Lactation stage-dependent protective effects of human milk exosomes against intestinal ischemia-reperfusion injury via lactoferrin/lysozyme-mediated Nrf2/HO-1 and NF-κB pathways

Abstract

BACKGROUND

Human milk exosomes have been reported to possess intestinal protective functions, but whether their protective effects vary by lactation stage remains unclear.

AIM

To investigate the differential protective effects of human milk exosomes from different lactation stages against intestinal ischemia-reperfusion (I/R) injury and elucidate the key protein components and underlying molecular mechanisms.

METHODS

Exosomes were isolated from colostrum (Col-Exo), transitional milk, and mature milk by ultracentrifugation and characterized. Their protective effects were compared in a Caco-2 cell hypoxia/reoxygenation model and in neonatal rat intestinal I/R. Western blot was used to detect lactoferrin (LTF) and lysozyme (LYZ) levels in exosomes and nuclear factor erythroid 2-related factor 2 (Nrf2)/heme oxygenase-1 (HO-1) and nuclear factor kappa B (NF-κB) pathway activities. Rescue experiments with ML385 and PMA were performed to verify pathway necessity. Correlation analysis was used to establish quantitative associations between protein abundance and pathway activity.

RESULTS

All three exosome types exhibited typical cup-shaped morphology with no significant differences in particle size, but Col-Exo showed the highest particle and protein concentrations. LTF and LYZ expression levels were significantly higher in Col-Exo than those isolated from transitional milk and mature milk. In hypoxia/reoxygenation cells and I/R rats, Col-Exo was most effective in enhancing cell viability, inhibiting apoptosis, reducing oxidative stress and inflammatory responses, restoring the expression of tight junction protein, and ameliorating intestinal mucosal injury. Mechanistic studies showed that Col-Exo significantly activated the Nrf2/HO-1 pathway and inhibited NF-κB p65 phosphorylation, and these protective effects were partially reversed by ML385 or PMA treatment. Correlation analysis revealed that LTF/LYZ abundance was positively correlated with Nrf2 expression and negatively correlated with p-p65/t-p65 ratio.

CONCLUSION

The protective effects of human milk exosomes against intestinal I/R injury are dependent on the lactation stage. Colostrum exosomes, rich in LTF and LYZ, exert optimal protection by activating the Nrf2/HO-1 pathway and inhibiting the NF-κB pathway.

Key Words: Human milk exosomes; Intestinal ischemia-reperfusion injury; Lactoferrin; Lysozyme; Nuclear factor erythroid 2-related factor 2/heme oxygenase-1 pathway; Nuclear factor kappa B pathway

Core Tip: This study demonstrates that human milk exosomes protect against intestinal ischemia-reperfusion injury in a lactation stage-dependent manner. Colostrum exosomes, rich in lactoferrin and lysozyme, exhibit optimal protection by activating the nuclear factor erythroid 2-related factor 2/heme oxygenase-1 antioxidant pathway and inhibiting the nuclear factor kappa B inflammatory pathway. These findings provide a basis for developing exosome-based therapies for neonatal gut injury.

INTRODUCTION

Intestinal ischemia-reperfusion (I/R) injury is a major pathological basis for intestinal dysfunction and necrotizing enterocolitis in neonates, particularly preterm infants[1]. This process involves oxidative stress, inflammatory cascades, and disruption of the intestinal barrier triggered by the restoration of blood flow following hypoxia. In severe cases, it may lead to systemic inflammatory response syndrome and multiple organ dysfunction[2-5]. At present, specific protective strategies for neonatal I/R in clinical practice are lacking, making the exploration of safe and effective interventions important.

Human milk is the optimal nutritional source for newborns, and it also contains abundant bioactive components. Among these, exosomes, key carriers of intercellular communication, have been recognized in recent years as possessing potential gut-protective functions[1]. Research indicates that human milk exosomes can modulate inflammatory responses, oxidative stress, and barrier function in recipient cells by delivering bioactive substances such as proteins and microRNAs[6,7]. A notable detail is that the composition of human milk undergoes dynamic changes throughout the lactation stages (colostrum, transitional milk, and mature milk), and the protein cargo of its exosomes may exhibit stage-specific variations[8,9]. However, whether exosomes from different lactation stages exhibit differential protective effects in I/R, and what their key functional components and molecular mechanisms are, remain unclear.

Existing research indicates that as a product of the early lactation phase, colostrum-derived exosomes are rich in immunologically active proteins such as lactoferrin (LTF) and lysozyme (LYZ). This suggests these proteins may constitute the essential material basis mediating the biological functions of colostrum exosomes[10-13]. As a multifunctional iron-binding glycoprotein, LTF has been shown in various models of intestinal inflammation to exert antioxidant effects by scavenging excess iron ions and inhibiting lipid peroxidation while downregulating the expression levels of pro-inflammatory cytokines such as tumor necrosis factor (TNF)-α and interleukin (IL)-1β[14,15]. In addition to its well-established antimicrobial activity, recent studies have shown that LYZ, a natural antimicrobial peptide, can reduce the infiltration of inflammatory cells, thereby alleviating damage to the intestinal mucosa[16,17]. On the basis of these functions, LTF and LYZ may confer exceptional intestinal barrier-protective potential to colostrum-derived exosomes by activating intracellular antioxidant defense systems [such as the nuclear factor erythroid 2-related factor 2 (Nrf2)/heme oxygenase-1 (HO-1) pathway][18-21] and inhibiting inflammatory cascades [such as the nuclear factor kappa B (NF-κB) pathway][14,22-25]. Previous studies have shown that activation of the Nrf2 pathway upregulates the expression levels of downstream antioxidant enzymes [such as HO-1, superoxide dismutase (SOD), and glutathione (GSH)], thereby alleviating oxidative stress in I/R injury[26]. Meanwhile, inhibition of the NF-κB pathway reduces the release of pro-inflammatory factors and protects the integrity of the intestinal mucosal barrier[27]. LTF and LYZ, which are enriched in colostrum-derived exosomes, may exert superior intestinal barrier protection by synergistically activating the Nrf2/HO-1 antioxidant pathway and inhibiting the NF-κB inflammatory pathway. However, whether this causal chain linking specific proteins, key pathways, and protective efficacy holds true in I/R remains to be experimentally verified.

Therefore, this study proposes the hypothesis that colostrum exosomes, owing to their enrichment in LTF and LYZ, exert optimal protection against intestinal I/R injury by activating the Nrf2/HO-1 pathway and inhibiting the NF-κB pathway. The protective effects of exosomes from different lactation stages were systematically evaluated in cellular and animal I/R models. The involvement of LTF/LYZ and the two signaling pathways were investigated. Correlation analyses were performed to link protein abundance with pathway activity. This study aims to elucidate the material basis and molecular mechanisms underlying the stage-dependent protective effects of human milk exosomes, thereby providing experimental evidence for understanding the physiological functions of human milk and developing novel strategies for protecting the neonatal gut.

MATERIALS AND METHODS

Collection of human milk sample and exosome isolation

Human milk samples were collected from healthy full-term mothers (gestational age of 37-41 weeks) aged ≥ 18 years with no history of smoking or alcohol consumption. Exclusion criteria were adopted from previous studies[28]. They included maternal illnesses, such as gestational hypertension, diabetes, autoimmune diseases, malignancy, acute or chronic infections, thyroid dysfunction, and severe malnutrition, and infants with congenital diseases or malformations. All participants provided written informed consent, and the study was approved by the Ethics Committee of Shanxi Medical University (Approval No. 2025009).

The human milk samples were collected in accordance with lactation stage: Colostrum milk exosome (Col-Exo, 1-5 days postpartum), transitional milk exosome (Tra-Exo, 6-14 days postpartum), and mature milk exosome (Mat-Exo, 15 days postpartum and beyond). Following collection, the samples were immediately aliquoted and stored at -80 °C in ultra-low temperature freezers for subsequent use.

The human milk samples were centrifuged at 2000 × g for 20 minutes at 4 °C to remove cells and large particulate debris. The supernatant was collected and centrifuged at 12000 × g for 40 minutes to eliminate casein micelles and similar components. The supernatant was then filtered through a 0.22 μm membrane filter and subjected to ultracentrifugation at 110000 × g (4 °C) for 70 minutes. The resulting pellet contained the crude exosome fraction. Finally, the exosome pellet was resuspended in sterile phosphate-buffered saline (PBS). Protein concentration was determined using a BCA protein assay kit. The solution was aliquoted and stored at -80 °C for future use.

Exosome characterization

Transmission electron microscopy characterization: Ten microliters of each exosome suspension (Col-Exo, Tra-Exo, and Mat-Exo) were added dropwise to copper grids. The grids were allowed to stand at room temperature for 10 minutes to permit exosome adsorption onto the copper grid surface. After excess liquid was aspirated, 2% uranyl acetate solution was applied for negative staining. The grids were air-dried at room temperature and examined via transmission electron microscopy to capture images and analyze the characteristic morphological features of the three exosome types.

Nanoparticle tracking analysis: Ten microliters of Col-Exo, Tra-Exo, and Mat-Exo were diluted in PBS at ratios ranging from 1:25 to 1:1000 to maintain a particle count of 50-200 per frame. This enabled real-time direct imaging and observation of exosomes within a diameter range of 50-1000 nm. By following the operating protocol for the NanoSight NS300 nanoparticle tracking analyzer (NAT 3.1, Malvern Panalytical, United Kingdom), the particle size distribution, peak particle size, and particle concentration of the three exosome suspensions were sequentially determined. Each sample was analyzed in triplicate.

Protein concentration assay: Exosomes were rapidly thawed at 37 °C, and 6 × RIPA lysis buffer was immediately added. The mixture was incubated on ice for 30 minutes to ensure thorough lysis. Standard samples were prepared for protein concentration determination using BCA assay. First, 5 μL of sample was added to the BCA working reagent and mixed thoroughly. After the sample was incubated for 30 minutes, absorbance at 562 nm was measured using a microplate reader. Protein concentrations were calculated on the basis of the standard curve.

Western blot analysis: Total proteins were extracted using RIPA lysis buffer (Beyotime, Shanghai, China) and quantified via BCA method (Thermo, MA, United States). Twenty micrograms of total protein was subjected to sodium-dodecyl sulfate gel electrophoresis and membrane transfer. After the membranes were blocked, they were incubated overnight at 4 °C with primary antibodies against CD63 (Abcam, 1:2000, United Kingdom), CD81 (CST, 1:2000, MA, United States), tumor susceptibility gene 101 (TSG101, Proteintech, 1:2000, Wuhan, Hubei Province, China), hypoxia-inducible factor-1α (Abcam, 1:1000, United Kingdom), zonula occludens-1 (ZO-1, CST, 1:1000, MA, United States), occludin (Abcam, 1:1000, United Kingdom), LTF (Proteintech, 1:1500, Wuhan, Hubei Province, China), LYZ (Santa Cruz, 1:1000, TX, United States), Nrf2 (CST, 1:1000, MA, United States), HO-1 (Proteintech, 1:1500, Wuhan, Hubei Province, China), p-p65 (CST, 1:1000, MA, United States), t-p65 (CST, 1:1000, MA, United States), and GAPDH (Beyotime, 1:5000, Shanghai, China). Following TBST washing, incubation with HRP-labelled secondary antibody (Beyotime, 1:5000 dilution, Shanghai, China) was conducted at room temperature for 1 hour. After ECL development, band grey values were analyzed, with relative expression levels calculated using GAPDH as an internal control. CD63, CD81, and TSG101 were employed to validate exosome purity and specificity, and the remaining proteins were utilized for function-related expression analysis.

Cell culture and establishment of hypoxia/reoxygenation injury model: The human colonic carcinoma cell line Caco-2 was procured from Wuhan Procell Life Technology Co., Ltd. The cells were cultured in MEM supplemented with 20% fetal bovine serum and 1% penicillin-streptomycin at 37 °C in a humidified incubator containing 5% CO₂. The cells were placed in a hypoxic chamber containing 5% CO₂, 1% O₂, and 94% N₂ for 12 hours to simulate hypoxia, followed by 4 hours of reoxygenation under normoxic conditions.

Cell viability and apoptosis assay: (1) Cell viability (CCK-8 assay): Following reoxygenation, the original medium was discarded, and fresh medium containing 10% CCK-8 reagent was added to each well. The culture plate was incubated in a 37 °C incubator, protected from light, for 2 hours. Subsequently, the optical density values at 450 nm for each well were measured using a multi-mode microplate reader. Cell viability was expressed as a percentage relative to the normoxic control group; and (2) Apoptosis [Annexin V-FITC/propidium iodide (PI) dual-staining flow cytometry]: Cells were collected using trypsin digestion without EDTA then washed two times with pre-chilled PBS. In accordance with the apoptosis assay kit protocol, the cells were resuspended in 1 × binding buffer, added with Annexin V-FITC and PI staining solutions, and incubated at room temperature in the dark for 15 minutes. Afterwards, the samples were analyzed within 1 hour by using flow cytometry. The total apoptosis rate was calculated by determining the proportion of Annexin V-positive/PI-negative cells (early apoptosis) and Annexin V-positive/PI-positive cells (late apoptosis or necrosis).

Detection of intracellular reactive oxygen species levels

Intracellular reactive oxygen species (ROS) levels were assessed using the DCFH-DA fluorescent probe via laser confocal microscopy. After reoxygenation, the culture medium was discarded, and 10 μM of DCFH-DA probe was added. The cells were then incubated at 37 °C in the dark for 30 minutes. After being washed with PBS, the cells were fixed with 4% paraformaldehyde, and nuclei were stained with DAPI. Imaging was performed using laser confocal microscopy: DCF green fluorescence (excitation of 488 nm and emission of 500-550 nm) and DAPI blue fluorescence (excitation of 405 nm and emission of 420-480 nm). ImageJ software was employed to quantitatively analyze the average green fluorescence intensity in the cytoplasmic region for semi-quantitative comparison of ROS levels across groups.

Animal models and grouping

All animal experiments were approved by the Animal Ethics Committee of Tianjin Jinke Bona Biotechnology Co., Ltd (Approval No. GENINK-20240067), and all procedures were conducted in accordance with its guidelines and regulations. Ten-day-old Sprague Dawley rats were randomly assigned to five groups (n = 8): Sham surgery (sham), I/R model, I/R + Col-Exo, I/R + Tra-Exo, and I/R + Mat-Exo. Human milk exosomes (5 μg/g) were administered via gastric gavage immediately before superior mesenteric artery occlusion in the I/R + exosomes groups. Surgery was performed under general anesthesia with isoflurane. All rats received sustained-release buprenorphine for analgesia. A midline incision was made under aseptic conditions to perform a laparotomy. Intestinal ischemia was induced by clamping superior mesenteric artery with a vascular clamp. After 30 minutes of ischemia, the clamp was removed, and intestinal color change confirmed reperfusion (Figure 1A). This duration of ischemia was selected to cause intestinal injury without inducing irreversible full-thickness necrosis[1]. After the abdomen was closed, the rats resumed normal activity. Six hours after reperfusion, the rats were euthanized using carbon dioxide, and the ileum (10 cm distal to the caecum) was harvested for further assessment. The sham-operated rats underwent the same surgical procedure as other groups but without intestinal ischemia.

Figure 1 Human milk exosomes alleviate intestinal injury in neonatal rats with ischemia/reperfusion. A: Neonatal rat intestinal ischemia-reperfusion model; B: Representative macroscopic images of intestinal segments; C: Hematoxylin and eosin staining of ileal sections (scale bar = 50 μm; × 400); D: Chiu’s score for intestinal mucosal injury; E and F: Serum levels of intestinal barrier injury markers diamine oxidase and intestinal fatty acid binding protein. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. I/R: Ischemia-reperfusion; iFABP: Intestinal fatty acid binding protein; DAO: Diamine oxidase.

Enzyme-linked immunosorbent assay

Enzyme-linked immunosorbent assay (ELISA) was employed to detect inflammatory cytokines (TNF-α, IL-1β, and IL-10) in cell supernatants, serum, and intestinal tissue, alongside serum intestinal barrier damage markers [intestinal fatty acid binding protein (iFABP) and diamine oxidase (DAO)]. Cell supernatants and animal serum samples were centrifuged after collection and stored at -80 °C. Intestinal tissue was homogenized in PBS and centrifuged to obtain supernatant. Protein concentration was determined using the BCA method for standardization. For the assays, human-specific kits (BioLegend, CA, United States) were employed for cellular inflammatory cytokines, and rat-specific kits (R&D Systems, Cusabio, Wuhan, Hubei Province, China) were used for serum and tissue inflammatory cytokines and serum iFABP and DAO. All procedures were conducted in accordance with the manual: Standards and samples were added to pre-coated microplates, incubated, and washed, followed by sequential addition of biotinylated detection antibody, horseradish peroxidase-labelled streptavidin, and TMB substrate for color development. After termination, optical density was measured at 450 nm and concentrations were calculated. The marker concentrations in cell supernatants and serum are expressed in pg/mL, and the inflammatory factor levels in intestinal tissue are reported as pg/mg protein.

Assessment of intestinal histopathological lesions

Fixed ileal tissue was paraffin-embedded, sectioned consecutively (4 μm thick), and subjected to hematoxylin and eosin (H&E) staining. Sections were dewaxed in xylene, hydrated with graded ethanol solutions, stained with hematoxylin for 5-8 minutes, rinsed under running water, differentiated with 1% hydrochloric acid in ethanol for 1-3 seconds, and counterstained with a weakly alkaline solution. Cytoplasmic counterstaining was performed with 0.5% eosin solution for 1-3 minutes, followed by dehydration with graded ethanol, clearing in xylene, and mounting with neutral resin. All samples within the same batch underwent identical procedures, with each step’s timing strictly controlled by the same laboratory technician, to ensure staining consistency. Following staining, two pathologists unaware of the experimental groups independently assessed the degree of intestinal mucosal damage under an optical microscope using the classical Chiu’s scoring system (0-5 points). The mean score was taken as the final pathological score for each sample.

Detection of oxidative stress markers

Oxidative stress markers were measured in serum and intestinal tissue homogenates. Animal serum and frozen ileum tissue (prepared as homogenates) were analyzed using appropriate biochemical assay kits to determine malondialdehyde (MDA) content, SOD activity, and reduced GSH levels. MDA content was measured using the thiobarbituric acid method (Nanjing Jiancheng, Nanjing, Jiangsu Province, China). Concentrations in serum (nmol/mL) and tissue homogenate (nmol/mg protein) were calculated via standard curves. SOD activity was measured using xanthine oxidase assay (Nanjing Jiancheng, Nanjing, Jiangsu Province, China), with enzyme activity units calculated on the basis of the inhibition rate of superoxide anion production. GSH content was measured using a microplate-based GSH assay kit, with concentrations calculated from a standard curve. All results from tissue homogenates were normalized using total protein concentration determined by the BCA method.

Correlation analysis of exosomal protein abundance and cellular pathway activity

Pearson correlation analysis was conducted to investigate the quantitative relationship between key functional protein content in human milk exosomes and the cellular signaling pathway activity they induce. Exosomal protein abundance data were derived from measurements of LTF and LYZ in exosomes from different stages. Western blot (WB) detection was performed, with band grey values quantified using ImageJ software and normalized against TSG101 as an internal control to calculate relative expression levels (target protein grey value/TSG101 grey value). All data were based on exosome preparations from ≥ 3 independent donor sources. Cellular pathway activity data were derived from Nrf2 pathway activity (expressed as Nrf2 protein levels) and NF-κB pathway activity (expressed as p-p65/t-p65 ratio). Pearson correlation analysis was performed between paired protein abundance and pathway activity data.

Statistical analysis

Experimental data are presented as mean ± SD. Intergroup comparisons were conducted using one-way ANOVA followed by Tukey’s post-hoc test. Furthermore, Pearson correlation analysis was conducted to assess the linear relationship between the abundance of key exosomal proteins (LTF and LYZ) and cellular pathway activity, with correlation coefficients (r) and their significance calculated. All statistical analyses were performed using GraphPad Prism (version 10.1.2) software. P < 0.05 was considered statistically significant.

RESULTS

Identification of human milk-derived exosomes

As shown in Figure 2A, the transmission electron microscopy results indicated that human milk exosomes extracted via ultracentrifugation exhibited a double-layered membrane structure, appearing circular or elliptical in shape, consistent with the morphological characteristics of exosomes. As shown in Figure 2B and C, the nanoparticle tracking analysis revealed that human milk exosomes from different stages exhibited a single peak. The average particle sizes of Col-Exo, Tra-Exo, and Mat-Exo were 141.4 ± 6.1 nm, 153.6 ± 4.4 nm, and 150.6 ± 5.4 nm, respectively, consistent with exosome size characteristics. The differences in peak particle size and average particle size among Col-Exo, Tra-Exo, and Mat-Exo were not statistically significant. This finding suggests that the particle size of human milk exosomes remains relatively stable across different lactation stages. The particle concentration of Col-Exo was significantly higher than that of Tra-Exo and Mat-Exo, suggesting that a greater number of exosomes could be isolated from colostrum (Figure 2D). The BCA protein quantification results indicated that the total protein concentrations of Col-Exo, Tra-Exo, and Mat-Exo were approximately 2.77 mg/mL, 1.91 mg/mL, and 1.63 mg/mL, respectively (Figure 2E). Statistical analysis indicated that the protein concentration of Col-Exo was significantly higher than that of Tra-Exo and Mat-Exo, consistent with the particle concentration trends observed in nanoparticle tracking analysis. This finding further corroborates the differential enrichment efficiency of human milk exosomes across distinct lactation stages. As depicted in Figure 2F, the WB analysis of human milk exosomes extracted via ultracentrifugation identified the marker proteins CD63, CD81, and TSG101, demonstrating that these exosomal marker proteins were detectable in human milk exosomes from different lactation stages. These findings confirmed the successful extraction of human milk exosomes by using the ultracentrifugation method, rendering them suitable for subsequent experimental applications.

Figure 2 Identification of extracellular vesicles derived from breast milk at different lactation stages. A: Representative transmission electron microscopy images showing Col-Exo, Tra-Exo, and Mat-Exo (scale bar = 200 nm); B: Nanoparticle tracking analysis showed the size distribution characteristics of extracellular vesicles at different stages; C: Peak particle size; D: Particle concentration; E: Protein concentration; F: Western blot analysis of extracellular vesicle markers CD63, CD81, and tumor susceptibility gene 101. ^bP < 0.01. NS: Not significant; TSG101: Tumor susceptibility gene 101.

Protective effects of exosomes at different stages against hypoxia/reoxygenation injury in Caco-2 cells

Determination of optimal exosome concentration and validation using the hypoxia/reoxygenation model: CCK-8 assay revealed that 50 μg/mL of exosomes exhibited the most pronounced enhancement of Caco-2 cell viability in the hypoxia/reoxygenation (H/R) model, with cell viability exceeding 90% and no observable toxicity (Figure 3A). Consequently, this concentration was employed in subsequent experiments. Protein immunoblotting analysis revealed significantly increased hypoxia-inducible factor-1α protein levels in cells from the H/R model group compared with those in the normoxic control group (Figure 3B), indicating successful establishment of the hypoxia model.

Figure 3 Protective effects of stage - specific exosomes derived from human milk on Caco-2 cells subjected to hypoxia/reoxygenation injury. A: Cell viability assessed by CCK-8 assay to determine the optimal exosome concentration; B: Western blot analysis of hypoxia-inducible factor-1α expression; C: Cell viability under hypoxia/reoxygenation injury with or without exosome treatment; D: Apoptosis rates detected by Annexin V-FITC/PI staining; E: Representative fluorescence images of intracellular reactive oxygen species levels (scale bar = 50 μm; × 400); F: Quantification of reactive oxygen species fluorescence intensity; G: Levels of tumor necrosis factor-α, interleukin-1β, and interleukin-10 in cell supernatants; H: Western blot analysis of tight junction proteins zonula occludens-1 and occludin. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. HIF-1α: Hypoxia-inducible factor-1α; IL: Interleukin; H/R: Hypoxia/reoxygenation; ROS: Reactive oxygen species; ZO-1: Zonula occludens-1.

Effects of exosomes on cell viability and apoptosis under H/R injury: CCK-8 assays revealed that H/R treatment resulted in a considerable decline in cell viability. Following pretreatment with exosomes from different stages, cell viability recovered to 90.5% ± 9.1% in the Col-Exo group, 75.3% ± 6.2% in the Tra-Exo group, and 65.2% ± 4.6% in the Mat-Exo group. The Col-Exo group exhibited the most pronounced recovery effect, significantly outperforming the Tra-Exo and Mat-Exo groups (Figure 3C). Flow cytometry analysis of apoptosis confirmed this trend, with Col-Exo pretreatment exhibiting the strongest anti-apoptotic effect (Figure 3D).

Regulatory effects of exosomes on H/R-induced oxidative stress and inflammatory responses: Fluorescence assays revealed that H/R stimulation caused a remarkable increase in intracellular ROS levels, with the relative fluorescence intensity increasing to (3.00 ± 0.15)-fold of the control group. Exosome pretreatment effectively suppressed excessive ROS accumulation, reducing the relative fluorescence intensity to (1.12 ± 0.13)-fold in the Col-Exo group, (1.65 ± 0.15)-fold in the Tra-Exo group, and (2.00 ± 0.13)-fold in the Mat-Exo group. The Col-Exo group exhibited the most pronounced inhibitory effect, significantly outperforming the Tra-Exo and Mat-Exo groups (Figure 3E and F). ELISA analysis revealed increased levels of pro-inflammatory factors (TNF-α and IL-1β) and reduced levels of anti-inflammatory factors (IL-10) in the supernatant of the H/R model group (Figure 3G). All exosome treatment groups partially corrected this imbalance, with the Col-Exo group exhibiting the most pronounced effect.

Effects of exosomes on the expression of tight junction protein following H/R injury: The protein expression levels of key intestinal epithelial barrier proteins were assessed via WB. As shown in Figure 3H, compared to the control group (1.00 ± 0.1 for ZO-1 and occludin), the H/R model group exhibited significantly reduced expression levels of ZO-1 (0.35 ± 0.03) and occludin (0.41 ± 0.03). Following pretreatment with human milk exosomes from different stages, both proteins were restored to varying degrees. The relative expression levels of ZO-1 and occludin recovered to 0.92 ± 0.05 and 0.95 ± 0.06, respectively, in the H/R + Col-Exo group; 0.70 ± 0.06 and 0.73 ± 0.03, respectively, in the H/R + Tra-Exo group; and 0.50 ± 0.08 and 0.52 ± 0.04, respectively, in the H/R + Mat-Exo group. The Col-Exo group exhibited the most pronounced recovery, significantly outperforming the Tra-Exo and Mat-Exo groups.

Alleviation of I/R injury by human milk-derived exosomes in the intestines of neonatal rats

Ameliorative effects of exosomes on histopathological damage in intestinal tissue: The extent of intestinal mucosal injury was assessed by observing gross intestinal morphology and H&E staining. As shown in Figure 1B, the I/R group exhibited marked hyperemia, oedema, and dark red discoloration of the intestinal tract. All exosome treatment groups demonstrated reduced intestinal injury, with the most pronounced improvement observed in the I/R + Col-Exo group. H&E staining revealed intact and neatly arranged intestinal villi in the sham group. In the I/R group, villous disruption and detachment were observable, accompanied by submucosal oedema. Following exosome intervention, these pathological alterations were alleviated to varying degrees, with the Col-Exo group exhibiting intestinal mucosal architecture most closely resembling that of the sham group (Figure 1C). Chiu’s quantitative analysis revealed that the I/R group had significantly increased pathological scores (4.00 ± 0.12) compared with the sham group (0.75 ± 0.10). Following exosome treatment, the scores decreased to 0.85 ± 0.10 in the Col-Exo group, 2.34 ± 0.06 in the Tra-Exo group, and 3.25 ± 0.12 in the Mat-Exo group, all significantly lower than that in the I/R group. The Col-Exo group exhibited the lowest score, significantly lower than the Tra-Exo and Mat-Exo groups (Figure 1D).

Effects of exosomes on serum markers of intestinal barrier impairment: The serum levels of the intestinal barrier damage markers DAO and iFABP were measured using ELISA. The results demonstrated significantly increased serum DAO and iFABP concentrations in the I/R group (455.00 ± 11.25 pg/mL and 210.23 ± 8.00 pg/mL) compared with the sham group. Compared with the I/R group, all exosome treatment groups demonstrated a decrease in the levels of both markers, with the most pronounced reduction observed in the Col-Exo group (365.11 ± 9.65 pg/mL and 156.56 ± 7.00 pg/mL), suggesting its strongest protective effect on intestinal barrier integrity (Figure 1E and F).

Effects of exosomes on serum and intestinal tissue inflammatory cytokine levels: The inflammatory cytokine levels in serum and intestinal tissue homogenates were measured using ELISA. The results demonstrated that compared with the sham group, the I/R group exhibited significantly increased levels of pro-inflammatory cytokines TNF-α and IL-1β in serum and intestinal tissue (TNF-α: 2400.00 ± 35.23 pg/mL and 610.00 ± 25.00 pg/mg protein; IL-1β: 435.01 ± 15.65 pg/mL and 258.21 ± 15.65 pg/mg protein), alongside a marked reduction in anti-inflammatory cytokine IL-10 (50.02 ± 10.65 pg/mL and 6.56 ± 1.00 pg/mg protein). Following intervention with human milk exosomes at different stages, the aforementioned inflammatory cytokine imbalance was ameliorated to varying degrees. Notably, the I/R + Col-Exo group exhibited the most pronounced suppression of increased TNF-α (701.21 ± 23.00 pg/mL and 180.02 ± 13.65 pg/mg protein) and IL-1β (102.21 ± 13.00 pg/mL and 85.32 ± 13.00 pg/mg protein), alongside the most significant restoration of IL-10 levels (302.35 ± 13.02 pg/mL and 80.00 ± 1.65 pg/mg protein), demonstrating markedly superior efficacy compared with the Tra-Exo and Mat-Exo groups (Figure 4A and B).

Figure 4 Effects of human milk exosomes on inflammation, oxidative stress, and tight junction proteins in ischemia-reperfusion rats. A: Levels of tumor necrosis factor-α, interleukin (IL)-1β, and IL-10 in serum; B: Levels of tumor necrosis factor-α, IL-1β, and IL-10 in intestinal tissues; C: Levels of malondialdehyde, superoxide dismutase, and glutathione in serum; D: Levels of malondialdehyde, superoxide dismutase, and glutathione in intestinal tissues; E: Western blot analysis of zonula occludens-1 and Occludin expression in intestinal tissues. ^bP < 0.01, ^cP < 0.001. TNF-α: Tumor necrosis factor-α; IL: Interleukin; I/R: Ischemia-reperfusion; MDA: Malondialdehyde; SOD: Superoxide dismutase; GSH: Glutathione; ZO-1: Zonula occludens-1.

Effects of exosomes on oxidative stress markers in serum and intestinal tissue: The analysis of oxidative stress-related markers in serum and intestinal tissue revealed significantly increased MDA levels in the serum and intestinal homogenates of the I/R group (125.00 ± 2.30 U/mg and 27.00 ± 1.10 U/mg protein) compared with those of the sham group, alongside markedly reduced SOD activity and GSH levels (SOD: 10.00 ± 0.30 U/mg and 2.50 ± 0.30 U/mg protein; GSH: 19.98 ± 1.00 U/mg and 6.20 ± 0.15 U/mg protein). Following exosome intervention, the aforementioned abnormal changes in oxidative stress markers exhibited a degree of reversal. Intergroup comparisons revealed that I/R + Col-Exo treatment demonstrated the most potent effects in reducing MDA content (38.01 ± 1.20 U/mg and 6.02 ± 0.30 U/mg protein), restoring SOD activity (62.99 ± 4.30 U/mg and 10.00 ± 0.23U/mg protein), and increasing GSH levels (120.00 ± 1.60 U/mg and 40.00 ± 1.60 U/mg protein), with statistically significant differences compared with the other exosome-treated groups (Figure 4C and D).

Effects of exosomes on the expression of tight junction proteins in intestinal tissue: The protein expression levels of the tight junction proteins ZO-1 and occludin in ileal tissue were assessed via WB. The results demonstrated that compared with the sham group, the I/R group had significantly downregulated ZO-1 and occludin protein expression levels (0.52 ± 0.03 and 0.46 ± 0.05). Following human milk exosome intervention at different stages, the expression levels of both tight junction proteins exhibited varying degrees of recovery. Quantitative analysis indicated that the relative expression levels of ZO-1 and occludin recovered most markedly in the I/R + Col-Exo group (0.85 ± 0.03 and 0.85 ± 0.04), being significantly higher than in the Tra-Exo and Mat-Exo groups (Figure 4E).

Validation of the mechanism by which human milk exosomes modulate the Nrf2/HO-1 and NF-κB pathways

Differential expression levels of LTF and LYZ in human milk exosomes across different stages: WB analysis was employed to detect the protein levels of LTF and LYZ in human milk exosomes at different stages. Following normalization using TSG101 as an internal reference, quantitative analysis revealed that the relative expression levels of LTF and LYZ were highest in the Col-Exo group (1.00 ± 0.02 and 1.00 ± 0.04), significantly exceeding those in the Tra-Exo (0.46 ± 0.05 and 0.45 ± 0.05) and Mat-Exo groups (0.23 ± 0.04 and 0.18 ± 0.04), as shown in Figure 5A. This finding suggests that colostrum exosomes may possess enhanced biological functions due to their enrichment in LTF and LYZ.

Figure 5 Differential expression of lactoferrin and lysozyme in stage-specific exosomes and their regulation of nuclear factor erythroid 2-related factor 2/heme oxygenase-1 and nuclear factor kappa B pathways. A: Western blot analysis of lactoferrin and lysozyme in Col-Exo, Tra-Exo, and Mat-Exo; B: Protein expression of nuclear factor erythroid 2-related factor 2 and heme oxygenase-1; C: Phosphorylation level of nuclear factor kappa B p65 (p-p65/t-p65); D and E: Levels of tumor necrosis factor-α and interleukin-1β in cell supernatants. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. NS: Not significant; LTF: Lactoferrin; LYZ: Lysozyme; H/R: Hypoxia/reoxygenation; Nrf2: Nuclear factor erythroid 2-related factor 2; HO-1: Heme oxygenase-1; TNF-α: Tumor necrosis factor-α; IL-1β: Interleukin-1β.

Effects of exosomes on Nrf2/HO-1 antioxidant pathway: WB analysis was employed to detect the protein expression levels of Nrf2 and its downstream product HO-1 in cells to investigate the regulatory role of human milk exosomes at different stages on the antioxidant pathway. The results indicated that compared with the control group (normalized to 1.00 ± 0.10 for Nrf2 and HO-1), the H/R group exhibited a slight increase in Nrf2 (1.50 ± 0.05) and HO-1 (2.00 ± 0.15) expression levels, potentially representing a compensatory response following cellular stress. Following exosome intervention, the Nrf2 and HO-1 expression levels exhibited an upward trend. The relative expression levels of Nrf2 and HO-1 increased to 6.00 ± 0.18 and 5.98 ± 0.13, respectively, in the Col-Exo group; 4.50 ± 0.06 and 4.35 ± 0.06, respectively, in the Tra-Exo group; and 3.50 ± 0.05 and 3.52 ± 0.05, respectively, in the Mat-Exo group. Col-Exo treatment demonstrated the most pronounced activation of the Nrf2/HO-1 pathway, with significantly higher Nrf2 and HO-1 protein levels than the Tra-Exo and Mat-Exo groups (Figure 5B).

Effects of exosomes on NF-κB inflammatory pathway: WB analysis was employed to assess inflammatory pathway activity by detecting the phosphorylation levels of NF-κB p65. As depicted in Figure 5C, the p-p65/t-p65 ratio in the H/R group (0.65 ± 0.08) exhibited a significant increase compared with that in the control group (0.10 ± 0.03), indicating activation of the inflammatory pathway. Following exosome intervention, the p-p65/t-p65 ratio exhibited varying degrees of reduction. Notably, the Col-Exo group (0.18 ± 0.05) demonstrated the most pronounced decrease in this ratio, with significantly lower p-p65 levels than the Tra-Exo (0.35 ± 0.06) and Mat-Exo groups (0.50 ± 0.07). This finding suggests that colostrum exosomes exert the strongest inhibitory effect on the NF-κB inflammatory pathway. ELISA was employed to measure downstream inflammatory cytokine levels in cell supernatants to further validate the functional outcome of pathway inhibition. The results demonstrated significantly increased secretion of TNF-α and IL-1β in the H/R group compared with that in the control group. Consistent with the trend observed for p-p65, the Col-Exo group exhibited the most pronounced inhibitory effect on increased TNF-α and IL-1β, outperforming the Tra-Exo and Mat-Exo groups (Figure 5D and E). This finding further confirmed that colostrum exosomes exert anti-inflammatory effects by suppressing the NF-κB pathway.

Functional rescue experiments verifying the necessity of Nrf2 pathway

Determination of ML385 treatment concentration: The effects of varying ML385 concentrations on cell viability were assessed using CCK-8 assay to determine the optimal concentration for the Nrf2-specific inhibitor ML385. Cells from the H/R model were treated with DMSO and ML385 at concentrations of 2 μM, 5 μM, and 10 μM. The results indicated that compared with H/R, treatment with 2 (63.00% ± 6.00%) and 5 μM (59.00% ± 5.00%) ML385 had no significant effect on cell viability, whereas 10 μM (48.00% ± 3.00%) ML385 treatment caused a marked decrease in cell viability, suggesting potential cytotoxicity at this concentration (Figure 6A). Consequently, all subsequent experiments employed 5 μM of ML385 for treatment.

Figure 6 Nuclear factor erythroid 2-related factor 2 pathway is required for the protective effects of colostrum exosomes. A: Cell viability under different concentrations of ML385; B: Western blot analysis of nuclear factor erythroid 2-related factor 2 and heme oxygenase-1 expression; C: Cell viability assessed by CCK-8; D: Apoptosis rates measured by flow cytometry; E: Reactive oxygen species levels; F: Malondialdehyde content; G: Superoxide dismutase activity; H: Glutathione levels; I: Western blot analysis of zonula occludens-1 and occludin. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. NS: Not significant; H/R: Hypoxia/reoxygenation; Nrf2: Nuclear factor erythroid 2-related factor 2; HO-1: Heme oxygenase-1; ROS: Reactive oxygen species; MDA: Malondialdehyde; SOD: Superoxide dismutase; GSH: Glutathione; ZO-1: Zonula occludens-1.

Effects of ML385 on Nrf2/HO-1 pathway activity: WB analysis was performed to detect the protein expression levels of Nrf2 and HO-1 in each group of cells. As shown in Figure 6B, compared with the control group, the H/R group exhibited a slight increase in Nrf2 (1.36 ± 0.03) and HO-1 (1.34 ± 0.03) expression levels; Col-Exo treatment significantly upregulated both, with Nrf2 increasing to 2.50 ± 0.03 and HO-1 to 2.45 ± 0.03. Following ML385 addition, the Col-Exo-induced upregulation of Nrf2 (1.68 ± 0.06) and HO-1 (1.69 ± 0.06) was markedly attenuated, whereas ML385 alone had no significant effect on the H/R group. This finding indicates that ML385 effectively blocked the activation of the Nrf2 pathway by colostrum exosomes.

Effects of ML385 on cell viability and apoptosis: CCK-8 assays revealed that Col-Exo treatment significantly enhanced cell viability following H/R injury. As shown in Figure 6C, the cell viability in the H/R group was reduced to 60.00% ± 8.00% of the control group. Col-Exo treatment significantly increased the viability to 84.30% ± 7.10%. Co-treatment with ML385 partially counteracted the protective effect of Col-Exo, resulting in a cell viability of 69.80% ± 8.00%, which was significantly lower than that of the H/R + Col-Exo group. ML385 alone had no significant effect on cell viability (55.00% ± 5.00%). Flow cytometry analysis of apoptosis revealed consistent results. As shown in Figure 6D, the apoptotic rate in the H/R group was 38.34% ± 1.50%, significantly higher than that in the control group (4.37% ± 1.12%). Col-Exo treatment significantly reduced the apoptotic rate to 15.82% ± 1.50%. ML385 partially reversed the anti-apoptotic effect of Col-Exo, increasing the apoptosis rate to 28.09% ± 2.00%. ML385 alone had no significant effect on apoptosis in the H/R group (39.22% ± 1.84%).

Effects of ML385 on oxidative stress markers: The intracellular ROS levels and biochemical indicators of oxidative stress were assessed. As shown in Figure 6E-H, compared with control (normalized to 1.00 ± 0.03 for all indicators), H/R injury significantly increased the ROS levels to (3.50 ± 0.05)-fold and the MDA content to (2.80 ± 0.05)-fold while significantly reducing the SOD activity to (0.61 ± 0.05)-fold and the GSH levels to (0.60 ± 0.01)-fold. Col-Exo treatment significantly reversed these changes, reducing ROS to (1.50 ± 0.03)-fold and MDA to (1.40 ± 0.05)-fold and increasing SOD to (1.80 ± 0.06)-fold and GSH to (1.81 ± 0.04)-fold. Following the addition of ML385, the antioxidant effects of Col-Exo were markedly attenuated: ROS increased to (2.35 ± 0.21)-fold and MDA to (2.00 ± 0.02)-fold, whereas SOD decreased to (1.12 ± 0.06)-fold and GSH to (1.10 ± 0.02)-fold. ML385 alone had no significant effect on these oxidative stress markers in the H/R group, with values remaining at (3.80 ± 0.08)-fold (ROS), (3.00 ± 0.04)-fold (MDA), (0.50 ± 0.03)-fold (SOD), and (0.50 ± 0.03)-fold (GSH).

Effects of ML385 on the expression of tight junction proteins: WB analysis was performed to assess the expression levels of ZO-1 and occludin, key proteins in intestinal epithelial barrier function. As shown in Figure 6I, H/R injury resulted in significant downregulation of ZO-1 and occludin expression levels (0.35 ± 0.05 and 0.36 ± 0.01); Col-Exo treatment markedly restored the expression levels of both proteins (0.90 ± 0.06 and 0.89 ± 0.04). The addition of ML385 markedly attenuated the upregulation of these tight junction proteins by Col-Exo (0.60 ± 0.06 and 0.60 ± 0.02), whereas ML385 treatment alone had no significant effect on the H/R group. This finding indicated that the activation of the Nrf2 pathway constitutes a crucial mechanism by which colostrum-derived exosomes maintain intestinal epithelial barrier integrity.

Validation of the necessity of NF-κB pathway via functional rescue experiments

Determination of PMA treatment concentration and duration: The effect of varying PMA concentrations on cell viability was assessed using CCK-8 assay to establish optimal conditions for the NF-κB activator PMA. The results indicated that compared with H/R group (65.00% ± 4.00%), 25 ng/mL (62.00% ± 6.00%) and 50 ng/mL (58.00% ± 5.00%) PMA treatment had no significant effect on cell viability, whereas 100 ng/mL (49.00% ± 3.00%) PMA treatment caused a marked decrease in cell viability, suggesting this concentration may exhibit cytotoxicity (Figure 7A). Consequently, 50 ng/mL was selected as the safe concentration for the subsequent experiments. Samples were treated with 50 ng/mL of PMA for 30 minutes, 1 hour, and 2 hours separately. The p-p65/t-p65 ratio was determined via WB analysis to establish the optimal activation duration. As depicted in Figure 7B, the p-p65 levels increased over time following PMA treatment, peaking at 1 hour, compared with H/R group. The levels declined slightly after 2 hours but remained increased. The subsequent experiments employed 50 ng/mL of PMA treatment for 1 hour to activate the NF-κB pathway.

Figure 7 Nuclear factor kappa B pathway inhibition mediates the anti-inflammatory effects of colostrum exosomes. A: Cell viability under different concentrations of PMA; B: Quantitative analysis of p-p65/t-p65 ratio after PMA stimulation at indicated time points; C: Western blot analysis of p-p65 and t-p65; D: Cell viability by CCK-8; E: Apoptosis rates by flow cytometry. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. NS: Not significant; H/R: Hypoxia/reoxygenation.

Effects of PMA on NF-κB pathway activity: WB analysis was conducted to assess the p-p65 and total p65 expression levels across cell groups. As shown in Figure 7C, the H/R group showed a significant increase in the p-p65/t-p65 ratio (0.51 ± 0.08) compared with the control group (0.10 ± 0.03), indicating activated inflammatory pathways. Col-Exo treatment markedly suppressed the p-p65 phosphorylation levels (0.19 ± 0.05). Following PMA addition, the inhibitory effect of Col-Exo on p-p65 was partially counteracted, with the p-p65/t-p65 ratio (0.64 ± 0.06) markedly recovering compared with that in the H/R + Col-Exo group. Concurrent PMA treatment further amplified the p-p65 levels in the H/R group. No significant differences in total p65 expression were observed across groups, suggesting PMA specifically enhances p65 phosphorylation.

Effects of PMA on cell viability and apoptosis: CCK-8 assays revealed that Col-Exo treatment significantly enhanced cell viability following H/R injury. As shown in Figure 7D, the cell viability in the H/R group was reduced to 60.00% ± 2.00% of the control group. Col-Exo treatment significantly increased viability to 85.10% ± 6.00%. Co-treatment with PMA partially counteracted the protective effect of Col-Exo, resulting in a cell viability of 70.00% ± 4.00%, which was significantly lower than that of the H/R + Col-Exo group. PMA alone had no significant effect on cell viability (55.00% ± 4.00%). Flow cytometry analysis of apoptosis revealed consistent results. As shown in Figure 7E, the apoptotic rate in the H/R group was 35.67% ± 1.35%, significantly higher than that in the control group (4.34% ± 1.01%). Col-Exo treatment significantly reduced the apoptotic rate to 15.86% ± 1.56%. PMA partially reversed the anti-apoptotic effect of Col-Exo, increasing the apoptotic rate to 28.13% ± 1.89%. Notably, compared with H/R group, PMA treatment alone showed a slight downward trend in cell viability and a slight upward trend in apoptotic rates, though neither reached statistical significance.

Effects of PMA on inflammatory cytokine levels: The levels of pro-inflammatory cytokines TNF-α and IL-1β in cell supernatants were measured using ELISA. As shown in Figure 8A, compared with the control group (normalized to 1.00 ± 0.05 for both cytokines), the H/R group demonstrated significantly increased TNF-α and IL-1β concentrations by (4.00 ± 0.06)-fold and (4.20 ± 0.05)-fold. Col-Exo treatment markedly reduced both levels to (1.50 ± 0.08)-fold (TNF-α) and (1.48 ± 0.03)-fold (IL-1β). Following PMA addition, the anti-inflammatory effect of Col-Exo was partially attenuated, with the TNF-α and IL-1β levels rebounding to (3.50 ± 0.06)-fold and (3.48 ± 0.05)-fold. PMA treatment alone further enhanced cytokine secretion in the H/R group, increasing TNF-α to (4.50 ± 0.04)-fold and IL-1β to (4.80 ± 0.05)-fold.

Figure 8 Lactoferrin and lysozyme abundance correlates with pathway activity and barrier protection. A: Levels of tumor necrosis factor-α and interleukin-1β in cell supernatants; B: Western blot analysis of zonula occludens-1 and occludin; C and D: Pearson correlation analysis between lactoferrin/Lysozyme protein abundance and nuclear factor erythroid 2-related factor 2 expression; E and F: Pearson correlation analysis between lactoferrin/Lysozyme protein abundance and p-p65/t-p65 ratio. ^cP < 0.001. TNF-α: Tumor necrosis factor-α; IL-1β: Interleukin-1β; H/R: Hypoxia/reoxygenation; Nrf2: Nuclear factor erythroid 2-related factor 2; ZO-1: Zonula occludens-1; LTF: Lactoferrin; LYZ: Lysozyme.

Effect of PMA on the expression of tight junction proteins: WB analysis was employed to assess the expression levels of the tight junction proteins ZO-1 and occludin. As depicted in Figure 8B, H/R injury resulted in a significant downregulation of ZO-1 (0.40 ± 0.05) and occludin (0.42 ± 0.01) expression levels; Col-Exo treatment markedly restored the expression levels of both proteins (0.90 ± 0.06 and 0.88 ± 0.04). Upon PMA addition, the tight junction protein upregulation by Col-Exo was markedly attenuated, and PMA alone further exacerbated the H/R-induced downregulation of ZO-1 (0.59 ± 0.06) and occludin (0.60 ± 0.02). This finding indicated that inhibition of the NF-κB pathway constitutes a crucial mechanism by which colostrum-derived exosomes maintain intestinal epithelial barrier integrity.

Correlation analysis of LTF/LYZ protein abundance in exosomes and pathway activity: Pearson correlation analyses were conducted between the relative protein abundance of LTF and LYZ in human milk exosomes from different stages and the Nrf2 protein expression levels and p-p65/t-p65 ratios in cellular models to further investigate whether LTF and LYZ enriched in colostrum exosomes correlate with the activity of pathways they regulate. The results indicated that the protein abundance of LTF and LYZ exhibited a highly significant positive correlation with Nrf2 expression (LTF: r = 0.9987, P < 0.001; LYZ: r = 0.9929, P < 0.001; Figure 8C and D) while exhibiting a highly significant negative correlation with the p-p65/t-p65 ratio (LTF: r = -0.9922, P < 0.001; LYZ: r = -0.9783, P < 0.001; Figure 8E and F). This finding provides further quantitative support that the enrichment of LTF and LYZ in colostrum exosomes may exert superior protective effects against I/R by activating the Nrf2 pathway and inhibiting the NF-κB pathway.

DISCUSSION

This study systematically revealed the lactation stage dependency of human milk exosome protective effects and elucidated their key molecular mechanisms. Dual validation in cellular and animal models showed that colostrum exosomes significantly outperformed transitional and mature milk exosomes in mitigating I/R. This superior protective action is closely associated with their enriched LTF and LYZ content. Mechanistic studies indicated that colostrum exosomes exert antioxidant, anti-inflammatory, and intestinal barrier protective functions by activating the Nrf2/HO-1 antioxidant pathway while inhibiting the NF-κB inflammatory pathway. Pathway inhibitor and activator rescue experiments further confirmed that regulation of these two pathways constitutes an essential component of colostrum exosome protective effects. Correlation analysis revealed strong associations between LTF/LYZ protein abundance and pathway activity, preliminarily establishing a stage-specific logical chain linking proteins, pathway activation, and protective efficacy.

The stage-specific functional differences in human milk exosomes represent one of the core findings of this study. The particle and protein concentrations in colostrum exosomes were significantly higher than those in transitional and mature milk, with LTF and LYZ exhibiting the highest expression levels in colostrum exosomes. This result is corroborated by multiple recent studies. Gao et al[8] compared the effects of colostrum and mature milk exosomes in an experimental necrotizing enterocolitis model and found that colostrum exosomes were more effective in promoting intestinal epithelial proliferation and gut regeneration. Proteomics analysis by Picáns-Leis et al[29] further indicated that extracellular vesicles from colostrum are enriched in proteins associated with immune protection and inflammatory regulation (e.g., TENA, TSP1, and OLF4), whereas those from mature milk exhibit a preference for proteins linked to metabolic processes. Hussain et al[30] further emphasized that the lactation stage constitutes a critical determinant of the biological activity of breast milk-derived exosomes, with significant differences observed in the immunomodulatory cargo between colostrum and mature milk exosomes. Collectively, these studies support the core proposition of the present research: The functional properties of human milk exosomes exhibit lactation stage dependency, with colostrum exosomes demonstrating enhanced gut protective potential due to their unique protein cargo. Notably, the superior efficacy of colostrum-derived exosomes may be attributed not only to their highest levels of LTF and LYZ but also to the synergistic effects of these proteins within the exosomes. Colostrum-derived exosomes are enriched with multiple immunomodulatory proteins, which may act synergistically to target the same core signaling network (such as Nrf2/NF-κB) and exert their effect. Furthermore, recent studies have shown that colostrum-derived exosomes outperformed mature milk-derived exosomes in terms of quantity, protein concentration, and their ability to inhibit apoptosis and inflammation[31]. They also exhibited greater stability in the gastrointestinal tract and more efficient at delivering bioactive cargo to the intestine[32]. More importantly, colostrum-derived exosomes have been shown to be efficiently taken up by human intestinal epithelial cells[33]. Their unique surface protein profile may make them more easily recognized and internalized by target cells, thereby enabling more effective delivery of bioactive cargo.

Regarding the roles of LTF and LYZ in gut protection, this study provides novel quantitative evidence through correlation analysis. The results revealed that LTF and LYZ, enriched in colostrum exosomes, exhibited a highly significant positive correlation with Nrf2 pathway activity and a highly significant negative correlation with NF-κB pathway activity. A notable detail that the correlation coefficient is close to 1, suggesting that LTF and LYZ may be closely associated with Nrf2/HO-1 activation and NF-κB inhibition. This study is the first quantitative report demonstrating such a strong association between exosomal protein abundance and pathway activity, providing an important statistical basis for further validation of the role of LTF/LYZ in the protective effects of colostrum exosomes. In the field of LTF research, studies have reported that LTF can mitigate lipopolysaccharide-induced inflammatory responses in intestinal epithelial cells, maintain cellular barrier integrity, and alleviate oxidative stress by regulating the NF-κB/mitogen-activated protein kinase pathway and activating Nrf2[34]. Research has demonstrated that LTF exerts therapeutic effects in a mouse model of Alzheimer’s disease combined with ulcerative colitis by modulating gut microbiota abundance, proving that LTF can regulate the intestinal microbiome via the brain-gut axis[35]. The research further confirmed that LTF performs a critical function in the intestinal mucosal immune system and inflammatory bowel disease, suppressing NF-κB-mediated inflammatory responses and restoring intestinal barrier integrity[36]. In LYZ research, recent studies have revealed that LYZ, as the most abundant protein in eggshell membranes (accounting for 47%), significantly alleviated weight loss, diarrhea, and fecal blood in DSS-induced colitis models while improving disease activity indices[37]. Another study in broiler chickens demonstrated that dietary LYZ supplementation markedly enhanced the intestinal mucosal barrier, increased villus height and the villus height/crypt depth ratio, promoted the expression levels of tight junction proteins occludin and claudin-1, and simultaneously boosted antioxidant capacity[38]. These studies, together with the findings of the present study, suggest that LTF and LYZ may represent key protein components mediating the protective effects of colostrum exosomes by regulating the Nrf2/HO-1 and NF-κB pathways.

The dual regulation of the Nrf2/HO-1 and NF-κB pathways constitutes the core mechanism of this study. The results showed that colostrum exosomes significantly activate the Nrf2/HO-1 antioxidant pathway while effectively suppressing NF-κB p65 phosphorylation levels. The expression levels of Nrf2 and HO-1 in the H/R group were slightly higher than those in the control group. This phenomenon may represent a compensatory stress response to oxidative damage, a common occurrence in I/R models, where the endogenous antioxidant defense system is transiently upregulated to counteract excessive ROS production. Multiple studies have demonstrated that in I/R models of the kidney, heart, brain, and liver, I/R or H/R treatment alone can induce upregulation of Nrf2 and HO-1 expression levels, which is considered a self-protective response of cells under stress[3,39,40]. However, this endogenous compensatory response is insufficient to fully protect cells from H/R-induced damage. By contrast, colostrum-derived exosomes further amplify the activation of Nrf2/HO-1, increasing it to a level sufficient to exert a potent protective effect. Functional rescue experiments further confirmed that blocking the Nrf2 pathway or activating the NF-κB pathway partially counteracted the protective effects of colostrum exosomes on cellular viability, apoptosis, oxidative stress, inflammatory responses, and tight junction proteins. This finding indicated that both pathways jointly contribute to the protective effects of colostrum exosomes. A complex interplay exists between Nrf2 and NF-κB. Studies have shown that activation of Nrf2 inhibited the nuclear translocation of NF-κB and the expression levels of its downstream pro-inflammatory genes, whereas activation of NF-κB suppressed the transcriptional activity of Nrf2[41,42]. Colostrum-derived exosomes may act at the intersection of these two pathways, simultaneously activating antioxidant defenses and suppressing inflammatory responses. This finding aligns with those of multiple recent studies in the field of intestinal injury protection: NADPH mitigated intestinal I/R injury by activating the Nrf2/HO-1 pathway[3]; angiotensin-(1-7) exerted protective effects by enhancing Nrf2/HO-1 while inhibiting the NF-κB/NLRP3 pathway[43]; and inhibition of the NF-κB pathway effectively mitigated intestinal I/R injury[44-46]. Unlike the aforementioned studies that predominantly focused on individual pathways, the present study systematically validated at the functional level that the Nrf2/HO-1 and NF-κB pathways play crucial roles in the protective effects of colostrum exosomes. This finding was achieved through bidirectional rescue experiments using ML385 and PMA.

This study has certain limitations. First, correlation analysis suggests a strong association between LTF/LYZ abundance and pathway activity. However, causal relationship has not been confirmed through gene interference or neutralizing antibody experiments. Knockdown or knockout strategies could be employed in the future to further validate this finding. Second, this study primarily focuses on protein cargo. The roles of other bioactive components in exosomes (such as microRNAs and lipids) remain to be explored in future research. Furthermore, the mechanistic exploration in this study primarily revolved around the Nrf2/HO-1 and NF-κB pathways. In the future, multi-omics analyses (such as proteomics, transcriptomics, and metabolomics) could be integrated to construct a more comprehensive molecular map, thereby systematically elucidating the multifaceted regulatory network of colostrum exosomes. Finally, the observation period in the animal model was relatively short (6 hours after reperfusion), making it impossible to assess long-term effects or potential long-term outcomes. Future studies should extend the follow-up period (e.g., to 7 days or 14 days) and include pharmacokinetic and toxicological studies.

CONCLUSION

This study demonstrates that the protective efficacy of human milk exosomes against intestinal I/R injury exhibits lactation stage dependency. Colostrum-derived exosomes, being rich in LTF and LYZ, more effectively activate the Nrf2/HO-1 pathway while inhibiting the NF-κB pathway, thereby exerting optimal protective effects. This finding provides experimental evidence for developing novel strategies to protect the neonatal gut. Colostrum-derived exosomes or bio-mimetic formulations enriched with LTF/LYZ hold promise as novel interventions for preventing and treating neonatal intestinal injury diseases such as necrotizing enterocolitis.