A total of forty-eight male Hu sheep with an average initial body weight of 17.74 ± 0.17 kg were randomly separated into a control group and a treatment group. The control group (CON) was fed a basal diet and the treatment group (GLU) was fed a basal diet + 3 g/head/d of L-glutamate. The L-glutamate was purchased from Shanghai Yuanye Biological Co. (Shanghai, China) and had a purity of 99%. The supplementation dosage of L-glutamate was estimated according to an in vitro pilot experiment (unpublished). The basal diet (Table 1) consisted of a complete mixed diet with a concentrate to forage ratio of 7:3 designed to meet the Chinese Feeding Standard of Sheep (NY/T816-2004) requirements for growing-sheep. Each treatment group contained twenty-four sheep divided into six replicates of four sheep.

The trial lasted for 90 days from 1 June to 29 August 2021. Animals were fed twice a day (8:00 and 17:00 h) with free access to water and feed. The feed intake of each replicate was recorded daily, and the body weight of sheep was recorded after fasting for 12 h on days 0, 30, 60, and 90. The feed efficiency was calculated using the F:G ratio based on the average daily feed intake (ADFI) and average daily gain (ADG).

Temperature and humidity were recorded by an Apresys^® instrument (179-TH, Apresys International Inc., San Ramon, CA, United States), and the temperature and humidity index (THI) was calculated according to NRC (1971) (20) as follows:

where AT is air temperature (^°C) and RH is relative humidity (%).

Blood samples were collected with non-anticoagulation vacuum blood vessels before morning feeding after fasting for 12 h on days 30, 60, and 90. Serum was harvested following centrifugation at 3,000 × g for 10 min at 4^°C and subsequently frozen at −80^°C until analysis.

The concentrations of serum glucose, blood urea nitrogen (BUN), total protein (TP), albumin (ALB), glutamate, and globulin (GLB) were determined using corresponding commercial kits (Zhongsheng Beikong Bio-technology and Science Inc., Beijing, China) and an automatic biochemical analyzer (BS-420, Shenzhen Mindray Bio-medical Electronics Co., Shenzhen, China). The concentrations of glutamate, immunoglobulin A (IgA), immunoglobulin G (IgG), and immunoglobulin M (IgM) were measured using corresponding commercial kits (Beijing Sino-UK Institute of Biological Technology, Beijing, China) and a semi-automatic biochemical analyzer (A6, Beijing Songshang Technology Co., Beijing, China). The concentrations of growth hormone (GH), heat shock protein 70 (HSP70), adrenocorticotrophic hormone (ATCH), corticosterone (CORT), triiodothyronine (T₃), and tetraiodothyronine (T₄) were determined using corresponding commercial ELISA kits (Beijing Sino-UK Institute of Biological Technology, Beijing, China) and a microplate reader (DR-200BS, Wuxi Hua Wei De Lang Instrument Co., Wuxi, China).

Rumen fluid pH was measured immediately after slaughter using a portable pH meter [Testo 206-pH1, Testo Instruments International (shanghai) CO., Shanghai, China]. Then samples were collected for the determination of microbial crude protein (MCP), ammonia nitrogen (NH₃-N), and volatile fatty acids (VFAs). The VFA concentration was analyzed by gas chromatography (GC-2014, Shimadzu, Tokyo, Japan) equipped with a capillary column (Stabilwax, Restek, Bellefonte, PA, United States). NH₃-N concentration was determined according to the method of Broderica and Kang (21). MCP concentration was analyzed according to the method of Makkar et al. (22).

Feces were collected for five consecutive days at the end of the trial. In the laboratory, all feces of each replicate were evenly mixed and 10% sulfuric acid was added to fix nitrogen. Fecal samples were stored at −20^°C for subsequent determination.

Dry matter (DM), crude protein (CP), ash, and ether extract (EE) of the feed samples and fecal samples were determined according to the Association of Analytical Communities (AOAC) (23) using the methods 967.03, 984.13, 924.05, and 920.39, respectively. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined using an Ankom A200i Fiber analyzer (ANKOM Technology Co., New York, NY, United States) according to Van Soest et al. (24). The content of hydrochloric acid insoluble ash (AIA) was determined according to Keulen and Young (25) described, and AIA was used as a digestibility marker. The formula was:

where AD represents the apparent digestibility of the nutrient content, a represents the nutrient content of feedstuffs, b represents the nutrient content of feces, c represents the AIA content of feedstuffs, and d represents the AIA content of feces.

One sheep was randomly selected from each replicate for slaughter, after fasting for 12 h. After the removal of head, feet, tail, and internal organs, the carcass, heart, spleen, liver, lungs, and kidneys were weighed. The dressing percentage was calculated as: carcass weight (kg)/live weight (kg) before slaughter × 100%. The organ index was calculated as organ index = organ weight (g)/live weight (kg) before slaughter. Backfat thickness, eye muscle area (EMA), and grade-rule (GR) tissue depth values were measured on the left side of the carcass according to Song et al. (26). Backfat thickness is the thickness between the 12th and 13th ribs. EMA is the cross-sectional area of the 12th and 13th ribs. GR is the total tissue thickness between the 12th and 13th rib and 11 cm from the dorsal midline of the carcass, which can represent the carcass fat content.

All data were analyzed using SPSS 17.0 software (SPSS Inc., Chicago, IL, United States). Data were analyzed using double-tailed t-tests after checking normal distribution. The data were considered significantly different if p ≤ 0.05, and a tendency was suggested if 0.05 < p < 0.10.

The Hu sheep was suffering from heat stress throughout the whole trial period, as shown in Figure 1. The days that the Hu sheep spent at different THI thresholds were: 72 ≤ THI < 78 for 2 days, 78 ≤ THI < 90 for 48 days, and THIs ≥ 90 for 40 days.

The growth performance of the Hu sheep is shown in Table 2. No differences were observed in body weight on days 30, 60, and 90 between CON and GLU (p > 0.05). No differences were observed in ADFI, ADG, and F:G ratio between GLU and CON during phase 1 (1–30 days) and phase 2 (31–60 days) (p > 0.05). However, during phase 3 (61–90 days), the ADG in GLU was higher than in CON (p < 0.05), and the ADFI and F:G ratio in GLU was lower, compared with those in CON (p < 0.05).

Table 3 shows no differences in carcass weight, dressing percentage, backfat thickness, eye muscle area, GR value, organ weight, and organ index between GLU and CON (p > 0.05). However, spleen weight (p = 0.078) and spleen index (p = 0.079) in GLU tended to increase (by 54.11 and 53.71%), respectively, as compared with CON (Table 4).

As shown in Table 5, the digestibility of DM, OM, and CP in GLU was higher than those in CON (p < 0.05), whereas no significant differences were observed in the digestibility of EE, NDF, ADF, and soluble carbohydrates (SCHO) between GLU and CON (p > 0.05).

Rumen fermentation characteristics are shown in Table 6. Ruminal pH, MCP, NH₃-N, and isovalerate concentration in GLU were all higher than those in CON (p < 0.05). The concentrations of TVFAs and VFAs in GLU, except for isovalerate, were similar to those in CON (p > 0.05).

The levels of serum glutamate, GLB, IgA, IgG, and IgM in GLU were higher than in CON on day 90 (p < 0.05), and the ALB/GLB in GLU was lower than that in CON on day 90 (p < 0.05) (Table 7). However, all the serum biochemical parameters-listed above were similar between GLU and CON on days 30 and 60 (p > 0.05).

The levels of serum hormones and HSP70 are shown in Table 8. The level of GH in GLU was higher than those in CON on day 90 (p = 0.001), and the levels of HSP70, ACTH, CORT, T₃, and T₄ in GLU were lower than those in CON on day 90 (p < 0.001). However, no differences were found between GLU and CON on days 30 and 60 (p > 0.05), expect that level of T₃ was lower in GLU on day 30 (p < 0.05).

It is known that the heat stress leads to decreases in feed intake, nutrient digestibility, and absorption capacity, resulting in reduced animal growth performance of livestock (29). The research results of glutamate on the growth performance of animals are inconsistent. Wang et al. (30) found that the glutamate was a key neurotransmitter in the hypothalamus, which could inhibit the feed intake in animals. Other researchers indicated that it could also increase the feed intake by activating taste receptors in the digestive system (31). Hu et al. (32) and Rezaei et al. (33) reported lower ADFI after glutamate supplementation in pigs. In growing-finishing pigs, glutamate supplementation of 1% had no effect on ADG and F:G for the 60 days feeding regimen (32). By contrast, it linearly increased ADG, and decreased ADFI and F:G with monosodium glutamate addition of 0.5–4% in post-weaning pigs (33). In dairy calves, a lower dosage of glutamate (0.1 or 0.3%) has no obvious effect on ADFI, ADG, and F:G within a 70 or 56 days feeding period (34, 35). In this study, similar results of glutamate supplementation were shown in Hu sheep during phase 1 (1–30 days) and phase 2 (31–61 days). However, glutamate resulted in a marked increase of ADG, and a decrease of ADFI and F:G during phase 3 (61–90 days), which might be because of the long experimental period of glutamate supplementation. The levels of serum glutamate, serum biochemical indices and serum hormones were similar during phases 1 and 2, whereas a higher level of serum glutamate, GLB, IgA, IgG, IgM, and GH, and a lower level of HSP70, ACTH, CORT, T₃, and T₄ in GLU were observed during phase 3 in this study. It suggested that a longer period of glutamate supplementation increased the serum glutamate to promote the growth performance, improve the metabolism, and enhance the immunity of Hu sheep.

Accounting for the highest proportion of AA composition in rumen microorganisms, glutamate plays an essential role in MCP synthesis and ruminant growth (36). In this study, a higher concentration of MCP was observed owing to the dietary supplementation of glutamate, which promoted rumen microbial growth (16, 17). The NH₃-N is the main source of nitrogen for rumen microbial protein synthesis, and its utilization rate reflects the balance between substrate nitrogen degradation and microbial synthesis utilization (37). In the present study, the NH₃-N concentration was notably increased by adding glutamate, which was consistent with a previous report demonstrating that dietary glutamate addition increased the NH₃-N concentration in dairy cows (38). In this study, higher levels of NH₃-N via the supplementation with glutamate provided a nitrogen source promoting microbial growth and reproduction, which supported the higher MCP in GLU. The higher NH₃-N concentration in GLU indicated a higher rumen fermentation rate; however, the NH₃-N concentration in this study did not reach the 235 mg/L required for the maximal fermentation rate (39), suggesting that the specific dosage of glutamate addition did not inhibit the rumen fermentation in Hu sheep. The concentrations of NH₃-N and VFAs are considered to be the main factors affecting ruminal pH. In the present study, dietary glutamate supplementation increased rumen pH, which might be attributed to the higher NH₃-N production in GLU. The results were consistent with the study of Dann et al. (38) who reported that glutamate supplementation could increase the rumen pH. VFAs are absorbed by the ruminal epithelial cells and provide the main energy substances for the ruminant production. These organic acids can improve the morphology of rumen mucosa, and reduce the infection of intestinal diseases by improving the intestinal microbial structure and morphology (40, 41, 42). Isovalerate is also an energy substance for the ruminal epithelial cells, and linear increases in isovalerate are positively correlated with the promotion of rumen development (40). In this study, the higher isovalerate in GLU might be associated with an improved feed degradation rate.

Oral intake of glutamate stimulates the salivation essential for mastication, swallowing, and digestion (43). A previous study demonstrated that free glutamate in digestive juices obtained from dietary digestion, absorption, or glutamine synthesis, could improve nutrient digestibility by stimulating the release of digestive enzymes and pancreatic hormones (44). Glutamate also could improve intestinal morphology and promote nutrient utilization in heat-stressed broilers (13). In the present study, the higher digestibility of DM, OM, and CP in GLU was probably due to higher serum glutamate, as mentioned above. The results of this study were similar to the research of Dann et al. (38) who reported that supplementation of 80 g glutamate per cow/day tended to increase the digestibility of DM and OM.

Heat stress reduces animal immune function, such as decreased immune antibody secretions and spleen developmental retardation (45, 46). As the largest secondary lymphoid organ, the spleen is rich in various immune cells, such as B cells and T cells, which play vital roles in immune function and are responsible for IgA, IgG, and IgM production (47, 48). In the present study, glutamate supplementation tended to increase the spleen weight and the spleen index by 54.11 and 53.71%, respectively, probably because glutamate was beneficial to spleen development through the activation of T cell function and protection against T cell apoptosis (49). It is critical to emphasize that glutamate is an important immunomodulator (50). A previous study reported that the glutamate receptors (GluRs) were highly expressed in T cells and B cells (51), suggesting that glutamate had a positive effect on targeting them. However, if serum glutamate concentration exceeds the pathological 10^–3 moles level, then T cell function and survival are inhibited (49). Serum glutamate in this study was within normal physiological concentrations. As mentioned above, higher levels of IgA, IgG, and IgM in GLU might result from the higher serum glutamate in GLU stimulating spleen development and activating T cell function. Furthermore, the decreased level of ALB/GLB in GLU might be because of the increased GLB concentration.

The hypothalamic–pituitary–adrenal (HPA) axis and hypothalamic–pituitary–thyroid (HPT) axis can be activated by heat stress, which in turn regulate the corresponding hormones secreted by endocrine glands (52). Higher secretion of serum ACTH and CORT has been reported in animals exposed to heat stress (53). Generally, CORT is an appropriate end-product of the HPA axis, and higher ACTH leads to increase CORT (54). However, excessive CORT would accelerate the heat stress-induced protein degradation (55), which partially explained the lower body weight in CON in this experiment. The lower levels of ACTH and CORT in GLU on day 90 might result from the higher level of serum glutamate, which reduced the secretion of ACTH and CORT and the protein degradation of muscle.

The T₃ and T₄ (the main thyroid hormones), are key regulators of energy metabolism in the body (56). It is essential for animals to maintain the thermal balance by lowering the secretion of T₃ and T₄ (56). Several studies have reported that monosodium glutamate has a cumulative effect on reducing thyroid function (57, 58). Similarly, in this study, a decrease in T₃ and T₄ after dietary glutamate supplementation was observed on day 90, suggesting that dietary glutamate supplementation might reduce heat stress responses by lowering heat production. A higher GH level often indicates a better anti-heat stress capability in goats (59). As an excitatory AA, glutamate contributes to the stimulation of GH secretion by regulating the hypothalamus (60). Serum GH concentrations are linearly correlated with serum glutamate (60), which can explain the higher level of GH in GLU in this study.

The HSP70, the most widely distributed and expressed HSP in the body, is involved in regulating the stability of cellular proteins, increasing the tolerance of cells to stressors, and reducing cell apoptosis (61). Higher HSP70 is usually positively related to the degree of heat stress (62). A previous study reported that glutamate was conducive to enhancing the thermodynamic stability of proteins against temperature stress (12). Another study showed that glutamine normalized metabolism by reducing the serum levels of HSP70 in atrial fibrillation patients (63). In the present study, HSP70 content in GLU decreased on day 90, suggesting that glutamate supplementation might reduce the heat stress response and normalize metabolism by reducing HSP70 levels.