized the encapsulation process, achieving a maximum embedding efficiency of 88.64% for the slow-release model. The quick-release model demonstrated a high cumulative release rate of 83.3%. Structural characterization revealed microcapsules with dense, smooth surfaces and internal porous structures. Storage stability tests confirmed that low temperature (4 °C) best preserved viability. In the DSS-induced murine colitis model, the quick-release model significantly alleviated disease symptoms, including weight loss, colon shortening, inflammatory cytokine imbalance, and mucosal damage. 16S rRNA analysis further showed that the quick-release system helped restore the gut microbiota of colitis mice to a state closer to that of healthy controls. This work establishes a novel technological platform for the controlled release and targeted delivery of probiotics, holding significant promise for the development of live biotherapeutic products.

Introduction

Probiotics, defined as live microorganisms conferring health benefits when administered in adequate amounts (≥10⁸ CFU/mL), primarily colonize the human digestive tract, particularly the small intestine and colon. They contribute to maintaining gut microbiome homeostasis and have shown potential in ameliorating various disorders, including obesity, inflammatory bowel disease, and metabolic syndromes^1,2. However, the efficacy of conventional liquid probiotic formulations is often compromised by poor stability during storage and transport, as well as substantial viability loss under harsh gastrointestinal conditions, limiting their colonization and therapeutic effect.

Limosilactobacillus reuteri (L. reuteri), formerly known as Lactobacillus reuteri³, is a kind of probiotic that naturally exists in the intestines of many mammals⁴, making it one of the most common probiotics in our daily life. It plays a probiotic role by secreting lactic acid and promoting the proliferation of intestinal epithelial cells, and has the role of immune regulation⁵. Studies have shown that L. reuteri can improve colitis in mice clinically and morphologically, significantly relieve colitis symptoms in mice, and prevent the occurrence and development of anxiety and depression⁶. L. reuteri is often used as an additive in livestock and poultry production, which can improve intestinal health and production performance of livestock and poultry⁷. In addition, studies have shown that L. reuteri can stimulate the production of anti-cancer cytokines in vivo, inhibit the migration and colonization of melanoma cells, and thus slow down the progress of melanoma^8,9. However, L. reuteri is easily affected by the gastrointestinal digestive environment, resulting in its low colonization efficiency, which seriously affects its effect.

Nowadays, oral delivery remains the preferred route for probiotic administration due to its simplicity, cost-effectiveness, and patient compliance¹⁰. Nevertheless, low oral bioavailability persists as a major challenge, largely due to degradation by gastric acid, bile salts, and digestive enzymes¹¹. Microencapsulation has emerged as a promising strategy to enhance probiotic survival by encapsulating viable cells within protective matrices that shield them from environmental stressors and enable controlled release¹². While most current studies focus on developing sustained-release systems using novel materials, such approaches often suffer from slow drug accumulation and incomplete release prior to excretion. Especially in some people with digestive system problems or diseases (such as the elderly, acute diarrhea), most of the embedded probiotics cannot be released rapidly in the intestine, resulting in low colonization efficiency.

The extrusion method is often used to prepare controlled-release drug microcapsules¹³. Microcapsules prepared using this method can maximize cell activity. Natural polymer materials have attracted much attention for their excellent biosafety, film-forming properties and storage stability. Sodium Aginate (SA) has been widely used in the biomedical field because of its excellent biocompatibility, non-toxicity and abundant storage¹⁴. When the pH value is neutral or alkaline, the viscosity is stable, and when the temperature rises above 80 °C, the viscosity decreases. The lower the pH, the stronger the viscosity of the sodium alginate solution. Free carboxyl groups in sodium alginate can react with divalent cations such as Ca²⁺ to form a stable gel¹⁵. As the pH increases, the structure continues to dissociate, and the formed gel will slowly swell and degrade, but the gel formed by SA is porous and large in size and is not tight in packaging, which allows gastric acid and digestive enzymes to penetrate the gel and degrade probiotics. Therefore, it is usually combined with chitosan (CS) to improve its stability and protection ability through electrostatic action. SA and CS are often used together and are good delivery materials¹⁶. CS is a biodegradable, highly biocompatible and non-toxic polysaccharide, and is one of the most commonly used biomaterials in the field of biomedical¹⁷. It also has antioxidant, antibacterial and antifungal properties¹⁸ and is widely used in the preparation of capsules containing substances such as drugs, proteins and cells¹⁹.

To overcome these limitations, L. reuteri A-1 was employed as a model probiotic. Following the characterization of its physiological properties and gastrointestinal tolerance, biomass was produced via high-density fermentation. Two distinct microcapsule systems were developed: a slow-release system optimized for encapsulation efficiency and a quick-release system engineered for enhanced disintegration triggered by pancreatic juice. Both were systematically evaluated for their morphological characteristics, storage stability, in vitro release profiles, and in vivo efficacy in a DSS-induced murine colitis model.

Results

Gastrointestinal fluid tolerance, growth curve and culture medium optimization

The pronounced acidity of gastric juice poses a significant survival challenge to most intestinal probiotics. When evaluated in simulated gastric juice, L. reuteri A-1 exhibited poor acid tolerance, with a viability of only 7.2 ± 3.18% after 2 h of exposure (Fig. 1A). In contrast, the neutral pH of pancreatic juice generally supports high probiotic survival. Consistent with this, L. reuteri A-1 showed strong tolerance in simulated pancreatic juice, retaining a viability of 93.08 ± 3.40% after 4 h (Fig. 1A). The growth kinetics of L. reuteri A-1 were monitored over 48 h by measuring OD₆₀₀ and pH at 6 h intervals. As shown in Fig. 1B, the strain reached a maximum OD₆₀₀ of 3.15 after 24 h of static culture at 37 °C in MRS medium, with the pH stabilizing thereafter. This suggests that the culture entered the stationary phase, during which bacterial proliferation slowed while metabolite secretion—such as short-chain fatty acids (SCFA, e.g., lactic acid and acetic acid)—may have continued. Accordingly, a 24 h culture is suitable for maximizing biomass, whereas extending the culture to 36–48 h may be beneficial for enhancing metabolite accumulation (e.g., SCFAs or bacteriocins).

Fig. 1. Gastrointestinal fluid tolerance, growth curve and culture medium optimization diagram of L. reuteri A-1.

A Tolerance result of L. reuteri A-1 in gastrointestinal fluid. B Growth curve of L. reuteri A-1 and its pH value changes; Growth curve (C) and pH (D) value changes of L. reuteri A-1 during optimization of different carbon source; growth curve (E) and pH value (F) changes of L. reuteri A-1 during optimization of different nitrogen sources; growth curve (G) and pH value (H) changes of L. reuteri A-1 during optimization of different carbon-nitrogen ratios.

The effects of different carbon sources, nitrogen sources, and carbon-to-nitrogen (C/N) ratios on the growth (OD₆₀₀) of L. reuteri A-1 and concomitant pH changes are shown in Fig. 1C–H. Among the carbon sources tested, glucose supported the most rapid growth, yielding the highest cell density (OD₆₀₀ = 3.17) at 21 h, followed by maltose and sucrose. While the final densities at 48 h were not significantly different (P > 0.05), glucose is recommended as the optimal carbon source for large-scale fermentation due to its economic efficiency. Regarding nitrogen sources, soybean peptone yielded the maximum cell density (OD₆₀₀ = 3.93) at 30 h, outperforming yeast extract, tryptone, peptone, and casein peptone. Consequently, soybean peptone was identified as the most suitable nitrogen source. Furthermore, a C/N ratio of 10:1 was found to be optimal, achieving the peak density (OD₆₀₀ = 3.93) at 30 h, superior to ratios of 5:1, 20:1, and 30:1. In all cases, bacterial density increased rapidly before stabilizing or declining, with pH trends inversely correlating with growth.

Construction of a slow-release model of L. reuteri A-1 microcapsules

The results of the single-factor experiments are presented in Fig. 2A–F. The embedding rate (ER) of L. reuteri A-1 microcapsules was influenced by each factor as follows: sodium alginate concentration (optimal at 3%, ER = 77.8%; Fig. 2A), chitosan concentration (optimal at 1%, ER = 80.05%; Fig. 2B), CaCl₂ concentration (optimal at 2%, ER = 82.02%; Fig. 2C), and curing time (optimal at 1 h, ER = 76.04%; Fig. 2D). Beyond these optimal points, further increases in concentration or time resulted in a gradual decline in ER. Based on these findings, the following conditions were identified as optimal for encapsulation and used as the central points for subsequent response surface methodology (RSM): 3% sodium alginate, 1% chitosan, 2% CaCl₂, and 1 h curing time.

Fig. 2. Single-factor experiment and response surface methodology (RSM) to optimize the microcapsule embedding (A–F) and release (G–L) efficiency of L. reuteri A-1.

A Sodium alginate concentration for embedding rate. B Chitosan concentration for embedding rate. C CaCl₂ concentration for embedding rate. D Curing time for the embedding rate. E, F 3D response surface and contour plots for embedding rate. G Sodium alginate concentration for release rate. H Chitosan concentration for the release rate. I CaCl₂ concentration for the release rate. J Curing time for the release rate. K, L 3D response surface and contour plots for release rate.

A Box-Behnken design was employed for the RSM optimization, with the embedding rate (Y), determined by the release amount in sodium citrate solution before lyophilization, as the response variable. The experimental data (Supplementary Table 1) were fitted to obtain a quadratic regression equation describing the relationship between Y and the independent variables: X₁ (sodium alginate concentration), X₂ (CaCl₂ concentration), X₃ (chitosan concentration), and X₄ (curing time)

$$\begin{array}{c}\mathrm{Y}={-6.84\mathrm{X}}_1^2-{11.91\mathrm{X}}_2^2-{4.27\mathrm{X}}_3^2-{4.82\mathrm{X}}_4^2-{6.24\mathrm{X}}_1{\mathrm{X}}_2+{0.8271\mathrm{X}}_1{\mathrm{X}}_3-{1.51\mathrm{X}}_1{\mathrm{X}}_4\hfill \\ -{2.48\mathrm{X}}_2{\mathrm{X}}_3-{1.58\mathrm{X}}_2{\mathrm{X}}_4+{0.0910\mathrm{X}}_3{\mathrm{X}}_4+{2.15\mathrm{X}}_1+{6.65\mathrm{X}}_2-{2.00\mathrm{X}}_3+{0.7997\mathrm{X}}_4+80.28\hfill \end{array}$$ 1

The regression analysis of variance (ANOVA) results (Supplementary Table 2) demonstrate that the regression model is highly significant (P < 0.0001), confirming its reliability. The relative influence of each factor on the ER, determined by comparing F-values, followed the order: CaCl₂ concentration (B) > sodium alginate concentration (A) > chitosan concentration (C) > curing time (D). The significant interaction between factors was visualized using 3D response surface and contour plots generated by Design-Expert software (Fig. 2E, F). The ER exhibited a trend of initial increase followed by a decrease with rising factor levels. The elliptical contours in Fig. 2F indicate a significant interaction between the corresponding factors. Based on the model, the predicted optimal conditions for maximizing EE were 3.78% sodium alginate, 2.09% CaCl₂, 0.76% chitosan, and a curing time of 0.97 h, yielding a predicted ER of 88.01%. Experimental verification under these conditions resulted in an actual ER of 88.64% for the slow-release model, closely matching the prediction and validating the model’s accuracy for process optimization.

Construction of a quick-release model of L. reuteri A-1 microcapsules

The influence of key formulation variables on the in vitro release rate of L. reuteri A-1 microcapsules was investigated (Fig. 2G–J). The release rate exhibited a parabolic relationship with sodium alginate concentration, reaching a maximum of 67.07% at 2% (Fig. 2G). Similarly, a maximum release rate of 48.89% was observed at 0.6% chitosan concentration (Fig. 2H). In contrast, the release rate decreased monotonically with increasing CaCl₂ concentration, with the highest value (66.85%) observed at 2% (Fig. 2I). Curing time also showed an optimal point, yielding a maximum release of 65.19% at 1 hour (Fig. 2J). Based on these single-factor experiments, the central points for the RSM were established. A Box-Behnken design was employed, with the release rate (Y) in simulated pancreatic juice as the response variable. The data from Supplementary Table 3 were fitted to obtain a quadratic regression equation relating Y to the independent variables (X₁: sodium alginate, X₂: CaCl₂, X₃: chitosan, X₄: curing time).

$$\begin{array}{c}\mathrm{Y}={-6.41\mathrm{X}}_2^1-{12.76\mathrm{X}}_2^2-{11.13\mathrm{X}}_3^2-{12.12\mathrm{X}}_4^2+{4.84\mathrm{X}}_1{\mathrm{X}}_2-{0.9955\mathrm{X}}_1{\mathrm{X}}_3-{0.9549\mathrm{X}}_1{\mathrm{X}}_4\hfill \\ -{1.24\mathrm{X}}_2{\mathrm{X}}_3+{0.6622\mathrm{X}}_2{\mathrm{X}}_4+{0.4236\mathrm{X}}_3{\mathrm{X}}_4+{6.65\mathrm{X}}_1+{5.61\mathrm{X}}_2+{1.08\mathrm{X}}_3-{1.64\mathrm{X}}_4+66.77\hfill \end{array}$$ 2

Regression ANOVA indicated that the model was highly significant (P < 0.0001, Supplementary Table 4), demonstrating excellent fit and reliability. The relative influence of each factor on the release rate, determined by comparing F values, followed the order: sodium alginate concentration (A) > CaCl₂ concentration (B) > curing time (D) > chitosan concentration (C). Interaction effects between factors were analyzed using response surface and contour plots generated by Design-Expert software (Fig. 2K, L). The release rate exhibited a parabolic trend, initially increasing and then decreasing with rising factor levels. The gentle slope of the response surface in Fig. 2K and the elliptical contours in Fig. 2L indicate that the concentrations of sodium alginate and CaCl₂ had a substantial and interactive impact on the release rate. The model predicted an optimal release rate of 79.64% under the following conditions: 3% sodium alginate, 2.5% CaCl₂, 0.59% chitosan, and a 0.95 h curing time. Experimental verification using this formulation yielded a release rate of 80.3% for the quick-release model, a result closely aligned with the prediction, thereby validating the model’s efficacy for optimizing the microencapsulation process.

Characterization of two different models of L. reuteri A-1 microcapsules

As shown in Fig. 3A, the microcapsules produced by the two models were comparable in appearance, size, and microstructure. Measurement of 100 randomly selected microcapsules revealed average diameters of 1.17 ± 0.04 mm for the slow-release model and 1.12 ± 0.08 mm for the quick-release model, with no statistically significant difference between them (Fig. 3B). Transmission electron microscopy (TEM) was employed to further examine the morphology and internal structure. Representative images (Fig. 3C) show that the freeze-dried microcapsules possessed a smooth surface. A cross-sectional view (Fig. 3D) revealed a loose, porous internal structure resulting from freeze-drying dehydration, which provides the necessary space for probiotic encapsulation. These results demonstrate that the optimized microencapsulation process successfully embeds the probiotics within a protective matrix, which is expected to facilitate their transit through the gastrointestinal tract.

Fig. 3. Morphological and in vitro evaluation characteristics of L. reuteri A-1 microcapsules.

A Appearance and morphology of L. reuteri A-1 microcapsules; B Size of two different models of L. reuteri A-1 microcapsules; C Surface morphology of L. reuteri A-1 microcapsules observed by TEM; D Cross-sectional image of L. reuteri A-1 microcapsules observed by TEM; E Changes in strain release during the sustained gastrointestinal fluid tolerance process of two different models of L. reuteri A-1 microcapsules as well as freeze-dried powder.; F Changes in the number of live bacteria in two different models of L. reuteri A-1 microcapsules stored at 4 °C and room temperature (25 °C).

In vitro evaluation of L. reuteri A-1 microcapsules in two different models

Based on the initial findings of poor gastric juice tolerance in L. reuteri A-1, this study employed microcapsulation to enhance its environmental resistance. The results of the simulated gastrointestinal tolerance assay (Fig. 3E) demonstrated the protective effect of microencapsulation. When microcapsules with an initial load of 8.7 lg(CFU/mL) were subjected to sequential gastrointestinal fluids, the release rate was only 13.5% after 2 h in simulated gastric juice (SGJ), increasing to 34.6% after 4 h in simulated pancreatic juice (SPJ), and reaching 80.45% after 18 h in simulated colonic fluid. In contrast, unencapsulated freeze-dried powder showed a higher release (41.0%) in SGJ, indicating significant bacterial inactivation, which led to markedly lower subsequent release values (24.4% in colonic fluid) compared to microcapsules. These results confirm the superior protective capacity of microencapsulation over freeze-dried powder.

A direct comparison between the slow-release and quick-release models under continuous gastrointestinal simulation is also shown in Fig. 3E. While both models started with a similar bacterial load, the quick-release microcapsules exhibited a nearly 1-log higher release in SGJ and a 2-log higher release in SPJ compared to the slow-release model. However, their release profiles converged in the neutral colonic fluid after 18-24 h. This indicates that the quick-release model facilitates rapid and substantial bacterial release in the upper GI tract, potentially enabling faster colonization and probiotic effects, though this may also increase susceptibility to rapid clearance.

Storage stability at different temperatures (4 °C and 25 °C) over 42 days was evaluated as a critical practical indicator (Fig. 3F). The slow-release model showed a 1.9% decrease in viable cell count after 42 days of storage at 4 °C compared to the initial viable cell count (0 d), and a 2.4% decrease in viable cell count after 42 days of storage at 25 °C compared to the initial viable cell count (0 d). The quick-release model showed a 2.8% decrease in viable cell count after 42 days of storage at 4 °C compared to the initial viable cell count (0 d), and a 4.2% decrease in viable cell count after 42 days of storage at 25 °C compared to the initial viable cell count (0 d). These results indicate that both formulations, particularly the slow-release model, maintain high viability under common storage conditions, supporting their potential for widespread application.

In vivo evaluation of L. reuteri A-1 microcapsules in two different models

Following one week of acclimatization, inflammatory bowel disease was induced in 6-week-old male mice (all groups except the CK group) by administering 2.5% DSS. Disease activity was monitored daily using a standard disease activity index (DAI) score, where a higher score indicates more severe colitis. On the final day of the modeling period, the DAI scores across all treatment groups were ~3, indicating a moderate level of disease severity, with no significant differences observed between groups (P > 0.05, Fig. 4B). These results confirm the successful establishment of the DSS-induced colitis model.

Fig. 4. The effects of two different models of L. reuteri A-1 microcapsules were evaluated in the DSS-induced mouse enteritis model.

A DSS-induced enteritis model in mice and experimental design flow chart. B DAI score trend during DSS-induced enteritis model in mice. C Weight change trend of DSS-induced enteritis mice during different treatments. D Comparative data bar graph of colon in mice under different treatments. E Comparative appearance of the colon of mice under different treatments. F Pathological observation of colon tissues of mice in different treatment groups; determination of serum inflammatory factors (IL-10, G; TNF-α, H; IL-1β, I) in mice under different treatment. J Determination of short-chain fatty acids in the intestine of mice in different treatment groups. K The number of L. reuteri A-1 colonization in different locations of the digestive system of mice in different treatment groups.

Effects of L. reuteri A-1 microcapsules in different models on body weight and colon length

Body weight was monitored weekly throughout the study (Fig. 4C). After one week of acclimatization, all groups showed comparable initial weights (P > 0.05). Following DSS induction, mice in the DSS, DSS + S (slow-release microcapsules), DSS + Q (quick-release microcapsules), DSS + P (probiotics), and DSS + B (prebiotics) groups exhibited a marked reduction in body weight, consistent with IBD symptoms, which differed significantly (P < 0.05) from the steady weight gain in the CK group. Subsequent intervention with probiotics or microcapsules (DSS + S, DSS + Q, DSS + P) promoted weight recovery, whereas the DSS + B group, receiving only prebiotics, showed limited improvement, underscoring the necessity of viable probiotics.

DSS-induced colitis typically results in colon shortening. As shown in Fig. 4D, E, all probiotic-treated groups (DSS + P, DSS + Q, DSS + S) showed significant restoration of colon length compared to the DSS group, approaching that of the CK group. Specifically, colon length increased by (46 ± 0.46)% in the DSS + P group, (42.8 ± 0.70)% in the DSS + Q group, and (31.7 ± 0.47)% in the DSS + S group (all P < 0.01). The DSS and DSS + B groups showed no such improvement. These results indicate that the probiotic formulation was more effective than either microencapsulation model in alleviating colitis-associated colon shortening.

Effects of different models of L. reuteri A-1 microcapsules on colon histopathology

Histopathological analysis of colon tissue was performed to evaluate the protective effects of various interventions against DSS-induced inflammatory damage (Fig. 4F). Compared to the CK group, which exhibited intact mucosal architecture, clear crypts, and an absence of inflammatory cell infiltration, the DSS group showed severe pathological changes, including tissue disruption, marked edema, and multiple epithelial ulcers. The DSS + B (prebiotic) group displayed similar but somewhat less severe damage, with reduced ulcer area and neutrophil infiltration, possibly due to prebiotic-mediated promotion of anti-inflammatory bacteria. In contrast, treatment with probiotics (DSS + P) or probiotic microcapsules (DSS + Q, DSS + S) resulted in notably attenuated pathology, characterized by relatively intact mucosa, significantly reduced edema and ulceration, and diminished inflammatory cell infiltration. These results demonstrate that both free and microencapsulated probiotics can ameliorate DSS-induced colitis, with the quick-release microcapsule formulation providing a protective effect superior to prebiotic intervention alone.

Effects of different models of L. reuteri A-1 microcapsules on inflammatory factors

The impact of DSS-induced colitis on systemic immunity was assessed by measuring serum cytokine levels (Fig. 4G–I). DSS challenge significantly reduced the anti-inflammatory cytokine IL-10 while markedly up-regulating the pro-inflammatory cytokines TNF-α and IL-1β, effects that were only marginally mitigated by prebiotics alone (DSS + B group). In contrast, all probiotic interventions (DSS + P, DSS + Q, DSS + S) significantly reversed these changes, normalizing the cytokine profile toward that of the healthy CK group. Specifically, these treatments elevated IL-10 levels (by 2.73-, 2.31-, and 1.51-fold, respectively) and substantially reduced TNF-α (by 80.56%, 74.7%, and 65.39%) and IL-1β (by 55.34%, 56.8%, and 45.09%) compared to the DSS group. The free probiotic (DSS + P) and quick-release microcapsule (DSS + Q) groups generally exhibited slightly superior efficacy in modulating these inflammatory factors compared to the slow-release microcapsules (DSS + S).

Effects of different models of L. reuteri A-1 microcapsules on SCFA

The concentrations of major short-chain fatty acids (SCFAs)—acetic, propionic, and butyric acid—were quantified (Fig. 4J). Acetic acid was the most abundant SCFA detected. Its concentration was highest in the DSS + B group (0.54 mg/mL), likely due to the direct provision of prebiotic substrates for bacterial fermentation, and lowest in the DSS group (0.25 mg/mL). The acetic acid content in the DSS + Q group exceeded that of the DSS + S and DSS + P groups. Although present in much lower quantities, the levels of propionic and butyric acid were also marginally higher in the DSS + Q group compared to the DSS + S group. These results suggest that the quick-release microcapsules may be more conducive to enriching SCFA-producing bacteria, leading to a slightly enhanced production of these beneficial metabolites.

Effects of different models of L. reuteri A-1 microcapsules on colonization

In order to explore the effects of different treatments on the L. reuteri A-1 colonization in various digestive system sites, qPCR method was used for characterization. As shown in Fig. 4K, the number of L. reuteri A-1 in various parts of the mouse digestive tract in DSS group was lower than that of other groups. In the small intestine and colon, the number of L. reuteri A-1 in the DSS + P, DSS + S, and DSS + Q groups was significantly higher than that in the CK group. The oral administration group (DSS + P) had more probiotics in the stomach but fewer in the small intestine and colon than the DSS + Q and DSS + S groups. Both microcapsule models can successfully pass through the extremely harsh environment of the stomach compared to the direct oral administration of probiotics. Moreover, the quick-release group (DSS + Q) more effectively colonized the small intestine and colon than the slow-release group (DSS + S), with a 1.3-fold higher probiotic count in the small intestine. It can be reasonably speculated that the quick-release model can more effectively alleviate DSS-induced enteritis.

Effects of different models of L. reuteri A-1 microcapsules on gut microbiota

As depicted in Fig. 5A, 58 OTUs were shared between the experimental and control groups (CK). Specifically, 9 OTUs were unique to the DSS-induced enteritis model group, 110 to the CK group, and 39, 57, 61, and 96 to groups B (DSS + B), P (DSS + P), Q (DSS + Q), and S (DSS + S), respectively. The coverage index, which assesses the completeness of community species representation based on sequencing depth, demonstrated high reliability of the sequencing data, with all sample values exceeding 0.998 (Fig. 5B).

Fig. 5. Effects of two different models on the intestinal microbiota structure and predicted function in mice with DSS-induced enteritis.

A OTU Venn diagram of intestinal microbiota of mice in different treatment groups. B Gut microbiome coverage of mice in different treatment groups. C Comparison of Shannon diversity index of the gut microbiome in mice in different treatment groups. D Comparison of Chao 1 diversity index of the gut microbiome in mice in different treatment groups. E Column chart of the community structure of gut microbiome of mice in different treatment groups at the genus level. F PCoA distribution of gut microbiome of mice in different treatment groups at the OTU level. G Typical species distribution of gut microbiome of mice in different treatment groups. H Comparison of the predicted functions of gut microbiome in mice in different treatment groups.

The Shannon index, an indicator of microbial diversity, was significantly lower in the DSS model group, indicating a disrupted gut microbiota structure (Fig. 5C). In contrast, the probiotic-treated groups (B, P, Q, and S) exhibited significantly higher Shannon indices than the DSS group (P < 0.05), approaching levels observed in the CK group. This suggests that probiotic microcapsule intervention effectively restored microbial diversity and improved community structure. Although not statistically significant (P > 0.05), the quick-release group (Q) showed a marginally higher Shannon index than the slow-release group (S). Similarly, the Chao index, reflecting species richness, was significantly reduced by DSS treatment (Fig. 5D). Probiotic and prebiotic supplementation (groups B, P, Q, and S) restored species richness to levels comparable to the CK group, with significant improvements relative to the DSS group (P < 0.05). A slight increase in the Chao index was also noted in the quick-release model compared to the slow-release model, though this difference was not significant (P > 0.05). Collectively, the Shannon and Chao index analyses confirm that probiotic treatments effectively mitigated DSS-induced reductions in microbial diversity and richness. Moreover, the quick-release model demonstrated a marginally superior effect on microbiota restoration compared to the slow-release model.

The analysis of gut microbiota in mice with DSS-induced enteritis at the genus level (Fig. 5E) showed that the genus levels of gut microbiota in the six treatment groups mainly included norank_f__Muribaculaceae, Bacteroides, Prevotellaceae, Ligilactobacillus, Parabacteroides, Escherichia-Shigella, Lachnospiraceae_NK4A136_group, Klebsiella, Lactobacillus, Akkermansia, Dubosiella, Blautia, etc. Compared with the DSS group, the abundance of Turicimonas and Escherichia-Shigella decreased in the CK group, while the abundance of norank_f__Muribaculaceae, unclassified_f__Lachnospiracea, and norank_f__Prevotellaceae increased. Compared with the DSS group, the abundance of Turicimonas and Escherichia-Shigella decreased in DSS + B, DSS + P, DSS + Q, and DSS + S, while the abundance of norank_f__Muribaculaceae, norank_f__Prevotellaceae, and Allobaculum increased, which was closer to the community composition of the CK group. In addition, the probiotic group significantly increased the abundance of unclassified_f__Lachnospiracea, which is a potential probiotic genus that participates in the metabolism of many carbohydrates, ferments to produce acetic acid and butyric acid, and provides the main energy source for the host. Among them, the abundance of Lactobacillus in the DSS + S and DSS + Q groups was higher than that in the DSS group, while the abundance of Lactobacillus in the DSS + P group was lower, but the abundance of Akkermansia was significantly higher than that in other groups, and the abundance of Lachnospiraceae_NK4A136_group in the DSS + B group was significantly increased. Prevotellaceae is a beneficial bacterium that can produce SCFAs. In summary, after mice were infected with DSS enteritis, the intestinal microbiota was disturbed, mainly manifested by the increase in the relative abundance of harmful microbiota and the decrease in the relative abundance of beneficial microbiota. The administration of probiotics and probiotic microcapsules can regulate the abnormal changes in the intestinal microbiota structure, making its intestinal microbiota structure more similar to that of the CK group. In terms of diversity, the Shannon index of the DSS + Q group was significantly higher than that of the DSS + S group, further confirming the advantage of the quick-release strategy in promoting the complexity of the microbial structure. The relative abundance of L. reuteri A-1 in the DSS + S group was higher, which may be related to the local colonization advantage caused by its slow-release characteristics, but the enrichment of this single genus did not translate into a wider microbial diversity or a stronger inhibitory effect on pathogens.

As shown in Fig. 5H, functional annotation via FAPROTAX revealed that pathways for “Human pathogens gastroenteritis” and “Fumarate respiration” were significantly elevated in the DSS group, but markedly ameliorated in the DSS + Q, DSS + S, and DSS + P groups. These pathways are closely associated with pathogen enrichment: fumarate respiration promotes an intestinal hypoxic environment favorable to pathogens, nitrite respiration supports pathogen energy metabolism during chronic infection, and nitrite ammonification detoxifies nitrites produced during host inflammation. Probiotic supplementation thus suppressed pathogen proliferation to varying degrees.

In contrast to the DSS group, the CK group exhibited upregulation of “Fermentation” and “Aerobic chemoheterotrophy.” Enhanced fermentation activity is conducive to the enrichment of SCFA-producing bacteria, which help maintain intestinal barrier integrity and an anti-inflammatory environment. All treatment groups partially restored these metabolic pathways, with the DSS + Q group most closely resembling the CK group. Specifically, in the “Fermentation” pathway, recovery was strongest in DSS + Q, weaker in DSS + S, and least in DSS + P, consistent with measured SCFA levels in mouse feces. Regarding the “Human pathogens” pathway, both DSS + Q and DSS + S groups showed significant pathogen reduction, with a more pronounced effect in DSS + Q, indicating superior antibacterial efficacy of the quick-release formulation. Although sulfate respiration remained elevated in both DSS + B and DSS groups, the DSS + B group exhibited partial recovery of fermentation and moderate mitigation of inflammation.

Discussion

This study constructed two different models of L. reuteri A-1 microcapsules based on a single-factor experiment and response surface methodology. The appearance, storage capacity, in vitro effect and in vivo effect of the two models' microcapsules were evaluated. As is known, L. reuteri A-1 can inhibit pathogens and maintain the balance of gut microecology by producing antimicrobial substances (such as reuterin)²⁰, increase the production of SCFA in the intestine, and promote the health of intestinal epithelial cells²¹. It can also enhance immune function by activating macrophages and increasing IgA secretion²². The bacterium can also regulate cytokine expression, reduce inflammatory responses, and has potential therapeutic effects on inflammatory bowel disease²³. However, L. reuteri A-1 has a very poor ability to resist gastric acid, which greatly limits its wide application and effectiveness. Therefore, it is necessary to develop targeted delivery and controlled release systems to improve these problems.

Targeted delivery refers to the precise delivery of drugs to target tissues, cells or organs through specific carriers or technologies, thereby increasing the concentration of drugs at the target site while reducing exposure to non-target tissues and side effects²⁴. The ability of targeted delivery systems to achieve precise drug delivery relies on targeting ligands (such as antibodies, peptides, nucleic acid aptamers) or responsive materials (such as pH-sensitive, temperature-sensitive materials)²⁵. Controlled-release systems refer to the use of specific materials or technologies to control the release rate, time and location of drugs, so that the drug concentration accumulates in the body, prolongs the drug’s duration of action, and reduces the frequency of administration²⁶. Controlled-release systems can achieve controlled release of drugs by adjusting material properties (such as degradation rate, diffusion properties) or external stimuli (such as pH, temperature, light)²⁷. The combination of targeted delivery and controlled release systems refers to the precise delivery of drugs to the target site through targeted technology and the sustained release of drugs through controlled release technology. This combination can further improve the therapeutic effect of drugs and reduce side effects²⁸.

Here, we chose microencapsulation technology to solve the problem of poor tolerance and colonization ability of L. reuteri A-1. The study found that from the single-factor results, when the sodium alginate concentration is too low, the cross-linked structure formed with CaCl₂ is loose and the embedding rate is low. After the optimal sodium alginate concentration, the concentration will continue to increase, and the solution will be too viscous, which is not conducive to the movement of Ca²⁺ and the combination of sodium alginate. The membrane structure is loose, and the density is poor, and the embedding rate is reduced²⁹. The presence of chitosan can improve the stability and tolerance of sodium alginate microcapsules. The concentration of chitosan can affect the thickness of the film. The higher the concentration, the tighter the membrane structure. The presence of too much chitosan increases the formation of alginate-chitosan bonds, which in turn causes the porosity of the alginate-CaCl₂ wall to decrease with the increase of chitosan concentration³⁰. This result is similar to the research results of Xiang Zhenghao et al.³¹. When the concentration of the cross-linking agent is too high, a large amount of Ca²⁺ will compete for binding sites, and excess Ca²⁺ will adhere to the capsule surface, further hindering the binding of chitosan and sodium alginate³². Therefore, the spatial network structure formed is loose, and the embedding rate decreases. When the curing time is too short, the cross-linking reaction has not been completed, the cross-linking structure has not been fully formed, the cross-linking structure is loose, and the embedding rate is low. When the curing reaction is complete, increasing the curing time will not have much effect on the cross-linking structure³³.

After intervention with L. reuteri A-1 microcapsules, mice with DSS-induced enteritis showed different degrees of weight and colon recovery, increased levels of anti-inflammatory factors, increased levels of pro-inflammatory factors, and intact intestinal tissue mucosal cells. This research result is consistent with the results of Ahl et al.³⁴. The gut microbiome structure of mice after oral administration of L. reuteri A-1 gradually recovered and tended to be similar to that of normal mice, and inter-group difference analysis confirmed that oral administration of L. reuteri A-1 changed the structure of the gut microbiome. In functional prediction, it was found that the L. reuteri A-1 mainly restored the Fermentation pathway related to the synthesis of SCFA. The DSS + Q group had a higher effect than the DSS + S group in the recovery of disease indicators caused by enteritis and the composition of the mouse gut microbiome, confirming that the rapid-release microcapsules were better than the slow-release microcapsules for DSS-induced inflammatory bowel disease.

The development of quick-release microcapsules for L. reuteri represents a significant departure from conventional slow-release formulations, offering a novel therapeutic strategy predicated on immediate probiotic action rather than sustained methods. While slow-release systems aim to enhance survival through the gastrointestinal (GI) tract, their gradual payload release may fail to provide the critical mass of viable cells necessary for time-sensitive interventions, such as acute diarrhea or the rapid modulation of the gut-brain axis. In contrast, quick-release capsules, engineered with pH-sensitive or enzyme-triggered polymers, ensure the immediate liberation of a high concentration of viable L. reuteri in the proximal small intestine, a key site for host-pathogen competition and immunomodulation. This approach is particularly innovative for exploiting the bacterium’s well-documented capacity for rapid niche competition and pathogen exclusion through mechanisms like bacteriocin production and reinforcement of tight junctions. The clinical implications are profound; such formulations could be superior for acute conditions where speed of action is paramount. For instance, in preventing Clostridioides difficile infection onset post-antibiotic therapy, a rapid probiotic “bolus” could immediately occupy ecological niches and produce antimicrobial compounds, effectively outcompeting pathogens before they establish a foothold³⁵. Furthermore, this technology aligns with the emerging paradigm of “next-generation probiotics,” where specific microbial functions, rather than general presence, are the therapeutic target. By guaranteeing the targeted and timely delivery of a functional payload, quick-release microcapsules for L. reuteri enhance the precision and efficacy of probiotic therapy, opening new avenues for treating acute gastrointestinal and even systemic disorders.

However, the clinical translation of quick-release L. reuteri microcapsules still faces several challenges. Scalability remains a hurdle, as laboratory-scale extrusion methods may not be cost-effective or easily transferable to industrial-scale production while maintaining capsule uniformity and probiotic viability. Long-term stability is another critical concern; ensuring high viability after months of storage under various temperature and humidity conditions requires further optimization of cryoprotectants and capsule matrices. Furthermore, the regulatory landscape for such advanced live biotherapeutic products (LBPs) is evolving but lacks clear, standardized guidelines defining what constitutes a “sufficient” release profile or stability endpoint, potentially delaying approval pathways. Finally, a comprehensive assessment of the ecological impact following rapid, high-dose delivery is needed to ensure it does not disrupt the resident microbiota.

This study constructed quick and slow-release microcapsules for L. reuteri A-1 using an alginate-chitosan matrix. In vitro assays confirmed the enhanced gastrointestinal tolerance and stability of both formulations. Crucially, in a DSS-induced colitis mouse model, the rapid-release microcapsules demonstrated superior efficacy in alleviating colonic inflammation and restoring gut microbiome homeostasis. This work has established a new technical system for controlled release and targeted probiotics, which has greatly promoted the development of LBPs.

Methods

Experimental strains, growth curve and pH determination

The L. reuteri A-1 used in this study was previously screened by our laboratory. L. reuteri A-1 was inoculated into MRS liquid medium for resuscitation. The composition and preparation method of the basic MRS medium (1 L) are as follows: 2.0 g of dipotassium hydrogen phosphate, 2.0 g of triammonium citrate, 5.0 g of anhydrous sodium acetate, 0.25 g of manganese sulfate heptahydrate, 0.58 g of magnesium sulfate heptahydrate, 20.0 g of glucose, 10.0 g of peptone, 10.0 g of beef extract, 5.0 g of yeast extract powder, and 1 mL of Tween 80. The volume is adjusted to 1 L, and the pH is adjusted to 6.2–6.4. When preparing the solid medium, 20 g of agar needs to be added and sterilized at 115 °C for 30 min.

The initial pH value of the medium was adjusted to 6.2. After washing with sterile water, it was inoculated into MRS liquid medium at a 5% (volume ratio) inoculum using aseptic operation and cultured in a 37 °C constant temperature incubator for 48 h. The corresponding blank medium was used as a control. The absorbance of the bacterial solution was measured at 600 nm with an ultraviolet spectrophotometer every 6 h, and the pH of the bacterial solution was also measured.

Preparation and tolerance test of artificial gastric juice, pancreatic juice and small intestinal juice

Artificial gastric juice, pancreatic juice, and small intestinal juice were prepared according to Part 4 of the 2020 edition of the Chinese Pharmacopoeia and sterilized with a 0.2 μm sterile microporous filter membrane before use³⁶. 1 mL of bacterial solution was mixed with 9 mL of artificial gastric juice with a pH of 1.3 and 9 mL of artificial intestinal juice with a pH of 6.8, respectively, and then placed in a 37 °C constant temperature incubator and shaken at 180 rpm for 2 h and 4 h. 1 mL of sample was taken at 0 h, 2 h and 4 h, respectively, and the samples were diluted, spread and cultured for 24 h. The counts were then calculated to calculate the survival rate of the strains³⁷.

$$\mathrm{Survival\; rate\; in\; artificial\; gastric\; fluid}/\%=\frac{N}{N_0}\times 100$$ 3

N: The number of viable bacteria after 2 h of artificial gastric juice treatment.

N₀: The number of viable bacteria after artificial gastric juice treatment at 0 h.

$$\mathit{Survival}\mathit{rate}\mathit{in}\mathit{artificial}\mathit{pancreatic}\mathit{juice}/\%=\frac{N_2}{N_0}\times 100$$ 4

N₂: The number of viable bacteria after 4 h of artificial pancreatic juice treatment.

N₀: The number of viable bacteria after artificial pancreatic juice treatment at 0 h.

Optimization of fermentation medium for L. reuteri A-1

Optimization of carbon source

Using MRS medium as the basic medium, L. reuteri A-1 used sucrose and maltose to replace the carbon source (glucose) in the basic medium, respectively. The added amount was the same as the carbon content (molar mass) in the original medium, and the other components remained unchanged. The total culture time was 48 hours, and OD₆₀₀ and pH were measured every 3 hours.

Optimization of nitrogen source

After the carbon source in the MRS medium was replaced with the optimal carbon source, the nitrogen source (peptone) in the basal medium was replaced with casein peptone, tryptone, soy peptone, and yeast extract powder, respectively, and the amount added was the same as the nitrogen content (molar mass) in the original medium, and the other components remained unchanged. The total culture time was 48 hours, and the OD₆₀₀ and pH were measured every 3 hours.

Optimization of carbon-nitrogen ratio

After replacing the carbon and nitrogen sources in the MRS medium with the optimal carbon and nitrogen sources, the medium was prepared according to the carbon-nitrogen ratios of 5:1, 10:1, 20:1, and 30:1, respectively, and the other components remained unchanged. The total culture time was 48 h, and the OD₆₀₀ and pH were measured every 3 h.

Preparation, embedding rate and release rate determination of L. reuteri A-1 microcapsules

The initial probiotic microcapsule preparation process formula was based on the method of Ma Hangyu et al.³⁷, with slight adjustments. After fermenting the probiotics in MRS liquid medium for 24 h, the mixture was centrifuged at 7106 g for 10 min, the cells were collected and fully diluted with an equal volume of sterile water. The diluted cells were then fully stirred with 2% (mass volume ratio) sodium alginate and lyophilization protective agent in a volume ratio of 2: 2: 1, wherein the composition of the lyophilization protective agent was 12% (mass volume ratio) skim milk, 10% (mass volume ratio) glycerol, 13% (mass volume ratio) sucrose, 4% (mass volume ratio) mannitol, and 0.8% (mass volume ratio) tyrosine. Chitosan dissolves at low pH and does not dissolve at high pH³⁸. In order to completely dissolve chitosan, 1% (mass volume ratio) glacial acetic acid was added when preparing a 0.8% (mass volume ratio) chitosan solution³⁹. CaCl₂ solution with a concentration of 3% (mass volume ratio) was prepared to prepare a 1:1 CaCl₂-chitosan solidification mixed solution. The bacterial mud-sodium alginate-protectant mixture was uniformly dripped into the solidified mixed solution through a 1 mL tubular syringe (flat-head needle) to form wet microcapsules, and then allowed to solidify for 1.5 h. The wet microcapsules were fully washed 3–5 times with sterilized distilled water to wash off the excess bacterial solution, solidifying solution and chitosan coating on the surface of the wet capsules. Then, the microcapsules were pre-frozen overnight in a -80 °C refrigerator and freeze-dried in a freeze dryer for 48 h to obtain probiotic microcapsule particles.

Embedding rate determination: Microcapsules with sodium alginate as the wall material can achieve rapid disintegration in sodium citrate solution⁴⁰. Take 1 mL of the pre-freeze-dried microcapsules made from bacterial mud and lyse them in 9 mL of 3% sodium citrate solution. The solution was placed in a 37 °C water bath for 30 min, then oscillated on a vortex oscillator for 5 s, and cycled 2–3 times in sequence until complete lysis. The lysate was taken and diluted appropriately for the plate live bacterial count. The calculation formula is as follows:

$$\mathit{Embedding}\mathit{rate}/\%=\frac{N}{M}\times 100$$ 5

N is the number of viable bacteria embedded in the microcapsules before freeze-drying, in CFU/mL.

M is the number of viable bacteria in the initial sludge, in CFU/mL.

Release rate determination: The release amount after 6 h of continuous gastrointestinal fluid treatment was taken as the release rate. 1 mL of freeze-dried microcapsules prepared from bacterial sludge was placed in 9 mL of simulated artificial gastric juice and artificial pancreatic juice, respectively. After 4 h of incubation at 180 rpm in a constant temperature incubator at 37 °C, samples were taken, and viable bacteria counts were performed on the dilution plate after appropriate dilution. The calculation formula is as follows:

$$\mathit{Release}\mathit{rate}/\%=\frac{Y}{M}\times 100$$ 6

Y is the number of viable bacteria released from microcapsules in pancreatic juice after 4 h of treatment, in CFU/mL.

M is the number of viable bacteria in the initial bacterial sludge, in CFU/mL.

Construction and optimization of two different models of L. reuteri A-1 microcapsules

The single-factor experimental method of preparing L. reuteri A-1 microcapsules (slow-release model) with embedding rate as an indicator is as follows: (1) Sodium alginate concentration gradient experiment The chitosan concentration (0.8%), CaCl₂ concentration (3%) and curing time (0.5 h) were fixed, and the effect of the microcapsules prepared at five gradients of sodium alginate concentration from 1% to 5% on the embedding rate was investigated. (2) CaCl₂ concentration gradient experiment. The sodium alginate concentration (2%), chitosan concentration (0.8%) and curing time (0.5 h) were fixed, and the effect of the microcapsules prepared at five gradients of CaCl₂ concentration from 1% to 5% on the embedding rate was investigated. (3) Chitosan concentration gradient experiment. The sodium alginate concentration (2%), CaCl₂ concentration (3%) and curing time (0.5 h) were fixed, and the effect of the microcapsules prepared at five gradients of chitosan concentration from 0.4% to 1.2% on the embedding rate was investigated. (4) Curing time gradient experiment. The concentrations of sodium alginate (2%), chitosan (0.8%) and CaCl₂ (3%) were fixed, and the effects of the microcapsules prepared at five curing time points (0.5–2.5 h) on the encapsulation efficiency were investigated.

Slow-release model: It was found that when the sodium alginate concentration was 3%, the CaCl₂ concentration was 2%, the chitosan concentration was 0.6%, and the curing time was 1 h, it could meet the optimal encapsulation condition of L. reuteri A-1 microcapsules. The Box-Behnken experimental design method was used to design 4 factors and 3 levels, and a regression equation was established to fit the relationship between each factor and the response value of microcapsule encapsulation rate. The response surface analysis was performed with Design-Expert 8.0.6 software (Stat-Ease, USA) to screen the optimal preparation parameters for microcapsule encapsulation rate. The experimental design factors and levels are shown in Table 1.

Table 1.

Box-Behnken embedding rate experiment parameter levels

Factors	Levels
	−1	0	1
X1 Sodium alginate concentration (%)	2	3	4
X2 CaCl2 concentration (%)	1	2	3
X3 Chitosan concentration (%)	0.4	0.6	0.8
X4 Curing time (h)	0.5	1	1.5

Quick-release model: Single-factor test of the release rate of L. reuteri A-1 microcapsules: The release rate of the microcapsules treated for 6 h in the continuous gastrointestinal fluid experiment was used as an indicator. The single-factor experimental method for the release rate of L. reuteri A-1 microcapsules was the same as the above method. According to the results of the single-factor test on release rate, when the concentration of sodium alginate was 2%, the concentration of CaCl₂ was 2%, the concentration of chitosan was 0.6%, and the curing time was 1 h, the Box-Benhnken experimental design method was used to design 4 factors and 3 levels, and a regression equation was established to fit the relationship between each factor and the response value of the microcapsule release rate. The response surface analysis was performed using the Design-Expert 8.0.6 software (Stat-Ease, USA) to screen out the optimal preparation parameters for the microcapsule release rate. The experimental design factors and levels are shown in Table 2.

Table 2.

Box-Behnken release rate experiment parameter levels

Factors	Levels
	−1	0	1
X1 Sodium alginate concentration (%)	1	2	3
X2 CaCl2 concentration (%)	1	2	3
X3 Chitosan concentration (%)	0.4	0.6	0.8
X4 Curing time (h)	0.5	1	1.5

Characterization of two different models of L. reuteri A-1 microcapsules

The particle size of 20 randomly selected two different models of L. reuteri A-1 microcapsules was measured with a vernier caliper, and the average value was taken to obtain the particle size distribution of the L. reuteri A-1 microcapsules. In order to further observe the surface characteristics and internal structure of the microcapsules, the microcapsules were characterized using a TEM (Hitachi HT7700, Japan).

In vitro evaluation of two different models of L. reuteri A-1 microcapsules

The continuous gastrointestinal fluid tolerance test of microcapsules was performed with reference to Ren Guangyu’s method with slight modifications⁴¹. Take 1 mL of bacterial mud made of two different models of L. reuteri A-1 microcapsules and freeze-dried powder (control) and add them to 9 mL of gastric juice. After fully shaking and mixing, place them in a 37 °C shaker and culture them at 180 rpm in for 2 hours. The mixed solution is aspirated every 1 hour. Collect the microcapsules treated with gastric juice and add them to 9 mL of pancreatic juice. Incubate them for 4 hours in the same way and aspirate the mixed solution every 2 hours. Collect the microcapsules treated with pancreatic juice and add them to 9 mL of colonic juice. Incubate them for 18 hours in the same way and aspirate the mixed solution every 3 hours. The mixed solution aspirated above is diluted and spread on a plate and cultured at 37 °C for 24 hours to count the viable bacteria. Storage stability test of L. reuteri A-1 microcapsules: The freeze-dried microcapsules were placed in centrifuge tubes and stored at 4 °C and 25 °C for 42 days. Every 7 days, 1 mL of the freeze-dried microcapsule particles made of bacterial mud was taken out and added to 9 mL of 3% sodium citrate solution for disintegration and viable bacterial count determination.

Experimental animal selection and design

All mice (C57BL/6) were selected from 6-week-old male mice weighing 20 ± 2 g and free of specific pathogens. Experimental animal ethics approval number: GENINK 20240026 (Tianjin Jinke Bona Biotechnology Co., LTD). All procedures conformed to the guidelines of the NIH Guide for the Care and Use of Laboratory Animals. All animals were raised in a specific pathogen-free environment with a temperature of 22 ± 3 °C, a humidity of 50 ± 15%, and a 12-h light–dark cycle. All mice were given a normal diet during the experiment. All mice were divided into 6 groups (CK, DSS, DSS + S, DSS + Q, DSS + P, and DSS + B), with six mice in each group. The specific grouping and animal experimental process are shown in Fig. 4A. During the adaptation period, all groups were given normal drinking water and feed. Starting from the second week, the CK group was given normal drinking water and feed, while the other groups were given 2.5% DSS (Dextran Sulfate Sodium) drinking water to construct a mouse model of enteritis. During the modeling period, the physical condition of the mice was observed and recorded every day, including the presence of bloody stools and weight loss. The comprehensive score was used to represent the DAI⁴². The presence of bloody stools and consistent DAI scores indicates that the modeling is complete. After the modeling is successful, the DSS group is given normal drinking water and a diet. The DSS + S group was given slow-release model microcapsules every day through the gavage method (10⁸ CFU/mL). The DSS + Q group was given rapid-release model microcapsules every day through the gavage method (10⁸ CFU/mL). The DSS + P group was directly given L. reuteri A-1 in free form daily by gavage (10⁸ CFU/mL). DSS + B group was given empty capsules without L. reuteri A-1 every day through gavage. During the entire experimental period (35 d), mice in different treatment groups were experimentally intervened on every day.

At the end of each week, the weight changes of the mice were recorded, and the feces of the mice were collected, stored in −80 °C freezer. At the end of the animal experiment, blood was collected from the eyeball. The supernatant was centrifuged at 4 °C, 1006.2 g rpm in for 20 minutes and stored in a −80 °C refrigerator. The stomach, small intestine, colon and their contents of the mouse were obtained by dissection, and the length of the mouse colon tissue (from the anus to the end of the cecum) was measured with a ruler, photographed and recorded. Then 1 cm of colon tissue was taken, placed in a sterile centrifuge tube, fixed with 4% paraformaldehyde solution and stored in a −80 °C refrigerator.

Determination of inflammatory factors and SCFA

ELISA kits were used to quantify the levels of inflammatory factors IL-10, IL-1β, and TNF-α in mouse serum, and the specific operation steps were strictly carried out in accordance with the kit instructions (Shanghai Enzyme-linked Biotechnology Co., Ltd., Shanghai, China). Take 20 mg of mouse feces and thaw it on ice, then resuspend it in 600 μL of 1% sulfuric acid solution, shake and mix for 2 min, and then centrifuge it at 7200 × g at 4 °C for 15 min. Transfer 500 μL of supernatant to a 2 mL sterile centrifuge tube. An appropriate amount of SCFA standard was weighed and mixed with pure water to prepare a 100 mg/mL stock solution. The stock solution was then diluted with pure water to obtain working standard solutions of different concentration gradients. The sample preparation and detection method were based on Hu Wenbing et al.⁴³. The standard solutions of different concentrations and the in vitro fermentation supernatant were acidified with sulfuric acid and mixed with 1 mL of anhydrous ether. After centrifugation, the supernatant was transferred to a centrifuge tube containing anhydrous sodium sulfate and allowed to stand to fully remove water. After centrifugation, the filter membrane was transferred to an injection vial and separated using an Agilent DB-WAX capillary column (30 m × 0.25 mm ID × 0.25 μm) gas chromatography system (GC9790, Fuli Instruments Co., Ltd., Zhejiang, China) with an inlet temperature of 220 °C, a column temperature of 160 °C, and a detector temperature of 220 °C for a total of 10 min.

Detection of L. reuteri A-1 colonization in different parts of the mouse digestive system

The DNA kit (DP712-01, Tiangen Biochemical Technology Co., Ltd., Beijing, China) was used to extract the DNA, and the specific primers of L. reuteri A-1 were used to amplify the target fragment. The specific primer sequences of the 16S rRNA of L. reuteri A-1 strain are: Lac-F: GAGTACGACCGCAAGGTTGA; Lac-R: CATGCACCACCTGTCATTGC; the amplified product was subjected to 1% (m/v) agarose gel electrophoresis detection, and the PCR product was purified by gel cutting. The purified product was used to connect and transform the target fragment using a ligation transformation kit and pick positive clones. The cultured bacteria of the above positive clones were collected, and plasmid DNA was extracted. After obtaining the standard plasmid DNA template, the corresponding copy number was calculated according to the concentration, and the calculation formula was:

$$\mathit{copy}=\frac{6.02\times {10}^{23}\times C\times {10}^{-9}}{\mathit{DNA\; lengt}\text{h}\times 660}$$ 7

C represents the concentration of plasmid DNA, the unit is ng/μL

DNA length represents the full length of plasmid DNA.

The total volume of the qPCR system used in the experiment was 20 μL, and the specific components were as follows: 2×ChamQ Universal SYBR qPCR Master Mix 10 μL, Forward Primers 1 μL, Reverse Primers 1 μL, Template DNA 1 μL, and then ddH₂O was added to 20 μL. The following qPCR program was used: denaturation at 95 °C 30 s, annealing to 60 °C, then 95 °C 15 s, 60 °C 60 s, and finally 95 °C 15 s, for 40 cycles. The 16S rRNA gene absolute quantitative standard curve of L. reuteri A-1 was constructed based on the copy number of the standard plasmids with different concentration gradients and the corresponding CT values. Equal amounts of the stomach, small intestine, and colon contents obtained by dissection were taken, and DNA was extracted using a DNA extraction kit. After the concentration was determined using Nanodrop 2000 (nd2000, Thermo Fisher Scientific, Waltham, USA), qPCR absolute quantification was performed using specific primers of the target strain.

HE staining

The fixed colon tissue was processed through the steps of tissue embedding, sectioning, paraffin section dewaxing, HE staining, dehydration and sealing, and then the colon tissue pathology was observed.

16S rRNA amplicons high-throughput sequencing analysis

The fecal samples at the end of the experiment, stored at −80 °C, were sent to Shanghai Major Biotechnology Co., Ltd. for 16S rRNA amplicon sequencing and cloud platform analysis with the variable V3-V4 region of 16S rRNA by the Illumina MiSeq sequencing platform. In this study, the bacterial universal primer sequences were F515 5’-GTG CCA GCM GCC GCG GTA A-3’ and R806 5’-GGA CTA CHV GGG TWT CTA AT-3’⁴⁴.

Statistical analysis

In this study, all study groups were conducted in parallel, with at least 3 independent samples in each group. All data were expressed as mean ± SD. Statistical significance of multiple comparisons in each study was determined by one-way ANOVA. If the ANOVA results showed significant differences (P < 0.05), Tukey’s post hoc test was used for pairwise comparison between groups. P value < 0.05 was defined as statistically significant. (When P < 0.05 (*), P < 0.01 (**), and P < 0.01 (***), it was considered statistically significant). All statistical analyses were performed using IBM SPSS Statistics (Version 26).

Author contributions

S.X.: investigation, conceptualization, formal analysis, writing–original draft, writing–review & editing. R.H.: formal analysis, data curation. J.W.: formal analysis. X.Z.: data curation. Z.Z.: resources. Z.H.: conceptualization, funding acquisition, supervision, writing–review & editing.

Data availability

No datasets were generated or analysed during the current study.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41538-025-00677-4.