The Protective Effects and Underlying Mechanisms of Taraxacum kok-saghyz Polysaccharides Against Intestinal Dysbiosis-Induced Mastitis Were Elucidated Using a Murine Model of the “Gut–Mammary” Axis

TL;DR

This study shows that polysaccharides from Taraxacum kok-saghyz leaves can reduce mastitis in mice by improving gut health and reducing inflammation.

Contribution

The novel finding is that TKP-L polysaccharides can target the gut–mammary axis to treat dysbiosis-induced mastitis.

Findings

TKP-L reduced mammary inflammation and improved barrier integrity in gut and mammary tissues.

TKP-L inhibited bacterial translocation and restored gut and milk microbiota balance.

Beneficial bacteria like Limosilactobacillus increased with TKP-L treatment.

Abstract

Mastitis in dairy cows is a costly inflammatory disease that adversely affects milk production and animal health. Although commonly associated with bacterial infections, emerging evidence indicates that dysbiosis of gut microbiota may contribute to mastitis via the “gut–milk axis,” a pathway through which intestinal disturbances influence the mammary gland. This study investigated whether polysaccharides derived from Taraxacum kok-saghyz leaves (TKP-L), a natural compound, could mitigate mastitis by targeting this gut–mammary connection. Using a murine model, we demonstrated that TKP-L significantly reduced mammary inflammation, enhanced barrier integrity in both intestinal and mammary tissues, and inhibited systemic translocation of pathogenic bacteria. Furthermore, TKP-L restored microbial homeostasis in the gut and milk by promoting beneficial genera such as Limosilactobacillus. These findings suggest that TKP-L represents a safe and environmentally sustainable strategy for preventing and treating mastitis through modulation of gut health, with potential implications for advancing sustainable dairy farming practices.

The gut–mammary axis represents a promising therapeutic target for mastitis. Although plant-derived polysaccharides exhibit immunomodulatory properties, their capacity to modulate this axis—and specifically to ameliorate dysbiosis-induced mastitis—remains unexplored. Here, we investigated the therapeutic potential of Taraxacum kok-saghyz leaf-derived polysaccharides (TKP-L) against mastitis in a murine model of gut dysbiosis, with dysbiosis induced by fecal microbiota transplantation (FMT) from donor cows. Pregnant mice (n = 60) with antibiotic-depleted microbiota received FMT suspensions prepared from the feces of healthy dairy cows or cows with clinical mastitis (based on somatic cell count). Mice were randomly divided into five groups: Control (vehicle), M-FMT (mastitis-cow FMT, disease model), H-FMT (healthy-cow FMT), TKP-L (M-FMT + oral TKP-L, 500 mg/kg/day), and Ciprofloxacin (M-FMT + ciprofloxacin, positive Control). After FMT establishment, TKP-L or ciprofloxacin was administered for 14 days. We assessed histopathology, pro-inflammatory mediators (IL-6, IL-1β, TNF-α, MPO), tight junction proteins (occludin, ZO-1, Claudin-3), and bacterial translocation using GFP-E. coli, and gut/milk microbiota via 16S rRNA sequencing. Compared to the M-FMT group, TKP-L treatment significantly alleviated mammary inflammation and pathology, inhibited pro-inflammatory cytokine expression, and enhanced the expression of tight junction proteins in both intestinal and mammary tissues, correlating with reduced bacterial translocation to the mammary gland. Microbiota analysis showed that TKP-L restored microbial homeostasis in the gut and milk, concurrently increasing the relative abundance of beneficial bacteria such as Limosilactobacillus. TKP-L alleviates gut dysbiosis-induced mastitis in mice by concurrently modulating the gut–mammary axis through microbial remodeling, enhancement of epithelial barriers, and anti-inflammatory actions. These findings highlight TKP-L as a promising gut microbiota-targeting candidate for mastitis intervention.

Figures

Full text

1. Introduction

Mastitis in dairy cows remains one of the major factors limiting the development of dairy farming and the dairy product industry [1], resulting in substantial economic losses. Traditionally, bovine mastitis has been attributed to pathogenic microorganisms such as Staphylococcus aureus, Escherichia coli, and Streptococcus dysgalactiae [2], as well as physical trauma or improper milking practices. Recent evidence further suggests that disturbances in the gut microbiota represent a key contributing factor in the development of mastitis. The idea that intestinal microbes may travel to the mammary gland via internal pathways—termed the “gut–mammary axis” [3]—has gained significant scientific attention in recent years. Under normal physiological conditions, the intestinal mucosal layer and epithelial tight junctions maintain barrier integrity and prevent pathogenic bacteria from disseminating to peripheral organs [4]. When dysbiosis occurs, excessive bacterial and endotoxin production impairs intestinal barrier integrity [5]. These bacterial components may then translocate through the bloodstream or lymphatic system to the mammary gland, compromise mammary epithelial integrity, and trigger mastitis [6]. Key tight junction proteins, including ZO-1, Claudin-3, and Occludin, play essential roles in maintaining intestinal barrier function. Damage to these proteins results in a “leaky gut,” facilitating microbial pr endotoxin entry into the mammary gland and promoting mastitis [7].

Once pathogenic bacteria such as Escherichia-Shigella, Streptococcus, and Desulfovibrio colonize mammary tissue, they trigger a cascade of pathological responses. In particular, lipopolysaccharide (LPS) produced by Gram-negative bacteria like E. coli triggers inflammation primarily through activation of the Toll-like receptor 4 (TLR4) pathway in mammary epithelial cells. This leads to NF-κB activation [8] and the elevated release of pro-inflammatory cytokines such as IL-1β, IL-6, TNF-α, and MPO [9]. TNF-α and IL-1β further amplify the inflammatory response via NF-κB-mediated cascades, increasing vascular permeability and degrading tight junction proteins such as ZO-1, Claudin-3, and Occludin, thereby exacerbating mammary gland injury [10]. Concurrently, the overgrowth of Desulfovibrio is often accompanied by reductions in beneficial bacteria such as Akkermansia and Bacteroides, which further aggravates gut dysbiosis and delays intestinal barrier repair [11]. Therefore, based on the “gut–mammary axis” theory, it is essential to develop novel feed additives that can regulate gut microbiota composition, strengthen intestinal barrier function, and attenuate systemic inflammation, which holds significant value for the prevention and treatment of mastitis.

Russian dandelion, scientifically named Taraxacum kok-saghyz (TKS), is a member of the Taraxacum genus in the Asteraceae family. It has recently attracted considerable industrial interest due to the substantial quantity and superior quality of the natural rubber in its roots [12]. Simultaneously, its pronounced anti-inflammatory properties highlight potential therapeutic applications [12]. Polysaccharides are among its principal bioactive constituents and are widely recognized for their ability to modulate immune responses and mitigate inflammatory processes [13]. The polysaccharides contained in significant amounts within dandelion stems and leaves [14] have been found to alleviate inflammation. This is achieved by downregulating key pro-inflammatory cytokines (TNF-α, IL-1β, and IL-6), while also suppressing the activation of both the NF-κB-p65 pathway and the NLRP3 inflammasome [15]. Previous research demonstrated that dandelion polysaccharides can reduce the relative abundance of pathogenic gut bacteria, including Prevotellaceae, Helicobacteraceae, and Oscillospiraceae, thereby mitigating the inflammation associated with microbial dysbiosis [16]. The overgrowth of harmful microbes is known to compromise intestinal barrier integrity [17]. Dandelion polysaccharides also promote intestinal barrier repair by restoring paracellular permeability in Caco-2 cell monolayers and upregulating Claudin-5 expression [18], consequently reducing the risk of endogenous infections. Given these anti-inflammatory, microbiota-modulating, and gut barrier-enhancing properties, dandelion polysaccharides represent a promising therapeutic candidate for mastitis-associated gut microbiota dysbiosis. Taraxacum kok-saghyz, a closely related species of medicinal dandelion and an important strategic resource for natural rubber production, shares high genetic similarity and comparable bioactive components with traditional medicinal dandelions [19]. Accordingly, we hypothesize that polysaccharides extracted from the stems and leaves of Taraxacum kok-saghyz (TKP-L) can effectively ameliorate mastitis by modulating the gut–mammary axis, reducing pro-inflammatory cytokine expression, enhancing intestinal barrier integrity, and reshaping the gut microbiota.

Therefore, we propose the following hypothesis: The polysaccharides extracted from the stems and leaves of Toraakamukazza grass (TKP-L) can effectively improve mastitis by regulating the gut–mammary axis, reducing the expression of pro-inflammatory cytokines, enhancing the integrity of the intestinal barrier, and reshaping the intestinal microbiota. To verify this hypothesis, this study aims to explore the protective effect of TKP-L on mastitis caused by intestinal flora imbalance through a mouse model via the gut–mammary axis, and to investigate its potential mechanism.

2. Materials and Methods

2.1. Detection of Polysaccharide Content, Monosaccharide Composition, and Infrared Spectroscopy

The aldose and total sugar content of TKP-L was determined by employing two methods: m-hydroxybenzidine and phenol-sulfuric acid, respectively. Then we distinguished the monosaccharide composition of TKP-L through liquid chromatography. For infrared spectroscopy, TKP-L (1 mg) was mixed with dry 98% potassium bromide and then ground into a fine powder. It was pressed into a tablet and scanned by an infrared spectrometer to detect the functional groups of TKP-L.

2.2. Collection of Donor Cow Feces and Microbiota Sequencing

According to the standardized criteria, six healthy dairy cows and six dairy cows clinically diagnosed with mastitis were selected. The criteria included: (i) whether they had typical clinical symptoms of mastitis (such as swelling, fever, pain, and abnormal milk secretion), and (ii) the somatic cell count (SCC) in their milk. For the cows with mastitis, this was ≥500,000 cells/mL (Table A1), while the value for the healthy Control cows was <500,000 cells/mL (Table A2). Fresh fecal samples of approximately 50 mL from each animal were collected using sterile gloves for sterile collection, immediately mixed with 30% (volume/volume) glycerol as a cryoprotectant, and rapidly frozen in liquid nitrogen within 5 min for storage. The samples were transported to the laboratory in liquid nitrogen and processed within 2–3 h: each sample was homogenized under sterile conditions, aliquoted into sterile cryovials, and stored at −80 °C until DNA extraction. Genomic DNA was extracted from the fecal samples using the QIAamp PowerFecal Pro DNA Kit (Qiagen, Hilden, North Rhine-Westphalia, Germany), quantified using the NanoDrop One spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), and the purity (A260/A280 ratio 1.8–2.0) and integrity were evaluated by 1% agarose gel electrophoresis. Primers with barcodes (341F/805R) were used to amplify the V3–V4 hypervariable region of the bacterial 16S rRNA gene by polymerase chain reaction (PCR); the amplification products were purified, standardized, and mixed for Illumina MiSeq sequencing. The sequencing data were analyzed by by Meiji Biotechnology Co., Ltd. (Shanghai, China) using QIIME2 and SILVA v138 reference databases for classification identification, α/β diversity analysis, and differential abundance analysis.

2.3. Animal Experiment

The sample size for this study was determined through a pre-set power analysis to ensure sufficient statistical power. Using the widely adopted G*Power 3.1 software, a one-way analysis of variance (ANOVA) fixed-effect test was selected [20]. The parameter settings were as follows: significance level (α) was 0.05, expected power (1 − β) was 0.80, and the number of groups (k) was 5. The effect size was set at Cohen’s f = 0.25, which, according to conventional standards, represents a moderate effect. This effect size (f) is defined as the standardized difference between the group means, including the expected TNF-α levels [21] (the main outcome) between groups and the within-group variability based on our pre-experiment data and previous studies on mouse mammary inflammation. The analysis indicated that at least 12 animals were required per group. Therefore, a total of 60 female KM mice were used.

Depletion of Gut Microbiota: After one week of acclimation, sixty female and twenty male Kunming (KM) mice (7–9 weeks old) were mated at a 3:1 female-to-male ratio. Successful mating was determined by vaginal plug detection, after which the females were isolated. To achieve intestinal microbiota depletion, pregnant mice were then treated via daily oral gavage for 7 days with antibiotics such as metronidazole (200 mg/kg), ampicillin (200 mg/kg), and neomycin sulfate (200 mg/kg) and vancomycin (100 mg/kg).

FMT Modeling Phase: Feces from cows with mastitis and healthy cows were dissolved in sterile PBS at a concentration of 50 mg/mL, and centrifuged at 4 °C and 1000× g for 3 min. The supernatant was used as FMT donor material.

Following a 24 h acclimatization period, 60 microbiota-depleted pregnant mice were randomly assigned to five experimental groups (n = 12 per group):

• Group 1: Control (no FMT),
• Group 2: Mastitis cow feces FMT (M-FMT),
• Group 3: Healthy cow feces FMT (H-FMT),
• Group 4: TKP-L treatment (M-FMT + 500 mg/kg TKP-L)
• Group 5: Ciprofloxacin treatment (M-FMT + 5 mg/kg CIP).

Model establishment and drug treatment stage: All mice except those in the Control group were administered 50 mg/mL of healthy or mastitis fecal suspension by gavage for 3 consecutive days, followed by once every other day until delivery. After delivery, 500 mg/kg TKP-L and 5 mg/kg ciprofloxacin were administered orally for 14 days.

Sample collection: Mice were anesthetized with ether, blood was collected from the orbital sinus, and the mice were sacrificed by cervical dislocation. The left mammary gland and part of the colon and liver were placed in 4% paraformaldehyde for fixation to prepare pathological sections. The right mammary gland and part of the colon were weighed and quickly frozen in liquid nitrogen and stored at −80 °C for Q-PCR detection. Intestinal contents and milk were collected for 16S microbiota sequencing.

2.4. Safety Detection

The blood samples were processed using a benchtop high-speed refrigerated centrifuge (Da Long, model D3024R), which was set at 2–8 °C and spun at 3000 rpm for 15 min. Subsequently, the levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), urea nitrogen (BUN), and creatinine (CREA) were rapidly measured from the separated serum using an automatic biochemical analyzer (Leju Life Science Technology, model Chemray 800, Shenzhen, China). The safety detection range for mice was as follows: ALT: 10.06–96.47 U/L, AST: 36.31–235.48 U/L, BUN: 10.81–34.74 mg/dL, CREA: 10.91–85.09 μmol/L.

2.5. HE Staining and Pathological Scoring

Sections of colon and mammary gland tissues measuring two to three centimeters were removed and washed with physiological saline. The samples underwent paraffin embedding and sectioning after having been soaked in 4% paraformaldehyde (Saiwei Technology Co., Ltd. Urumqi, China). Using an optical microscope, tissue morphology was investigated. The previously reported approach served as the basis for the pathological assessment of mammary gland inflammation (Table 1) [22].

2.6. Tight Junction Immunofluorescence

To evaluate the expression and distribution of tight junction proteins (ZO-1, Occludin, Claudin-3) in breast and colon tissues, we performed immunofluorescence staining on paraffin sections. After dewaxing, hydration, and antigen retrieval, the sections were incubated at room temperature for 1 h with PBS containing 10% normal serum. Subsequently, the sections were incubated with the following primary antibodies overnight at 4 °C: anti-ZO-1 (CN catalog number: 21773-1-AP, company: Proteintech, Wuhan, China), anti-Occludin (catalog number: 27260-1-AP, company: Proteintech, Wuhan, China), or anti-Claudin-3 (CN 16456-1-AP, Proteintech, Wuhan, China). After PBS washing, the sections were incubated with the corresponding fluorescent secondary antibodies at room temperature in the dark for 1 h. After another wash, the sections were sealed with mounting medium containing DAPI. All images were captured using a fluorescence microscope under the same parameters. The average fluorescence intensity of the specific protein fluorescence channel was quantitatively analyzed using the ImageJ software (Version:1.54p).

2.7. Bacterial Migration Frozen Section

A plasmid-producing green fluorescent protein was used to convert the Escherichia coli BL21 (DE3) strain, which was then cultivated in LB medium with 50 μg/mL kanamycin. Mice in the Control, M-FMT, H-FMT, TKP-L, and CIP groups received 10^9^ CFU of BL21(DE3) orally on the thirteenth postpartum day. Mice were put to sleep after 24 h, and their mammary glands were detached and preserved in 4% formaldehyde. After that, these issues were kept at −80 °C for 48 h after being implanted in the OCT embedding agent.

2.8. RT-qPCR Detection of Mammary Gland and Intestine

The colon and mammary gland tissues of mice that had been previously frozen in liquid nitrogen were subjected to RT-PCR to detect gene transcription in the tissues. Total RNA was extracted from the mammary glands and intestines of mice using TRIzol (Fujigene Biotechnology Co., Ltd, Urumqi, China), and the RNA content was determined by spectrophotometry. RT-qPCR was performed using 2× RT OR-Easy TM mixture and Real Time PCR Easy TM (Fujigene Biotechnology Co., Ltd.) premix. It is important to add the internal reference gene, GAPDH. This study included the genes of IL-1β, IL-6, TNF-α, MPO, ZO-1, Occludin and Claudin-3 (Table 2).

2.9. 16s RNA Gut Microbiota Sequencing

On the 14th day postpartum, the mice were euthanized, and then intestinal fecal contents and milk samples were collected and rapidly frozen in liquid nitrogen. Genomic DNA was extracted from each sample and quantified using a nanodrop spectrophotometer. The bacterial 16S rRNA gene was amplified by PCR, followed by purification and fluorescence-based quantification. Sequencing libraries were prepared, and high-throughput sequencing was performed. The analysis of the intestinal and milk microbiota-related indicators and the False Discovery Rate (FDR) correction using the Benjamini–Hochberg method were all completed by Majorbio Bio-pharm Technology Co., Ltd (Shanghai, China).

2.10. Statistical Analysis

All data are presented as the mean ± standard deviation (SD). Prior to statistical analysis, the normality of each dataset was assessed. The majority of the data satisfied the normality assumption. For the few datasets that did not, an appropriate transformation [e.g., y′ = ln(y)] was applied to approximate a normal distribution before proceeding with parametric tests; such cases are explicitly noted in their respective figure legends. Inter-group comparisons were performed using one-way analysis of variance (ANOVA). When a significant overall effect was identified by ANOVA, pairwise comparisons were conducted using Tukey’s honestly significant difference (HSD) post hoc test. In the figures, bars labeled with different lowercase letters (e.g., a, b, c) indicate statistically significant differences between groups (p < 0.05). All statistical analyses were performed using GraphPad Prism software (version 10.0).

3. Results

3.1. Monosaccharide of TKP-L

The monosaccharide composition of TKP-L is summarized in Figure 1. TKP-L was primarily composed of galacturonic acid (Gal-UA), which accounted for 74.51% of the total monosaccharide. The remaining components included Arabinose (Ara, 9.38%), galactose (Gal, 7.02%), rhamnose (Rha, 5.34%), xylose (Xyl, 1.21%), glucuronic acid (Glc-UA, 1.05%), fucose (Fuc, 0.91%), and glucose (Glc, 0.58%). Our results indicate that TKP-L is a galacturonic acid-rich acidic polysaccharide.

3.2. Polysaccharides Content and FTIR Spectroscopy Analysis of TKP-L

Figure 2A shows the polysaccharides extracted from the stems and leaves of Taraxacum kok-saghyz. The structural features of TKP-L were further analyzed using Fourier Transform Infrared Spectroscopy (FTIR), as presented in Figure 2B. A broad and strong absorption peak was observed at 3317.52 cm^−1^ in TKP-L, which has stretching vibration for O-H bonds within and between molecules. A weak absorption peak appeared at 2932.89 cm^−1^ in TKP-L, characteristic of the asymmetric stretching vibration of C-H bonds (mainly -CH_2_-) on the sugar ring. A typical pair of absorption peaks for carboxylate ions appeared at 1638.92 cm^−1^ and 1415.50 cm^−1^, providing direct structural evidence that TKP-L is an acidic polysaccharide containing uronic acids. The results indicate that TKP-L possesses typical polysaccharide structural characteristics and is rich in uronic acid components.

Quantitative analysis (Figure 2C) showed that the total polysaccharide content of TKP-L was 31.37%, while the uronic acid content reached 68.63%, further supporting the FTIR results that TKP-L is predominantly an acidic polysaccharide.

3.3. Safety Evaluation of Polysaccharides from Taraxacum kok-saghyz Leaves

As shown in Figure 3, panel A illustrates the liver histopathological sections from the five groups of mice. Following oral administration of TKP-L, the overall liver architecture remained intact, and hepatocytes retained normal morphology, with no observable signs of cell necrosis, apoptosis, or inflammatory cell infiltration.

Consistent with histopathological findings, the serum biochemical markers depicted in Figure 3B–E. It included different physiological ranges such as alanine aminotransferase (ALT), blood urea nitrogen (BUN), creatinine (CREA), and aspartate aminotransferase (AST). These results collectively demonstrate that TKP-L does not induce hepatic or renal toxicity under the experimental conditions.

3.4. Sequencing Results of Donor Cow Fecal Microbiota

Figure 4A shows the schematic diagram of FMT modeling. PCoA analysis (Figure 4E) revealed a clear separation between the M and H groups, indicating that the overall intestinal microbiota composition of donor cows with mastitis (M) differed markedly from that of healthy donor cows (H). To further explore these differences, the top 10 bacteria that underwent specific changes at the phylum level were initially determined. The predominant bacterial phyla in bovine intestinal microbiota were Firmicutes (Bacillota), Bacteroidetes (Bacteroidota), and Spirochaetes (Spirochaetota) (Figure 4C), whereas the dominant genera included “UCG-005, Rikenellaceae RC9, and norank f_UCG-010” (Figure 4D). Figure 4E–J further show the trends of microbiota variation in both donor cows and FMT recipient mice. The results indicate that the microbial communities from donor cows successfully colonized the recipient mice. At the phylum level, the mastitis group (M) exhibited a significantly reduced relative abundance of Pseudomonodota (Figure 4E) and Thermodesulfobacteria (Figure 4F) relative to the healthy group (H) (p < 0.05). Correspondingly, at the genus level, the abundance of Escherichia-Shigella was markedly elevated in the M group compared to the H group (p < 0.05; Figure 4G). Following FMT intervention, the gut microbiota profiles of the recipient mice closely mirrored those of their corresponding donors. As shown in Figure 4H–J, mice in the healthy transplantation group (H-FMT) displayed a significantly lower relative abundance of Pseudomonodota (Figure 4H), Thermodesulfobacteria (Figure 4I), and Escherichia-Shigella (Figure 4J) compared with mice receiving microbiota from mastitis cows (M-FMT) (p < 0.05). These results show sustained colonization of donor-derived populations and effective microbial translocation.

3.5. Therapeutic Effects of Polysaccharides from Taraxacum kok-saghyz Leaves

The tissues of the Control and H-FMT groups exhibited normal morphology, with pale pink coloration and intact structural integrity (Figure 5A). Histopathological observations of the mammary gland (Figure 5B) and intestinal tissues (Figure 5C) also revealed complete and undamaged tissue architecture. In contrast, mice in the M-FMT group displayed pronounced redness and swelling of the tissues, disrupted alveolar structures, and visible damage to the intestinal mucosa. Following treatment with TKP-L or CIP, the mammary tissue of mice showed near-normal morphology, pale-red coloration, well-preserved alveolar structures, and significantly reduced inflammatory cell infiltration. Tissue damage scores (Figure 5D,E) further supported these findings. The M-FMT group exhibited significantly elevated injury scores in both mammary and intestinal tissues (p < 0.05). However, TKP-L and CIP treatment markedly reduced tissue damage scores compared with the M-FMT group (p < 0.05). These results indicate that TKP-L effectively attenuates pathological injury induced by FMT from mastitis cows. Furthermore, the expression of TNF-α, MPO, IL-1β, and IL-6 was significantly higher in the M-FMT group (Figure 5G–I) compared to the Control group. By contrast, both TKP-L and CIP treatments significantly downregulated the expression of these cytokines as compared to the M-FMT group.

3.6. “Intestine–Milk” Axis Microbial Migration

The expression of GFP signals in the mammary glands of mice in different treatment groups was detected using frozen sections and immunofluorescence. As shown in Figure 6A,B, only weak fluorescence signals were observed in the Control and H-FMT groups, indicating minimal bacterial translocation. In contrast, the M-FMT group exhibited strong GFP fluorescence, demonstrating substantial accumulation of fluorescent E. coli in the mammary tissue. Mice treated with TKP-L and CIP displayed only low levels of GFP-labeled bacteria, suggesting that both treatments effectively inhibited microbial migration.

Fluorescence intensity quantification (Figure 6B) further validated the observations. A significant rise in fluorescent E. coli was detected in the M-FMT group, in contrast to the H-FMT group, which showed no statistical difference. Both TKP-L and CIP interventions, however, significantly reduced fluorescent E. coli counts versus the M-FMT group (p < 0.05), restoring them to levels indistinguishable from the Control.

3.7. Taraxacum kok-saghyz Leaves Polysaccharide Restores the Integrity of the Mammary Gland Barrier

Immunofluorescence staining results (Figure 7A,B) showed that the tight junction proteins ZO-1, Claudin-3 and Occludin exhibited continuous and intact membrane-associated distribution in the Control and the H-FMT groups. However, the M-FMT group showed disrupted, discontinuous localisation of these proteins, characterized by diffuse, weakened signals, indicating severe impairment of the mammary epithelial barrier. TKP-L and CIP treatment effectively restored the continuity and structural integrity of tight junction proteins ZO-1, Claudin-3, and Occludin. Quantitative analysis (Figure 7B–D) indicated that fluorescence intensity for these proteins was downregulated signaificantly in the M-FMT group as compared to Control group. In contrast, both the TKP-L and CIP groups exhibited a non-significant increase in protein expression levels compared to the M-FMT group.

As shown in Figure 7E–G, M-FMT group exhibited a significant downregulation such as ZO-1, Claudin-3, and Occludin. Treatment with TKP-L reversed this effect, upregulating the expression of these tight junction genes to levels significantly higher than those in M-FMT group (p < 0.05). Therefore, these results reinforce that TKP-L effectively alleviates mammary gland barrier impairment.

3.8. Taraxacum kok-saghyz Leaves Polysaccharide Restores Intestinal Barrier Integrity

Immunofluorescence staining and quantitative analysis of ZO-1, Claudin-3, and Occludin in the colon are shown in Figure 8A. In the Control and H-FMT groups, all three tight junction proteins exhibited continuous, well-defined membrane localisation, indicating intact intestinal epithelial barrier function. In contrast, the M-FMT group exhibited weak, discontinuous, and irregular fluorescence signals, reflecting significant disruption of the intestinal barrier structure. Quantitative fluorescence assessment (Figure 8B–D) indicated significantly higher expression intensities of ZO-1, Claudin-3, and Occludin in the Control group compared to the M-FMT group (p < 0.05). Treatment with TKP-L and CIP effectively enhanced the continuity and intensity of these proteins. In particular, the TKP-L group showed significantly elevated fluorescence versus the M-FMT group, confirming partial restoration of intestinal barrier integrity. The mRNA levels of proteins, i.e., ZO-1, Claudin-3, and Occludin, in colon tissue were significantly elevated in the TKP-L group (Figure 8E–G). Finally, it is important to know that neither the H-FMT versus Control comparison nor the TKP-L versus CIP comparison was significant.

3.9. Taraxacum kok-saghyz Leaves Polysaccharide Regulates Intestinal Flora Disorders

Compared with the Control group, the M-FMT group showed decreased Sobs, Shannon, and Ace indices along with increased Simpson index, indicating reduced microbial richness and diversity. Following TKP-L intervention, the Sobs and Shannon indices increased, and the Simpson index decreased, demonstrating that TKP-L intervention effectively alleviated the flora imbalance induced by M-FMT treatment. The α-diversity indices of the H-FMT group (Figure 9A) showed no significant difference from the Control group, and its flora richness and diversity remained within normal levels. PCoA analysis (Figure 9B) revealed clear separation between the M-FMT and Control groups, indicating significant alterations in the overall microbial community structure after M-FMT treatment. Notably, the TKP-L group clustered closer to the Control group and was significantly separated from the M-FMT group, suggesting that TKP-L treatment helped restore the disordered microbial structure toward a more balanced state. As shown in the Venn diagram (Figure 9C), 281 OTUs were shared among the five groups.

To further investigate TKP-L’s influence on microbiota structure, the ten phyla with the most pronounced abundance changes were evaluated. Firmicutes (Bacillota), Bacteroidetes (Bacteroidota), and Actinobacteria (Actinomycetota) were the dominant phyla in all groups (Figure 9D). The relative abundance of Firmicutes was significantly higher in the M-FMT, H-FMT, TKP-L, and CIP groups versus the Control (p < 0.05; Figure 9E). Notably, Thermodesulfobacteriota levels were significantly elevated in the M-FMT group relative to the H-FMT, TKP-L, and CIP groups (p < 0.01; Figure 9F).

At the genus level, the dominant taxa comprised Lactobacillus, Limosilactobacillus, and norank_f__Muribaculaceae (Figure 9G). Compared to the Control, the M-FMT group displayed a significant increase in the relative abundance of Desulfovibrio (p < 0.05), an effect that was substantially reversed by TKP-L and CIP treatment (p < 0.05; Figure 9I). Additionally, the abundance of Limosilactobacillus was significantly higher in the TKP-L group than in the M-FMT group (p < 0.05; Figure 9H), indicating a beneficial shift in gut microbiota composition.

3.10. Polysaccharides from Taraxacum kok-saghyz Leaves Regulate Dysbiosis of Microbiota in Mouse Milk

The microbial diversity of milk samples from each group is shown in Figure 10A. Compared with the Control group, the Ace index of the M-FMT group remained unchanged, while the Shannon, Sobs, and Simpson indices significantly increased. These findings indicate that although microbial richness increased in the M-FMT group, microbial diversity significantly decreased, and evenness slightly declined, suggesting an ecological imbalance caused by the abnormal proliferation of specific bacterial genera. This over-proliferation may increase species richness while simultaneously causing highly uneven species distribution. In the H-FMT group, the Simpson, Shannon, Ace, and Sobs indices did not differ significantly from those in the Control group, suggesting that the richness, diversity, and evenness of the milk microbiota were largely preserved. By contrast, both the TKP-L and CIP groups exhibited higher Simpson and Shannon indices relative to the M-FMT group, while the Ace index remained comparable, and the Sobs index displayed a decreasing tendency. The overall diversity profiles of the TKP-L and CIP groups shifted toward those of the Control group, suggesting that both treatments effectively restored the microbial richness and diversity. PCoA analysis (Figure 10B) showed clear separation between the M-FMT and Control groups, indicating a significant change in the overall intestinal microbiota of the M-FMT group. The TKP-L group was closer to the Control group and clearly separated from the M-FMT group, suggesting that TKP-L treatment helped restore the dysbiotic microbiota structure to a normal state. As shown in the Venn diagram (Figure 10C), the number of shared OTUs among the six groups was 76.

To further evaluate the impact of TKP-L on milk bacterial composition, the top 10 bacteria that showed specific phylum-level changes were initially identified. At the phylum level, results indicated that the bacterial flora in the milk of mice across all groups were mainly composed of Bacillota, Pseudomonadota, and Actinomycetota (Figure 10D). Compared with the M-FMT group, the relative abundance of Bacillota in the TKP-L and CIP groups both showed an upward trend. At the genus level, the intestinal bacterial flora in mice from all groups was mainly Lactobacillus, Limosilactobacillus, and Ligilactobacillus (Figure 10G). Compared with the M-FMT group, the relative abundance of Limosilactobacillus in the TKP-L group significantly increased (p < 0.05) (Figure 10F), and compared with the Control group, the abundance of Limosilactobacillus in the TKP-L group also significantly increased (p < 0.05) (Figure 10E). Based on the changes in the bacterial flora of mouse feces and milk, these results suggest that Limosilactobacillus may migrate from the intestine to the mammary gland through the “gut–milk axis”. Furthermore, the increased abundance of Limosilactobacillus in the intestine can protect the intestinal mucosa and prevent harmful bacteria such as Desulfovibrio from translocating into the mammary gland. However, this correlation does not constitute direct evidence of bacterial migration, which remains a limitation of the current study.

4. Discussion

Mastitis in dairy cattle is a complex, multifactorial inflammatory disease that causes major economic losses in livestock [23]. Traditionally, exogenous invasion of pathogenic microorganisms through the teat canal has been regarded as the primary etiological factor. However, the recently proposed “gut–milk” axis hypothesis offers a novel framework for understanding mastitis pathogenesis. Growing research indicates that disruption of the gut microbiota can endogenously trigger mastitis. This occurs mainly through impaired intestinal barrier function, which allows bacterial translocation into circulation. Subsequent migration to the mammary gland undermines the blood–milk barrier and promotes mastitis development [24]. Guided by this concept, the present study evaluated the therapeutic efficacy of Taraxacum kok-saghyz leaf Polysaccharide (TKP-L) in a murine mastitis model induced by gut microbiota imbalance through fecal microbiota transplantation (FMT), with TKP-L administered orally.

Polysaccharides represent major bioactive constituents of traditional Chinese herbal medicines and have been widely recognized for their anti-inflammatory, antioxidant, and immunomodulatory properties, as well as their capacity to reshape gut microbial communities [25]. Recent investigations have identified acidic polysaccharides as key anti-inflammatory agents in numerous medicinal plants. For instance, Song et al. demonstrated that acidic polysaccharides extracted from Lentinula edodes attenuate inflammation by suppressing the NF-κB and MAPK pathways and reducing TNF-α and IL-6 secretion [26]. In the current study, infrared spectroscopy analyses and aldaric acid quantification (Figure 2A–C) confirmed that TKP-L is an acidic polysaccharide, supporting the hypothesis that its anti-inflammatory effects are partly attributable to its acidic molecular characteristics. Our findings further confirmed TKP-L’s anti-inflammatory efficacy by modulating the “gut–milk” axis, highlighting its potential as a therapeutic agent for mastitis.

Taxa such as Pseudomonadota, Thermodesulfobacteriota, and Escherichia-Shigella successfully colonized the murine intestines (Figure 4E–J), closely mirroring the donor microbial composition, and confirmed the FMT-induced mastitis.

The “gut–mammary” connection showed therapeutic potential for induced mastitis via modulation of the intestinal microbiota. In our study, TKP-L effectively ameliorated mastitis-like symptoms. As shown in Figure 3 and Figure 4B, TKP-L treatment significantly reduced erythema, edema, inflammatory cell infiltration, and alveolar damage. The reductions in MPO, IL-1β, IL-6 and TNF-α levels with respect to mRNA expression, as shown in Figure 5F–I, further indicate potent anti-inflammatory activity. These findings demonstrate that TKP-L enhances tissue repair and exerts therapeutic effects against mastitis via systemic and microbiota-mediated mechanisms.

Additionally, the intestinal mammary epithelial cells have essential barriers for maintaining mammary gland health [27]. Both the mammary gland and intestinal barriers comprise ZO-1, Claudin-3, and Occludin tight junctions, produced by epithelial cells of the mammary gland and intestine. These proteins play a pivotal role in stopping the translocation of bacteria and endotoxin in systemic circulation [28].

Within the context of the “gut–mammary” axis, disruption of tight junction proteins in intestinal and mammary epithelia increases intestinal permeability, commonly referred to as “leaky gut,” which subsequently triggers inflammatory responses in the mammary gland [6]. In accordance with the “gut–mammary” axis theory, our findings showed that the M-FMT group significantly reduced tight junction protein expression in both intestinal and mammary tissues (Figure 7E–G and Figure 8E–G), accompanied by increased penetration of fluorescently labeled E. coli into the mammary gland (Figure 6). These changes indicate compromised barrier integrity and enhanced bacterial translocation. Contrastingly, TKP-L markedly restored the expression, continuity, and distribution of tight junction proteins in both tissues, suggesting that TKP-L effectively reinforces barrier integrity and limits hematogenous transfer of pathogenic bacteria to the mammary gland.

Modulation of the gut microbiota is another key mechanism through which TKPL exerts protective effects. 16S rRNA sequencing revealed reduced microbial diversity and significant dysbiosis in the M-FMT group, characterized by increased levels of the pathogenic genus Desulfovibrio and decreased abundance of the beneficial genus Limosilactobacillus (Figure 9D–G). Prior studies have demonstrated that Desulfovibrio impairs intestinal epithelial barrier function and activates intrinsic inflammatory pathways [29], whereas Limosilactobacillus and Lactobacillus species enhance mucosal immunity, inhibit pathogens through acid production, and strengthen epithelial barriers [30]. Moreover, TKP-L intervention significantly decreased Desulfovibrio abundance and increased Limosilactobacillus abundance, thereby restoring microbial homeostasis. We speculate that TKP-L can specifically promote the growth of Limosilactobacillus, and the mechanism may be related to its unique chemical structure. As an acidic polysaccharide rich in galacturonic acid, TKP-L can become the preferred fermentable substrate for Limosilactobacillus. This fermentation process may produce short-chain fatty acids (SCFAs), such as butyric acid and acetic acid, which have been proven to have the effects of enhancing the intestinal epithelial barrier, regulating immune responses, and inhibiting the growth of pathogenic bacteria. Therefore, the observed microbial changes are not merely correlations but may represent a causal relationship: TKP-L promotes the proliferation of beneficial fermenting bacteria, increasing the production of short-chain fatty acids, thereby enhancing the integrity of the barrier and the formation of an anti-inflammatory environment, which in turn inhibits mastitis. He et al. reported that Limosilactobacillus inhibits Staphylococcus aureus-induced mastitis by promoting oxytocin release [31], highlighting its relevance in mammary defense. Thus, TKP-L appears to exert multi-target therapeutic effects by repairing tissues and beneficially modulating the intestinal microbial ecosystem.

In the present study, alterations in milk microbiota further support TKP-L’s ability to restore mammary microecology. The M-FMT group exhibited reduced milk microbiota diversity, whereas TKP-L treatment increased Shannon and Pielou indices and normalized community composition. Notably, Limosilactobacillus abundance increased substantially in the TKP-L group, while Desulfovibrio declined in both intestinal and mammary ecosystems. Chauhan et al. reported that Limosilactobacillus exhibits strong antibacterial activity against major mastitis pathogens, including E. coli, S. aureus, and S. agalactiae [32], underscoring its therapeutic potential.

Analysis of both intestinal and mammary microbiota in murine models revealed that TKP-L intervention significantly increased the relative abundance of Limosilactobacillus in both compartments compared to the M-FMT and Control groups. In contrast, the relative abundance of Desulfovibrio was markedly decreased in both the intestinal and mammary tracts. Previous studies have indicated that intestinal Limosilactobacillus can enhance the thickness of the intestinal mucus layer by upregulating mucin gene expression (e.g., MUC2 and MUC13), thereby reinforcing the physical barrier, preventing pathogen invasion, and improving intestinal barrier function [33]. Based on these findings, we hypothesize that TKP-L enhances intestinal barrier function by elevating Limosilactobacillus levels, thereby restricting the entry of harmful bacteria into the mammary gland and mitigating mastitis through the “gut–milk” axis.

5. Conclusions

Our research results indicate that TKP-L alleviates mastitis caused by intestinal flora imbalance through its action on the gut–mammary axis. Mechanistically, TKP-L reshapes the intestinal flora by promoting the growth of beneficial bacteria (e.g., Limosilactobacillus), enhances intestinal barrier function, and thereby may mitigate the pathological impact associated with dysbiotic bacteria such as Desulfovibrio within the gut–mammary axis framework. Additionally, the increase in Lactococcus lactis numbers inhibits the release of pro-inflammatory cytokines in the mammary gland, limits tissue damage, and helps regulate the blood–milk barrier. These results not only confirm the key role of intestinal flora imbalance as an endogenous inducer of mastitis but also fully demonstrate the therapeutic potential of TKP-L in multi-target intervention of the “gut–mammary” axis. In conclusion, TKP-L is a natural active substance that can regulate intestinal microecology, effectively alleviate mammary inflammation in mice, and has good biological safety.