Glutamine alleviates feed restriction-induced barrier function injury in yak rumen epithelium

Abstract

The yak (Bos grunniens) is a crucial species for herders on the Qinghai–Tibet Plateau. Cold-season forage scarcity often leads to starvation, impairing the rumen epithelial barrier function and growth performance. Glutamine (Gln) has been shown to mitigate injury to the gastrointestinal tract barrier function, but its efficacy and mechanism against feed restriction (FR)-induced barrier injury in yaks remains unclear. This study combined in vivo and in vitro experiments to investigate this. Twenty-four healthy yaks (31 months, 265.35 ± 25.81 kg) were randomly divided into three groups (n = 8 per group) and fed for 60 d : the control (Con, ad libitum access to the basal diet), FR (50% of the basal diet), and FR + Gln (50% of the basal diet in the first month, with 1% Gln added to the 50% basal diet in the second month). Yak rumen epithelial cells (YRECs) were treated under four conditions: Control (normal medium), Gln (normal medium + 10 mmol/L Gln), Gln deficiency (Gln-D, Gln-free medium), and Gln-D + Gln (Gln-free medium + 10 mmol/L Gln). Results showed that FR significantly decreased the average daily gain (ADG) and the Gln concentration in ruminal epithelial tissue and serum (P < 0.05) and increased serum levels of permeability parameters (diamine oxidase, D-lactate, lipopolysaccharide, and histamine) compared to the Con group (P < 0.05). In addition, FR in vivo or Gln-D in vitro significantly down-regulated the protein expression of tight junction (TJ) proteins (ZO-1, occludin, and claudin-1) (P < 0.05) and enhanced the phosphorylation of nuclear factor-κB p65 (NF-κB p65), p38 mitogen-activated protein kinase (p38 MAPK), and c-junN-terminal kinase (JNK) (P < 0.05). Gln supplementation effectively reversed these changes, significantly reducing serum permeability (P < 0.001), restoring TJ protein expression, and suppressing the phosphorylation of NF-κB p65, p38 MAPK, and JNK compared to the FR or Gln-D groups (P < 0.01). Crucially, the ability of Gln to restore TJ integrity was abolished by p38 MAPK and JNK activators in YRECs (P < 0.001). Overall, this suggests that dietary Gln supplementation alleviates FR-induced rumen epithelial barrier damage in yaks through the MAPK/NF-κB pathway.

1. Introduction

Yak (Bos grunniens), primarily found on the Qinghai-Tibet Plateau at altitudes of 3000 to 6000 m, is crucial for local herders, providing meat, milk, wool, skin, and fuel (Ma et al., 2021a,b). China has the largest population of yaks worldwide, accounting for more than 90% of yaks globally (Yi et al., 2023). There are over 16 million existing yaks worldwide, with approximately 14 million distributed in China (Zhang et al., 2025a). Research indicates that yaks have better fiber digestibility, ruminal volatile fatty acid concentration, and energy metabolism than low-altitude cattle, allowing them to utilize low-energy diets more effectively (Hu et al., 2021; Liu et al., 2023). However, despite the ability of yaks to endure extreme conditions such as hypoxia, low temperatures, and intense ultraviolet radiation, malnutrition occurs during the cold season each year compared to feedlot cattle (Liu et al., 2022b). This phenomenon is primarily due to the unique natural environment of the Qinghai–Tibet Plateau, which results in reduced pasture yield and quality during the cold season (Liu et al., 2020). Therefore, grazing yaks always encounter undernutrition and starvation, while the nutritional requirements of yaks significantly elevate because of the low ambient temperatures (Ma et al., 2021a). This leads to negative weight gain, high mortality in the cold season, and an extended growth cycle for yaks (Yi et al., 2023). Previous research has demonstrated that starvation decreased the average daily gain (ADG) of yaks and disrupted lipid metabolism homeostasis by inhibiting lipogenesis and accelerating lipolysis in the yak liver (Yu et al., 2016). Moreover, severe starvation of yaks during early life caused growth-retarded yaks (Ma et al., 2021a). It has been reported that starvation damaged ruminal epithelial barrier function via down-regulated zonula occludens-1 (ZO-1) and claudin-1 mRNA expression levels in pregnant ewes (Hu et al., 2018). However, the impact of starvation on the rumen epithelial barrier function in yaks has been scarcely investigated.

The rumen is a unique digestive organ that influences the growth and health of ruminants (Xue et al., 2020). The ruminal epithelium is crucial for nutrient digestion and serves as a barrier to prevent the entry of endotoxins into the body (Yang et al., 2022a). Impairment of the rumen epithelium barrier function leads to increased epithelial permeability and the translocation of endotoxins into the bloodstream, resulting in systemic inflammation (Ri-gele et al., 2023). Tight junctions (TJs) between the epithelial cells are one of the critical structures of epithelial barrier function (Hagen et al., 2018). Previous research has demonstrated that yaks experiencing growth retardation due to chronic cold-season starvation exhibit compromised rumen epithelium (Ma et al., 2020). However, the mechanisms by which starvation disrupts the rumen epithelial barrier function have not been thoroughly investigated. A study on rats has shown that starvation reduced the intestinal glutamine synthetase activity, thereby decreasing glutamine (Gln) concentration (Kong et al., 2000). In addition, Gln has been found to mitigate TJ damage in methotrexate-treated human colorectal adenocarcinoma (Caco-2 cells) (Beutheu et al., 2013). Consequently, this study aimed to investigate whether Gln could alleviate TJ disruption induced by feed restriction in the yak rumen epithelium.

As the most abundant metabolic substrate in animal serum (Zhang et al., 2024), Gln has been demonstrated to promote gastrointestinal development and enhance the barrier function of the rumen and intestinal epithelium (Ma et al., 2021a; Singh et al., 2020). Research showed that Gln reduced the secretion of inflammatory cytokines and increased the expression of TJ proteins in the intestinal epithelial tissue of broiler chickens under stress (Wu et al., 2018). Furthermore, a previous study found that dietary supplementation with Gln increased the TJ protein expression in rumen epithelium, promoting the growth performance of growth-retarded yaks (Ma et al., 2021a). Nevertheless, the underlying mechanism of Gln augmenting rumen epithelium TJ protein expression in feed-restricted yaks remains unclear. It has been shown that activation of the nuclear factor-κB (NF-κB) pathway could induce inflammatory response and TJ injury in porcine jejunal epithelial cells (IPEC-J2 cells) (Ge et al., 2020). Furthermore, mitogen-activated protein kinase (MAPK) activation has been reported to induce the secretion of NF-κB-mediated inflammatory cytokines (Hsia et al., 2018). Therefore, we hypothesized that Gln could promote TJ protein expression in the rumen epithelium of feed-restricted yaks by inhibiting the activation of the MAPK/NF-κB signaling pathway. To evaluate this hypothesis, we investigated the mechanisms through which Gln enhances TJ protein expression in the rumen epithelium of feed-restricted yaks, employing both in vivo and in vitro assays. This study provided a theoretical basis for the mitigation of rumen epithelial damage of yaks under starvation during the cold season on the Qinghai–Tibet Plateau using Gln.

2. Materials and methods

2.1. Animal ethics statement

The animal trial and all experimental procedures were approved by the Animal Care and Use Committee of Sichuan Agricultural University (approval No.: SCAUAC-2021-114-009). According to State Council Decree No. 676, Regulations on the Administration of Laboratory Animals (2017 Revision) were followed for the animal experiment.

2.2. Animal, experimental design, and diet

This study was implemented at the Ya'an experimental site of the Animal Nutrition Institute, Sichuan Agricultural University, Ya'an, Sichuan, China. A total of 24 healthy female yaks (age: 31 months old, body weight: 265.35 ± 25.81 kg) were selected and randomly divided into three groups. All yaks were raised in a formulated diet on the modern farm, and each yak was housed in an individual pen (1.2 m × 2.2 m) with an individual feed trough and water trough. Before the experiment, all yaks were marked with ear tags, immunized, and dewormed. The feeding experiment lasted for 75 d, including the adaptation period (−15 to 0 d) and formal study period (1 to 30 d and 31 to 60 d two stages). From −15 to 0 d, all yaks were given a basal diet ad libitum. From 1 to 30 d, yaks in the control (Con) group had ad libitum access to the basal diet, and yaks in the feed restriction (FR) group and feed restriction + glutamine (FR + Gln) group were offered a 50% basal diet of the dry matter intake (DMI) of the adaptation period. From 31 to 60 d, yaks in the Con group had ad libitum access to the basal diet, yaks in the FR group were offered a 50% basal diet of the DMI of the adaptation period, and yaks in FR + Gln group were offered a 50% basal diet of the DMI of the adaptation period with 1% added Gln (99% purity, Cat. # chx-51, Fufeng Biotechnologies Co., Ltd., Binzhou, Shandong, China) to the basal diet. Yaks were fed twice daily (08:00 and 16:00) with free access to water.

The basal diet was formulated according to the Feeding Standard of Beef Cattle (NY/T 815-2004). The feed composition and nutrient levels of the basal diets are shown in Table S1 (Yue et al., 2025). All yak feeding was performed regularly twice a day by total mixed ration (TMR). The crude protein (CP) content was determined by the Kjeldahl method (GB/T 6432-2018; China National Standard, 2018b) using an automatic Kjeldahl nitrogen analyzer (OLB9870A, Biobase Biodustry Co., Ltd., Ji'nan, Shandong, China). The neutral detergent fiber (NDF) (GB/T 20806-2022; China National Standard, 2022) and acid detergent fiber (ADF) (NY/T 1459-2022) contents were determined by the filtration method using a reflux-digestion device and suction filtration apparatus. The calcium (Ca) content was determined by the potassium permanganate titration method (GB/T 6436-2018; China National Standard, 2018a). The phosphorus (P) content was determined by spectrophotometry (GB/T 6437-2018; China National Standard, 2018c) using an ultraviolet–visible spectrophotometer (400 nm, DU 730, Beckman Coulter, Pasadena, CA, USA). Net energy for maintenance (NEm) and total digestible nutrients (TDN) were calculated according to the guidelines of NASEM (2016) with the specific formulas as follows:

$$\text{NEm}=1.37\text{ME}–0.138{\text{ME}}^2+0.0105{\text{ME}}^3–1.12\text{,}$$

where ME = metabolizable energy;

$$\text{TDN}{}_{\text{Roughage}}=0.98\times \left\{100–\left[\left(\text{NDF}–\text{NDICP}\right)+\text{CP}+\text{EE}+\text{Ash}\right]\times \text{PAF}\right\}+\left[\text{CP}\times {\mathrm{e}}^{-1.2\times \left(\text{ADICP}/\text{CP}\right)}\right]+\left(\text{EE}–1\right)\times 2.25+0.75\times \left[\left(\text{NDF}–\text{NDICP}\right)–\text{Lignin}\right]\times \left\{1–{\left[\text{Lignin}/\left(\text{NDF}–\text{NDICP}\right)\right]}^{0.667}\right\}–7\text{,}$$

where NDICP = neutral detergent insoluble crude protein; EE = ether extract; PAF = processing adjustment factor; ADICP = acid detergent insoluble crude protein;

$$\text{TDN}{}_{\text{Concentrate}}=0.98\times \left\{100–\left[\left(\text{NDF}–\text{NDICP}\right)+\text{CP}+\text{EE}+\text{Ash}\right]\times \text{PAF}\right\}+\left[\text{CP}\times \left(1–0.4\times \text{ADICP}/\text{CP}\right)\right]+\left(\text{EE}–1\right)\times 2.25+0.75\times \left[\left(\text{NDF}–\text{NDICP}\right)–\text{Lignin}\right]\times \left\{1–{\left[\text{Lignin}/\left(\text{NDF}–\text{NDICP}\right)\right]}^{0.667}\right\}–7\text{.}$$

2.3. Growth performance

Before the morning feeding, all yaks were weighed on d 1, 30, and 60. Feed consumption was recorded daily for each yak. Then, these values were used to calculate the ADG and DMI as per the following formula:

ADG (kg/d) = [Final body weight (kg) – Initial body weight (kg)]/Days;

DMI (kg/d) = The total of dry matter intake/Days.

2.4. Serum sample collection and measurement

Before the morning feeding, blood samples from the caudal vein were collected on d 1, 30, and 60 using the vacuum blood collection tube. The blood samples were centrifuged (3000 × g, 4 °C, 15 min) and stored at −20 °C pending further analysis. Then, the serum concentrations of Gln (Cat. # MM-5115701, Jiangsu enzyme immunoassay Industry Co., Ltd., Nanjing, Jiangsu, China), lipopolysaccharide (LPS; Cat. # RXJ1600917B, Ruixin Biological Technology Co., Ltd., Quanzhou, Fujian, China), D-lactate (D-LA; Cat. # KTB1111, Abbkine Scientific Co., Ltd., Wuhan, Hubei, China), and histamine (HIS; Cat. # RXJ1600735B, Ruixin Biological Technology Co., Ltd., Quanzhou, Fujian, China), and the activity of diamine oxidase (DAO; Cat. # KTB1220, Abbkine Scientific Co., Ltd., Wuhan, Hubei, China), were detected using biochemical kits and enzyme-linked immunosorbent assay (ELISA) kits.

2.5. Rumen epithelial tissue sample collection and measurement

After the trial period, yaks were humanely slaughtered following the Operating procedure of livestock and poultry slaughtering-Cattle (GB/T 19477-2018; China National Standard, 2018d). The rumen epithelial tissue was collected after the digesta was removed. For transmission electron microscopy (TEM) analysis, the ruminal epithelial samples (1 cm × 0.5 cm) were placed into 2.5% glutaraldehyde. Then, samples were postfixed in osmium for 2 h, dehydrated with acetone, and embedded in epoxy resin for sectioning, followed by standard staining in uranyl acetate and lead salts. Then, the TJs were observed using TEM (H7700 Transmission Electron Microscope, Hitachi, Tokyo, Japan). In addition, the ruminal epithelial samples were packed into tubes and stored at −80 °C for Gln concentration assays and molecular determination. The total proteins in ruminal epithelial samples were detected using the bicinchoninic acid (BCA) method using a biochemical kit (Cat. # KTD3001, Abbkine Scientific Co., Ltd., Wuhan, Hubei, China).

2.6. Cell culture and treatment

Feed restriction affects more than just the Gln concentration in yak rumen epithelium. To ascertain the critical role of Gln in this process, this study further investigated the effects of feed restriction-induced insufficient endogenous Gln synthesis on yak rumen epithelial cells, as well as the protective effect of exogenous Gln supplementation. Thus, yak rumen epithelial cells were selected for the subsequent experiments.

The yak rumen epithelial cell line (YRECs) used in this study was established in-house and authenticated by a Chinese patent (Patent Number: LU502505) (Wang et al., 2024). The YRECs were cultured in RPMI-1640 medium (Cat. #C22400500BT, Thermo Fisher Scientific Co., Ltd., Shanghai, China) supplemented with 10% fetal bovine serum (FBS; Cat. # 10099-141, Thermo Fisher Scientific Co., Ltd., Shanghai, China) and 1% penicillin-streptomycin-amphotericin (Cat. # G4015, Servicebio Technology Co., Ltd., Wuhan, Hubei, China) held at 37 °C in a 5% CO₂ incubator. Yak rumen epithelial cells used in this study were between passages 20 and 30.

In order to investigate the suitable glutamine concentration and culture time, the following experiments were conducted. First, the medium was prepared. Control medium: RPMI-1640 medium (with Gln) supplemented with 10% serum replacement (SR; Gln-free, Cat. # C10828028BT, Thermo Fisher Scientific Co., Ltd., Shanghai, China) and 1% penicillin-streptomycin-amphotericin. Gln-free medium: Gln-free RPMI-1640 medium (Cat. #C21870076BT, Thermo Fisher Scientific Co., Ltd., Shanghai, China) + 10% SR + 1% penicillin-streptomycin-amphotericin + 4 mmol/L L-methionine-DL-sulfoximine (MSO; Cat. # HY-B1692, MedChemExpress Co., Ltd., Shanghai, China). For YRECs, they were cultured in the control medium for 6, 12, 24, and 48 h, respectively. For Gln deficiency YRECs, they were cultured in Gln-free medium for 12 h and then supplemented with 0, 5, 10, 20, or 40 mmol/L Gln (99% purity, Cat. # G8230, Solarbio Science & Technology Co., Ltd., Beijing, China) treatment for 6, 12, 24, and 48 h, respectively. Cell viability was detected, by which the most suitable treatment concentration and time of Gln were determined. In addition, the treatment concentration of Gln was determined by testing the cell proliferation and the mRNA expression of TJs and inflammatory factors.

To investigate the effects of Gln on inflammation and TJ in YRECs, a 36-h experiment was conducted with four treatment groups. All groups underwent an initial 12-h culture period, followed by a 24-h treatment period in fresh medium with the respective modifications: the Con group was cultured in complete medium during both periods; the Gln group was cultured in complete medium, with Gln added only during the 24-h treatment period to assess its effect under normal conditions; the Gln deficiency (Gln-D) group was cultured in Gln-free medium during both periods to induce deficiency; and the Gln deficiency + Gln (Gln-D + Gln) group was cultured in Gln-free medium for the first 12 h to establish deficiency, then switched to Gln-free medium supplemented with Gln for the subsequent 24 h to examine recovery.

To explore the mechanism by which Gln alleviates Gln-D-induced TJ injury, the experimental design consisted of the following groups. The Con group, in which YRECs were cultured in a control medium for 12 h and cultured for another 24 h in a fresh control medium. The Gln deficiency (Gln-D) group, in which YRECs were cultured in Gln-free medium for 12 h and cultured for another 24 h in fresh Gln-free medium. The Gln deficiency + Gln (Gln-D + Gln) group, in which YRECs were cultured in Gln-free medium for 12 h and cultured for another 24 h in Gln-free medium supplemented with Gln. The Gln deficiency + Gln + p38 MAPK/c-Jun N-terminal kinase (JNK) activator (metformin) (Gln-D + Gln + metformin) group, in which YRECs were cultured in Gln-free medium for 12 h and then cultured for another 24 h in Gln-free medium supplemented with Gln and metformin. To determine the treatment concentration of metformin, the mRNA expression of p38 MAPK and JNK in YRECs treated with metformin (0, 2.5, 5, and 10 μmol/L) was detected.

2.7. Cell viability analysis

Based on instruction provided by the manufacturer, YRECs viability was tested using a cell counting kit-8 (CCK-8) Cell Proliferation and a Cytotoxicity Assay Kit (Cat. # KTA1020, Abbkine Scientific Co., Ltd., Wuhan, Hubei, China). Briefly, after YRECs were given different treatments, 10 μL of CCK-8 solution was poured into each well, and the solution was incubated at 37 °C for 1 h, then, a microplate reader was used to measure the optical density value at 450 nm.

2.8. Cell proliferation

According to the manufacturer's instruction, YRECs proliferation was tested using 5-ethynyl-2′-deoxyuridine (EdU) Cell Proliferation Assay Kit (Cat. # KTA2030, Abbkine Scientific Co., Ltd., Wuhan, Hubei, China). Briefly, EdU solution was added to each well, followed by 1 h at 37 °C with 5% CO₂. Using 0.2% Triton X-100, cells were permeabilized after 4% paraformaldehyde (Cat. #P1110, Solarbio Science & Technology Co., Ltd., Beijing, China) fixation. As directed by the supplier, the mixture solution was prepared. At room temperature, YRECs were incubated with the mixture solution for 1.5 h. Afterward, the nucleus of YRECs was stained with 4′,6-diamidino-2-phenylindole (DAPI) and observed using an inverted fluorescence microscope (Leica, Wetzlar, Germany).

2.9. RNA extraction and reverse-transcription quantitative PCR (RT-qPCR)

A SteadyPure Quick RNA Extraction kit (Cat. # AG21101, Accurate Biology Co., Ltd., Changsha, Hunan, China) was used to extract the total RNA of the rumen epithelial tissues or culture-derived homogenized YRECs. The total RNA was reverse transcribed into cDNA using a Reverse Transcription Kit (Cat. # A502-01, Exongen Biotechnology Co., Ltd., Chengdu, Sichuan, China). Then, a real-time q-PCR system (QuantStudio 5, Applied Biosystem, Foster City, CA, USA) performed q-PCR with the SYBR Green q-PCR Master Mix kit from Servicebio Technology Co., Ltd. (Cat. #G3328, Wuhan, Hubei, China). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was chosen as the internal control. Table 1 illustrates the primer sequences used for quantitation of target genes. The primers were designed and synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). Then, relative gene expression levels were calculated from the RT-qPCR data using the 2^−ΔΔCt method.

Table 1.

Real-time polymerase chain reaction primer sequences.

Gene	Primer sequence (5′-3′)	Length, bp	Accession number
GAPDH	F: CGGCACAGTCAAGGCAGAGAAC	116	XM_014482068.1
	R: CCACATACTCAGCACCAGCATCAC
IL-1β	F: CTCCGACGAGTTTCTGTGTGACG	120	NM_174093.1
	R: GAGAGGAGGTGGAGAGCCTTCAG
IL-6	F: CACTGACCTGCTGGAGAAGATGC	115	XM_005901249.2
	R: CCGAATAGCTCTCAGGCTGAACTG
TNF-α	F: TGAAGGAAGAGGAGAGGCTCATCG	107	XM_005904178.1
	R: GTGGTCATCGGAGTTGCTGGTG
TLR4	F: CTTCAGGAACGCCACTTGTCAG	136	XM_005891938
	R: GCAGCCAGCAAGAAGCATCAG
NF-κB	F: GCCTGCTGAATGCTCTGTCTGAC	143	XM_005887214.2
	R: CTCTGTTTCCTGTTCCACCGACTG
IL-10	F: GAACCACGGGCCTGACATCAAG	127	XM_005891650
	R: CTTCTCCACCGCCTTGCTCTTG
ZO-1	F: CCGAATGAAACCGCACACAAACC	107	XM_014476599.1
	R: GTCTCCACGCCACTGTCAAACTC
Occludin	F: GCCTGTGTTGCCTCCACTCTTG	143	XM_005889348.2
	R: CCATAGCCATAACCGTAGCCATAGC
Claudin-1	F: CCCGTGCCTTGATGGTGATTGG	110	XM_005897671.2
	R: CATCTTCTGTGCCTCGTCGTCTTC
JAM-A	F: GTGCCTCCATCCAAGCCTACAATC	134	XM_010802736.3
	R: GGCATCTCTACTCCATCCTTGAACC
p38 MAPK	F: TGCTGGAGAAGATGCTTGTATTGG	94	XM_005890506.2
	R: TCGTCGTCAGGATCGTGGTAC
JNK	F: CAGAAGCAAGCGTGACAGCA	130	XM_005900620.1
	R: TTCCTTGGGCTCCTGAACCT
ERK1/2	F: AGGTGTGGTGTTCAAGGTCTCC	100	XM_005893283.2
	R: TGATCTCGCCGTCGCTGTAG

GAPDH = glyceraldehyde-3-phosphatedehydrogenase; IL-1β = interleukin-1β; IL-6 = interleukin-6; TNF-α = tumor necrosis factor-alpha; TLR4 = Toll like receptor 4; NF-κB = nuclear factor-κB; IL-10 = interleukin-10; ZO-1 = zonula occludens-1; JAM-A = junctional adhesion molecule-A; p38 MAPK = p38 mitogen-activated protein kinase; JNK = c-Jun N-terminal kinase; ERK1/2 = extracellular signal-regulated 1/2.

2.10. Western blot

From rumen epithelial tissues or culture-derived homogenized YRECs, 20 μg of proteins were isolated for western blot analysis. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was used to separate proteins, and polyvinylidene difluoride (PVDF) membranes were electrophoretically transferred. At 4 °C, primary antibodies were applied overnight, while secondary antibodies were applied for 1.5 h. After that, an enhanced chemiluminescence developer (ECL; Cat. # PD203, Oriscience Biological Technology Co., Ltd., Chengdu, Sichuan, China) was used for membrane visualization, and the analysis of the protein bands grayscale values was performed by Image J software (Wang et al., 2016). The antibody information is given in Table S2. GAPDH was chosen as the internal control.

2.11. Immunofluorescence

The rumen epithelial tissues were fixed in 4% paraformaldehyde. The section was prepared using hematoxylin and eosin staining, which was dewaxed and rehydrated with stepwise alcohol. The YRECs were fixed in 4% paraformaldehyde and then treated in 1% Triton X-100. Then, the sections and YRECs were blocked with 5% goat serum (Cat. # E510009-0100, Sangon Biotech Co., Ltd., Shanghai, China) for 1 h and incubated overnight at 4 °C with an appropriate primary antibody for ZO-1 (Cat. # A0659, ABclonal Biotech Co., Ltd., Wuhan, Hubei, China) (1:100). The appropriate secondary antibodies (Cat. # AS079, ABclonal Biotech Co., Ltd., Wuhan, Hubei, China) were incubated at room temperature for 1 h with a dilution of 1:500. Each step was followed by three phosphate buffered saline (PBS) washes (5 min each). Nuclei were counterstained with DAPI for 5 min. Finally, the sections and YRECs were observed by an inverted fluorescence microscope (Leica, Wetzlar, Germany) and fluorescence intensity was analyzed using Image J software (Wang et al., 2016).

2.12. Statistical analysis

The results of all experiments were analyzed by one-way analysis of variance (ANOVA) followed by Duncan's multiple comparisons (SPSS 27.0, IBM, Chicago, IL, USA). Subsequently, the significance of growth performance, permeability parameters, and Gln concentration between FR and FR + Gln groups within the in vivo study were analyzed by the F-test (SPSS 27.0, IBM, Chicago, IL, USA). The GraphPad Prism 8.0 (GraphPad, La Jolla, CA, USA) software was used to create all graphs. The data were expressed as mean and standard error of the mean. Statistical significance was defined as P < 0.05.

The mathematical model for one-way ANOVA is expressed as:

$$Y_{ij}=\mu +T_i+e_{ij}\text{,}$$

where Y_ij is the observation of dependent variables; μ is the overall mean; T_i is the treatment effect; e_ij is the random error.

3. Results

3.1. Growth performance

The growth performance for feed restriction and Gln supplementation is summarized in Table 2. The body weight (BW) of yaks in the FR and FR + Gln groups was significantly lower than that in the Con group at 60 d (P = 0.012). Average daily gain was significantly decreased in FR and FR + Gln groups than that in the Con group from 1 to 30 d and 1 to 60 d (P < 0.001). From 31 to 60 d, FR significantly decreased the ADG, and supplementation with Gln significantly alleviated it (P < 0.001).

Table 2.

Effects of feed restriction and Gln supplementation on growth performance of yaks.

Item	Treatment1			SEM	P-value
	Con	FR	FR + Gln		ANOVA	FR vs (FR + Gln)
BW, kg
1 d	265.38	268.38	262.31	5.486	0.904	0.657
30 d	262.06	249.14	245.06	5.324	0.401	0.761
60 d	262.56a	229.14b	228.75b	5.989	0.012	0.984
ADG, kg/d
1–30 d	−0.11a	−0.61b	−0.57b	0.059	<0.001	0.686
31–60 d	0.02a	−0.67c	−0.54b	0.068	<0.001	0.047
1–60 d	−0.05a	−0.64b	−0.56b	0.060	<0.001	0.193
DMI, kg/d
1–30 d	2.92a	1.52b	1.43b	0.061	<0.001	0.399
31–60 d	3.02a	1.52b	1.43b	0.036	<0.001	0.399
1–60 d	2.97a	1.52b	1.43b	0.036	<0.001	0.399

SEM = standard error of the mean; BW = body weight; ADG = average daily gain; DMI = dry matter intake.

Results were expressed as mean and SEM (n = 8). Within a row, means with different superscripts indicated significantly different (P < 0.05).

¹Con = control group; FR = feed restriction group; FR + Gln = feed restriction + Gln group.

3.2. Permeability parameters in serum

Serum permeability parameters were measured to reflect gastrointestinal damage. As shown in Table 3, the 30-d and 60-d FR groups showed higher activity of DAO and higher concentrations of LPS and D-LA in serum of yaks (P < 0.001), while FR only significantly increased the concentration of HIS in serum in 60 d (P < 0.001). With d 60, dietary Gln supplementation moderately improved the negative effect of FR on the activity of DAO and the concentrations of LPS, D-LA, and HIS in serum (P < 0.001).

Table 3.

Effects of feed restriction and Gln supplementation on permeability parameters in the serum of yaks.

Item	Treatment1			SEM	P-value
	Con	FR	FR + Gln		ANOVA	FR vs (FR + Gln)
DAO, U/mL
1 d	12.47	13.93	15.09	1.144	0.648	0.705
30 d	11.13b	26.21a	24.44a	1.880	<0.001	0.590
60 d	11.60b	16.20a	12.68b	1.575	<0.001	0.003
D-LA, μmol/mL
1 d	0.625	0.545	0.541	0.0223	0.209	0.953
30 d	0.695b	0.882a	0.854a	0.0246	0.001	0.543
60 d	0.718b	1.143a	0.757b	0.0554	<0.001	<0.001
LPS, EU/L
1 d	125.08	122.89	126.11	2.347	0.860	0.595
30 d	108.71b	129.72a	129.40a	3.070	0.001	0.955
60 d	93.29b	106.29a	96.96b	1.624	<0.001	<0.001
HIS, ng/mL
1 d	2.640	2.249	2.326	0.0614	0.055	0.628
30 d	2.028	1.961	1.964	0.0212	0.397	0.953
60 d	1.613c	2.875a	2.241b	0.1424	<0.001	0.015

SEM = standard error of the mean; DAO = diamine oxidase; D-LA = D-lactate; LPS = lipopolysaccharide; HIS = histamine.

Results were expressed as mean and SEM (n = 8). Within a row, means with different superscripts indicated significantly different (P < 0.05).

¹Con = control group; FR = feed restriction group; FR + Gln = feed restriction + Gln group.

3.3. Gln concentration in serum and ruminal epithelial

The effect of FR and dietary Gln supplementation on Gln concentration in serum and rumen epithelial of yaks was further evaluated. In the present study, as shown in Table 4, compared with the Con group, FR and FR + Gln groups significantly reduced the serum Gln on d 30 (P < 0.001). Moreover, FR highly significantly decreased the Gln concentration in serum on d 60 and rumen epithelial compared to the Con group (P < 0.01), while dietary Gln supplementation improved them (P < 0.01). Therefore, Gln deprivation was used for cell experiments in vitro.

Table 4.

Effects of feed restriction and Gln supplementation on Gln concentration in serum and ruminal epithelium of yaks.

Item	Treatment1			SEM	P-value
	Con	FR	FR + Gln		ANOVA	FR vs (FR + Gln)
Serum, mmol/L
1 d	0.329	0.336	0.330	0.0177	0.986	0.908
30 d	0.339a	0.204b	0.208b	0.0179	<0.001	0.870
60 d	0.323a	0.137b	0.247a	0.0231	0.001	<0.001
Ruminal epithelium, μmol/g protein	22.19a	11.12b	18.82a	1.247	<0.001	<0.001

Gln = glutamine; SEM = standard error of the mean.

Results were expressed as mean and SEM (n = 8). Within a row, means with different superscripts indicated significantly different (P < 0.05).

¹Con = control group; FR = feed restriction group; FR + Gln = feed restriction + Gln group.

3.4. Inflammatory response and TJ in rumen epithelial

To observe the effects of feed restriction and Gln supplementation on barrier function in the rumen epithelium of yaks, the mRNA and protein expression of inflammatory cytokines and TJs-related proteins were determined by TEM, RT-qPCR, western blot, and Immunofluorescence. As shown in Fig. 1A, FR caused damage to the TJs ultrastructure in the yak rumen epithelium, while FR + Gln treatment alleviated the damage. In addition, FR significantly down-regulated the mRNA expression of ZO-1, occludin, claudin-1, claudin-4, and junctional adhesion molecule-A (JAM-A) in yak rumen epithelium. At the same time, FR + Gln treatment alleviated them (P < 0.001) (Fig. 1B). Moreover, FR significantly up-regulated the mRNA expression of IL-1β, IL-6, and TNF-α (P < 0.001). In contrast, FR + Gln treatment alleviated them in yak rumen epithelium (P < 0.001) (Fig. 1C). Further research found that FR significantly decreased the protein levels of occludin, claudin-1, and JAM-A, and the fluorescence intensity of ZO-1 in yak rumen epithelium, while Gln supplementation alleviated them (P < 0.05) (Fig. 1D–G).

Fig. 1.

Effects of feed restriction and glutamine (Gln) Gln supplementation on inflammation reaction and tight junction (TJ) in yak rumen epithelium. (A) The ultrastructure of yak rumen epithelium (magnification 15,000 × and 30,000 × , scale bars represent 2 and 1 μm, respectively). (B) The mRNA expression of ZO-1, occludin, claudin-1, claudin-4, and JAM-A. (C) The mRNA expression of IL-1β, IL-6, TNF-α, and IL-10. (D and E) The protein expression levels of occludin, claudin-1, and JAM-A. (F and G) Representative images of ZO-1 distribution (magnification 200 ×, scale bars represent 200 μm) and fluorescence intensity of ZO-1 protein expression levels. Results were expressed as means and SEM, n = 6 independent experiments. Different letters indicate significant differences (P < 0.05). Con = control group; FR = feed restriction group; FR + Gln = feed restriction + Gln group. ZO-1 = zonula occludens-1; JAM-A = junctional adhesion molecule-A; DAPI = 4′,6-diamidino-2-phenylindole; IL-1β = interleukin-1β; IL-6 = interleukin-6; TNF-α = tumor necrosis factor-α; IL-10 = interleukin-10; GAPDH = glyceraldehyde-3-phosphatedehydrogenase.

3.5. MAPK/NF-κB signaling pathway in yak rumen epithelium

The signaling pathway regulating the barrier function of yak rumen epithelium was further explored. It was found that p38 MAPK, JNK, and NF-κB mRNA expression levels were significantly up-regulated in FR group and these were down-regulated following dietary FR + Gln treatment (P < 0.05) (Fig. 2A). The extracellular signal-regulated 1/2 (ERK1/2) transcriptional level did not significantly change in FR and FR + Gln groups compared to the Con group (P = 0.499) (Fig. 2A). Furthermore, the western blot results showed that FR significantly increased the phosphorylation levels of p38 MAPK, JNK, nuclear factor-κB p65 (NF-κB p65), and inhibitor of NF-κB (IκB) and protein expression of Toll-like receptor 4 (TLR4) (P < 0.05). In contrast, FR + Gln treatment decreased the phosphorylation levels of p38 MAPK, JNK, and IκB (P < 0.05) (Fig. 2B–E). These results suggested that Gln alleviated the damage of FR on the barrier function of yak rumen epithelium, and the MAPK/NF-κB signaling pathway might be involved.

Fig. 2.

Effects of feed restriction and Gln supplementation on MAPK/NF-κB signaling pathway in yak rumen epithelium. (A) The mRNA expression of p38 MAPK, JNK, ERK1/2, and NF-κB. (B and D) Phosphorylated p38 MAPK and phosphorylated JNK expression levels. (C and E) TLR4 protein expression level and phosphorylated NF-κB p65 and phosphorylated IκB expression levels. Results were expressed as means and SEM, n = 6 independent experiments. Different letters indicate significant differences (P < 0.05). Con = control group; FR = feed restriction group; FR + Gln = feed restriction + Gln group. p38 MAPK = p38 mitogen-activated protein kinase; p-p38 MAPK = phosphorylated p38 mitogen-activated protein kinase; JNK = c-Jun N-terminal kinase; p-JNK = phosphorylated c-Jun N-terminal kinase; ERK1/2 = extracellular signal-regulated 1/2; NF-κB = nuclear factor-κB; p–NF–κB = phosphorylated nuclear factor-κB; TLR4 = Toll-like receptor 4; IκB = inhibitor of NF-κB; p-IκB = phosphorylated inhibitor of NF-κB; GAPDH = glyceraldehyde-3-phosphatedehydrogenase.

3.6. Cell viability and cell proliferation in YRECs

To explore the effects of Gln on the YRECs viability, the cell viability was determined by CCK-8 assay. As shown in Fig. 3A, Gln-D significantly decreased the YRECs viability (P < 0.001). In contrast, different Gln concentrations significantly increased the cell viability in the Gln-D group at 6, 12, 24, and 48 h (P < 0.001), among which, 10 mmol/L Gln treatment for 24 h had the best effect. The cell proliferation assay further indicated that Gln-D significantly inhibited the YRECs proliferation, while 5, 10, 20, and 40 mmol/L Gln improved it at 24 h (P < 0.001), among which the 10 mmol/L Gln group had the best effect (Fig. 3B).

Fig. 3.

Glutamine (Gln) alleviated Gln deficiency-induced reduction of cell viability and cell proliferation in yak rumen epithelial cells (YRECs). (A) Cell viability of YRECs upon Gln deficiency (12 h) treated with serial concentrations (0, 5, 10, 20, and 40 mmol/L) of Gln assayed by cell counting kit-8 (CCK-8) at 6, 12, 24, and 48 h. (B) Cell proliferation of YRECs upon Gln deficiency (12 h) treated with serial concentrations (0, 5, 10, 20, and 40 mmol/L) of Gln assayed by 5-ethynyl-2′-deoxyuridine (EdU), and the images were taken under a fluorescence microscope (magnification 100×, scale bars represent 400 μm). Results were expressed as means and SEM, n = 6 independent experiments. Different letters indicate significant differences (P < 0.05). Con = control group; Gln = glutamine group; Gln-D = Gln deficiency group; Gln-D + Gln = Gln deficiency + Gln group; DAPI = 4′,6-diamidino-2-phenylindole.

3.7. Inflammatory response and TJ in YRECs

The effects of Gln on inflammatory factors and TJ-related mRNA expression in YRECs were determined by RT-qPCR. As shown in Fig. 4A, compared with the control group, Gln-D significantly up-regulated IL-1β and IL-6 mRNA expression (P < 0.05), while supplementation with 5, 10, 20, and 40 mmol/L Gln alleviated them (P < 0.05). Gln-D did not affect the mRNA expression of TNF-α, but supplementation with 10 mmol/L Gln significantly down-regulated TNF-α mRNA expression compared with the Gln-D group (P = 0.041). Among them, the effect of 10 mmol/L Gln was optimal. Surprisingly, Gln-D and Gln supplementation did not change the mRNA expression of IL-10 in YRECs (P = 0.076). Similarly, compared with the control group, Gln-D significantly increased the protein expression of IL-1β, IL-6, and TNF-α (P < 0.05), while supplementation with 10 mmol/L Gln alleviated them (P < 0.05) (Fig. 4B).

Fig. 4.

Inflammation reaction and tight junction (TJ) status in yak rumen epithelial cells (YRECs). (A) The mRNA expression of IL-1β, IL-6, TNF-α, and IL-10 in YRECs upon Gln deficiency (12 h) treated with serial concentrations (0, 5, 10, 20, and 40 mmol/L) of Gln. (B) The protein expression levels of IL-1β, IL-6, and TNF-α. (C) The mRNA expression of ZO-1, occludin, claudin-1, and JAM-A in YRECs upon Gln deficiency (12 h) treated with serial concentrations (0, 5, 10, 20, and 40 mmol/L) of Gln. (D) The protein expression levels of ZO-1, occludin, claudin-1, and JAM-A. (E) Detection of ZO-1 protein expression level by immunofluorescence staining (magnification 200×, scale bars represent 200 μm). Results were expressed as means and SEM, n = 6 independent experiments. Different letters indicate significant differences (P < 0.05). Con = control group; Gln = glutamine group; Gln-D = Gln deficiency group; Gln-D + Gln = Gln deficiency + Gln group. IL-1β = interleukin-1β; IL-6 = interleukin-6; TNF-α = tumor necrosis factor-α; IL-10 = interleukin-10; ZO-1 = zonula occludens-1; JAM-A = junctional adhesion molecule-A; GAPDH = glyceraldehyde-3-phosphatedehydrogenase; DAPI = 4′,6-diamidino-2-phenylindole.

Compared with the control group, Gln-D significantly down-regulated ZO-1, occludin, claudin-1, and JAM-A mRNA expression (P < 0.05), while supplementation with Gln alleviated them (P < 0.05), among which 10 mmol/L Gln affected the mRNA expression of TJs-related (P < 0.05) (Fig. 4C). As shown in Fig. 4D, Gln-D significantly decreased the protein expression levels of ZO-1, occludin, claudin-1, and JAM-A compared with the Con group (P < 0.05). In contrast, supplementation with 10 mmol/L Gln alleviated them in YRECs (P < 0.05). The immunofluorescence analysis further indicated that Gln-D decreased the ZO-1 fluorescence integrity, while supplementation with 10 mmol/L alleviated it in YRECs (P < 0.001) (Fig. 4E).

3.8. MAPK/NF-κB signaling pathway in YRECs

NF-κB and MAPK signaling pathway-related mRNA and protein expression were detected to investigate the mechanism of Gln alleviating TJ injury. As shown in Fig. 5A, compared with the Con group, Gln-D significantly up-regulated the mRNA expression of TLR4, NF-κB, p38 MAPK, and JNK (P < 0.001), while supplementation of 10 mmol/L Gln alleviated them in YRECs (P < 0.001). Gln-D and supplementation with Gln did not affect the mRNA expression of ERK1/2 (P = 0.060). Furthermore, compared with the Con group, Gln-D significantly increased the protein expression of TLR4 and phosphorylation levels of p38 MAPK, JNK, NF-κB p65, and IκB (P < 0.001), while supplementation with Gln alleviated them (P < 0.001) (Fig. 5B and C).

Fig. 5.

Glutamine (Gln) alleviated Gln deficiency-induced the activation of MAPK and NF-κB signaling pathways in yak rumen epithelial cells (YRECs). (A) The mRNA expression of p38 MAPK, JNK, ERK1/2, NF-κB, and TLR4. (B) Phosphorylated p38 MAPK and phosphorylated JNK expression levels. (C) TLR4 protein expression level and phosphorylated NF-κB p65 and phosphorylated IκB expression levels. Results were expressed as means and SEM, n = 6 independent experiments. Different letters indicate significant differences (P < 0.05). Con = control group; Gln = glutamine group; Gln-D = Gln deficiency group; Gln-D + Gln = Gln deficiency + Gln group. p38 MAPK = p38 mitogen-activated protein kinase; p-p38 MAPK = phosphorylated p38 mitogen-activated protein kinase; JNK = c-Jun N-terminal kinase; p-JNK = phosphorylated c-Jun N-terminal kinase; ERK1/2 = extracellular signal-regulated 1/2; TLR4 = Toll-like receptor 4; NF-κB p65 = nuclear factor-κB p65; p–NF–κB = phosphorylated nuclear factor-κB; IκB = inhibitor of NF-κB; p-IκB = phosphorylated inhibitor of NF-κB; GAPDH = glyceraldehyde-3-phosphatedehydrogenase.

Metformin, a p38 MAPK/JNK activator, was used to further examine the mechanism by which Gln regulates TJs through the p38 MAPK and JNK pathways. At first, as shown in Fig. S1A, 0.2% and 0.4% dimethyl sulfoxide (DMSO) significantly decreased the YRECs viability (P < 0.001). In addition, 10 μmol/L (the concentration of DMSO in medium was 0.1%) metformin significantly increased the mRNA expression of p38 MAPK and JNK (P < 0.001) (Fig. S1B). And 10 μmol/L metformin blocked Gln-induced p38 MAPK and JNK mRNA expression decreases and ZO-1, occludin, claudin-1, and JAM-A mRNA expression increases (P < 0.001) (Fig. S1C–D). As shown in Fig. 6A, compared with the Gln-D + Gln group, metformin significantly increased the phosphorylation levels of p38 MAPK and JNK (P < 0.001). Further testing revealed the phosphorylation levels of NF-κB and IκB were significantly increased (P < 0.001), and the protein expression of ZO-1, occludin, and claudin-1 was decreased in the Gln-D + Gln + metformin group (P < 0.01) (Fig. 6B and C).

Fig. 6.

Glutamine (Gln) alleviated Gln deficiency-induced tight junction (TJ) damage through MAPK signal in yak rumen epithelial cells (YRECs). (A) Phosphorylated p38 MAPK and phosphorylated JNK expression levels. (B) Phosphorylated NF-κB p65 and phosphorylated IκB expression levels. (C) The protein expression levels of ZO-1, occludin, and claudin-1. Results were expressed as means and SEM, n = 6 independent experiments. Different letters indicate significant differences (P < 0.05). Con = control group; Gln-D = Gln deficiency group; Gln-D + Gln = Gln deficiency + Gln group; Gln-D + Gln + metformin = Gln deficiency + Gln group + p38 MAPK/JNK activator (metformin). p38 MAPK = p38 mitogen-activated protein kinase; p-p38 MAPK = phosphorylated p38 mitogen-activated protein kinase; JNK = c-Jun N-terminal kinase; p-JNK = phosphorylated c-Jun N-terminal kinase; ERK1/2 = extracellular signal-regulated 1/2; NF-κB p65 = nuclear factor-κB p65; p–NF–κB = phosphorylated nuclear factor-κB; IκB = inhibitor of NF-κB; p-IκB = phosphorylated inhibitor of NF-κB; ZO-1 = zonula occludens-1; GAPDH = glyceraldehyde-3-phosphatedehydrogenase.

4. Discussion

The scarcity of forage during the cold season on the Qinghai-Tibet Plateau leads to nutrient deficiencies and subsequent damage to the rumen epithelium in yaks, thereby hindering their growth (Ma et al., 2020). Glutamine is an animals' most abundant amino acid (Deters and Saleem, 2021). A study reported that starvation decreased the concentration of Gln in rat intestines (Kong et al., 2000). A previous study has found that Gln deprivation damaged the integrity of TJs in yak rumen epithelial cells (Yue et al., 2024). Moreover, in vivo studies have demonstrated that dietary Gln supplementation could promote gastrointestinal morphology, mitigate inflammatory responses, and improve gastrointestinal barrier function, thereby promoting compensatory growth in growth-retarded yaks (Ma et al., 2021a). Glutamine is considered a “conditionally” necessary amino acid. In normal physiological circumstances, the liver and skeletal muscle can synthesize sufficient glutamine (Tomaszewska et al., 2025). However, under certain stress conditions, such as malnutrition, disease, and environmental changes, glutamine synthesis in tissues and organs decreases and fails to cover the physiological needs of the organism (Cruzat et al., 2018; Han et al., 2018). The present results found that feed restriction significantly decreased Gln concentration in serum and rumen epithelium tissue. Therefore, this study investigated the effects of dietary Gln supplementation on the TJs of rumen epithelium in restricted-feeding yaks and further used YRECs as a model to explore the effects of Gln deprivation and Gln supplementation on TJs and their mechanism.

4.1. Gln alleviated the inhibition of growth performance induced by feed restriction in yaks

Adequate nutritional intake is critical for animal growth performance. In the present study, the DMI for each yak was recorded at 2.97 kg/d, with the net energy for maintenance (NEm) of the dry matter measured at 6.38 MJ/kg. Consequently, the total NEm intake of each yak was 18.95 MJ/d, which sufficiently meets the requirements for maintaining body weight. In the cold season, a scarcity of forage often results in a weight loss of 10%–20% in yaks (Jing et al., 2022; Qi et al., 2023). In this study, feed restriction decreased 14.62% of the body weight in yaks at d 60, consistent with findings from previous research (Zou et al., 2019). The ruminal epithelium, as the primary site of digestion in ruminants, plays a vital role in nutrient absorption and transport, while also serving as a robust barrier against endotoxins from ruminal microbes and feed (Lin et al., 2019). Gln has been demonstrated to facilitate the repair of the rumen epithelial barrier following injury (Ma et al., 2021a; Zhang et al., 2025b). Rumen epithelial barrier integrity contributes to the uptake of nutrients, thereby promoting animal growth. In addition, rumen epithelial barrier integrity could effectively avoid LPS translocating into the bloodstream, thereby avoiding a systemic inflammation response (Yang et al., 2022b; Zhao et al., 2023). Research on dairy cows has shown that higher LPS concentrations in serum could increase the expression of pro-inflammatory cytokines (IL-1β, IL-6, and TNF-α) in the mammary glands, thereby promoting the onset of mastitis (Guo et al., 2020; Zhao et al., 2023). Thus, damage to rumen epithelial barrier function may precipitate systemic inflammation, adversely impacting the health of yaks. It was previously shown that Gln could restore the rumen epithelial barrier function in growth-retarded yaks, thereby promoting growth performance (Ma et al., 2021a). These findings are consistent with the findings obtained from this study. Therefore, it was theorized that Gln alleviated the adverse effects of feed restriction on growth performance in yaks, potentially through its influence on rumen epithelial barrier function.

4.2. Gln inhibited the increase of permeability parameters in serum induced by feed restriction in yaks

The integrity of the rumen epithelial barrier is crucial for preventing the translocation of harmful substances into the bloodstream (Li et al., 2022). Rumen epithelium barrier function could be assessed by monitoring serum levels of DAO, D-LA, LPS, and HIS (Aschenbach and Gäbel, 2000; Chen et al., 2024; Liu et al., 2013; Ma et al., 2021a). In the present study, the DAO activity and D-LA and LPS concentrations in serum significantly increased after feed restriction on d 30, and HIS concentration significantly increased after feed restriction on d 60. The results showed that feed restriction may impair rumen epithelium barrier function. Previous research has established that the integrity of the rumen epithelial barrier is crucial for sustaining growth performance in ruminants. For example, under stress conditions, dietary feed additive (Pogostemon cablin essential oil and Fermented Chinese herbal medicines combined with probiotics) supplementation improved barrier function in rumen epithelium and increased the ADG of ruminants (Chen et al., 2024; Zhao et al., 2025). Additionally, a previous study has found that DAO activity and the concentrations of LPS and D-LA in the serum of growth-retarded yaks were significantly increased compared to growth-normal yaks, indicating impaired rumen epithelium barrier function. However, dietary Gln supplementation was found to mitigate these effects (Ma et al., 2021a). Similarly, this research showed that dietary Gln supplementation significantly decreased DAO activity and D-LA, LPS, and HIS concentrations in FR-yaks serum at d 60. These results indicated that Gln might alleviate the barrier function damage of rumen epithelium in yaks induced by feed restriction.

4.3. Gln alleviated the damage of barrier function in rumen epithelium induced by feed restriction in vivo and in vitro

The integrity of TJs is the primary determinant of rumen epithelial barrier function (Aschenbach et al., 2019). It was reported that rumen epithelial inflammation is closely related to abnormal expression of TJs (Wang et al., 2022). The present results showed that Gln effectively attenuates the upregulation of pro-inflammatory cytokines and the downregulation of TJs-related mRNA in rumen epithelial tissue or YRECs subjected to feed restriction or Gln deprivation. In addition, Gln mitigated the decrease in proliferative activity of YRECs induced by Gln deprivation. Previous studies found that starvation increased the mRNA expression of pro-inflammatory cytokines while reducing that of anti-inflammatory cytokines and TJs in rumen epithelial cells, thereby compromising the barrier function of the rumen epithelium (Hu et al., 2018; Ma et al., 2020). Furthermore, feed restriction decreased the proliferation of rumen epithelial cells and damaged the integrity of rumen epithelium (Steele et al., 2015). Dietary Gln supplementation significantly decreased the mRNA expression of pro-inflammatory cytokines (IL-6 and TNF-α) and increased the mRNA expression of anti-inflammatory cytokines (IL-10) and claudin-1 in the rumen epithelium of Hu Lambs (Wu et al., 2022). In addition, at the cellular level, previous results revealed that Gln could effectively repair HIS-induced barrier damage in YRECs (Zhang et al., 2025b). Therefore, these findings imply that Gln alleviates feed restriction-induced damage to rumen epithelial TJs in yaks by suppressing the inflammatory response and enhancing cell proliferation, thereby maintaining barrier function integrity.

4.4. Gln alleviated the activation of MAPK/NF-κB signaling pathway in rumen epithelium induced by feed restriction in vivo and in vitro

The primary signaling pathways of MAPKs include p38 MAPK, JNK, and ERK1/2, which are instrumental in regulating the expression of TJ-associated proteins (Ritch et al., 2014). Within the NF-κB transcription complex, NF-κB p65 is pivotal for gene regulation and plays a crucial role in mediating inflammatory and immune responses (Ji et al., 2011). The IκB protein serves as a key regulator of the NF-κB pathway and is utilized as a marker for NF-κB-mediated inflammation (Kim et al., 2019). In addition, in the inflammatory response, TLR4, NF-κB, and MAPK are essential pathways for modulating the expression and secretion of inflammatory cytokines (Kessel et al., 2008; Liu et al., 2022a). Studies have shown that starvation could activate the MAPK signaling pathway, leading to increased phosphorylation levels of p38 MAPK, JNK, and ERK1/2 (Back et al., 2018; Moruno-Manchón et al., 2013; Ryu et al., 2019). In previous research, Gln supplementation could reduce the pro-inflammatory factor levels in the rumen epithelium of growth-retarded yaks (Ma et al., 2021a). Similarly, in this study, Gln mitigated the inflammatory response and TJ damage by inhibiting feed restriction or Gln deprivation-induced activation of TLR4, p38 MAPK, JNK, NF-κB, and IκB in rumen epithelial tissues and cells. Further studies on cells showed that the addition of p38 MAPK and JNK activators could negate the inhibitory effects of Gln on the MAPK/NF-κB pathways in feed restriction-induced yak rumen epithelial cells, thereby impeding the protective effects of Gln on TJs in Gln deprivation-induced yak rumen epithelial cells. TLR4 and MAPK are recognized to be upstream regulators of the NF-κB signaling pathway and are pivotal in the regulation of intestinal inflammation (Zhou et al., 2018). Furthermore, Gln has been shown to mitigate the LPS-induced increase in TLR4, p38 MAPK, and JNK protein levels in mouse lung tissue (Huang et al., 2021). And Gln could downregulate the TLR4 expression in intestinal epithelial cells infected with gram-negative bacteria (Abreu et al., 2001). Moreover, previous research indicated that Gln could increase the expression of TJ-related proteins by reducing the phosphorylation levels of p38 MAPK induced by HIS in yak rumen epithelial cells (Zhang et al., 2025b). These results demonstrated that Gln improved rumen epithelial barrier function in yaks by inhibiting the activation of the MAPK/NF-κB signaling pathway.

Nonetheless, metformin did not completely inhibit the upregulation of ZO-1 and claudin-1 mRNA expression, nor the claudin-1 protein expression. These findings indicate the potential involvement of additional signaling pathways in the mitigation of Gln-D-induced barrier function injury by Gln. For example, Gln could regulate TJ integrity through AMP-activated protein kinase (AMPK) signaling, thereby contributing to the improvement of intestinal mucosal barrier function (Wang et al., 2016). Furthermore, deprivation of exogenous and endogenous Gln decreased transepithelial electrical resistance and increased permeability in Caco-2 cells, an effect that was reversed by Gln supplementation through the phosphoinositide 3-kinase (PI3K)/Akt pathway (Li and Neu, 2009). Thus, although this study is the first study to demonstrate that exogenous Gln can repair the barrier function injury induced by insufficient endogenous Gln synthesis via the MAPK/NF-κB signaling pathway in yaks, it does have certain limitations. Specifically, the study predominantly focused on the role of Gln, without considering the effects of other amino acids. Additionally, an in vivo inhibitor experiment was not conducted.

5. Conclusion

In summary, the present results showed that feed restriction significantly decreased the Gln concentration in serum and rumen epithelium tissue of yaks, increased the expression of pro-inflammatory cytokines, and down-regulated the expression of TJ-related mRNA and protein, thereby compromising the integrity of the rumen epithelial barrier function. Nevertheless, Gln alleviated the decrease of ZO-1, occludin, and claudin-1 protein expression levels in yak rumen epithelial cells via the MAPK/NF-κB signaling pathway. This study provided evidence for potential links between Gln and the barrier function of rumen epithelial tissues.

CRediT authorship contribution statement

Ziqi Yue: Writing – original draft, Investigation, Formal analysis. Shanpeng Ke: Writing – original draft, Investigation, Formal analysis. Ali Mujtaba Shah: Writing – review & editing, Project administration. Junmei Wang: Formal analysis. Zhisheng Wang: Conceptualization. Rui Hu: Writing – review & editing, Project administration, Conceptualization. Quanhui Peng: Methodology. Huawei Zou: Formal analysis. Jianxin Xiao: Formal analysis. Yahui Jiang: Investigation. Fali Wu: Investigation. Yiping Tang: Investigation.

Declaration of competing interest

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, and there is no professional or other personal interest of any nature or kind in any product, service, and/or company that could be construed as influencing the content of this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: