Dietary α-Mangostin alleviates lipopolysaccharide-induced stress and enhances growth performance in broilers via improved intestinal immunity and integrity Authors: Kaibo Fu, Yuan Yue, Jian Zhao, Xianlei Li, Bingkun Zhang Published: 2026 ABSTRACT α-Mangostin (α-MG), a bioactive xanthone from mangosteen, possesses anti-inflammatory and antioxidant properties, but its efficacy and optimal dosage in broilers remain unclear. This study aimed to evaluate the effects of α-MG on lipopolysaccharide (LPS)-challenged and normally reared broilers to determine the optimal supplementation level. Two experiments were conducted. In Exp. 1, 396 one-d-old Cobb broilers (initial body weight: 42.45 ± 0.46 g) were divided into six groups, each with six replicates of 11 birds. The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS and fed diets containing 0 (positive control), 100, 150, 200, and 250 mg/kg α-MG for 21 d. In Exp. 2, 792 one-d-old Arbor Acres broilers (initial body weight: 42.54 ± 0.44 g) were divided into six groups, each with six replicates of 22 birds, and fed diets containing 0, 50, 100, 150, 200, and 250 mg/kg α-MG for 42 d. The results showed that in Exp. 1, supplementation with 150 mg/kg α-MG improved ileal morphology (villus height/crypt depth), immune cell proportions (CD4+CD8-/CD3+), and secretory immunoglobulin A (IgA) concentration compared with the positive control (P < 0.05). In Exp. 2, supplementation with 150 mg/kg α-MG enhanced body weight and average daily gain on d 21, increased muscle glutathione (GSH) and superoxide dismutase (SOD) content, reduced serum interleulin-1 beta (IL-1β) and diamine oxidase (DAO) activities, improved ileal morphology (villus height), and upregulated ERK1 expression level (P < 0.05). In conclusion, dietary supplementation with α-MG at 150 mg/kg effectively mitigates LPS-induced intestinal damage and immune stress, and promotes growth performance in broilers under normal conditions. The benefits are mediated through the enhancement of antioxidant capacity, downregulation of nuclear factor kappa-B (NF-κB) pathway-related gene expression, and reinforcement of intestinal mucosal immunity and integrity. Introduction The high rearing densities used in the modern intensive poultry industry are confronted with substantial challenges from environmental and pathogenic stressors, such as microbial infections, dietary toxins, and management issues (Jin et al., 2025; Liu et al., 2023a). These stressors can disrupt intestinal homeostasis, resulting in impaired growth performance, reduced meat quality, and substantial economic losses (Eratalar et al., 2022; Shaukat et al., 2025). The intestinal tract is a critical immune organ that not only facilitates nutrient digestion and absorption but also serves as a first line of defense against external harmful factors, thereby preserving intestinal homeostasis and physical health (Bian et al., 2020; Bogatyrev et al., 2020; Parrish et al., 2023). Under pathological stress conditions, this barrier is highly susceptible to disruption (Balakin et al., 2025; Mohammadzadeh et al., 2025), as it is composed of epithelial cells that are bound by tight junction proteins, a chemical mucus layer, and a sophisticated local immune system (Zhang et al., 2024). The poultry industry's long-term reliance on antibiotics for growth promotion and disease control (Looft et al., 2014; Tang et al., 2025) may result in intestinal dysbiosis, antibiotic-resistant bacteria, and drug accumulation (Beyari et al., 2024; Huang et al., 2024). Finding efficient natural substitutes that can improve resilience and preserve gut health is urgently needed to reduce dependence on antibiotics (Pan et al., 2022). Phytochemicals—bioactive compounds derived from plants—have emerged as promising candidates due to their multifaceted properties, including antioxidant (Kong et al., 2023; Liu et al., 2025), anti-inflammatory (Liu et al., 2023a; Wang et al., 2025), and microbiota improvement (Martin et al., 2024; Zhang et al., 2025). The isoprenylated xanthone α-mangostin (α-MG) was isolated from the pericarps of the mangosteen (Garcinia mangostana, Clusiaceae) and has been used for clinical application in traditional Chinese medicinal practices for centuries (Jo et al., 2022). Previous in vitro and rodent studies have demonstrated its potential anti-inflammatory capacity, largely attributed to the inhibition of the nuclear factor kappa-B (NF-κB) signaling pathway (Majdalawieh et al., 2025; Zhang et al., 2022), as well as its remarkable antioxidant activity (John et al., 2022). However, there is a current lack of clarity regarding the optimal supplementation dosage and impact of α-MG in broiler diets. The current study aimed to evaluate the efficacy and determine the optimal supplementation level of α-MG as a broiler feed additive in both normal and stressed conditions. Additionally, changes in the integrity of the intestinal barrier, systemic and mucosal immune responses, and critical inflammatory signaling pathways were observed. Reference data and a theoretical foundation for the application of dietary α-MG as a novel nutritional intervention in poultry production will be provided by the results of this research. Materials and methods Animal ethics statement The experimental use of animals and related procedures was approved by the Chinese Agricultural University Laboratory Animal Welfare and Experimental Ethical Committee at the China Agricultural University, Hebei Zhuozhou Experimental Base (approval number AW50206202-1-05, approved on February 5, 2026). The management of chicks was consistent with the “Arbor Acres Broiler Pocket Guide” and “Cobb Broiler Management Guide”, respectively (Arbor Acres, 2026; Cobb, 2026). Reagents Lipopolysaccharide (LPS) (Escherichia coli O55: B5) was purchased from Sigma–Aldrich Corporation (St. Louis, MO, USA) and prepared as a 1 mg/mL solution with 0.9% saline solution. The α-MG (purity 98.5%) was purchased from Guangzhou Leader Biotechnology Co., Ltd. (Guangzhou, Guangdong, China) and added by replacing the wheat bran in the basal diet. Animals, diets, and experimental design All the broilers were fed in a fully enclosed chicken house with 3-tier battery cages (200 cm × 80 cm × 60 cm). Feed in pellet form and water were provided ad libitum. The temperature was set at 34 °C at the beginning of bird rearing, then decreased by 2 to 3 °C every week to adjust to a final temperature of 23 °C. A standard lighting program was followed: 23 h of light and 1 h of darkness for the first 5 d, followed by 20 h of light and 4 h of darkness from d 6 until the end of the trial. The basal diets were formulated based on the Feeding Standard of Chicken (NY/T 33–2004) recommended by the Ministry of Agriculture of the People's Republic of China (2004) to meet the nutritional requirements (Table 1).Table 1Composition and nutrient levels of experimental diets (%, air-dry basis).Table 1IngredientsContents, %Nutrients2Contents, %Exp. 1 and Exp. 2 0–21 dExp. 2 22–42 dExp. 1 and Exp. 2 0–21 dExp. 2 22–42 dCorn47.2048.80AME, MJ/kg12.3512.98Wheat flour6.0010.00Crude protein22.6220.50Soybean-meal (43% CP)31.8020.65Calcium0.730.65Soybean-meal (46% CP)3.505.50Total phosphorus0.660.52Corn protein powder2.504.50Available phosphorus0.350.23Soybean oil3.855.40Lysine1.581.33Vitamin and mineral premix11.091.68Methionine0.610.54Limestone1.001.20Methionine + cystine0.960.93Calcium hydrogen phosphate1.300.60Threonine1.100.97Sodium chloride0.200.20Tryptophan0.240.20L-Lysine HCl (98.5%)0.680.68Arginine1.451.25DL-Methionine (98%)0.370.30L-Threonine (98.5%)0.330.32L-Tryptophan (98.5%)0.030.02Wheat bran0.150.15Total100.00100.00AME = apparent metabolizable energy.1Vitamin and mineral premix (provided per kilogram of diet): vitamin A, 9000 IU; vitamin D3, 4200 IU; vitamin E, 40 IU; vitamin K3, 3 mg; vitamin B1, 2 mg; vitamin B2, 6.4 mg; vitamin B12, 0.02 mg; biotin, 0.1 mg; folic acid, 2 mg; pantothenic acid, 13 mg; niacin, 40 mg; Cu, 10 mg; Fe, 50 mg; Mn, 80 mg; Zn, 80 mg; Se, 0.3 mg; I, 0.8 mg.2The AME and tryptophan contents were calculated according to Zheng et al. (2021) and NRC (1994), respectively. The available phosphorus contents were calculated based on Han et al. (2025) for the 0 to 21 d phase and on Tay-Zar et al. (2019) for the 22 to 42 d phase, the other values are measured values. Table 1Composition and nutrient levels of experimental diets (%, air-dry basis).Table 1IngredientsContents, %Nutrients2Contents, %Exp. 1 and Exp. 2 0–21 dExp. 2 22–42 dExp. 1 and Exp. 2 0–21 dExp. 2 22–42 dCorn47.2048.80AME, MJ/kg12.3512.98Wheat flour6.0010.00Crude protein22.6220.50Soybean-meal (43% CP)31.8020.65Calcium0.730.65Soybean-meal (46% CP)3.505.50Total phosphorus0.660.52Corn protein powder2.504.50Available phosphorus0.350.23Soybean oil3.855.40Lysine1.581.33Vitamin and mineral premix11.091.68Methionine0.610.54Limestone1.001.20Methionine + cystine0.960.93Calcium hydrogen phosphate1.300.60Threonine1.100.97Sodium chloride0.200.20Tryptophan0.240.20L-Lysine HCl (98.5%)0.680.68Arginine1.451.25DL-Methionine (98%)0.370.30L-Threonine (98.5%)0.330.32L-Tryptophan (98.5%)0.030.02Wheat bran0.150.15Total100.00100.00AME = apparent metabolizable energy.1Vitamin and mineral premix (provided per kilogram of diet): vitamin A, 9000 IU; vitamin D3, 4200 IU; vitamin E, 40 IU; vitamin K3, 3 mg; vitamin B1, 2 mg; vitamin B2, 6.4 mg; vitamin B12, 0.02 mg; biotin, 0.1 mg; folic acid, 2 mg; pantothenic acid, 13 mg; niacin, 40 mg; Cu, 10 mg; Fe, 50 mg; Mn, 80 mg; Zn, 80 mg; Se, 0.3 mg; I, 0.8 mg.2The AME and tryptophan contents were calculated according to Zheng et al. (2021) and NRC (1994), respectively. The available phosphorus contents were calculated based on Han et al. (2025) for the 0 to 21 d phase and on Tay-Zar et al. (2019) for the 22 to 42 d phase, the other values are measured values. Composition and nutrient levels of experimental diets (%, air-dry basis). AME = apparent metabolizable energy. Vitamin and mineral premix (provided per kilogram of diet): vitamin A, 9000 IU; vitamin D3, 4200 IU; vitamin E, 40 IU; vitamin K3, 3 mg; vitamin B1, 2 mg; vitamin B2, 6.4 mg; vitamin B12, 0.02 mg; biotin, 0.1 mg; folic acid, 2 mg; pantothenic acid, 13 mg; niacin, 40 mg; Cu, 10 mg; Fe, 50 mg; Mn, 80 mg; Zn, 80 mg; Se, 0.3 mg; I, 0.8 mg. The AME and tryptophan contents were calculated according to Zheng et al. (2021) and NRC (1994), respectively. The available phosphorus contents were calculated based on Han et al. (2025) for the 0 to 21 d phase and on Tay-Zar et al. (2019) for the 22 to 42 d phase, the other values are measured values. A total of 396 one-d-old Cobb Broiler chicks (initial body weight: 42.45 ± 0.46 g) and 792 Arbor Acres Broiler chicks (initial body weight: 42.54 ± 0.44 g) healthy chicks were selected in Exp. 1 and Exp. 2, respectively. In Exp. 1, chicks were allocated into six groups, each with six replicates and 11 birds. The entire experimental period was 21 d. The negative control group chicks were fed the corn-soybean meal without α-MG supplementation, and the experimental chicks (positive control, α-MG100, α-MG150, α-MG200, and α-MG250) were fed the corn-soybean meal diets supplemented with α-MG at 0, 100, 150, 200, and 250 mg/kg, respectively. On d 11 and 13, all chicks in the experimental groups received an injection of LPS at 1 mg/kg body weight (BW), and the chicks in the negative control group received an injection of the same volume of regular saline. In Exp. 2, chicks were allocated into six groups, each with six replicates and 22 birds. The entire experimental period lasted 42 d, divided into two feeding phases: phase 1 (d 1-21) and phase 2 (d 22-42). The experimental chicks (control, α-MG50, α-MG100, α-MG150, α-MG200, and α-MG250) were fed the corn-soybean meal diets supplemented with α-MG at 0, 50, 100, 150, 200, and 250 mg/kg, respectively. The crude protein and amino acid contents were analyzed according to GB/T 6432–2018 (China National Standard, 2018b) and GB 5009.124–2016 (China National Standard, 2016), respectively. The calcium and total phosphorus contents were determined according to GB/T 6436–2018 (China National Standard, 2018a) and GB/T 6437–2018 (China National Standard, 2018c), respectively. The apparent metabolizable energy was calculated based on values from the feed composition tables published by Institute of Animal Sciences et al. (2024). Growth performance In Exp. 1, chicks were starved for 12 h before and after LPS challenge and at d 21. The BW and feed intake (FI) of each replicate was recorded. The BW, average daily gain (ADG), average daily feed intake (ADFI), feed conversion ratio (FCR), and phase FI were calculated according to the Chinese agricultural standard (NY/T 823–2020) (Ministry of Agriculture of the People's Republic of China, 2020). In Exp. 2, BW and FI of each replicate were recorded at d 0, 21, and 42 after starving for 12 h. The calculation methods were the same as the Exp. 1. Carcass performance In Exp. 2, at 42 d of age, one chick was selected and euthanized from each replicate. The carcass cutting operation was carried out according to GB/T 19478–2018 (China National Standard, 2018d) and GB/T 24864–2010 (China National Standard, 2010), while dressed weight, half-eviscerated weight (including giblet), eviscerated weight, leg muscle weight, pectoral muscle weight, and abdominal fat weight were measured and their yield calculated following the definitions of NY/T 823–2020 (Ministry of Agriculture of the People's Republic of China, 2020). Organ indices The liver, spleen, bursa of Fabricius, and thymus were weighed, and the relative organ weights (expressed as percentages of BW) were calculated according to the methods described by Gorenz et al. (2024) and Islam et al. (2023). Meat quality In Exp. 2, the pectoral muscle was used in the same way as the carcass performance. White striping grade was measured using the methods of Kuttappan et al. (2016) (Fig. 1). A section (5 cm × 2 cm × 1 cm) from the same position on each pectoral muscle was removed, weighed, and stored at a 4 °C for 24 h to measure drip loss. The activities of malondialdehyde (MDA; A003-1-2), glutathione (GSH; A006-2-1), total antioxidant capacity (T-AOC; A015-1-2), superoxide dismutase (SOD; A001-1-2), and glutathione peroxidase (GSH-Px; A005-1-2) in the pectoral muscle were analyzed by using assay kits (Nanjing Jiancheng Biotechnology Institute, Nanjing, Jiangsu, China) and measured with a microplate reader (SpectraMax i3x, Molecular Devices, LLC., San Jose, CA, USA).Fig. 1Scoring standard for white striping (cited from Kuttappan et al., 2016). Normal–no distinct white lines; moderate–small white lines, generally <1 mm thick, but apparently visible on the fillet surface; severe–large white lines (1–2 mm thick) very visible on the fillet surface; extreme–thick white bands (>2 mm thickness) covering almost entire surface of fillet. 0 = normal; 1 = moderate; 2 = severe; 3 = extreme. Scoring standard for white striping (cited from Kuttappan et al., 2016). Normal–no distinct white lines; moderate–small white lines, generally <1 mm thick, but apparently visible on the fillet surface; severe–large white lines (1–2 mm thick) very visible on the fillet surface; extreme–thick white bands (>2 mm thickness) covering almost entire surface of fillet. 0 = normal; 1 = moderate; 2 = severe; 3 = extreme. Serum antioxidant enzyme, pro-inflammatory factors, and intestinal barrier-related parameters Blood was collected from the wing vein and serum was obtained by centrifuged (1500 × g, 15 min, 4 °C) and stored at −20 °C. In Exp. 2. the activities of T-AOC, SOD, and GSH-Px were analyzed using the assay kits described above. For Exp. 1, the concentrations of tumor necrosis factor α (TNF-α; ml002790), interleukin-1β (IL-1β; ml059835), interferon-γ (INF-γ; ml042758), interleukin-6 (IL-6; ml059839), and diamine oxidase (DAO; ml036981) were quantified. For Exp. 2, the concentrations of inducible nitric oxide synthase (iNOS; ml036925), lysozyme (LYS; ml036945), and immunoglobulin G (IgG; SEKCN-0126) were quantified. All enzyme linked immunosorbent assay (ELISA) kits were purchased from Shanghai Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China), and the optical density was measured using a microplate reader (SpectraMax i3x, Molecular Devices, LLC., San Jose, CA, USA). Ileum immune parameters In Exp. 1, according to the method of Liu et al. (2023b), immune cells from the ileal lamina propria were isolated by mechanical dissociation and then enzymatically digested to obtain a suspension. The ileal content suspension was filtered, centrifuged, washed, and finally resuspended in Roswell Park Memorial Institute-1640 medium (Thermo Fisher Scientific Inc., Waltham, MA, USA) containing 10% fetal bovine serum (FBS) (Thermo Fisher Scientific Inc., Waltham, MA, USA). Single-cell suspensions were stained with fluorescently labeled monoclonal antibodies against chicken cell surface markers: cluster of differentiation 45 (CD45; pan-leukocyte; 8270–02, Southernbiotech, Birmingham, AL, USA), CD3 (T cells; 8200–27, Southernbiotech, Birmingham, AL, USA), CD4 (8210–26, Southernbiotech, Birmingham, AL, USA), CD8α (8220–09, Southernbiotech, Birmingham, AL, USA), CD25 (for Treg identification; HCA173F, Bio-Rad Laboratories, Inc., Hercules, CA, USA), bursal disease virus receptor 1 (Bu-1; B cells; 8395–26, Southernbiotech, Birmingham, AL, USA), major histocompatibility complex class II (MHC class II; 8350–02, Southernbiotech, Birmingham, AL, USA), CD11c (for dendritic cells; 12–0116–42, eBioscience, San Diego, CA, USA), and monocyte (Mon; 8420–31, SouthernBiotech, Birmingham, AL, USA). Data were acquired on a BD LSR II flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA) and analyzed using FlowJo software (Tree Star Inc., Ashland, OR, USA). The ileal mucosa was scraped, snap-frozen in liquid nitrogen, homogenized, and centrifuged for secretory immunoglobulin A (sIgA) analysis using ELISA kit (ml002778, Shanghai Enzyme-linked Biotechnology Co., Ltd., Shanghai, China). Histological analysis A 1-cm segment from the midpoint of the ileum (Exp. 1) or jejunum (Exp. 2) was collected and flushed with 0.9% saline, and then fixed in 4% polyformaldehyde solution. Fixed intestinal segments were embedded in paraffin blocks, cut into sections, and stained with hematoxylin-eosin (H&E). After dehydration, the sections were observed using a Pannoramic MIDI (3DHISTECH Ltd., Budapest, Hungary); the villus height (VH) and crypt depth (CD) were measured using CaseViewer 2.4 (3DHISTECH Ltd., Budapest, Hungary), and the ratio of VH to CD (VH/CD) was calculated. RNA isolation and quantitative real-time PCR Total RNA was extracted from the ileum and muscle tissues using TRIzol reagent (P118–05, GenStar Biosolutions Co., Ltd., Beijing, China) according to the manufacturer's instructions at the end of the trial, and RNA purity and concentration were measured using a spectrophotometer (NanoDrop, 2000; Thermo Fisher Scientific Inc., Waltham, MA, USA). Reverse transcription was performed using StarScript II First-strand cDNA Synthesis Kit (A212–02, GenStar Biosolutions Co., Ltd., Beijing, China). Real-time quantitative PCR was carried out with an Applied Biosystems 7500 Real-time PCR system (Thermo Fisher Scientific Inc., Foster City, CA, USA) and 2 × RealStar SYBR Mixture (A301–10, GenStar Biosolutions Co., Ltd., Beijing, China). The results were normalized to that of β-actin expression, and the relative gene expression levels were calculated via the 2−ΔΔCt method by Livak and Schmittgen (2001). The primer sequences for GCLC, HO-1, GPX, NRF2, MUC2, ZO-1, CLDN, OCLN, IL-1β, IFN-γ, NF-κB, ERK1, and β-actin are given in Table 2. The full names of the above genes are presented in Table S1.Table 2The primer sequences for the target and reference genes.Table 2GenesForward/reverse prime (5′-3′)AccessionGCLCATCCTCCAGTTCCTCCACATCTACCXM_053938427.1AACTTCTTCGCCACTCAGCACTAAGHO-1CAAGAAGTTCACCGTGTGCGXM_046921508.1CCACCGCACCAGGGGAGAGGPXACGGCACCAACGAGGAGATCCXM_032189139.1CTTCCCGTTCACCTGGCACTTCNRF2GGGACGGTGACACAGGAACAACXM_046943489.1GCTCTCCACAGCGGGAAATCAGIL-1βACTGGGCATCAAGGGCTAXM_052702322.1GGTAGAAGATGAAGCGGGTCIFN-γAAAGCCGCACATCAAACACAXM_042872704.1GCCATCAGGAAGGTTGTTTTTCNF-κBTGGAGAAGGCTATGCAGCTTXM_061192591.1CATCCTGGACAGCAGTGAGAERK1CAAGGATATGGTTAAGCTGGTXM_038891758.1AGATTTGTGACTTCGGACTGGMUC2CGGGTGAATGGAGGACAGAGXM_040673077.2CATACACAGTCCCTTCGGGGZO-1TGTAGCCACAGCAAGAGGTGXM_035553853.2CTGGAATGGCTCCTTGTGGTOCLNTCATCGCCTCCATCGTCTACXM_067312817.1TCTTACTGCGCGTCTTCTGGCLDNTGGAGGATGACCAGGTGAAGANM_001013611.2CGAGCCACTCTGTTGCCATAβ-ActinACCGTTCTTTCCACTCTGGCXM_042044413.1AGATACTAGCCCACCCCTCG Table 2The primer sequences for the target and reference genes.Table 2GenesForward/reverse prime (5′-3′)AccessionGCLCATCCTCCAGTTCCTCCACATCTACCXM_053938427.1AACTTCTTCGCCACTCAGCACTAAGHO-1CAAGAAGTTCACCGTGTGCGXM_046921508.1CCACCGCACCAGGGGAGAGGPXACGGCACCAACGAGGAGATCCXM_032189139.1CTTCCCGTTCACCTGGCACTTCNRF2GGGACGGTGACACAGGAACAACXM_046943489.1GCTCTCCACAGCGGGAAATCAGIL-1βACTGGGCATCAAGGGCTAXM_052702322.1GGTAGAAGATGAAGCGGGTCIFN-γAAAGCCGCACATCAAACACAXM_042872704.1GCCATCAGGAAGGTTGTTTTTCNF-κBTGGAGAAGGCTATGCAGCTTXM_061192591.1CATCCTGGACAGCAGTGAGAERK1CAAGGATATGGTTAAGCTGGTXM_038891758.1AGATTTGTGACTTCGGACTGGMUC2CGGGTGAATGGAGGACAGAGXM_040673077.2CATACACAGTCCCTTCGGGGZO-1TGTAGCCACAGCAAGAGGTGXM_035553853.2CTGGAATGGCTCCTTGTGGTOCLNTCATCGCCTCCATCGTCTACXM_067312817.1TCTTACTGCGCGTCTTCTGGCLDNTGGAGGATGACCAGGTGAAGANM_001013611.2CGAGCCACTCTGTTGCCATAβ-ActinACCGTTCTTTCCACTCTGGCXM_042044413.1AGATACTAGCCCACCCCTCG The primer sequences for the target and reference genes. Statistical analysis All the data were analyzed using one-way analysis of variance (ANOVA), and the means were compared using Duncan's multiple range test in SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Spearman's correlation analysis was conducted to evaluate the relationships between muscle antioxidant indices and meat drip loss. The linear and quadratic effects of the supplemental α-MG concentration were assessed using regression analysis, as below. Differences were considered statistically significant at P < 0.05. The data are presented as the mean and pooled standard error of mean (SEM). The regression model was as follows:Yij=β0+β1Xi+eij(linearregression);Yij=β0+β1Xi+β2Xi2+eij(quadraticregression),where Yij is the response variable; β0 is the intercept; β1 and β2 are regression coefficients; Xi is the fixed effect that included α-MG (i = 0, 50, 100, 150, 200, and 250 mg/kg); and eij is the observational error for the ij-th observation. Results Growth performance The growth performance of broilers subjected to LPS challenge and dietary α-MG supplementation is presented in Table 3. In the LPS challenge model (Exp. 1), there was no significant difference in growth performance among all groups from d 0 to d 11 (P > 0.05). The LPS challenge combined with dietary α-MG supplementation decreased body weight gain (BWG) (P < 0.001) and FI (P = 0.004), and increased FCR (P = 0.031) compared with the negative control group from d 12 to d 14, But there was no significant difference among the LPS challenge groups supplemented with different concentrations of α-MG (P > 0.05). Over the entire 21-d period, LPS-challenged birds showed significantly lower overall BWG compared with the negative control group (P < 0.05), and BWG in the α-MG100 group was significantly higher than that in the positive control group (P < 0.05). Effects of α-MG supplementation on growth performance are shown in Table 4. Under normal rearing conditions (Exp. 2), BW and ADG from d 0 to d 21 were higher in the α-MG250 group than in the control group (P < 0.05), and with increasing dietary α-MG levels, both of the indices above showed a linear increment (P = 0.001).Table 3Effects of LPS challenge combined with dietary α-MG supplementation on growth performance in broilers1 (Exp. 1).Table 3ItemsBWG, gFI, gFCR0–11 dNegative control261.71291.681.11Positive control255.42294.861.15α-MG100259.46301.981.16α-MG150258.35303.351.16α-MG200262.65301.801.15α-MG250258.73299.701.16SEM1.2881.4950.006P-value (ANOVA)20.6750.1500.110P-value (linear)0.3270.2680.863P-value (quadratic)0.5210.1310.71812–14 dNegative control113.03a148.99a1.32bPositive control53.10b123.98ab2.47aα-MG10053.23b106.86b2.03abα-MG15039.63b94.99b2.55aα-MG20042.82b102.89b2.33aα-MG25038.48b112.45b2.70aSEM4.9174.5890.138P-value (ANOVA)<0.0010.0040.031P-value (linear)0.0630.2800.555P-value (quadratic)0.8970.0960.5530–21 dNegative control914.76a1205.971.32Positive control838.96c1169.621.38α-MG100883.39ab1177.701.35α-MG150845.15bc1172.601.39α-MG200861.62bc1151.681.34α-MG250851.60bc1164.921.37SEM6.5787.5780.009P-value (ANOVA)0.0020.4680.225P-value (linear)0.7260.8680.422P-value (quadratic)0.2020.7290.262LPS = lipopolysaccharide; α-MG = α-mangostin; BWG = body weight gain; FI = feed intake; FCR = feed conversion ratio; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.2A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups.Table 4Effects of dietary α-MG supplementation on growth performance in broilers1 (Exp. 2).Table 4ItemsBW, gADG, gADFI, gFCR0–21 dControl911.61b41.38b50.721.22α-MG50920.89b41.83b51.211.22α-MG100932.80ab42.39ab50.881.20α-MG150931.43ab42.34ab51.941.22α-MG200929.54ab42.23ab51.081.20α-MG250951.41a43.28a52.101.20SEM3.4030.1620.6430.004P-value (ANOVA)0.0140.0130.2710.409P-value (linear)0.0010.0010.5140.251P-value (quadratic)0.9070.9090.3610.83222–42 dControl2812.2290.50148.331.61α-MG502892.2893.81150.231.59α-MG1002848.9091.25150.891.61α-MG1502928.5995.09151.511.57α-MG2002923.9494.96150.061.57α-MG2502961.1795.70153.781.61SEM16.1600.7480.9780.007P-value (ANOVA)0.0670.2130.7510.340P-value (linear)0.0050.6470.1970.394P-value (quadratic)0.8600.4910.9770.2020–42 dControl2812.2265.9497.641.48α-MG502892.2867.82100.071.48α-MG1002848.9066.8298.511.48α-MG1502928.5968.72100.221.46α-MG2002923.9468.6099.561.45α-MG2502961.1769.50102.541.48SEM16.1600.3860.5190.016P-value (ANOVA)0.0670.0670.1070.418P-value (linear)0.0050.0050.4430.288P-value (quadratic)0.8600.8640.3380.231α-MG = α-mangostin; BW = body weight; ADG = average daily gain; ADFI = average daily feed intake; FCR = feed conversion ratio; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 3Effects of LPS challenge combined with dietary α-MG supplementation on growth performance in broilers1 (Exp. 1).Table 3ItemsBWG, gFI, gFCR0–11 dNegative control261.71291.681.11Positive control255.42294.861.15α-MG100259.46301.981.16α-MG150258.35303.351.16α-MG200262.65301.801.15α-MG250258.73299.701.16SEM1.2881.4950.006P-value (ANOVA)20.6750.1500.110P-value (linear)0.3270.2680.863P-value (quadratic)0.5210.1310.71812–14 dNegative control113.03a148.99a1.32bPositive control53.10b123.98ab2.47aα-MG10053.23b106.86b2.03abα-MG15039.63b94.99b2.55aα-MG20042.82b102.89b2.33aα-MG25038.48b112.45b2.70aSEM4.9174.5890.138P-value (ANOVA)<0.0010.0040.031P-value (linear)0.0630.2800.555P-value (quadratic)0.8970.0960.5530–21 dNegative control914.76a1205.971.32Positive control838.96c1169.621.38α-MG100883.39ab1177.701.35α-MG150845.15bc1172.601.39α-MG200861.62bc1151.681.34α-MG250851.60bc1164.921.37SEM6.5787.5780.009P-value (ANOVA)0.0020.4680.225P-value (linear)0.7260.8680.422P-value (quadratic)0.2020.7290.262LPS = lipopolysaccharide; α-MG = α-mangostin; BWG = body weight gain; FI = feed intake; FCR = feed conversion ratio; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.2A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Effects of LPS challenge combined with dietary α-MG supplementation on growth performance in broilers1 (Exp. 1). LPS = lipopolysaccharide; α-MG = α-mangostin; BWG = body weight gain; FI = feed intake; FCR = feed conversion ratio; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Table 4Effects of dietary α-MG supplementation on growth performance in broilers1 (Exp. 2).Table 4ItemsBW, gADG, gADFI, gFCR0–21 dControl911.61b41.38b50.721.22α-MG50920.89b41.83b51.211.22α-MG100932.80ab42.39ab50.881.20α-MG150931.43ab42.34ab51.941.22α-MG200929.54ab42.23ab51.081.20α-MG250951.41a43.28a52.101.20SEM3.4030.1620.6430.004P-value (ANOVA)0.0140.0130.2710.409P-value (linear)0.0010.0010.5140.251P-value (quadratic)0.9070.9090.3610.83222–42 dControl2812.2290.50148.331.61α-MG502892.2893.81150.231.59α-MG1002848.9091.25150.891.61α-MG1502928.5995.09151.511.57α-MG2002923.9494.96150.061.57α-MG2502961.1795.70153.781.61SEM16.1600.7480.9780.007P-value (ANOVA)0.0670.2130.7510.340P-value (linear)0.0050.6470.1970.394P-value (quadratic)0.8600.4910.9770.2020–42 dControl2812.2265.9497.641.48α-MG502892.2867.82100.071.48α-MG1002848.9066.8298.511.48α-MG1502928.5968.72100.221.46α-MG2002923.9468.6099.561.45α-MG2502961.1769.50102.541.48SEM16.1600.3860.5190.016P-value (ANOVA)0.0670.0670.1070.418P-value (linear)0.0050.0050.4430.288P-value (quadratic)0.8600.8640.3380.231α-MG = α-mangostin; BW = body weight; ADG = average daily gain; ADFI = average daily feed intake; FCR = feed conversion ratio; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on growth performance in broilers1 (Exp. 2). α-MG = α-mangostin; BW = body weight; ADG = average daily gain; ADFI = average daily feed intake; FCR = feed conversion ratio; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Carcass performance Carcass performance of broilers is presented in Table 5 (Exp. 2). The dressed percentage was higher in the α-MG100 and α-MG150 groups than that of the α-MG50 group (P = 0.05). However, there was no difference in half-eviscerated yield, eviscerated yield, breast muscle yield, leg muscle yield, or abdominal fat yield among the different α-MG supplementation groups (P > 0.05; Fig. 1).Table 5Effects of dietary α-MG supplementation on carcass performance (%) in broilers1 (Exp. 2).Table 5ItemsDressed percentageHalf-eviscerated yieldEviscerated yieldBreast muscle yieldLeg muscle yieldAbdominal fat yieldControl93.71ab87.0375.3030.1720.281.22α-MG5093.38b87.3375.3630.4919.951.56α-MG10094.61a88.1476.0229.4420.231.33α-MG15094.49a88.1976.3729.6519.691.44α-MG20094.01ab88.2375.7129.2919.751.37α-MG25094.13ab88.6376.2729.0520.451.37SEM0.1290.1760.1780.2550.1360.050P-value (ANOVA)0.0500.0660.3680.5840.5100.586P-value (linear)0.1360.8710.6410.9280.3750.445P-value (quadratic)0.1010.8580.5650.8330.5100.369α-MG = α-mangostin; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 5Effects of dietary α-MG supplementation on carcass performance (%) in broilers1 (Exp. 2).Table 5ItemsDressed percentageHalf-eviscerated yieldEviscerated yieldBreast muscle yieldLeg muscle yieldAbdominal fat yieldControl93.71ab87.0375.3030.1720.281.22α-MG5093.38b87.3375.3630.4919.951.56α-MG10094.61a88.1476.0229.4420.231.33α-MG15094.49a88.1976.3729.6519.691.44α-MG20094.01ab88.2375.7129.2919.751.37α-MG25094.13ab88.6376.2729.0520.451.37SEM0.1290.1760.1780.2550.1360.050P-value (ANOVA)0.0500.0660.3680.5840.5100.586P-value (linear)0.1360.8710.6410.9280.3750.445P-value (quadratic)0.1010.8580.5650.8330.5100.369α-MG = α-mangostin; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on carcass performance (%) in broilers1 (Exp. 2). α-MG = α-mangostin; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Meat quality and muscle antioxidant As shown in Table 6, dietary α-MG supplementation tended to reduce drip loss (P = 0.059) (Exp. 2). As shown in Table 7, GSH activity was significantly increased in the α-MG150 and α-MG250 groups compared with the control group (P = 0.033). The α-MG150 group had greater SOD activity than the α-MG100 and α-MG200 groups (P = 0.033). However, there were no significant differences on relative mRNA expression level of pectoral muscle antioxidant genes, including GCLC, HO-1, GPX, and NRF2 (P > 0.05; Fig. 2).Table 6Effects of dietary α-MG supplementation on breast muscle quality in broilers1 (Exp. 2).Table 6ItemsDrip loss, %White strip scoreControl5.591.28α-MG504.571.16α-MG1004.461.20α-MG1504.471.05α-MG2004.491.31α-MG2504.841.02SEM0.1140.079P-value (ANOVA)0.0590.793P-value (linear)0.0800.493P-value (quadratic)0.7340.992α-MG = α-mangostin; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.Table 7Effects of dietary α-MG supplementation on muscle antioxidant enzyme activity in broilers1 (Exp. 2).Table 7ItemsMDA, nmol/mg protGSH, μmol/g protT-AOC, mmol/gSOD, U/mg protGSH-Px, U/mg protControl5.0145.17c0.8426.09ab170.73α-MG505.1569.50abc0.8128.49ab152.33α-MG1003.8854.19bc1.0219.00b89.16α-MG1504.8373.88ab0.9238.22a203.07α-MG2006.2267.78abc0.8117.88b136.53α-MG2505.5582.39a0.9026.79ab162.37SEM0.3853.7020.3961.99514.709P-value (ANOVA)0.6630.0330.6460.0330.348P-value (linear)0.4010.2870.8150.8030.241P-value (quadratic)0.4610.1820.4400.7580.166α-MG = α-mangostin; MDA = malondialdehyde; GSH = glutathione; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.Fig. 2Effects of dietary α-mangostin (α-MG) supplementation on muscle relative mRNA expression level of antioxidants in broilers. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Means without a superscript letter are not significantly different (P > 0.05). Table 6Effects of dietary α-MG supplementation on breast muscle quality in broilers1 (Exp. 2).Table 6ItemsDrip loss, %White strip scoreControl5.591.28α-MG504.571.16α-MG1004.461.20α-MG1504.471.05α-MG2004.491.31α-MG2504.841.02SEM0.1140.079P-value (ANOVA)0.0590.793P-value (linear)0.0800.493P-value (quadratic)0.7340.992α-MG = α-mangostin; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on breast muscle quality in broilers1 (Exp. 2). α-MG = α-mangostin; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 7Effects of dietary α-MG supplementation on muscle antioxidant enzyme activity in broilers1 (Exp. 2).Table 7ItemsMDA, nmol/mg protGSH, μmol/g protT-AOC, mmol/gSOD, U/mg protGSH-Px, U/mg protControl5.0145.17c0.8426.09ab170.73α-MG505.1569.50abc0.8128.49ab152.33α-MG1003.8854.19bc1.0219.00b89.16α-MG1504.8373.88ab0.9238.22a203.07α-MG2006.2267.78abc0.8117.88b136.53α-MG2505.5582.39a0.9026.79ab162.37SEM0.3853.7020.3961.99514.709P-value (ANOVA)0.6630.0330.6460.0330.348P-value (linear)0.4010.2870.8150.8030.241P-value (quadratic)0.4610.1820.4400.7580.166α-MG = α-mangostin; MDA = malondialdehyde; GSH = glutathione; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on muscle antioxidant enzyme activity in broilers1 (Exp. 2). α-MG = α-mangostin; MDA = malondialdehyde; GSH = glutathione; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-mangostin (α-MG) supplementation on muscle relative mRNA expression level of antioxidants in broilers. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Means without a superscript letter are not significantly different (P > 0.05). As shown in Table 8 (Exp. 2), Spearman's rank correlation analysis showed that there were no significant correlations between drip loss and any of the antioxidant indices (P > 0.05). A significant positive correlation was observed between GSH-Px and SOD activities (ρ = 0.341; P = 0.049).Table 8The correlation analysis between breast muscle antioxidant indices and drip loss1 (Exp. 2).Table 8ItemsDrip lossMDAGSHT-AOCSODGSH-PxDrip loss1.000MDA−0.0021.000GSH−0.004−0.1911.000T-AOC0.036−0.0770.1131.000SOD−0.0370.239−0.069−0.0741.000GSH-Px−0.1120.1100.110−0.2510.341∗1.000MDA = malondialdehyde; GSH = glutathione; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase.All values represent Pearson correlation coefficients; positive values indicate positive associations, whereas negative values indicate inverse associations.∗Indicates a significant correlation between two indices (P < 0.05, 2-tailed). Table 8The correlation analysis between breast muscle antioxidant indices and drip loss1 (Exp. 2).Table 8ItemsDrip lossMDAGSHT-AOCSODGSH-PxDrip loss1.000MDA−0.0021.000GSH−0.004−0.1911.000T-AOC0.036−0.0770.1131.000SOD−0.0370.239−0.069−0.0741.000GSH-Px−0.1120.1100.110−0.2510.341∗1.000MDA = malondialdehyde; GSH = glutathione; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase.All values represent Pearson correlation coefficients; positive values indicate positive associations, whereas negative values indicate inverse associations.∗Indicates a significant correlation between two indices (P < 0.05, 2-tailed). The correlation analysis between breast muscle antioxidant indices and drip loss1 (Exp. 2). MDA = malondialdehyde; GSH = glutathione; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase. All values represent Pearson correlation coefficients; positive values indicate positive associations, whereas negative values indicate inverse associations. Indicates a significant correlation between two indices (P < 0.05, 2-tailed). Organ indices Organ indices for the two studies are shown in Table 9. In Exp. 1, compared with negative control, LPS challenge (positive control) acutely increased the relative weights of the liver, spleen, and thymus at d 14 (P < 0.05), and the liver and thymus had returned to normal levels by d 21 (P > 0.05). However, the spleen weight in the LPS-challenged groups supplemented with different concentrations of α-MG was significantly higher than that in the negative control group at d 21 (P = 0.006). Compared with the negative control, the addition of 150, 200, and 250 mg/kg of α-MG under LPS challenge had no significant effect on thymus weight at d 14 (P > 0.05). Under normal conditions (Exp. 2), α-MG supplementation showed a tendency to increase the bursa of Fabricius index (P = 0.054).Table 9Effects of LPS challenge or dietary α-MG supplementation on organ indexes in broilers (%).Table 9ItemsLiverSpleenBursa of FabriciusThymusExp. 1,14 d1Negative control3.58c0.08b0.200.13cPositive control4.75ab0.13a0.160.19aα-MG1004.38b0.11a0.210.17abα-MG1504.92a0.14a0.160.16bcα-MG2004.51ab0.12a0.200.16bcα-MG2504.82ab0.12a0.140.15bcSEM0.0980.0060.0100.006P-value (ANOVA)2<0.0010.0230.1480.015P-value (linear)0.7770.6190.6270.007P-value (quadratic)0.3620.9580.0830.540Exp. 1,21 dNegative control3.240.08b0.150.36Positive control2.880.12a0.150.39α-MG1003.250.12a0.130.35α-MG1503.280.11a0.140.38α-MG2003.140.11a0.170.41α-MG2503.130.12a0.130.34SEM0.0560.0050.0080.015P-value (ANOVA)0.3590.0060.7810.798P-value (linear)0.1950.8480.9890.734P-value (quadratic)0.0540.5790.9750.866Exp. 2,42 d 3Control22.780.941.381.85α-MG5022.480.841.622.25α-MG10022.850.921.632.12α-MG15022.080.821.821.81α-MG20021.450.782.282.19α-MG25022.221.081.532.13SEM0.2590.0420.0900.093P-value (ANOVA)40.6770.3850.0540.695P-value (linear)0.2300.6150.0920.637P-value (quadratic)0.8250.1240.0930.893α-MG = α-mangostin; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG in Exp. 1.2For Exp. 1, a one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups.3The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG in Exp. 2.4For Exp. 2, one-way analysis of variance, as well as linear and quadratic trend analyses, were performed across all groups. Table 9Effects of LPS challenge or dietary α-MG supplementation on organ indexes in broilers (%).Table 9ItemsLiverSpleenBursa of FabriciusThymusExp. 1,14 d1Negative control3.58c0.08b0.200.13cPositive control4.75ab0.13a0.160.19aα-MG1004.38b0.11a0.210.17abα-MG1504.92a0.14a0.160.16bcα-MG2004.51ab0.12a0.200.16bcα-MG2504.82ab0.12a0.140.15bcSEM0.0980.0060.0100.006P-value (ANOVA)2<0.0010.0230.1480.015P-value (linear)0.7770.6190.6270.007P-value (quadratic)0.3620.9580.0830.540Exp. 1,21 dNegative control3.240.08b0.150.36Positive control2.880.12a0.150.39α-MG1003.250.12a0.130.35α-MG1503.280.11a0.140.38α-MG2003.140.11a0.170.41α-MG2503.130.12a0.130.34SEM0.0560.0050.0080.015P-value (ANOVA)0.3590.0060.7810.798P-value (linear)0.1950.8480.9890.734P-value (quadratic)0.0540.5790.9750.866Exp. 2,42 d 3Control22.780.941.381.85α-MG5022.480.841.622.25α-MG10022.850.921.632.12α-MG15022.080.821.821.81α-MG20021.450.782.282.19α-MG25022.221.081.532.13SEM0.2590.0420.0900.093P-value (ANOVA)40.6770.3850.0540.695P-value (linear)0.2300.6150.0920.637P-value (quadratic)0.8250.1240.0930.893α-MG = α-mangostin; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG in Exp. 1.2For Exp. 1, a one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups.3The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG in Exp. 2.4For Exp. 2, one-way analysis of variance, as well as linear and quadratic trend analyses, were performed across all groups. Effects of LPS challenge or dietary α-MG supplementation on organ indexes in broilers (%). α-MG = α-mangostin; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG in Exp. 1. For Exp. 1, a one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG in Exp. 2. For Exp. 2, one-way analysis of variance, as well as linear and quadratic trend analyses, were performed across all groups. Serum immune, antioxidant, and pro-inflammatory In Exp. 1, as illustrated in Table 10, the LPS challenge acutely decreased LYS activity in the positive control group compared with the other groups (P < 0.05), and the IgG concentration in the α-MG200 and α-MG250 groups was lower than that in the negative group on d 14 (P < 0.05). On d 21, compared with the negative group, LPS challenge decreased iNOS activity significantly (P < 0.001). Compared with the positive group, the α-MG100 and α-MG150 groups increased iNOS activity (P < 0.001); the α-MG150, α-MG200, and α-MG250 groups increased LYS activity (P < 0.001); and the α-MG200 group increased IgG concentration (P = 0.040). In addition, iNOS activity on d 21 showed a quadratic response (P < 0.001), and LYS activity on d 14 and d 21 and IgG concentration d 21 showed both linear and quadratic responses with increasing of dietary α-MG supplementation (P < 0.05).Table 10Effects of LPS challenge combined with dietary α-MG supplementation on serum immune indices in broilers1 (Exp. 1).Table 10ItemsiNOS, pg/mLLYS, ng/mLIgG, ng/mL14 dNegative control64.08747.98a36.43aPositive control61.12627.32b32.14abα-MG10063.65746.98a34.84abα-MG15059.92792.74a35.44abα-MG20061.03785.15a31.65bα-MG25062.05742.29a31.01bSEM0.71712.4780.620P-value (ANOVA)20.540<0.0010.045P-value (linear)0.9020.0010.161P-value (quadratic)0.801<0.0010.38421 dNegative control64.55a841.31b27.44abPositive control54.53c837.15b22.08bα-MG10061.38ab843.70b24.85abα-MG15062.79a978.42a21.96bα-MG20057.86bc953.90a29.78aα-MG25054.73c933.14a23.80bSEM0.80713.2360.856P-value (ANOVA)<0.001<0.0010.040P-value (linear)0.4500.0010.040P-value (quadratic)<0.0010.0280.026LPS = lipopolysaccharide; α-MG = α-mangostin; iNOS = inducible nitric oxide synthase; LYS = lysozyme; IgG = immunoglobulin G; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.2A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Table 10Effects of LPS challenge combined with dietary α-MG supplementation on serum immune indices in broilers1 (Exp. 1).Table 10ItemsiNOS, pg/mLLYS, ng/mLIgG, ng/mL14 dNegative control64.08747.98a36.43aPositive control61.12627.32b32.14abα-MG10063.65746.98a34.84abα-MG15059.92792.74a35.44abα-MG20061.03785.15a31.65bα-MG25062.05742.29a31.01bSEM0.71712.4780.620P-value (ANOVA)20.540<0.0010.045P-value (linear)0.9020.0010.161P-value (quadratic)0.801<0.0010.38421 dNegative control64.55a841.31b27.44abPositive control54.53c837.15b22.08bα-MG10061.38ab843.70b24.85abα-MG15062.79a978.42a21.96bα-MG20057.86bc953.90a29.78aα-MG25054.73c933.14a23.80bSEM0.80713.2360.856P-value (ANOVA)<0.001<0.0010.040P-value (linear)0.4500.0010.040P-value (quadratic)<0.0010.0280.026LPS = lipopolysaccharide; α-MG = α-mangostin; iNOS = inducible nitric oxide synthase; LYS = lysozyme; IgG = immunoglobulin G; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.2A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Effects of LPS challenge combined with dietary α-MG supplementation on serum immune indices in broilers1 (Exp. 1). LPS = lipopolysaccharide; α-MG = α-mangostin; iNOS = inducible nitric oxide synthase; LYS = lysozyme; IgG = immunoglobulin G; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. In Exp. 2, serum T-AOC activity in the α-MG200 and α-MG250 groups was greater than that in the control group at d 21 (P = 0.034; Table 11). Compared with the control group, greater serum GSH-Px activity was observed in the α-MG100 and α-MG200 groups (P = 0.019), and there was a quadratic response among different treatments at d 21 (P = 0.025). At d 42, there were no significant differences in these indices (P > 0.05). Serum IL-1β concentration was significantly lower in all α-MG supplemented groups than in the control group (P < 0.001; Table 12), and increasing dietary α-MG supplementation linearly and quadratically decreased serum IL-1β concentration at d 21 and d 42, with 150 mg/kg α-MG being the most effective (P < 0.001).Table 11Effects of dietary α-MG supplementation on serum antioxidant enzyme activity in broilers1 (Exp. 2).Table 11ItemsT-AOC, mmol/LSOD, U/mLGSH-Px, U/mL21 dControl0.74b84.161947.32bcα-MG500.75b81.492673.03abα-MG1000.93ab59.932830.04aα-MG1500.97ab60.811973.12abcα-MG2001.06a58.692823.14aα-MG2501.02a59.391804.33cSEM0.0374.006125.449P-value (ANOVA)0.0340.1700.019P-value (linear)0.7480.0210.622P-value (quadratic)0.7200.2880.02542 dControl0.9954.912513.28α-MG500.8666.432665.37α-MG1000.8967.292236.21α-MG1501.0142.752825.30α-MG2001.0243.102652.23α-MG2500.9669.602763.33SEM0.0303.87899.833P-value (ANOVA)0.5690.1290.182P-value (linear)0.4500.0790.206P-value (quadratic)0.7130.0550.595α-MG = α-mangostin; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.Table 12Effects of dietary α-MG supplementation on serum pro-inflammatory factors in broilers1 (Exp. 2).Table 12ItemsTNF-α, pg/mLIL-1β, pg/mLIFN-γ, pg/mLIL-6, pg/mL21 dControl55.13100.08a63.5726.71α-MG5053.5089.37b62.5424.67α-MG10051.2469.87c59.8523.43α-MG15049.0958.63d62.2222.79α-MG20051.5169.98c60.0223.21α-MG25054.0974.98c60.3024.42SEM1.5272.4351.4360.679P-value (ANOVA)0.899<0.0010.9700.587P-value (linear)0.689<0.0010.4970.248P-value (quadratic)0.280<0.0010.8060.14742 dControl53.01124.40a63.9428.40α-MG5062.92106.42b67.5926.85α-MG10057.5588.08c67.6427.27α-MG15056.3568.00d63.0725.07α-MG20054.9190.49c62.0025.95α-MG25055.5887.37c63.9125.52SEM1.9282.9641.4490.759P-value (ANOVA)0.803<0.0010.8410.850P-value (linear)0.763<0.0010.4890.240P-value (quadratic)0.504<0.0010.7010.682α-MG = α-mangostin; TNF-α = tumor necrosis factor-alpha; IL-1β = interleukin-1beta; IFN-γ = interferon-gamma; IL-6 = interleukin-6; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 11Effects of dietary α-MG supplementation on serum antioxidant enzyme activity in broilers1 (Exp. 2).Table 11ItemsT-AOC, mmol/LSOD, U/mLGSH-Px, U/mL21 dControl0.74b84.161947.32bcα-MG500.75b81.492673.03abα-MG1000.93ab59.932830.04aα-MG1500.97ab60.811973.12abcα-MG2001.06a58.692823.14aα-MG2501.02a59.391804.33cSEM0.0374.006125.449P-value (ANOVA)0.0340.1700.019P-value (linear)0.7480.0210.622P-value (quadratic)0.7200.2880.02542 dControl0.9954.912513.28α-MG500.8666.432665.37α-MG1000.8967.292236.21α-MG1501.0142.752825.30α-MG2001.0243.102652.23α-MG2500.9669.602763.33SEM0.0303.87899.833P-value (ANOVA)0.5690.1290.182P-value (linear)0.4500.0790.206P-value (quadratic)0.7130.0550.595α-MG = α-mangostin; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on serum antioxidant enzyme activity in broilers1 (Exp. 2). α-MG = α-mangostin; T-AOC = total antioxidant capacity; SOD = superoxide dismutase; GSH-Px = glutathione peroxidase; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 12Effects of dietary α-MG supplementation on serum pro-inflammatory factors in broilers1 (Exp. 2).Table 12ItemsTNF-α, pg/mLIL-1β, pg/mLIFN-γ, pg/mLIL-6, pg/mL21 dControl55.13100.08a63.5726.71α-MG5053.5089.37b62.5424.67α-MG10051.2469.87c59.8523.43α-MG15049.0958.63d62.2222.79α-MG20051.5169.98c60.0223.21α-MG25054.0974.98c60.3024.42SEM1.5272.4351.4360.679P-value (ANOVA)0.899<0.0010.9700.587P-value (linear)0.689<0.0010.4970.248P-value (quadratic)0.280<0.0010.8060.14742 dControl53.01124.40a63.9428.40α-MG5062.92106.42b67.5926.85α-MG10057.5588.08c67.6427.27α-MG15056.3568.00d63.0725.07α-MG20054.9190.49c62.0025.95α-MG25055.5887.37c63.9125.52SEM1.9282.9641.4490.759P-value (ANOVA)0.803<0.0010.8410.850P-value (linear)0.763<0.0010.4890.240P-value (quadratic)0.504<0.0010.7010.682α-MG = α-mangostin; TNF-α = tumor necrosis factor-alpha; IL-1β = interleukin-1beta; IFN-γ = interferon-gamma; IL-6 = interleukin-6; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on serum pro-inflammatory factors in broilers1 (Exp. 2). α-MG = α-mangostin; TNF-α = tumor necrosis factor-alpha; IL-1β = interleukin-1beta; IFN-γ = interferon-gamma; IL-6 = interleukin-6; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Intestinal barriers In Exp. 1, as presented in Table 13, CD was lower in the α-MG150, α-MG200, and α-MG250 groups than in the positive control group (P = 0.016), VH/CD was higher in the α-MG150 group than in the negative and positive control groups; and the positive control group showed the lowest VH/CD among all groups (P < 0.001). With the increment of α-MG supplementation, the CD decreased linearly (P = 0.001), while VH/CD showed a significant linear increase (P < 0.001). In Exp. 2, CD in the α-MG50 and α-MG100 groups tended to decrease compared with the control group at d 21 (P = 0.059). At d 42, VH in the α-MG100 and α-MG150 groups increased significantly compared with the control group (P < 0.001), and with the increment of α-MG supplementation, the VH increased linearly (P = 0.007; Table 14).Table 13Effects of LPS challenge combined with dietary α-MG supplementation on ileal morphology in broilers1 (Exp. 1).Table 13ItemsVH, μmCD, μmVH/CDNegative control758.33158.04abc4.92bPositive control743.14174.65a4.262cα-MG100803.20162.57ab4.95bα-MG150782.85135.09c5.80aα-MG200770.90145.36bc5.32abα-MG250800.19146.88bc5.51abSEM11.4613.6390.114P-value (ANOVA)0.6460.016<0.001P-value (linear)0.2910.001<0.001P-value (quadratic)0.5340.0950.005LPS = lipopolysaccharide; α-MG = α-mangostin; VH = villus height; CD = crypt depth; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.Table 14Effects of dietary α-MG supplementation on jejunal morphology in broilers1 (Exp. 2).Table 14ItemsVH, μmCD, μmVH/CD21 dControl1020.66220.944.68α-MG50912.16168.455.63α-MG100906.37160.525.77α-MG1501067.14187.715.79α-MG2001003.01187.905.36α-MG250948.57198.874.97SEM23.5826.1830.159P-value (ANOVA)0.2920.0590.235P-value (linear)0.8790.8350.825P-value (quadratic)0.9440.2420.91542 dControl793.83c158.955.48α-MG501109.10b185.625.90α-MG1001415.38a231.106.33α-MG1501428.01a208.377.06α-MG2001228.84ab181.457.00α-MG2501381.72ab216.786.59SEM48.2688.7500.215P-value (ANOVA)<0.0010.1830.276P-value (linear)0.0070.1960.692P-value (quadratic)0.1150.2060.886α-MG = α-mangostin; VH = villus height; CD = crypt depth; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 13Effects of LPS challenge combined with dietary α-MG supplementation on ileal morphology in broilers1 (Exp. 1).Table 13ItemsVH, μmCD, μmVH/CDNegative control758.33158.04abc4.92bPositive control743.14174.65a4.262cα-MG100803.20162.57ab4.95bα-MG150782.85135.09c5.80aα-MG200770.90145.36bc5.32abα-MG250800.19146.88bc5.51abSEM11.4613.6390.114P-value (ANOVA)0.6460.016<0.001P-value (linear)0.2910.001<0.001P-value (quadratic)0.5340.0950.005LPS = lipopolysaccharide; α-MG = α-mangostin; VH = villus height; CD = crypt depth; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of LPS challenge combined with dietary α-MG supplementation on ileal morphology in broilers1 (Exp. 1). LPS = lipopolysaccharide; α-MG = α-mangostin; VH = villus height; CD = crypt depth; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Table 14Effects of dietary α-MG supplementation on jejunal morphology in broilers1 (Exp. 2).Table 14ItemsVH, μmCD, μmVH/CD21 dControl1020.66220.944.68α-MG50912.16168.455.63α-MG100906.37160.525.77α-MG1501067.14187.715.79α-MG2001003.01187.905.36α-MG250948.57198.874.97SEM23.5826.1830.159P-value (ANOVA)0.2920.0590.235P-value (linear)0.8790.8350.825P-value (quadratic)0.9440.2420.91542 dControl793.83c158.955.48α-MG501109.10b185.625.90α-MG1001415.38a231.106.33α-MG1501428.01a208.377.06α-MG2001228.84ab181.457.00α-MG2501381.72ab216.786.59SEM48.2688.7500.215P-value (ANOVA)<0.0010.1830.276P-value (linear)0.0070.1960.692P-value (quadratic)0.1150.2060.886α-MG = α-mangostin; VH = villus height; CD = crypt depth; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on jejunal morphology in broilers1 (Exp. 2). α-MG = α-mangostin; VH = villus height; CD = crypt depth; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. In Exp. 2, as shown in Table 15, serum DAO activity was lower in the α-MG100 and α-MG150 groups than that in the control group, whereas it was higher in the α-MG50 and α-MG200 groups at d 21 (P < 0.001). At d 42, serum DAO activity in the α-MG100, α-MG150, α-MG200, and α-MG250 groups was lower than that in the control and α-MG50 groups (P < 0.001). With the increment of dietary α-MG supplementation, serum DAO activity changed linearly and quadratically at d 21 and d 42 (P < 0.05). Fig. 3 showes that the addition of α-MG, particularly at 150 mg/kg, tended to upregulate the mRNA expression level of MUC2 in the ileum (P = 0.084).Table 15Effects of dietary α-MG supplementation on serum DAO activity in broilers1 (Exp. 2).Table 15ItemsDAO, U/mL21 d42 dControl157.38c250.02aα-MG50176.94b256.48aα-MG10086.12d204.31bα-MG15078.97d147.01cα-MG200193.04a160.84cα-MG250168.35bc159.46cSEM7.5477.484P-value (ANOVA)<0.001<0.001P-value (linear)0.027<0.001P-value (quadratic)<0.001<0.001α-MG = α-mangostin; DAO = diamine oxidase; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.Fig. 3Effects of dietary α-mangostin (α-MG) supplementation on ileal relative mRNA expression level of intestinal barrier in broilers. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Means without a superscript letter are not significantly different (P > 0.05). Table 15Effects of dietary α-MG supplementation on serum DAO activity in broilers1 (Exp. 2).Table 15ItemsDAO, U/mL21 d42 dControl157.38c250.02aα-MG50176.94b256.48aα-MG10086.12d204.31bα-MG15078.97d147.01cα-MG200193.04a160.84cα-MG250168.35bc159.46cSEM7.5477.484P-value (ANOVA)<0.001<0.001P-value (linear)0.027<0.001P-value (quadratic)<0.001<0.001α-MG = α-mangostin; DAO = diamine oxidase; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-MG supplementation on serum DAO activity in broilers1 (Exp. 2). α-MG = α-mangostin; DAO = diamine oxidase; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Effects of dietary α-mangostin (α-MG) supplementation on ileal relative mRNA expression level of intestinal barrier in broilers. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. Means without a superscript letter are not significantly different (P > 0.05). Intestinal immune The LPS challenge (Exp. 1) significantly decreased the CD4+CD8-/CD3+ ratio (P = 0.021), and dietary supplementation with 150 mg/kg α-MG reversed it (Table 16). The α-MG150 group significantly increased the proportion of MHC-II+CD11c+/CD45+ compared with the positive group (P = 0.042). The LPS challenge acutely decreased sIgA concentration in the ileum mucosa at d 14 (P = 0.040; Table 17), and dietary supplementation with 150 mg/kg α-MG reversed it at d 21 (P = 0.011). With the increasing dietary α-MG supplementation, the sIgA concentration increased quadratically on d 21 (P = 0.004).Table 16Effects of LPS challenge combined with dietary α-MG supplementation on intestinal immune index (%) in broilers1 (Exp. 1).Table 16ItemsCD3+/CD45+CD4+CD8-/CD3+Bu-1+/CD45+Mon+/CD45+MHC-II+CD11c+/CD45+Negative control18.4030.37a10.484.971.24abPositive control28.6013.77b11.906.220.57bα-MG15031.8323.53ab13.034.992.60aSEM2.9812.8357.5892.5180.369P-value0.1570.0210.8870.9240.042LPS = lipopolysaccharide; α-MG = α-mangostin; CD = cluster of differentiation; Bu-1 = bursal disease virus receptor 1; Mon = monocyte; MHC = major histocompatibility complex; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, the positive control and α-MG150 groups were challenged with LPS, and fed diets containing 0 and 150 mg/kg α-MG, respectively.Table 17Effects of LPS challenge combined with dietary α-MG supplementation on intestinal mucosal sIgA concentration in broilers1 (Exp. 1).Table 17ItemssIgA, ng/mL14 d21 dNegative control436.63a389.80bPositive control379.92b375.58bα-MG100381.45b397.84bα-MG150365.89b459.06aα-MG200379.22b416.80abα-MG250369.86b392.58bSEM7.1507.612P-value (ANOVA)20.0400.011P-value (linear)0.9050.321P-value (quadratic)0.7640.004LPS = lipopolysaccharide; α-MG = α-mangostin; sIgA = secretory immunoglobulin A; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.2A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Table 16Effects of LPS challenge combined with dietary α-MG supplementation on intestinal immune index (%) in broilers1 (Exp. 1).Table 16ItemsCD3+/CD45+CD4+CD8-/CD3+Bu-1+/CD45+Mon+/CD45+MHC-II+CD11c+/CD45+Negative control18.4030.37a10.484.971.24abPositive control28.6013.77b11.906.220.57bα-MG15031.8323.53ab13.034.992.60aSEM2.9812.8357.5892.5180.369P-value0.1570.0210.8870.9240.042LPS = lipopolysaccharide; α-MG = α-mangostin; CD = cluster of differentiation; Bu-1 = bursal disease virus receptor 1; Mon = monocyte; MHC = major histocompatibility complex; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, the positive control and α-MG150 groups were challenged with LPS, and fed diets containing 0 and 150 mg/kg α-MG, respectively. Effects of LPS challenge combined with dietary α-MG supplementation on intestinal immune index (%) in broilers1 (Exp. 1). LPS = lipopolysaccharide; α-MG = α-mangostin; CD = cluster of differentiation; Bu-1 = bursal disease virus receptor 1; Mon = monocyte; MHC = major histocompatibility complex; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The negative control group received a basal diet and was injected with saline, the positive control and α-MG150 groups were challenged with LPS, and fed diets containing 0 and 150 mg/kg α-MG, respectively. Table 17Effects of LPS challenge combined with dietary α-MG supplementation on intestinal mucosal sIgA concentration in broilers1 (Exp. 1).Table 17ItemssIgA, ng/mL14 d21 dNegative control436.63a389.80bPositive control379.92b375.58bα-MG100381.45b397.84bα-MG150365.89b459.06aα-MG200379.22b416.80abα-MG250369.86b392.58bSEM7.1507.612P-value (ANOVA)20.0400.011P-value (linear)0.9050.321P-value (quadratic)0.7640.004LPS = lipopolysaccharide; α-MG = α-mangostin; sIgA = secretory immunoglobulin A; SEM = standard error of the mean.Within the same row with no common superscript differ significantly (P < 0.05); n = 6.1The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG.2A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Effects of LPS challenge combined with dietary α-MG supplementation on intestinal mucosal sIgA concentration in broilers1 (Exp. 1). LPS = lipopolysaccharide; α-MG = α-mangostin; sIgA = secretory immunoglobulin A; SEM = standard error of the mean. Within the same row with no common superscript differ significantly (P < 0.05); n = 6. The negative control group received a basal diet and was injected with saline, and the other groups were challenged with LPS, and fed diets containing 0 (positive control), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. A one-way analysis of variance was performed across all groups, while linear and quadratic trend analyses were performed across positive control, α-MG100, α-MG150, α-MG200, and α-MG250 groups. Changes in the pro-inflammatory factors and ERK1 expression in the ileum by dietary α-MG are presented in Fig. 4. Under normal rearing conditions (Exp. 2), α-MG supplementation downregulated the mRNA expression of pro-inflammatory genes, including IL-1β, IFN-γ, and NF-κB (P < 0.05). The expression of ERK1 was significantly upregulated in the α-MG150 group (P = 0.001).Fig. 4Effects of dietary α-mangostin (α-MG) supplementation on ileal relative mRNA expression level of pro-inflammatory factors and ERK1 in broilers. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. The different superscript letters denote significant differences among experimental treatments (P < 0.05). Effects of dietary α-mangostin (α-MG) supplementation on ileal relative mRNA expression level of pro-inflammatory factors and ERK1 in broilers. The control group received a basal diet, the other groups were fed diets containing 50 (α-MG50), 100 (α-MG100), 150 (α-MG150), 200 (α-MG200), and 250 (α-MG250) mg/kg α-MG. The different superscript letters denote significant differences among experimental treatments (P < 0.05). Discussion The intensification of poultry production has increased the prevalence of various stressors, which have a detrimental impact on intestinal health and growth, resulting in significant economic losses (Sindiyo et al., 2018; Tompkins et al., 2023). Identifying safe, effective, and natural alternatives to reduce reliance on antibiotics is essential for the poultry industry (Chen et al., 2025; Pham et al., 2023). This study demonstrated that dietary supplementation with α-MG effectively alleviated LPS-induced immune stress and improved growth performance, intestinal health, and antioxidant status in broilers under normal rearing conditions. Under the conditions of this study, 150 mg/kg was considered the optimal supplemental dose. The significant impairment of growth performance following LPS challenge has been widely reported, where nutrients are diverted from chicken growth to support immune system activation (Fu et al., 2023; Tan et al., 2023). The findings showed that dietary α-MG supplementation mitigated this growth suppression, aligning with the reported effects of other flavonoids (Shi et al., 2022; Song et al., 2025; Yang et al., 2021). Moreover, under normal rearing conditions, α-MG improved ADG and BW in a dose-dependent manner. This indicated that α-MG not only alleviated LPS-induced damage, but also improved growth performance. Both meat yield and the economic benefits of the broiler industry were both significantly influenced by slaughter performance (Huang et al., 2024; Yang et al., 2025). The quality of fresh poultry meat was largely determined by its water-holding capacity, which was linked to consumer acceptance and processed meat yield (Kaić et al., 2021; Pu et al., 2025). The improvement in processing yields and the notable reduction of breast muscle drip loss may provide further commercial relevance. Notably, previous studies have shown that drip loss is closely linked to muscle membrane integrity, which was vulnerable to oxidative damage (Ibrahim et al., 2019; Malila et al., 2023; Zhu et al., 2024). In this study, dietary α-MG supplementation significantly increased breast muscle GSH activity, a key indicator of antioxidant capacity. Serum T-AOC was markedly elevated in α-MG-supplemented groups at d 21, with the α-MG200 and α-MG250 groups showing a significant increase compared with the control group. Concurrently, serum GSH-Px activity was also enhanced, particularly in the α-MG100 and α-MG200 groups. These results indicate that α-MG supplementation effectively boosts systemic and muscle antioxidant defense, which may play a crucial role in mitigating oxidative stress induced by metabolic challenge. However, no significant correlations were found between drip loss and the measured antioxidant indices, suggesting that the enhanced antioxidant capacity may not directly translate into improved meat quality under the current experimental conditions. Further studies with more sensitive meat quality indicators are warranted to explore this relationship. In summary, dietary α-MG supplementation improved growth performance in both quantitative (growth and yield) and qualitative (antioxidant capacity) aspects. The results of this study showed that dietary α-MG supplementation could enhance immunity and regulate inflammation. Changes in relative organ weights are sensitive indicators of toxicity disturbance in toxicity studies (Tavares et al., 2023). In the LPS-challenged model, a transient but significant rise was observed the relative of the liver, spleen, and thymus at 14 d post-inoculation. It's a classic marker of systemic immune activation; the liver engages in acute-phase protein synthesis and detoxification, while the spleen and thymus undergo lymphocyte proliferation and activation to mount a defense (Choi et al., 2023; Gong et al., 2025). The subsequent return of these indices to basal levels indicates that the acute inflammation had been alleviated by d 21. Notably, α-MG suppressed the relative weight of the thymus compared with LPS challenge on d 14, suggesting that it alleviates the adverse immune reactions without excessively impairing host defense. Under normal rearing conditions, dietary α-MG supplementation showed a tendency to increase the bursa of Fabricius index. The bursa is a humoral immune organ that plays a role in B lymphocyte maturation and diversification (Dong et al., 2024; Fu et al., 2024). An enlarged bursa index is often associated with enhanced B-cell development and antibody-producing capacity within the physiological range (Sastré-Calderón et al., 2024; Tang et al., 2021). This finding was in accordance with the increment in serum IgG concentration in the α-MG200 group and supports the concept that α-MG positively modulated humoral immunity. Additionally, the lack of significant effects on liver and spleen indices under normal rearing conditions further indicates that the recommended dose of α-MG (150 mg/kg) does not induce pathological organ hypertrophy, affirming its safety in broiler diets. Furthermore, under normal rearing conditions, serum IL-1β concentration was significantly and dose-dependently reduced, and ileal IL-1β, IFN-γ, and the master regulator NF-κB mRNA expression levels were downregulated significantly. Excessive production of pro-inflammatory factors consumes a large amount of nutrients for the immune response rather than for growth (Yang et al., 2021). It also could lead to tissue and organ dysfunction and progressive development of pathology, which may explain the observed changes in organ index (Bajgar and Krejčová, 2023). The concurrent reduction in serum IL-1β concentration and increase in ileal IL-1β and ERK1 mRNA expression levels were observed in the α-MG150 group. This may imply the existence of a regulatory or compensatory feedback mechanism at the transcriptional level. The upregulation of ERK1 mRNA expression level suggests the potential engagement of the mitogen-activated protein kinase (MAPK) signaling pathway, which could be involved in tissue repair and immune modulation. The ERK1 is a key component of the MAPK cascade that regulates cell proliferation, differentiation, and survival (Noorbakhsh et al., 2024). In the intestinal epithelium, ERK1 activation has been shown to promote epithelial cell migration and wound healing, thereby contributing to the restoration of mucosal integrity following injury (Roth et al., 2021). Additionally, ERK signaling can modulate immune responses by influencing cytokine production and immune cell function (Lucas et al., 2022). Thus, the upregulation of ERK1 expression level in the ileum of α-MG supplemented broilers may represent a protective mechanism that facilitates tissue repair and immune homeostasis. However, further investigation at the protein (e.g., phosphorylated ERK) and functional levels is necessary to elucidate this mechanism. These alterations contributed to dietary 150 mg/kg α-MG supplementation, which may inhibit the mRNA expression of NF-κB pathway to alleviate intestinal inflammation and prevent pathological organ hypertrophy. The status of gut health can be determined by intestinal morphology (Recharla et al., 2021), and increased VH and a higher VH/CD ratio are indicative of healthy intestinal morphology (Wang et al., 2024). The LPS challenge induced deepened crypts and a reduced VH/CD ratio, which indicates rapid enterocyte turnover and increased mucosal tissue maintenance requirements (Ayoola et al., 2015; Yu et al., 2024). Similar to other flavonoid additives (Shen et al., 2025; Zhi et al., 2024), the results showed that the supplementation with α-MG, particularly at 150 mg/kg, effectively restored ileal morphology. Under normal rearing conditions, α-MG supplementation significantly increased jejunal VH, indicating that α-MG supplementation could improve the intestinal absorptive capacity rather than repairing intestinal damage, providing a direct mechanistic explanation for the improved nutrient utilization and growth performance. Moreover, the significantly changed in serum DAO activity, an enzyme localized in mature intestinal villus that leaks upon mucosal damage, provide direct biochemical support for the morphological evidence (Tang et al., 2022). The mRNA expression levels of key tight junction proteins (CLDN, OCLN, and ZO-1) and mucin (MUC2) in the ileum was not pronounced; this suggests that α-MG enhances the intestinal barrier primarily through post-translational mechanisms, likely by mitigating the inflammatory and oxidative stress that disrupts junctional complexes, consistent with the observed anti-inflammatory and antioxidant activities. Overall, dietary 150 mg/kg α-MG improved the absorption capacity and integrity of the intestines. The cecal tonsil is the largest intestinal lymphoid tissue in poultry, and its indispensable immune function indirectly contributes to intestinal integrity (Guo et al., 2023). Mucosal sIgA plays a critical role in the immune system (Min et al., 2023). The present study showed that LPS challenge may disturb the cellular immune landscape in the ileal lamina propria, markedly reducing the proportion of CD4+ T helper cells and suppressing the secretion of sIgA for 14 d. Dietary 150 mg/kg α-MG supplementation restored these parameters and increase the proportion of MHC-II+CD11c+ dendritic cells, and also improved sIgA secretion, suggesting a role in re-establishing immune homeostasis and potentiating antigen presentation. These changes were consistent with the intestinal morphology, indicating that the improved intestinal integrity may be linked to improved function of the cecal tonsil and sIgA secretion. Conclusion In conclusion, as illustrated in Fig. 5, dietary supplementation with 150 mg/kg α-MG improved growth performance and intestinal health under both LPS-challenge and normal conditions. Additionally, an improved water-holding capacity and increase in antioxidant capacity were observed, suggesting potential benefits in meat quality improvement. The main benefits are achieved through increasing antioxidant capacity, downregulation of NF-κB pathway-related genes expression, enhancing intestinal mucosal immune function, and improving intestinal integrity.Fig. 5Dietary α-MG alleviates LPS-induced stress and enhances growth performance in broilers via improved intestinal immunity and integrity. LPS = lipopolysaccharide; α-MG = α-mangostin; IL-1β = interleukin-1 beta; IFN-γ = interferon-gamma; NF-κB = nuclear factor kappa-B; sIgA = secretory immunoglobulin A. Dietary α-MG alleviates LPS-induced stress and enhances growth performance in broilers via improved intestinal immunity and integrity. LPS = lipopolysaccharide; α-MG = α-mangostin; IL-1β = interleukin-1 beta; IFN-γ = interferon-gamma; NF-κB = nuclear factor kappa-B; sIgA = secretory immunoglobulin A. Credit Author Statement Kaibo Fu: Writing – original draft, Conceptualization. Yuan Yue: Data curation. Jian Zhao: Funding acquisition. Xianlei Li: Methodology. Bingkun Zhang: Writing – review & editing, Supervision, Resources. 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.