Transfer factor alleviates bovine mastitis and protects mammary epithelial barrier via the TAK1/NF-κB/MLCK signaling axis
Jinyou Zhang, Lingyu Xin, Aobo Zhang, Yaning Shi
TL;DR
Transfer factor, a natural immune booster, helps treat cow mastitis without antibiotics by reducing inflammation and protecting the mammary barrier.
Contribution
This study reveals that transfer factor alleviates bovine mastitis by inhibiting the TAK1/NF-κB/MLCK signaling pathway, offering a non-antibiotic treatment.
Findings
Transfer factor reduced pro-inflammatory cytokines IL-1β and IL-6 in mastitis-affected cows.
TF upregulated tight junction genes and inhibited MLCK activity in mammary epithelial cells.
In vivo, TF improved mastitis recovery with a 64.7% negative conversion rate compared to 13.3% in controls.
Abstract
Bovine mastitis is a prevalent and economically devastating disease in the global dairy industry. Antibiotic overuse leads to increased antimicrobial resistance and reduced milk quality, becoming major bottlenecks in clinical treatment. Transfer factor (TF), a safe, low-cost, and readily available immunomodulator that enhances cell-mediated immunity, has emerged as a promising antibiotic alternative. This study aimed to investigate TF’s alleviative effect on bovine mastitis and its underlying molecular mechanisms. We found that clinical mastitis cows had significantly higher mRNA levels of interleukin-1β (IL-1β) and interleukin-6 (IL-6) and markedly lower expression of tight junction (TJ) genes (nectin cell adhesion molecule 4 [NECTIN4], tight junction protein 1 [ZO-1], occludin) in mammary tissue and milk somatic cells compared to healthy controls. In vitro experiments, TF pretreatment…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14| Reaction grade | Tested milk appearance | Reaction state |
|
|---|---|---|---|
|
| Liquid | Flows smoothly without clots when tilting the test plate | 0–200 |
|
| Liquid with trace sediments | Trace sediments at the plate bottom, disappear upon shaking | 200–500 |
|
| Viscous sediment | Minimal viscous sediment at the plate bottom (not fully gelled), disperses with shaking | 500–800 |
|
| Gel-like consistency | Entire mixture forms a gel, rotates centripetally without dispersing | 800–5000 |
|
| Pronounced gelatinous sediment | Mixture forms a firm gel, adheres to the plate bottom, concentrates centrally upon rotation | >5000 |
| Gene | Sequence number | Primer sequence, 5′ to 3′ | Product size, bp |
|---|---|---|---|
|
|
F: CCTATTCTCTCCAGCCAACCT R: CTCATTCTCGTCACTGTAGTAAGC | 107 | |
|
|
F: TGTGAAAGCAGCAAGGAGACA R: CATCCGTCCTTTTCCTCCATT | 71 | |
|
|
F: GTGGCAGGTTGGCAGAGAAGG R: CGTCCAGCCGTGTCCAGTTG | 86 | |
|
|
F: GCCTGTGTTGCCTCCACTCTTG R: ACCGTAGCCATAGCCGTAGCC | 144 | |
|
|
F: CCGAATGAAACCGCACACAAACC R: GTCTCCACGCCACTGTCAAACTC | 107 | |
|
|
F: CCATCGGCAATGAGCGGTTCC R: CGTGTTGGCGTAGAGGTCCTTG | 146 |
| Pretreatment somatic cell count (103/mL) | Day 5 post-treatment somatic cell count (103/mL) |
| |
|---|---|---|---|
|
| 3769 ± 2144 | 1859 ± 1829 | 0.14 |
|
| 3680 ± 2388 | 355 ± 304 | <0.001 |
|
| 0.91 | 0.007 |
| Variables |
|
|
|
|---|---|---|---|
|
| 13 (86.7) | 2 (13.3) | 0.01 |
|
| 6 (35.3) | 11 (64.7) |
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMilk Quality and Mastitis in Dairy Cows · Barrier Structure and Function Studies · Microbial infections and disease research
Introduction
Bovine mastitis is universally recognized as one of the most prevalent and economically devastating diseases in the global dairy industry (Morales-Ubaldo et al. 2023). Its pathogenesis is multifactorial: pathogens ascend the teat canal, trigger the toll-like receptor 4 (TLR4)/TAK1/p65 signaling cascade, and elicit the release of pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and IL-6. These mediators disrupt the epithelial barrier assembled by TJ proteins (including occludin and ZO-1), precipitating apoptotic cell death and compromising barrier integrity (Wu et al. 2015). During lactation, the teat canal remains physiologically dilated for 1–2 h post-milking; this transient widening, compounded by environmental stressors, further attenuates barrier function. High-yielding animals, constrained by genetic trade-offs between productivity and immunity, exhibit heightened susceptibility to mastitis (Melvin et al. 2019).
Tight junctions (TJs) of bovine mammary epithelial cells constitute a primary component of the blood–milk barrier; their principal function is to preserve the structural integrity of the mammary gland and to prevent paracellular mixing of milk and blood constituents (Jiang et al. 2022). As the critical intercellular adhesive apparatus of epithelial monolayers, TJs are composed of the transmembrane protein families (occludin and claudins) and cytoplasmic scaffolding ZO proteins (Stelwagen and Singh 2014). These proteins anchor to the cytoskeleton and assemble into continuous belt-like strands that restrict the paracellular passage of water molecules, ions, and microorganisms (Liang and Weber 2014)—serving as the first physical barrier against external insults, whose integrity is essential for maintaining epithelial barrier function (Camilleri 2019). NECTIN4, an immunoglobulin superfamily adhesion molecule, contributes to inter-epithelial cohesion (Kedashiro et al. 2019). Its extracellular region comprises two Ig-like domains that engage in homophilic or heterophilic trans-interactions with other nectins, while its cytoplasmic tail couples to the actin cytoskeleton via afadin, thereby stabilizing cell–cell contacts and reinforcing the epithelial barrier (Harrison et al. 2012). In our previous transcriptomic study, NECTIN4 was found to be significantly down-regulated in mastitic mammary tissue compared with healthy tissue, suggesting that inflammatory challenge may impair intercellular adhesion or junction (Zhang et al. 2025a). Upon pathogen recognition, bovine mammary epithelial cells secrete potent chemokine signals that recruit immune effectors from the bloodstream into the infected gland and simultaneously up-regulate the synthesis of bactericidal molecules (Günther et al. 2009). Consequently, preserving the structural integrity of the mammary epithelial tight-junction barrier represents an indispensable therapeutic strategy for the prevention and control of bovine mastitis.
Transfer factor (TF) is a low-molecular-weight immunomodulatory complex composed of small peptides that confer the ability to transfer cell-mediated immunity from an immune donor to a naïve recipient (Kirkpatrick 2000). By conveying antigen-specific immunological information, TF orchestrates the T helper 1 (Th1)/Th2 balance; at the innate immune level, it activates macrophages and amplifies neutrophil chemotaxis (Orozco et al. 2004). Empirical studies have further demonstrated that TF up-regulates the TJ proteins claudin and occludin in the intestinal epithelium of laying hens (Ma et al. 2023). Endogenous TF exhibits physiological metabolic safety, is orally bioavailable without adverse effects, and lacks mutagenic potential (Polonini et al. 2021). Previous studies have reported that TF enhances natural killer (NK) cell activity and exerts immunomodulatory effects in a rat peritonitis model (Habar et al. 2021). In a murine asthma model, Imuno TF reduced peribronchial inflammatory cell infiltration, collagen deposition, and mucus secretion, thereby attenuating lung tissue injury (Oliveira et al. 2022). Nevertheless, whether TF can alleviate mammary inflammation and preserve tight-junction barrier integrity of mammary epithelial cells in dairy cows with mastitis remains entirely unexplored.
This study was therefore undertaken to systematically evaluate the capacity of TF to curb pro-inflammatory cytokine release within inflamed mammary tissue and to fortify the tight-junction barrier of bovine mammary epithelial cells, while elucidating the underlying molecular mechanisms. The findings are expected to furnish both a mechanistic framework and a practical rationale for developing non-antibiotic formulations aimed at the prevention or therapy of bovine mastitis.
Materials and methods
Ethical approval
All animal experiments were conducted in accordance with the Guidelines for the Ethical Treatment and Utilization of Experimental Animals of Heilongjiang Bayi Agricultural University and approved by the university’s Animal Ethics Committee (Approval Number: DWKJXY2023066).
Animals and sample collection
Mammary gland tissues were collected from lactating Holstein cows slaughtered at licensed local abattoirs. Experimental cows were sourced from a commercial dairy farm (housing 2,000 cattle) in Qiqihar City, Heilongjiang Province, China, and fed in accordance with the National Research Council (NRC) recommendations (2001).
Healthy mammary tissue: confirmed by Dairy Herd Improvement (DHI) reports (somatic cell count, SCC < 100,000 cells/mL), normal body temperature, soft and elastic mammary tissue (no palpable masses, erythema, swelling, heat, or pain in all four quarters), negative California mastitis test (CMT, Table 1) results, and post-mortem examination (mammary tissue filled with white milk, normal coloration).
Clinical mastitis mammary tissue: confirmed by DHI reports (SCC > 200,000 cells/mL), reduced tissue elasticity with palpable masses and visible erythema in affected quarters, positive CMT (Table 1) results (yellowish/pinkish milk), and post-mortem dissection (firm, inelastic, spongy-degenerated tissue with yellowish/pinkish milk secretion).
Post-slaughter, deep mammary tissue samples (diameter < 1 cm) were immediately collected into cryovials, rapidly immersed in liquid nitrogen, and stored at −80 °C until analysis.
Transfer factor teat canal infusion protocol
Porcine spleen-derived TF solution (provided by Harbin Shengtai Biopharmaceutical Co., Ltd.) complied with the 2020 edition of the Chinese Pharmacopoeia, with the following specifications: pH 6.2–7.3, peptide content ≥ 2 mg/mL, free amino acid content ≥ 14 mg/mL, T-cell activity (via de-E receptor assay) ≥ 10%, total aerobic bacteria ≤ 10 CFU/mL, mold and yeast ≤ 10 CFU/mL, and no detection of Escherichia coli (per mL) or Salmonella (per 10 mL). Because TF is a mixture, its immunological activity is attributed mainly to the peptide fraction. Therefore, the TF dose used in all experiments is expressed as peptide concentration (2 mg/mL).
Following the Technical Guidelines for Bovine Mastitis Prevention and Control (Trial) issued by China’s Ministry of Agriculture and Rural Affairs, 23 cows with subclinical mastitis (lactation days: 20–150; body condition score: 3.0–3.25), sourced from the herd, had a documented history of mastitis cases primarily caused by Gram-negative bacteria (mainly E. coli), and were screened via CMT (Table 1) and veterinary clinical diagnosis. Cows were divided into two groups:
Normal saline (NS) group: 11 cows with 15 infected quarters, infused with 50 mL of sterile saline per infected quarter via the teat canal.
TF group: 12 cows with 17 infected quarters, infused with 50 mL of TF solution (peptide concentration: 2 mg/mL) per infected quarter via the teat canal.
Infusion protocol: complete milk evacuation → 0.5% iodophor triple disinfection of teats → slow infusion via sterile catheters → 10-second circular massage of the mammary gland base. Treatments were administered daily at 3:00 PM for four consecutive days, with no milking within 6 h post-infusion.
Milk sample collection and somatic cell isolation
Milk samples were collected before treatment and at 120 h post-treatment and divided into three aliquots:
Aliquot 1: immediately subjected to CMT to assess subclinical mastitis status of each quarter.Aliquot 2: collected into DHI vials with preservatives, transported on ice to the Daqing Branch of the Heilongjiang Dairy Herd Improvement Laboratory, and analyzed for SCC.Aliquot 3: transferred to two sterile 50 mL centrifuge tubes for somatic cell extraction, with harvested cells cryopreserved in liquid nitrogen for subsequent analysis.
Teats were disinfected with 75% ethanol, and the first three milk squirts were discarded. Approximately 40 mL of milk was collected into sterile containers, mixed with 0.5 mM EDTA (Sangon Biotech, Shanghai, China), and transferred to 50 mL centrifuge tubes. Sequential centrifugation was performed at 4 °C: 1500×g for 15 min, 1000×g for 20 min, and 2000×g for 10 min. The supernatant (fat layer) and middle layer (skim milk) were removed, and residual fat on tube walls was wiped with alcohol-soaked cotton swabs. The cell pellet was resuspended in 20 mL PBS-EDTA solution, washed, and resuspended in ∼10 mL PBS (gentle pipetting to ensure dispersion). A final centrifugation (1500×g for 10 min at 4 °C) was performed, followed by supernatant aspiration and residual fat removal. Cells were resuspended in 2 mL PBS-EDTA and immediately cryopreserved in liquid nitrogen.
MAC-T cell culture and treatment
Bovine mammary epithelial cells (MAC-T) were provided by the Veterinary Molecular Diagnostics and Veterinary Medicine Research Laboratory at Heilongjiang Bayi Agricultural Reclamation University. Cells were cultured in DMEM/F12 (Cellmax, Beijing, China) medium supplemented with 10% fetal bovine serum (FBS, Cellmax, Beijing, China) at 37 °C in a humidified atmosphere with 5% CO_2_. Cells were seeded at 1 × 10^6^ cells/well in 6-well plates, and when reaching 70% confluence, washed three times with PBS and refreshed with fresh medium.
LPS optimization: LPS concentrations (1, 5, 10, 20 µg/mL, Sigma-Aldrich) and treatment durations (3, 6, 9, 12 h) were optimized to establish an effective inflammatory model.TF dose-response: TF concentrations (20, 60, 120, 180 µg/mL) were tested via 2 h pretreatment to determine anti-inflammatory efficacy.Pathway inhibition: cells were pretreated with MLCK inhibitor ML-7 (2 µmol/L), NF-κB inhibitor Bay 11-7082 (10 µmol/L), NF-κB activator 1 (1 µmol/L), or TAK1 inhibitor Takinib (10 µmol/L) for 1 h, followed by co-treatment with 10 µg/mL LPS for 6 h.
Experimental groups included: control (Ctrl), LPS, LPS+TF, LPS+ML-7, LPS+Bay 11-7082, NF-κB activator 1, NF-κB activator 1+TF, and LPS+Takinib.
To investigate or simulate the preventive effect of TF against inflammatory injury, MAC-T cells were pre-treated 2 h with TF before LPS or NF-κB activator 1 challenge in the in vitro experiments.
Cell viability assay (CCK-8)
MAC-T cell suspensions (1 × 10^5^ cells/mL) were seeded into 96-well plates (100 µL/well) and pre-incubated for 24 h. After removing the original medium, fresh medium containing gradient LPS concentrations (1, 5, 10, 20 µg/mL) was added for 3, 6, 9, or 12 h. For viability detection, 10 µL of CCK-8 (Apexbio, Shanghai, China) working solution was added per well, followed by 2 h incubation in the dark. Absorbance (OD) was measured at 450 nm using a full-wavelength microplate reader. Cell viability was calculated as: Viability (%) = (OD_treatment/OD_control) × 100%.
Detection of inflammatory cytokine secretion (ELISA)
After cell treatment, culture supernatants were collected into sterile EP tubes and centrifuged at 1000 rpm for 20 min at 4 °C to remove cellular debris. ELISA kits (ZCIBIO, Shanghai, China) were equilibrated to room temperature for ≥ 30 min. A 96-well microplate was prepared with standard curve wells (serial dilutions), sample wells, and blank controls (duplicate design):
Standard wells: 50 µL of gradient-concentration calibrator solutions.
Sample wells: 50 µL of processed supernatants.
All wells (except blanks) received 100 µL of HRP-conjugated detection antibody working solution. The plate was sealed with adhesive film and incubated at 37 °C for 60 min. After aspiration, wells were washed five times with pre-chilled washing buffer (1×PBS + 0.05% Tween-20). Substrate solutions A and B (50 µL each) were added, followed by 15 min dark incubation at 37 °C. The reaction was terminated with 50 µL of 2 M H_2_SO_4_ per well, and absorbance was measured at 450 nm within 15 min.
RNA extraction and RT-qPCR
Mammary gland tissue: 50 mg of minced tissue was mixed with 1 mL of Total RNA Extraction Reagent (Trizol, ABclonal, Beijing, China) and homogenized. MAC-T cells: Residual medium was removed, 1 mL of Trizol was added per 6-well plate well, and cells were lysed by gentle shaking (3–5 times) and pipetting (2–3 times). Lysates were vortexed for 30 s, incubated at room temperature for 5 min, and mixed with 0.2 mL of pre-chilled chloroform (per 1 mL Trizol). After vigorous shaking (15 s) and room temperature incubation (2–3 min), samples were centrifuged at 12,000 rpm for 10 min at 4 °C. The upper aqueous phase (containing RNA) was transferred to a new tube, mixed with an equal volume of pre-chilled (−20°C) isopropanol, and incubated at room temperature for 10 min. RNA was precipitated by centrifugation (12,000 rpm for 10 min at 4 °C), washed with 1 mL of pre-chilled 75% ethanol, and air-dried for 3 min. RNA pellets were dissolved in 30 µL of RNase-free ultrapure water.
Total RNA (2 µg) was reverse transcribed into cDNA using the Prime Script™ RT kit (ABclonal, Beijing, China). qRT-PCR was performed with 1 µg of cDNA and 2× Universal SYBR Green Fast qPCR Mix (ABclonal, Beijing, China) on a real-time fluorescence quantitative amplifier. PCR conditions: Pre-denaturation at 95 °C for 3 min; 40 cycles of denaturation at 95 °C for 15 s, annealing/extension at 60 °C for 1 min. β-actin was used as the reference gene, and relative gene expression was calculated via the 2^-ΔΔ^Ct method or 2^-Δ^Ct (only for Figure 3). Primers (designed via NCBI, synthesized by Sangon Biotech, Shanghai, China) are listed in Table 2.
Protein extraction and western blotting
Protein samples were adjusted to a uniform concentration with lysis buffer (Beyotime, Beijing, China), mixed with 5× SDS-PAGE (Beyotime, Beijing, China) loading buffer, and denatured at 100 °C for 7 min. Proteins (25 µg per sample) were separated by 10% SDS-PAGE (5% stacking gel) at 80 V (stacking gel) and 120 V (resolving gel), then transferred to methanol-activated PVDF membranes (Beyotime, Beijing, China) at 200 mA for 60 min. Membranes were rinsed with 1× TBST (0.1% Tween-20) and blocked with 5% skim milk for 60 min at room temperature.
Membranes were incubated overnight at 4 °C with primary antibodies: NECTIN4 (1:2000, A16149, Abclonal), Occludin (1:5000, A2601, Abclonal), ZO-1 (1:5000, A0659, Abclonal), p65 (1:2000, A2547, Abclonal), p-p65 (1:10,000, AP0124, Abclonal), MLCK (1:1000, A12840, Abclonal), MLC (1:1000, A8742, Abclonal), p-MLC (1:3000, AP0955, Abclonal), TAK1 (1:500, A12022, Abclonal), p-TAK1 (1:1000, AP0071, Abclonal), and β-actin (1:80,000, AC026, Abclonal; internal control). After four washes with 1× TBST (10 min each), membranes were incubated with HRP-conjugated Goat anti-Rabbit IgG secondary antibody (1:10,000, AS014, Abclonal) for 1 h at room temperature. Chemiluminescence was detected using ECL substrate (Beyotime, Beijing, China) in a BIO-RAD GelDoc XR+ imaging system (Bio-Rad, USA), and band gray values were quantified with ImageJ software.
Statistical analysis
Data are presented as mean ± SEM or mean ± SD from at least three independent biological replicates. Statistical analyses were performed using SPSS Statistics 27.0 (IBM, USA). Two-sample comparisons: independent samples t-test. Multiple group comparisons: One-way analysis of variance (ANOVA) followed by Duncan’s multiple range test. Association between CMT negative conversion rate and TF treatment: Pearson’s chi-square test. Significance was defined as *P *< 0.05 (significant difference) and *P *< 0.01 (extremely significant difference).
Results
Upregulation of pro-inflammatory cytokines and downregulation of tight junction transcripts in mastitic bovine mammary tissue
To characterize the transcriptional profile of mastitic mammary tissue, RT-qPCR was used to quantify mRNA expression of pro-inflammatory cytokines and TJ components. Compared to healthy cows, clinically mastitic cows exhibited a significant 2.0–2.5-fold upregulation of IL-1β and IL-6 mRNA (*P *< 0.01, Figure 1A, B). Concomitantly, the mRNA levels of TJ-related genes (NECTIN4, Occludin, ZO-1) were markedly reduced by 40–60% in mastitic tissue (*P *< 0.01, Figure 1C–E). These results confirm a distinct transcriptional signature of bovine mastitis: pro-inflammatory cytokine induction coupled with TJ transcript depletion.
*Expression profiles of pro-inflammatory cytokines and tight junction genes in mammary gland tissue of healthy and clinically mastitic cows. (A, B) RT-qPCR was used to detect the transcription levels of IL-1β and IL-6 (n = 5). (C–E) RT-qPCR was used to detect the transcription levels of NECTIN4, Occludin and ZO-1 (n = 5). Data are presented as mean ± SEM, *P < 0.01 vs. healthy cows. Bar graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression levels in mammary gland tissue of healthy and clinically mastitic cows.
Consistent expression patterns of inflammatory cytokines and TJ genes in milk somatic cells of mastitic cows
Milk somatic cells from clinically mastitic cows displayed expression patterns consistent with mammary tissue: IL-1β and IL-6 mRNA levels were significantly elevated (*P *< 0.01, Figure 2A, B), while NECTIN4, Occludin, and ZO-1 transcripts were drastically decreased (*P *< 0.01, Figure 2C–E) relative to healthy cows. This congruence validates milk somatic cells as a reliable surrogate for assessing mastitis-associated molecular changes. Subsequent in vitro experiments used MAC-T cells to dissect the underlying mechanisms.
*Expression patterns of pro-inflammatory cytokines and tight junction genes in milk somatic cells of healthy and clinically mastitic cows. (A, B) RT-qPCR was used to detect the transcription levels of IL-1β and IL-6 (n = 5). (C–E) RT-qPCR was used to detect the transcription levels of NECTIN4, Occludin and ZO-1 (n = 5). Data are presented as mean ± SEM, *P < 0.01 vs. healthy cows. Bar graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression levels in milk somatic cells of healthy and clinically mastitic cows.
TF restores TJ transcripts and suppresses pro-inflammatory cytokines mRNA in milk somatic cells of subclinical mastitic cows
To evaluate TF’s therapeutic potential, subclinical mastitic cows were treated with intramammary infusions of TF or NS. Compared to the NS control group, TF treatment significantly suppressed IL-1β and IL-6 mRNA expression in milk somatic cells (*P *< 0.01, Figure 3A, B) and restored the transcript levels of NECTIN4, occludin, and ZO-1 to near-healthy levels (*P *< 0.01, Figure 3C–E). These findings demonstrate that TF reverses the mastitis-associated transcriptional imbalance in subclinical cases.
*Effects of transfer factor (TF) treatment on gene expression in milk somatic cells of subclinical mastitic cows. (A, B) RT-qPCR was used to detect the transcription levels of IL-1β and IL-6 (n = 11 for NS group, n = 12 for TF group). (C–E) RT-qPCR was used to detect the transcription levels of NECTIN4, Occludin and ZO-1 (n = 11 for NS group, n = 12 for TF group). Cows were treated with normal saline (NS) or TF for 4 consecutive days. Relative expression levels were normalized independently within each treatment group (NS or TF), with values representing within-group expression ratios relative to respective internal reference genes (β-actin), rather than fold changes between groups. Data are presented as mean ± SEM, *P < 0.01 vs. NS group. Bar graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression levels in milk somatic cells of subclinical mastitic cows.
TF infusion ameliorates subclinical mastitis in dairy cows
At baseline, milk SCC did not differ between the NS and TF groups (*P *= 0.91, Table 3). On day 5 post-infusion, SCC in the NS group declined by approximately 50% (not significant, *P *= 0.14), while the TF group exhibited a > 90% reduction in SCC (*P *< 0.001 vs. baseline; *P *= 0.007 vs. NS group).
CMT results (Table 4) showed that on day 5, the negative conversion rate of inflamed mammary quarters in the TF group reached 64.7%, significantly higher than the 13.3% in the NS group. Collectively, TF rapidly improves mammary gland health and increases the negative conversion rate of inflamed quarters in subclinical mastitic cows.
Effect of LPS on MAC-T cell viability
MAC-T cells were treated with 1, 5, 10, or 20 µg/mL LPS for 3, 6, 9, or 12 h. LPS at 1, 5, and 10 µg/mL had no significant effect on cell viability (*P *> 0.05), while 20 µg/mL LPS reduced viability by 20–30% (*P *< 0.01, Figure 4). Thus, 10 µg/mL LPS was selected as the non-cytotoxic pro-inflammatory concentration for subsequent experiments.
*Cytotoxicity of lipopolysaccharide (LPS) on bovine mammary epithelial cells (MAC-T) at different concentrations and treatment durations. Cell viability was detected by CCK-8 assay. Data are presented as mean ± SEM (n = 3 independent experiments), *P < 0.01 vs. 0 h group. Bar graphs showing the cell viability of MAC-T treated by LPS.
Temporal effects of LPS on inflammatory cytokine and TJ gene expression in MAC-T cells
MAC-T cells were treated with 10 µg/mL LPS for 3, 6, 9, or 12 h. IL-1β mRNA was significantly upregulated at 3 h and remained elevated through 12 h (*P *< 0.01, Figure 5A), while IL-6 mRNA peaked at 6 h (*P *< 0.01, Figure 5B). For TJ genes, NECTIN4, occludin, and ZO-1 transcripts progressively decreased, reaching their lowest levels at 6 h (*P *< 0.01, Figure 5C–E). A 6-h LPS treatment was chosen to recapitulate the maximal inflammatory and barrier-disruptive response.
*Temporal effects of LPS (10 μg/mL) on pro-inflammatory cytokine and tight junction gene expression in MAC-T cells. (A, B) RT-qPCR was used to detect the transcription levels of IL-1β and IL-6 at 3, 6, 9, and 12 h post-treatment (n = 3 independent experiments). (C–E) RT-qPCR was used to detect the transcription levels of NECTIN4, Occludin and ZO-1 at 3, 6, 9, and 12 h post-treatment (n = 3 independent experiments). Data are presented as mean ± SEM. *P < 0.05, *P < 0.01 vs. control (Ctrl) group. Bar graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression levels in MAC-T cells treated by LPS (10 μg/mL).
Graded effects of TF (20-180 µg/mL) on LPS-induced inflammation and TJ dysfunction in MAC-T cells
MAC-T cells were pretreated with 20, 60, 120, or 180 µg/mL TF before LPS stimulation. LPS alone significantly elevated IL-1β and IL-6 mRNA levels and supernatant protein concentrations (*P *< 0.01, Figure 6A–D) while reducing NECTIN4, occludin, and ZO-1 mRNA expression (*P *< 0.01, Figure 6E–G). TF pretreatment reversed these effects in a concentration-dependent manner: 60, 120, and 180 µg/mL TF significantly downregulated cytokine expression (*P *< 0.05), with 180 µg/mL TF achieving the strongest suppression (60–70% reduction vs. LPS alone). Similarly, 180 µg/mL TF maximally restored TJ gene expression (*P *< 0.01), while 20 µg/mL TF had no significant effect. Thus, 180 µg/mL was identified as the optimal TF concentration.
*Graded effects of TF across tested concentrations on LPS-induced inflammation and tight junction gene/protein expression in MAC-T cells. (A, B) ELISA was used to detect the concentrations of IL-1β and IL-6 in cell supernatants (n = 3 independent experiments). (C, D) RT-qPCR was used to detect the transcription levels of IL-1β and IL-6 (n = 3 independent experiments). (E–G) RT-qPCR was used to detect the transcription levels of NECTIN4, Occludin and ZO-1 (n = 3 independent experiments). Cells were pretreated with TF (20, 60, 120, 180 μg/mL) for 2 h before LPS stimulation. Data are presented as mean ± SEM, *P < 0.01 vs. Ctrl group; #P < 0.05, ##P < 0.01 vs. LPS group. Graphs showing the effects of TF on LPS-Induced Inflammation and TJ Dysfunction in MAC-T Cells. A and B are graphs showing the concentrations of IL-1β and IL-6 in cell supernatants. C-G are graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression in MAC-T cells pretreated with graded TF (20, 60, 120, 180 μg/mL, 2 h) prior to LPS challenge.
TF inhibits LPS-induced activation of the TAK1/NF-κB/MLCK pathway in MAC-T cells
Western blot analysis showed that LPS significantly increased phosphorylation of TAK1, NF-κB p65, and MLC (*P *< 0.01, Figure 7) compared to the Ctrl group. TF pretreatment suppressed phosphorylation of these pathway components in a concentration-dependent manner, with 180 µg/mL TF reducing phosphorylation to near-baseline levels (*P *< 0.01 vs. LPS alone). These data confirm that TF inhibits LPS-induced activation of the TAK1/NF-κB/MLCK pathway.
*Effects of TF on LPS-induced activation of the TAK1/NF-κB/MLCK pathway in MAC-T cells. After total protein was extracted, the protein expression levels in MAC-T from each experimental group were analyzed by Western blot. (A, D) Western blot images of p-TAK1, TAK1, p-p65, p65, p-MLC, MLC and MLCK. (B, C, E, F) Quantitative analysis of relative phosphorylation levels (normalized to β-actin or total protein) of MLCK, p-MLC, p-p65 and p-TAK1. Cells were pretreated with TF (20, 60, 120, 180 μg/mL) for 2 h before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments), *P < 0.01 vs. Ctrl group; ##P < 0.01 vs. LPS group. Graphs showing TF modulates LPS-triggered TAK1/NF-κB/MLCK signaling in MAC-T cells. A and D are the Western blot images of p-TAK1, TAK1, p-p65, p65, p-MLC, MLC and MLCK in MAC-T cells pretreated with graded TF (20, 60, 120, 180 μg/mL, 2 h) prior to LPS challenge. B, C, E and F are graphs showing protein relative levels of MLCK, p-MLC, p-p65 and p-TAK1 in MAC-T cells pretreated with graded TF (20, 60, 120, 180 μg/mL, 2 h) prior to LPS challenge.
TF blocks MLCK-dependent downregulation of TJ components in LPS-challenged MAC-T cells
LPS (10 µg/mL) significantly downregulated NECTIN4, occludin, and ZO-1 mRNA in MAC-T cells (*P *< 0.05). Co-treatment with ML-7 (MLCK inhibitor) restored transcript levels to near-Ctrl levels (*P *< 0.05 vs. LPS, Figure 8A–C). Pretreatment with 180 µg/mL TF exerted a similar effect, with TJ gene expression comparable to the ML-7 group (*P *> 0.05 vs. ML-7).
Effects of TF and MLCK inhibitor ML-7 on LPS-induced tight junction gene downregulation in MAC-T cells. (A–C) RT-qPCR was used to detect the transcription levels of NECTIN4, Occludin and ZO-1. Cells were pretreated with 180 μg/mL TF or 2 μmol/L ML-7 before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments). Differences denoted by the same letter are not statistically significant (P > 0.05), differences indicated by different letters are considered statistically significant (P < 0.05). Graphs showing NECTIN4, Occludin and ZO-1 mRNA expression in MAC-T cells pretreated with 180 μg/mL TF or 2 μmol/L ML-7 before LPS stimulation.
Western blot analysis confirmed these findings: LPS reduced NECTIN4, occludin, and ZO-1 protein levels and upregulated MLCK expression and MLC phosphorylation (*P *< 0.05, Figure 9A–F). Both ML-7 and 180 µg/mL TF reversed these effects (*P *< 0.05), with no significant difference in efficacy (*P *> 0.05). These results indicate that TF blocks MLCK-dependent transcriptional and post-translational downregulation of TJ components.
Effects of TF and ML-7 on LPS-induced tight junction protein downregulation and MLCK/MLC activation in MAC-T cells. (A) Western blot images of NECTIN4, Occludin, ZO-1, p-MLC, MLC and MLCK. (B–F) Quantitative analysis of relative protein levels (normalized to β-actin or total protein) of MLCK, p-MLC, NECTIN4, Occludin and ZO-1. Cells were pretreated with 180 μg/mL TF or 2 μmol/L ML-7 before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments). Differences denoted by the same letter are not statistically significant (P > 0.05), differences indicated by different letters are considered statistically significant (P < 0.05). Graphs showing the effects of TF on MLCK/MLC activation and tight junction protein levels in MAC-T cells pretreated with 180 μg/mL TF or 2 μmol/L ML-7 before LPS stimulation. A is the Western blot images of NECTIN4, Occludin, ZO-1, p-MLC, MLC and MLCK in MAC-T cells. B-F are graphs showing protein relative levels of MLCK, p-MLC, NECTIN4, Occludin and ZO-1 in MAC-T cells.
TF interrupts NF-κB-driven MLCK activation to preserve TJ integrity
LPS significantly increased IL-1β and IL-6 secretion/gene expression and reduced TJ gene mRNA (*P *< 0.05, Figure 10A–G). Bay 11-7082 (NF-κB inhibitor) reversed these effects (*P *< 0.05), while NF-κB activator 1 mimicked LPS-mediated changes (*P *< 0.05). Pretreatment with 180 µg/mL TF blocked activator-induced inflammation and barrier disruption, with efficacy comparable to Bay 11-7082 (*P *< 0.05).
Effects of TF and NF-κB modulators on LPS-induced cytokine secretion and tight junction gene expression in MAC-T cells. (A, B) Concentrations of IL-1β and IL-6 in cell supernatants. (C–G) Relative mRNA levels of IL-1β, IL-6, NECTIN4, Occludin and ZO-1. Cells were pretreated with 180 μg/mL TF, 10 μmol/L Bay 11-7082 (NF-κB inhibitor), or 1 μmol/L NF-κB activator 1 before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments). Differences denoted by the same letter are not statistically significant (P > 0.05), differences indicated by different letters are considered statistically significant (P < 0.05). Graphs showing the effects of TF on cytokine secretion of IL-1β and IL-6 and mRNA expression of IL-1β, IL-6, NECTIN4, Occludin and ZO-1 by regulating NF-κB activation in MAC-T cells pretreated with 180 μg/mL TF, 10 μmol/L Bay 11-7082 (NF-κB inhibitor), or 1 μmol/L NF-κB activator 1 before LPS stimulation. A and B are graphs showing the concentrations of IL-1β and IL-6 in cell supernatants. C-G are graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression in MAC-T cells.
Western blot analysis further showed that LPS and NF-κB activator 1 downregulated TJ proteins and upregulated MLCK, p-p65, and p-MLC (*P *< 0.05, Figure 11A–H), while TF pretreatment reversed these changes (*P *< 0.05). These results demonstrate that TF interrupts NF-κB-driven MLCK activation to preserve TJ integrity and suppress cytokine release.
Effects of TF and NF-κB modulators on LPS-induced tight junction protein downregulation and NF-κB–MLCK pathway activation in MAC-T cells. (A) Western blot images of p-p65, p65, MLCK, p-MLC, and MLC. (B–D) Quantitative analysis of relative protein levels (normalized to β-actin or total protein) of MLCK, p-MLC and p-p65. (E) Western blot images of NECTIN4, Occludin and ZO-1. (F–H) Quantitative analysis of relative protein levels (normalized to β-actin) of NECTIN4, Occludin and ZO-1. Cells were pretreated with 180 μg/mL TF, Bay 11-7082, or NF-κB activator 1 before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments). Differences denoted by the same letter are not statistically significant (P > 0.05), differences indicated by different letters are considered statistically significant (P < 0.05). Graphs showing the effects of TF on tight junction protein downregulation and NF-κB–MLCK pathway activation in MAC-T cells pretreated with 180 μg/mL TF, Bay 11-7082, or NF-κB activator 1 before LPS stimulation. A and E are graphs showing the Western blot images of p-p65, p65, p-MLC, MLC, MLCK, NECTIN4, Occludin and ZO-1 in MAC-T cells. B-D and F-H are graphs showing protein relative levels of MLCK, p-MLC, p-p65, NECTIN4, Occludin and ZO-1 in MAC-T cells.
TF blocks the TAK1-mediated NF-κB/MLCK cascade in LPS-challenged MAC-T cells
LPS significantly increased IL-1β and IL-6 secretion/gene expression and reduced TJ gene mRNA (*P *< 0.05, Figure 12A–G). Pretreatment with 180 µg/mL TF or Takinib (TAK1 inhibitor) significantly reversed these effects (*P *< 0.05), with no significant difference in efficacy (*P *> 0.05).
Effects of TF and TAK1 inhibitor Takinib on LPS-induced cytokine secretion and tight junction gene expression in MAC-T cells. (A, B) Concentrations of IL-1β and IL-6 in cell supernatants. (C–G) Relative mRNA levels of IL-1β, IL-6, NECTIN4, Occludin and ZO-1. Cells were pretreated with 180 μg/mL TF or 10 μmol/L Takinib before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments). Differences denoted by the same letter are not statistically significant (P > 0.05), differences indicated by different letters are considered statistically significant (P < 0.05). Graphs showing the effects of TF on cytokine secretion of IL-1β and IL-6 and mRNA expression of IL-1β, IL-6, NECTIN4, Occludin and ZO-1 by regulating TAK1 activation in MAC-T cells pretreated with 180 μg/mL TF or 10 μmol/L Takinib before LPS stimulation. A and B are graphs showing the concentrations of IL-1β and IL-6 in cell supernatants. C-G are graphs showing IL-1β, IL-6, NECTIN4, Occludin and ZO-1 mRNA expression in MAC-T cells.
Western blot analysis revealed that LPS increased phosphorylation of TAK1, p65, and MLC and reduced TJ protein levels (*P *< 0.05, Figure 13A–H), while both TF and Takinib abolished these effects (*P *< 0.05). These results confirm that TF blocks the TAK1-mediated NF-κB/MLCK cascade to suppress inflammation and preserve TJ integrity.
Effects of TF and Takinib on LPS-induced tight junction protein downregulation and TAK1/NF-κB/MLCK pathway activation in MAC-T cells. (A) Western blot images of p-TAK1, TAK1, p-p65, p65, p-MLC, MLC and MLCK. (B–E) Quantitative analysis of relative protein levels (normalized to β-actin or total protein) of MLCK, p-MLC, p-p65 and p-TAK1. f) Western blot images of NECTIN4, Occludin and ZO-1. (G–I) Quantitative analysis of relative protein levels (normalized to β-actin) of NECTIN4, Occludin and ZO-1. Cells were pretreated with 180 μg/mL TF or 10 μmol/L Takinib before LPS stimulation. Data are presented as mean ± SEM (n = 3 independent experiments). Differences denoted by the same letter are not statistically significant (P > 0.05), differences indicated by different letters are considered statistically significant (P < 0.05). Graphs showing the effects of TF on tight junction protein downregulation and TAK1/NF-κB/MLCK pathway activation in MAC-T cells pretreated with 180 μg/mL TF or 10 μmol/L Takinib before LPS stimulation. A and F are graphs showing the Western blot images of p-TAK1, TAK1, p-p65, p65, p-MLC, MLC, MLCK, NECTIN4, Occludin and ZO-1 in MAC-T cells. B-E and G-I are graphs showing protein relative levels of MLCK, p-MLC, p-p65, p-TAK1, NECTIN4, Occludin and ZO-1 in MAC-T cells.
TF suppresses the TAK1/NF-κB/MLCK cascade in milk somatic cells of subclinical mastitic cows
To verify the in vivo mechanism by which TF targets the TAK1/NF-κB/MLCK axis, we isolated somatic cells from the milk of subclinically infected quarters after intramammary infusion of TF or saline and subjected them to Western blot analyses. These results showed that TF treatment significantly reduced MLCK protein expression (*P *< 0.05, Figure 14A, B) and suppressed phosphorylation of MLC, p65, and TAK1 (*P *< 0.01, Figure 14C–F) compared to the NS group.
*Effects of TF treatment on TAK1/NF-κB/MLCK pathway activation in milk somatic cells of subclinical mastitic cows. Milk somatic cells were isolated from the milk of subclinically infected quarters after intramammary infusion of TF or normal saline (NS) and subjected them to Western blot analyses. (A) Western blot images of MLCK, p-MLC, and MLC. (B, C) Quantitative analysis of relative protein levels (normalized to β-actin or total protein) of MLCK and p-MLC. (D) Western blot images of p-TAK1, TAK1, p-p65, and p65. (E, F) Quantitative analysis of relative protein levels (normalized to β-actin or total protein) of p-TAK1 and p-p65. Data are presented as mean ± SEM (n = 11 for NS group, n = 12 for TF group). *P < 0.05, *P < 0.01 vs. NS group. Graphs showing the effects of TF on TAK1/NF-κB/MLCK pathway activation in milk somatic cells from the milk of subclinically infected quarters after intramammary infusion of TF or normal saline (NS). A and D are graphs showing the Western blot images of p-TAK1, TAK1, p-p65, p65, p-MLC, MLC, MLCK in milk somatic cells of subclinical mastitic cows. B, C, E and F are graphs showing protein relative levels of MLCK, p-MLC, p-p65 and p-TAK1 in milk somatic cells of subclinical mastitic cows.
Discussion
Bovine mastitis, the primary disease constraining global dairy production, is initiated by bacterial ascension into the mammary parenchyma—most notably by E. coli (Tong et al. 2025). LPS from the outer membrane of Gram-negative bacteria binds to TLR4 on mammary epithelial cells, triggering an innate immune cascade that culminates in excessive pro-inflammatory mediator production and concurrent disruption of the blood–milk barrier (Gilbert et al. 2013). While antibiotics remain the clinical mainstay, their use is associated with high antimicrobial resistance rates and persistent recurrence (Ghumman et al. 2025). Thus, clarifying the molecular mechanisms of barrier injury and developing non-antibiotic interventions are urgently needed. In this context, natural products like TF have gained attention due to their low resistance risk, favorable safety profile, and immunomodulatory properties. TF can be prepared from the spleens of various animal species; because its bioactive components are highly conserved, cross-species use is common in clinical practice. In this study we used porcine-spleen-derived TF, owing to its low cost and ready availability.
Bovine mastitis is a cascading process involving pathogen invasion, barrier disruption, and metabolic-immunologic dyshomeostasis (Li et al. 2024). At the inflammatory level, LPS-TLR4 ligation recruits the MyD88 adaptor and activates the NF-κB axis, leading to upregulation of IL-1β, IL-6, and other pro-inflammatory mediators—a mechanism consistently validated in mastitis models (Che et al. 2023). The present study corroborates these findings, showing pronounced elevation of inflammatory cytokines in mastitic tissue and milk somatic cells, further confirming the TLR4/NF-κB pathway as a central driver of mastitic pathology. In vivo, TF administration significantly reduced IL-1β and IL-6 mRNA levels in milk somatic cells while enhancing expression of NECTIN4, occludin, and ZO-1—collectively promoting reconstruction and functional restoration of the mammary epithelial barrier. We used milk somatic cells because they are directly exposed to the local mammary microenvironment and thus capture gene-expression patterns that most accurately reflect mammary-specific immune activity. Moreover, these cells can be collected non-invasively, allowing long-term tracking of immune dynamics without compromising animal welfare. Notably, as we did not profile systemic leukocyte subsets or quantify acute-phase proteins, the data imply that TF exerts its anti-inflammatory action locally within the mammary gland rather than through systemic immunosuppression—an attribute that should reduce the risk of opportunistic infections occasionally seen with broad-spectrum anti-inflammatories.
TJ transcripts are systemically downregulated under inflammatory stress (Wu et al. 2025). In the current study, NECTIN4, occludin, and ZO-1 mRNA levels were significantly lower in both mastitic mammary tissue and milk somatic cells, with an identical decline reproduced in LPS-challenged MAC-T cells. The direct interaction between occludin and the ZO-1 C-terminus—essential for correct TJ localization—has been established in vitro (Zhang et al. 2025b). Notably, NECTIN4 (an immunoglobulin-superfamily adhesion molecule previously overlooked in mammary epithelial TJ research) was synchronously reduced with canonical TJ proteins. By incorporating NECTIN4 into the functional “TJ protein ensemble” of the blood–milk barrier, our data propose a novel, non-invasive biomarker for mastitis monitoring. Nevertheless, the diagnostic sensitivity and specificity of milk NECTIN4 mRNA remain to be validated across divergent herds, lactation stages, and bacterial species; future work should establish receiver-operating-characteristic curves and define threshold values that distinguish transient fluctuations from pathologically relevant declines.
Mechanistically, TF directly suppressed the TAK1/NF-κB/MLCK axis, as evidenced by reduced phosphorylation of TAK1, p65, and MLC and downregulated MLCK protein levels. TF mirrored the effects of the NF-κB inhibitor Bay 11-7082 and the MLCK inhibitor ML-7, interrupting NF-κB activation and downstream MLCK responses to preserve junctional integrity. These findings align with Yuan et al. (2021), who reported that oral allicin suppressed the TLR/MyD88/NF-κB cascade, restored occludin, claudin-1, and ZO-1 levels, and mitigated acrylamide-induced intestinal barrier injury (Yuan et al. 2021). Compared with the TAK1 inhibitor Takinib, TF produced nearly identical suppression of TAK1, p65, and MLC phosphorylation and MLCK downregulation—indicating TF directly or indirectly inhibits TAK1 activity, thereby blocking NF-κB signaling, restraining MLCK induction, and preserving TJ architecture. TAK1 is a critical node relaying LPS-TLR4 signals to the NF-κB and MAPK cascades, and its pharmacological inhibition markedly attenuates LPS-driven inflammation (Chen et al. 2022). Consistent with our findings, Nighot et al. (2019) reported that LPS induces TLR4/MyD88-dependent TAK1 phosphorylation, sequentially triggering IKK complex activation, IκB-α degradation, and NF-κB p50/p65 activation—ultimately enhancing MLCK activity/expression and MLCK-mediated intestinal TJ barrier disruption (Nighot et al. 2019). Despite these congruencies, we cannot exclude the possibility that TF additionally modulates parallel MAPK modules (p38, JNK, ERK) that crosstalk with NF-κB and influence TJ stability; targeted gene silencing or CRISPR-Cas9 knockout of TAK1 in MAC-T cells would provide unequivocal evidence that TAK1 blockade is both necessary and sufficient for TF-mediated protection.
The regulatory role of the TAK1/NF-κB/MLCK axis further clarifies TF’s dual anti-inflammatory and barrier-reparative functions. This mechanism closely resembles the effects of Acanthopanax senticosus polysaccharides (ASPS) in a murine model of LPS-induced intestinal barrier injury. Han et al. (2017) showed that ASPS pretreatment suppressed NF-κB/MLCK pathway activation (reduced p-MLC2, MLCK, and IκB-α phosphorylation), preventing ZO-1 and occludin degradation/redistribution and reducing intestinal epithelial permeability (Han et al. 2017). Additionally, MLCK plays a pivotal role in LPS-induced TJ permeability increases (Nighot et al. 2017). Therefore, by reducing MLCK abundance and MLC phosphorylation, TF further stabilizes the cytoskeleton, favoring TJ protein retention and repair. Yet, it should be acknowledged that MLCK-independent mechanisms (eg, claudin endocytosis, occludin oxidation, or actin-severing proteins such as cofilin) also contribute to barrier leakiness during mastitis; whether TF influences these parallel pathways awaits comprehensive phospho-proteomic profiling.
In vivo validation in subclinical mastitic cows confirmed TF’s therapeutic potential: intramammary TF perfusion rapidly reduced milk SCC, improved mammary gland health, and achieved a 64.7% CMT negative conversion rate (vs. 13.3% in the NS group). TF treatment also reduced MLCK protein expression and phosphorylation of MLC, p65, and TAK1 in milk somatic cells—consistent with in vitro findings. These results align with Ma et al. (2023), who demonstrated that TF upregulates occludin and ZO-1 expression in the intestinal epithelium of laying hens (Ma et al. 2023). Collectively, our findings confirm that TF restores the mammary epithelial barrier, reinstates its protective function, limits pathogen invasion, and prevents excessive inflammatory mediator secretion. Compared to conventional antibiotics, TF—of natural origin with low toxicity—holds promise for large-scale application in sustainable livestock production systems. An important methodological consideration is that our assessment of barrier repair relies on the measurement of TJ protein expression and phosphorylation states. While the upregulation of NECTIN4, occludin, and ZO-1, coupled with the inhibition of the MLCK/p-MLC pathway, provides strong molecular evidence and is a well-established correlate of improved barrier integrity, direct functional assays such as transepithelial electrical resistance (TEER) or paracellular permeability measurements were not conducted in this study. Therefore, our conclusions regarding the restoration of the blood-milk barrier, while strongly supported by the concordant molecular and clinical (SCC reduction) data, would be further strengthened by future functional validation in an appropriate epithelial monolayer model.
Nonetheless, several limitations merit emphasis. First, the cow trial was performed in a single herd fed a uniform TMR and milked under identical routines; multi-region, multi-breed studies are required to exclude herd- or parity-specific confounders. It is noteworthy that the in vivo mastitis model, while occurring in a herd with a high prevalence of E. coli infections, was defined by phenotypic markers (elevated SCC, positive CMT) rather than individual pathogen identification. This common practical approach validates TF’s efficacy against the overarching inflammatory pathology of mastitis, which our in vitro LPS model specifically mimics. Second, the seven-day follow-up precluded evaluation of recurrence and long-term impacts on milk yield, composition, or lactation persistency; trials extending ≥ 60 days should monitor SCC rebound and bacteriological cure. Third, although MAC-T cells are widely used, they are immortalized and non-lactating; confirmation in primary bovine mammary epithelial cells or ex vivo explants would improve translational relevance. Fourth, porcine spleen-derived TF is a < 10 kDa peptide-nucleotide complex devoid of species-specific immunogenicity, yet batch-to-batch heterogeneity in peptide fingerprints or oligonucleotide length could alter potency; industrial production should incorporate LC-MS/MS peptide mapping and UV quantification of ribonucleotides to ensure consistency. Fifth, although the anti-inflammatory and barrier-protective effects of TF at 180 µg/mL were robust and comparable to specific pharmacological inhibitors, we did not perform formal cell viability assays (eg, CCK-8, LDH release) for TF itself in this study. While our experimental design (TF pretreatment followed by LPS challenge) and the observed reversal of LPS-induced damage argue against nonspecific cytotoxicity at this concentration, future studies should comprehensively assess the cytotoxicity profile of TF across a wider concentration range and in primary mammary epithelial cells to establish its therapeutic window more precisely. Sixth, this study attributes the therapeutic effects to the TF mixture as a whole. As TF is a complex of peptides and nucleotides, the specific active component(s) mediating the observed inhibition of the TAK1/NF-κB/MLCK axis remain unidentified. Future studies to fractionate and characterize the constituents of TF are necessary to pinpoint the principal bioactive molecules. Finally, despite TF’s low molecular weight and rapid renal clearance—which have prevented tissue residues in swine and poultry—its intramammary pharmacokinetics (milk/plasma partitioning, half-life, intraglandular proteolysis) remain uncharacterized; a residue-depletion study under VICH GL50 guidelines is nevertheless warranted to formally establish zero milk withdrawal and to meet regulatory expectations. Closing these gaps will advance TF from a proof-of-concept immunomodulator to a rigorously validated, non-antibiotic mastitis-control tool for modern dairy production. Based on our findings and practical considerations, we propose that teat infusion of TF has strong potential for both preventing and treating subclinical mastitis, while in clinical mastitis TF could serve as a valuable adjunct to conventional veterinary therapy.
Conclusion
This study demonstrates that cows with mastitis exhibit inflammatory damage in mammary tissue, upregulated pro-inflammatory cytokines, downregulated TJ proteins, and activated TAK1/NF-κB/MLCK signaling. Through untargeted analysis and functional validation, TF was identified as a key mediator capable of alleviating LPS-induced inflammation and TJ injury. Mechanistically, TF targets the TAK1/NF-κB/MLCK signaling axis to inhibit inflammation and promote the restoration of key TJ components—exerting “anti-inflammation” and “barrier repair” dual effects. This study elucidates TF’s role and mechanism in bovine mastitis, providing a theoretical basis for TF as a dietary additive or therapeutic agent in mastitis prevention and treatment.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Camilleri M. 2019. Leaky gut: mechanisms, measurement and clinical implications in humans. Gut. 68:1516–1526. 10.1136/gutjnl-2019-31842731076401 PMC 6790068 · doi ↗ · pubmed ↗
- 2Che HY et al 2023. Allicin alleviated LPS-induced mastitis via the TLR 4/NF-κB signaling pathway in bovine mammary epithelial cells. Int J Mol Sci. 24:3805. 10.3390/ijms 2404380536835218 PMC 9962488 · doi ↗ · pubmed ↗
- 3Chen X et al 2022. FAK mediates LPS-induced inflammatory lung injury through interacting TAK 1 and activating TAK 1-NFκB pathway. Cell Death Dis. 13:589. 10.1038/s 41419-022-05046-735803916 PMC 9270420 · doi ↗ · pubmed ↗
- 4Ghumman NZ et al 2025. Bovine mastitis and antimicrobial resistance in Pakistan’s dairy sector: current status and future prospects. Vet Res Commun. 50:31. 10.1007/s 11259-025-10951-141219618 PMC 12605563 · doi ↗ · pubmed ↗
- 5Gilbert FB et al 2013. Differential response of bovine mammary epithelial cells to Staphylococcus aureus or Escherichia coli agonists of the innate immune system. Vet Res. 44:40. 10.1186/1297-9716-44-4023758654 PMC 3686618 · doi ↗ · pubmed ↗
- 6Günther J et al 2009. Assessment of the immune capacity of mammary epithelial cells: comparison with mammary tissue after challenge with Escherichia coli. Vet Res. 40:31. 10.1051/vetres/200901419321125 PMC 2695127 · doi ↗ · pubmed ↗
- 7Habar TR et al 2021. Transfer factor treatment in management of peritonitis condition: an experimental study in rat. Ann Med Surg (Lond). 69:102755. 10.1016/j.amsu.2021.10275534527231 PMC 8433118 · doi ↗ · pubmed ↗
- 8Han J et al 2017. Acanthopanax senticosus polysaccharides-induced intestinal tight junction injury alleviation via inhibition of NF-κB/MLCK pathway in a mouse endotoxemia model. World J Gastroenterol. 23:2175–2184. 10.3748/wjg.v 23.i 12.217528405145 PMC 5374129 · doi ↗ · pubmed ↗