Hindgut-targeted delivery alpha-ketoglutarate pellets improves broiler intestinal redox and growth performance by activating uncoupling protein 2

Abstract

Glutamate metabolism is crucial for intestinal epithelial metabolism, but its availability is limited in the hindgut. We used a hindgut-targeted delivery system to deliver L-glutamate and its metabolite α-ketoglutarate (AKG) to investigate the effects of glutamate metabolites on broiler growth and to reveal the potential underlying mechanism. In a three-week feeding trial, fifty-four broilers were randomly divided into three treatments, basal diet (control), basal diet supplemented with 0.1 % L-glutamate (L-Glu) pellets and basal diet supplemented with 0.1 % AKG pellets. The growth performance, intestinal health, microbiota and ammonia nitrogen metabolism were examined. Then, RNA sequencing was adopted to reveal transcriptomic profile of cecal mucosal epithelial cells. Finally, the potential mechanism of AKG in regulating intestinal redox homeostasis was also investigated by in vitro cell model. The results demonstrated that hindgut-targeted delivery of AKG significantly increased average daily gain and decreased the feed-to-gain ratio (F/G), which is better than the effect of L-Glu. Although AKG did not significantly affect intestinal barrier function, inflammation, or microbiota composition, it markedly reduced serum levels of phenylalanine, tryptophan and valine, while significantly reduced ammonia content in cecal contents. Transcriptomic sequencing showed that AKG significantly downregulated oxidative phosphorylation (OXPHOS) in the cecal mucosa and reduced oxidative damage. In vitro experiments confirmed that AKG significantly reduced OXPHOS and alleviated oxidative damage in colonocytes. Further, cellular thermal shift assay (CETSA) and pharmacological study revealed that uncoupling protein 2 (UCP2) mediated the effect of AKG on redox status of colonocytes. Thus, hindgut-targeted delivery of AKG enhances broiler growth performance by activating UCP2 in intestinal epithelial cells, reducing OXPHOS and alleviating oxidative damage in the intestinal mucosa. These findings support precise strategies for regulating growth in poultry production.

Introduction

Intestinal epithelial cells perform essential functions for animal growth, including nutrient absorption (Suzuki, 2020), physical and biochemical barrier maintenance (Macura et al., 2024; Peterson and Artis, 2014), and immune-metabolic regulation (Dmytriv et al., 2024; Haque et al., 2024). These cells utilize various metabolic substrates preferentially, with the foregut epithelial cells prefer to glucose and glutamate (He et al., 2022; He and Wu, 2022), while the hindgut favorite glutamine and butyrate (Cherbuy et al., 1995). Insufficient availability of these substrates compromised the intestinal epithelial barrier, disrupt immune homeostasis, and ultimately impair animal growth (Montoro-Huguet et al., 2021; Yu et al., 2025). Barrier dysfunction is often accompanied by oxidative damage, which increases intestinal permeability, triggers inflammatory factor release, and may lead to leaky gut (Camilleri, 2019). Both oxidative stress and inflammation in intestinal epithelial cells impaired animal growth performance (Jiang et al., 2025; Zhang et al., 2022).

Compared with other amino acids, intestinal epithelial cells prefer to metabolize glutamate (He et al., 2022; He and Wu, 2022), with approximately 95 % of dietary glutamate is metabolized within these cells (Reeds et al., 1996). Glutamate is involved in key intestinal functions such as energy metabolism (Tomé, 2018), barrier maintenance (Vermeulen et al., 2011), and immune regulation (Ruth and Field, 2013). Glutamate serves not only as a precursor for various important mucosal molecules but can also be metabolized to the bioactive metabolite α-ketoglutarate (AKG) (Macelline et al., 2025; Tomé, 2018). AKG exhibits multiple functions, including delaying aging (Asadi Shahmirzadi et al., 2020), modulating immune function (Wang et al., 2025a), promoting bone growth (Tomaszewska et al., 2021), and alleviating colitis (Si et al., 2022). However, nearly all dietary glutamate is absorbed and degraded by the small intestinal mucosa (Wu, 1998). The availability of glutamate in hindgut is limited and the physiological function of glutamate and its metabolites in the hindgut is often overlooked.

Therefore, to investigate the role and the underlying mechanism of glutamate and its metabolites in hindgut, we employed hindgut-targeted controlled-release delivery of glutamate and AKG and dietary supplemented these coated pellets to delivery glutamate and AKG in the hindgut of broilers. Then, the effects of hindgut-targeted delivery of glutamate and AKG on growth performance, nitrogen metabolism and intestinal health were examined. To reveal the potential mechanisms for AKG, we also analyzed the transcriptome and focused on oxidative phosphorylation (OXPHOS) and redox balance. Further, based on in vitro Caco-2 cell model, we revealed the crucial role of uncoupling protein 2 (UCP2) in mediating the function of AKG on colonocytes metabolism, including oxygen consumption rate (OCR) and mitochondrial membrane potential, by molecular docking and pharmacological inhibition. Together, these results firstly revealed a novel role of hindgut delivery AKG in maintaining intestinal redox homeostasis and broiler growth performance, which highlighting a new strategy for targeting delivery nutrients or metabolites to regulate intestinal health and growth performance.

Material and method

This study followed the guidelines of and was approved by the Animal Care and Use Committee of South China Agricultural University (Guangzhou, China; approval code: 2026A003).

Animals, housing, experimental design and diets

A total of 54 male white-feathered broilers (21 days of age) with uniform body weight (Arbor Acre male  ×  Ross female cross broilers, Qingnong Technology Co., Ltd., Guangdong, China,) with were housed individually in single cages and randomly allocated, based on body weight, to three dietary treatments, with 18 replicates per treatment and one bird per replicate. Initial body weight did not differ significantly among groups (P > 0.05). The pellets were prepared using the extrusion–spheronization method as previously described Jiang et al. (2025) Briefly, the active material was blended with microcrystalline cellulose (Henan Wanbang Chemical Technology Co., Ltd.) at a 1:2 ratio and then extruded; blank pellets contained only microcrystalline cellulose. The extrudates were subsequently rounded in a pelletizer (LBLX-360, Jiujiang Dongsheng Machinery Manufacturing Co., Ltd., China). The resulting pellets were dried at 55 °C for 12 h to obtain an optimal and uniform moisture content. To meet the particle-size requirements for poultry feed ingredients, pellets with a size of 30-40 mesh were collected by sieving, with an average pellet weight of approximately 0.03 mg. Finally, an acrylic resin (Youcare Pharmaceutical Group, Anhui, China) was applied to the pellet surface using a fluidized-bed coater (LDP-1.5, Changzhou Jintian Drying and Granulating Equipment Co., Ltd., China). The basal diet was formulated to meet the nutritional standards of National Research Council (NRC) (1994), NY/T33-2004, and the broiler feeding manual. The basal diet was provided by Huayang Feed Co., Ltd (Foshan, China). Its composition and nutritional levels are presented in the Table 1. Two experimental groups received the basal diet supplemented with 0.1 % micro-pellets coated with either AKG (328-50-7, Jing Jing Pharmaceutical Co., Ltd, Xingtai, China) or L-glutamate (L-Glu) (G2604, Macklin, Shanghai, China). The experiment ran for 21 days, with feed and water provided ad libitum.

Table 1.

Composition of basal diet (as fed).

Items	Diet %
Corn	65.80
Soybean meal (46 %)	23.40
Corn gluten meal	5.00
Soybean Oil	2.10
Limestone	1.20
Calcium hydrogen phosphate	1.00
NaCl	0.1650
Propionic acid antifungal agent d	0.0518
Premix of trace elements a	0.1037
Vitamin premix b	0.0311
L-Lysine	0.3628
DL Methionine	0.1451
L-Threonine	0.0311
Choline chloride	0.1037
Antioxidants e	0.0104
Heat-resistant phytase d	0.0104
AB Enzyme Complex c	0.0031
Sodium humate	0.2073
Mannanase (K302) d	0.0104
Sodium bicarbonate	0.1555
Zeolite d	0.1161
Total	100
Nutrient content
Metabolizable energy (kcal/kg)	3050
Crude protein (%)	19.187
Calcium (%)	0.760
Available phosphorus (%)	0.487
Lysine (%)	1.103
Methionine (%)	0.411
Methionine + Cystine (%)	0.839
Threonine (%)	0.794
Tryptophan (%)	0.215

Note: ^a Trace element premix (mg/kg feed): Mn 100, Fe 80, Cu 8, Zn 75, Se 0.15, I 0.35. ^b Vitamin premix (provided per kilogram of feed): Vitamin A 12500 IU, Vitamin B1 2 mg, Vitamin B2 6 mg, Vitamin B12 0.025 mg, Vitamin D3 2500 IU, Vitamin E 30 IU, Vitamin K3 2.65 mg, pantothenic acid 12 mg, biotin 0.0325 mg, niacin 50 mg, folic acid 1.25 mg. ^c AB Enzyme Complex: Acid pro-tease, Neutral proteases, Xylanase, Cellulase, Glucoamylase. ^d Heat-resistant phytase, Man-nanase (K302) and Zeolite are provided by SUNHY Co (Wuhan, China), propionic acid antifun-gal agent is provided by KEMIN (China) Technology Co (Guangzhou, China). ^e A mixer of ethoxyquin (EQ), butylated hydroxytoluene (BHT) and butyl hydroxy anisd (BHA).

Growth performance

Body weight was recorded at 21- and 42- days of age as the initial and final body weight, respectively, and feed intake was measured throughout the entire experimental period. We then calculated average daily gain, average daily feed intake, and feed-to-gain ratio (F/G) for the entire experimental period.

Sample collection

At 42 days of age, nine healthy broilers were randomly selected from each treatment group, live weight before slaughter were measured and then euthanized. Dressed weight, eviscerated weight, abdominal fat weight, and liver weight were measured, from which the dressed percentage, percentage of eviscerated yield, percentage of abdominal fat yield (Zhang et al., 2024a), and liver index were calculated. The calculation formula is as follows: dressed percentage = (dressed weight / live weight before slaughter) × 100 %, percentage of eviscerated yield = (eviscerated weight / live weight before slaughter) × 100 %, percentage of abdominal fat yield = (abdominal fat weight / eviscerated weight) × 100 %, liver index = (liver weight / live weight before slaughter) × 100 %.

Blood samples were collected and allowed to clot at room temperature for 2 hours. The samples were then centrifuged at 4 °C (3,000 rpm for 15 min) to obtain serum, which was stored at −20 °C until analysis. The cecum was gently straightened to its natural length and measured, then rinsed with 0.9 % saline. Both the cecal mucosa and luminal contents were collected, snap-frozen in liquid nitrogen, and stored at −80 °C for subsequent analysis.

Cell culture and treatment

Caco-2 cells were obtained from iCell Bioscience and cultured in Minimum Essential Medium. The medium was supplemented with 20 % fetal bovine serum, 1 % non-essential amino acid solution, 100 U/mL penicillin, and 100 μg/mL streptomycin. Cells were maintained at 37 °C in a 5 % CO₂ incubator. Cells were seeded at appropriate densities in uncoated plastic flasks (25, 60, or 100 cm²) or multi-well plates (12- or 24-well). Upon reaching 100 % confluence, the culture medium was replaced with complete medium containing AKG and incubated for 12 hours. Both cells and culture media were then harvested for analysis. Reactive oxygen species (ROS) (BB470515, Bestbio, Shanghai, China), glucose (E-BC-K234-M, Elabscience, Wuhan, China), L-lactic acid (E-BC-K044-M, Elabscience, Wuhan, China) were measured using commercial kits. Mitochondrial membrane potential was detected using JC-1 (C2006, Beyotime, Shanghai, China).

Hematoxylin and eosin staining

Cecal tissue samples were fixed in 4 % paraformaldehyde, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E) staining. Morphological features were examined using light microscopy. Villus height and crypt depth were measured, and the villus height-to-crypt depth ratio (VH/CD) was calculated using ImageJ software.

Analysis of serum, cecal mucosa, and cecal content

Serum, cecal contents, and mucosa were collected. Enzyme activities and contents were measured using commercial kits: superoxide dismutase (SOD) (BC0170, Solarbio, Beijing, China), glutathione peroxidase (GPx) (BC1190, Solarbio, Beijing, China), secretory immunoglobulin A (SIgA) (ml002778, Shanghai Enzyme-linked Biotechnology, Shanghai, China), malonaldehyde (MDA) (S0131S, Beyotime, Shanghai, China), ammonia test kit (K-URAMR, Megazyme, Bray, Ireland). Serum was mixed with an equal volume of 5 % sulfosalicylic acid. After centrifugation (15 min, 12,000 × g, 4 °C), the supernatant was collected and passed through a 0.22 μm filter.

High-performance liquid chromatography analysis

The concentrations of adenosine triphosphate (ATP) (A832633, Macklin, Shanghai, China), and adenosine diphosphate (ADP) (A119474, Aladin, Shanghai, China) were analyzed using high-performance liquid chromatography (HPLC) (Agilent 1260, Santa Clara, California).

For sample preparation, Caco-2 cells were treated with PBS or AKG for 12 h, washed with PBS, and lysed using RIPA buffer. An equal volume of the supernatant was mixed with ice-cold 0.4 mol/L HClO₄. The sample was centrifuged at 3,500 × g for 10 min. The supernatant was collected and combined with an equal volume of ice-cold 1 mol/L K₂HPO₄. The mixture was adjusted to pH 6.5 and centrifuged again under the same conditions. The final supernatant was collected and passed through a 0.22 μm filter. And all steps were performed at 4 °C. Individual stock solutions of ATP and ADP standards were prepared in ultrapure water and serially diluted to obtain a range of concentrations. Standard solutions were mixed as required and stored protected from light. Chromatographic separation was performed on a Zorbax SB-Aq column (4.6 mm × 250 mm, 5 μm; Agilent) at 25 °C. The mobile phase used 50 mmol/L potassium phosphate buffer (pH 6.5) at a flow rate of 1.0 mL/min. Detection occurred at 254 nm, with a 20 µL injection volume.

Free amino acid detecting

For the detection of free amino acid content in broiler serum, first prepare a 5 % sulfosalicylic acid solution as a stock solution. Add sulfosalicylic acid to broiler serum at a 1:1 ratio, mix thoroughly, and let it sit on ice for 2 min. Centrifuge to collect the supernatant for later use. Then, take at least 100 μL of the diluted serum homogenate supernatant and filter it (filter pore size: 0.22 μm). And auto-assay meter of amino acid content is for detecting amino acid. The method uses a Hitachi 855-4507 ion exchange column (60 mm × 4.6 mm, 3 μm). The elution pump flow rate is 0.35 mL/min and the derivative pump flow rate is 0.3 mL/min with a column temperature of 135°C. The chromatographic column uses a citric acid (lithium) PF buffer gradient elution, with a detection wavelength of 570 nm + 440 nm and an analysis time of 148 min.

Quantitative real-time PCR analysis

The absolute PCR experiment was conducted according to previously described methods (Jiang et al., 2025). Total DNA was extracted from cecal contents using a commercial kit (D3141-03, Magen, Guangzhou, China). Thermal cycling conditions included initial denaturation at 95 °C for 1 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s. The specific PCR amplification product was purified by electrophoresis and gel extraction by using an agarose gel recovery kit (Magen, D2111-02) to generate a standard curve for copies of total bacteria. Briefly, the gene copy numbers were determined using the following equation: (DNA concentration (μg/μL)  ×  6.0233  ×  1023 copies/mol) / (DNA size (bp)  ×  660  ×  106). Total bacterial copy number was normalized to the sample weight.

We extracted total RNA from cecal mucosal tissues using the Universal RNA Extraction Kit (Magen, D7107-03, Guangzhou, China) following the manufacturer's instructions. RNA concentration was measured using a NanoDrop ND-1000 spectrophotometer. One μg total RNA was reverse-transcribed into complementary DNA (cDNA, A0010CGQ, EZB, Roseville, Roseville). Quantitative real-time PCR (qRT-PCR) was performed on a QuantStudio 3 system (Applied Biosystems) using TB Green Premix Ex Taq (RR820A, Takara, Tokyo, Japan) Primer sequences are listed in Table 2. Relative gene expression was normalized to β-actin and calculated using the 2−ΔΔCt method.

Table 2.

Primer sequences used for quantitative PCR analysis, species: Gallus gallus (F: Forward; R: Reverse).

Gene name	Primer sequence	References
β-actin	F: CTGTGCCCATCTATGAAGGCTA R: ATTTCTCTCTCGGCTGTGGTG	NM_205518.2
Claudin-1	F: TGGAGGATGACCAGGTGAAG R: TGTGAAAGGGTCATAGAAGG	NM_001013611.2
Claudin-2	F: CGCTCGTATCTCTTGCTTGG R: AGAGTATGGCTGTGACGAGG	NM_001277622.1
Occludin	F: CTTCAGGTGTTTCTCTTCCTCCTC R: CTGTGGTTTCATGGCTGGA	XM_040680624.2
ZO-1	F: TCATCCTTACCGCCGCATAT R: GTTGACTGCTCGTACTCCCT	XM_046925214.1
ZO-2	F: AGTCCACCTCCAGCATTCAA R: CACAGAAACAGGTGGTGGTG	XM_046934796.1
IL-1β	F: GCTTCATCTTCTACCGCCTG R: ACTTAGCTTGTAGGTGGCGA	XM_046931582.1
IL-2	F: CAAGAGTCTTACGGGTCTAAATCAC R: GTTGGTCAGTTCATGGAGAAAATC	NM_204153.2
IL-4	F: GTGCCCACGCTGTGCTTAC R: AGGAAACCTCTCCCTGGATGTC	NM_001007079.1
IL-6	F: AAATCCCTCCTCGCCAATCT R: CCCTCACGGTCTTCTCCATAAA	NM_204628.2
IL-8	F: ATTCAAGATGTGAAGCTGAC R: AGGATCTGCAATTAACATGAGG	NM_205498.2
IL-10	F: CGCTGTCACCGCTTCTTCA R: TCCCGTTCTCATCCATCTTCTC	NM_001004414.4
IFN-α	F: TTCAGCTGCCTCCACACCTT R: TTGTGGATGTGCAGGAACCA	XM_046936231.1
IFN-β	F: CAGCTCTCACCACCACCTTCTC R: GGAGGTGGAGCCGTATTCTG	NM_001024836.2
All bacteria (target V3-V4 region)	F: ACTCCTACGGGAGGCAGCAG R: ATTACCGCGGCTGCTGG

16S rRNA gene sequencing and analysis of cecal microbiota

Intestinal content samples (approximately 0.3 g) were homogenized in sterile PBS. Total genomic DNA was extracted using the cetyltrimethylammonium bromide method. DNA quality was verified by 1 % agarose gel electrophoresis, and concentration and purity were measured with an ultraviolet spectrophotometer. Subsequently, the V3–V4 hypervariable regions of the bacterial 16S rRNA gene were amplified using primers 341F (5′-CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACNNGGGTATCTAAT-3′), with the extracted DNA as a template. The resulting amplicons were confirmed by 1 % agarose gel electrophoresis to verify the expected size (∼550 bp). They were then purified with Agencourt AMPure XP beads (Beckman Coulter), and subjected to paired-end sequencing (2 × 300 bp) on an Illumina NovaSeq 6000 platform.

Transcriptomic analysis of cecal mucosa

Total RNA was extracted from tissue samples, and eukaryotic mRNA was enriched using oligo(dT) magnetic beads. The enriched mRNA was then fragmented using a fragmentation buffer. First-strand cDNA was synthesized using random hexamers. Second-strand cDNA synthesis followed, using buffer, dNTPs, DNA polymerase I, and RNase H. The resulting double-stranded cDNA was purified with AMPure XP beads. The purified cDNA underwent end repair, A-tailing, and adapter ligation. Fragments of the desired size were selected using AMPure XP beads, followed by PCR amplification. The final PCR product was purified using AMPure XP beads to construct the sequencing library. Library quality was assessed, and qualified libraries were sequenced on an Illumina HiSeq 2500 platform to generate PE150 reads. Raw image data were processed through base calling to generate raw reads, which were stored in FASTQ format files. These files contain sequence information and corresponding quality scores. The resulting raw reads then underwent bioinformatic analysis.

Oxygen consumption rate analyses

The effect of AKG on the oxygen consumption rate (OCR) was determined in Caco-2 cells. Briefly, cells were treated with the control solvent or 0.25 mM AKG, and dissolved oxygen was continuously monitored for 60 min using a liquid-phase oxygen measurement system (Hansatech). The OCR was calculated as the rate of oxygen consumption per minute between the start and end of the measurement period.

Molecular docking analysis

The protein sequence was obtained from NCBI website. We used AlphaFold3 to model the complex between UCP2 and AKG, constructing a binary complex prediction model and outputting the structural coordinate file (PDB).

Cellular thermal shift assay

Cellular thermal shift assay (CETSA) was performed as previously described (Li et al., 2024). Caco-2 cells were treated with PBS or AKG for 12 h, washed with PBS, and harvested. Protein concentration was determined using a BCA assay. Equal amounts of cell suspensions were aliquoted into 0.2 mL PCR tubes and heated at different temperatures for 3 min using a thermal cycler (BIO-GENER, GE9612T). The samples were then cooled at room temperature for 5 min. After heating, soluble proteins were isolated by centrifugation and analyzed by SDS-PAGE and western blot. These values were then plotted against temperature to generate protein UCP2 (A4178, Abclonal, Wuhan, China) protein melt curves for each treatment.

Statistical analysis

Statistical analysis was conducted using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA). A non-paired two-tailed student's t-test were used to evaluate differences between the two groups. Data are presented as mean ± SEM (standard error of the mean). Statistical significance is displayed as P < 0.05.

Results

Hindgut delivery AKG increased growth performance of broilers

Compared with the control group, dietary supplementation of hindgut delivery AKG significantly increased the final body weight (P < 0.05) and the average daily gain (P < 0.05), and showed a decreased the F/G (P < 0.05). However, hindgut delivery L-Glu only have slightly growth-promoting effect. In addition, hindgut delivery AKG did not significantly affect broiler's eviscerated carcass weight rate, dressing percentage, abdominal fat ratio, liver ratio, cecum length, and rectum length (Table 3). These results demonstrated that hindgut delivery L-Glu metabolite, AKG, is more effective than L-Glu itself in promoting broiler growth.

Table 3.

Effect of hindgut delivery AKG on growth performance of broilers.

Items	Treatments			P value
	Control	L-Glu	AKG	Control vs L-Glu	Control vs AKG
Initial body weight (g)	421.58 ± 10.31	411.93 ± 10.71	422.24 ± 10.39	0.52	0.96
Final body weight (g)	1361.83 ± 37.25	1423.38 ± 33.90	1480.46 ± 36.48*	0.23	0.03
Average food intake (g/d)	94.04 ± 2.89	94.97 ± 2.11	98.16 ± 10.66	0.79	0.29
Average daily gain (g/d)	44.77 ± 1.60	48.16 ± 1.43	50.39 ± 1.50*	0.12	0.02
F/G	2.12 ± 0.07	2.00 ± 0.07	1.96 ± 0.04*	0.20	0.05
percentage of eviscerated yield (%)	76.70 ± 0.51	75.80 ± 1.03	77.67 ± 0.68	0.45	0.27
Dressed percentage (%)	90.33 ± 0.25	90.13 ± 0.25	90.88 ± 0.30	0.57	0.18
percentage of abdominal fat yield (%)	0.88 ± 0.12	0.92 ± 0.12	0.86 ± 0.13	0.80	0.88
liver index (%)	2.74 ± 0.31	2.29 ± 0.09	2.43 ± 0.15	0.19	0.39
Cecum length (cm)	16.09 ± 0.56	14.97 ± 0.55	15.01 ± 0.36	0.17	0.12
Rectum length (cm)	6.10 ± 0.25	6.16 ± 0.21	6.17 ± 0.23	0.87	0.82

The values are presented as the mean ± SEM (n = 9- 18). * represents a significant difference between the control group and the AKG group. * and ** indicate differences of P ≤ 0.05 and P ≤ 0.01, respectively.

Effect of hindgut delivery AKG on intestinal health of broilers

We further investigated the effects of hindgut delivery L-Glu and AKG on cecal morphology, barrier function, and inflammation associated gene expression in cecal mucosa of broilers. The results showed that neither L-Glu nor AKG administration significantly altered cecal morphology (Fig. 1A), villus height, crypt depth, the VH/CD (Fig. 1B), or muscular thickness (Fig. 1C). Both AKG and L-Glu significantly increased the mRNA expression of Occludin (P < 0.05) (Fig. 1D). In addition, L-Glu markedly elevated the mRNA levels of IL-4 and IFN-α (P < 0.05), while AKG failed to change any inflammation associated gene expression (Fig. S1A). Meanwhile, neither L-Glu nor AKG changed cecal mucosal SIgA content (Fig. S1B). Collectively, these results suggested that L-Glu or AKG increase broiler growth might be attributed to the beneficial effects on intestinal barrier, but not inflammation or specific immunity.

Fig. 1.

Effect of hindgut delivery α-ketoglutarate (AKG) on intestinal morphology and barrier function of broilers. (A) hematoxylin and eosin (H&E) staining of the cecum in broilers (magnification 40 ×). Scale bar is 100 μm. (B) Statistical analysis of cecal villus height, crypt depth, and cecal muscular thickness. (C) Statistical analysis of villus height-to-crypt depth ratio (VH/CD). (D) mRNA expression of tight junction proteins in broiler mucosa. ^# indicates a significant difference between the control group and the L-Glu group, * represents a significant difference between the control group and the AKG group. * or ^# indicates differences of P ≤ 0.05. ^## and ^### indicate differences of P ≤ 0.01 and P ≤ 0.001, respectively.

Effect of hindgut delivery AKG on the intestinal microbiota of broilers

Previous studies demonstrated that AKG may influence the gut microbiota (Chen et al., 2017; Li et al., 2019). Here, we performed microbial sequencing to exam if intestinal microbiota involves in maintaining intestinal barrier by hindgut delivery AKG. The results showed that AKG had no impact on total bacterial copy number (Fig. 2A), phylum-level bacterial copy numbers (Fig. 2B), α-diversity (Fig. 2C-F) or β-diversity (Fig. 2G), and cecum pH value (Fig. 2H). The results showed that AKG altered the abundance of microbiota such as Clostridiales and Clostridia (Fig. 2I and J). These results suggest that AKG does not alter the intestinal barrier through microbial mechanisms.

Fig. 2.

Effect of hindgut delivery AKG on the intestinal microbiota of broilers. (A) Total bacterial copy number in cecal content of broilers. (B) Phylum-level bacterial copy numbers in cecal content of broilers. (C) Shannon index. (D) Simpson index. (E) Chao1 index. (F) ACE index. (G) Principal coordinate analysis (PCoA). (H) Cecum pH value. (I) Cladogram of species composition. (J) Bar plot of linear discriminant analysis (LDA) scores. The values are presented as the mean ± SEM (n = 10).

Hindgut delivery AKG decreased cecal ammonia and serum amino acid content

AKG scavenges ammonia and generates glutamate and glutamine. Therefore, we hypothesized that AKG could regulate serum amino acid composition. To test whether AKG promotes broiler growth by modulating serum amino acid levels, we measured indicators related to serum ammonia nitrogen metabolism. The results showed that AKG significantly reduced ammonia content in cecal contents (P < 0.05, Fig. 3A), but tended to increase serum ammonia levels (Fig. 3B). Further analysis of serum amino acid composition revealed that AKG significantly decreased the levels of phenylalanine and tryptophan among aromatic amino acids (P < 0.05, Fig. 3C), as well as valine, a branched-chain amino acid (P < 0.05, Fig. 3D). No significant differences were observed in amino acids belonging to the other categories (Fig. S2A - D). Therefore, these evidences indicated that AKG promotes broiler growth may not attribute to serum amino acid levels.

Fig. 3.

Effect of hindgut delivery AKG on the amino acid metabolism of broilers. (A) Ammonia in cecum content. (B) Ammonia in serum. (C) Aromatic amino acids. (D) Branched chain amino acid. The values are presented as the mean ± SEM (n = 7 - 8). * represents a significant difference between the control group and the AKG group. ** indicates differences of P ≤ 0.01.

Hindgut delivery AKG regulated the OXPHOS and redox of broilers cecal mucosa

To investigate the mechanism of AKG, we analyzed the mRNA transcriptional profiles of cecal mucosal genes in broilers. Transcriptomic analysis identified 175 differentially expressed genes between the AKG-treated and control groups: 69 were upregulated and 106 were downregulated (Fig. 4A). KEGG analysis showed that hindgut-delivered AKG significantly enriched several pathways, including OXPHOS and Ribosome (Fig. 4B). Cluster analysis of differentially expressed genes related to OXPHOS revealed that hindgut-delivered AKG significantly downregulated mitochondrial metabolism genes, including ATP6V0E1, ATP5L, ATP5J2, and COX7C (Fig. 4C).

Fig. 4.

Hindgut delivery AKG regulated the oxidative phosphorylation (OXPHOS) and redox of broilers cecal mucosa. (A) Differential gene volcano map. (B) KEGG enrichment volcano map. (C) Heat map of the OXPHOS differentially expressed genes. (D) malonaldehyde (MDA) in cecum mucosa. (E) superoxide dismutase (SOD) in cecum mucosa. (F) glutathione peroxidase (GPx) in cecum mucosa. The values are presented as the mean ± SEM (n = 3 - 9). * represents a significant difference between the control group and the AKG group. * indicates differences of P ≤ 0.05. ** and *** indicate differences of P ≤ 0.01 and P ≤ 0.001, respectively.

Mitochondrial OXPHOS is a major source of free radicals and is closely associated with oxidative damage to the mucosa (Wen et al., 2023). To investigate this connection, we examined redox-related indicators in the cecal mucosa. Hindgut delivery of AKG significantly reduced MDA content and SOD enzyme activity in the cecal mucosa (P < 0.05), while GPx enzyme activity showed a trend toward increase. (Fig. 4D - F). These findings suggest that hindgut delivery AKG may promote broiler growth by reducing OXPHOS and oxidative damage, thereby improve the intestinal mucosal barrier.

AKG decreased OXPHOS and maintained the redox in Caco-2 cells

To validate AKG's mechanism in vitro, we used Caco-2 cells to investigate its effects on OXPHOS and redox balance in intestinal cells. Results showed that AKG at concentrations of 0.1, 0.25, and 0.5 mM increased cellular OCR (Fig. 5A). Meanwhile, AKG reduced glucose consumption and lactate production in a dose-dependent manner (P < 0.05) (Fig. 5B and C). Additionally, AKG at concentrations of 0.1, 0.25, and 0.5 mM decreased cellular ATP (P < 0.05) and ADP levels (Fig. 5D and E). To assess mitochondrial membrane potential, we selected the intermediate dose of AKG. AKG significantly lowered the mitochondrial membrane potential in cells (P < 0.05) (Fig. 5F and G). AKG also reduced cellular ROS levels at concentrations of 0.1, 0.25, and 0.5 mM (P < 0.05) (Fig. 5H). These results indicate that AKG suppresses OXPHOS and improves redox status in intestinal epithelial cells while reducing energy production. This suggests that AKG may influence energy utilization efficiency, specifically the phosphorus-to-oxygen (P/O) ratio.

Fig. 5.

AKG regulated OXPHOS and redox in Caco-2 cells. (A) Average oxygen consumption rate (OCR). (B) Glucose consumption in culture. (C) Lactic acid concentration in culture. (D) Adenosine triphosphate (ATP) content. (E) Adenosine diphosphate (ADP) content. (F) Mitochondrial membrane potential was detected by staining with JC-1 dye: red, aggregates; green, monomers. Scale bar is 50 μm. (G) Mitochondrial membrane potential statistical plots (relative to the control). (H) Reactive oxygen species (ROS) level. The values are presented as the mean ± SEM (n = 3 - 8). * represents a significant difference between the control group and the AKG group. * indicates differences of P ≤ 0.05. ** and *** indicate differences of P ≤ 0.01 and P ≤ 0.001, respectively.

Effect of AKG on the redox in Caco-2 cells was mediated by UCP2

Uncoupling proteins are crucial for regulating the P/O ratio and OXPHOS (Brand et al., 1993). To determine whether UCP2 mediates AKG's effect on redox processes, we treated cells with both AKG and the UCP2 inhibitor genipin. The results showed that adding genipin completely blocked all AKG-induced effects: it prevented the increase in OCR (Fig. 6A), reversed the reduction in glucose consumption and lactate production (Fig. 6B and C), negated the changes in cellular ATP and ADP levels (Fig. 6D and E), and eliminated the elevation in mitochondrial membrane potential (Fig. 6F and G). The AKG-induced decrease in ROS levels was also eliminated by this inhibitor (Fig. 6H). To confirm the interaction between AKG and UCP2, we performed molecular docking analysis. The binding energy was −5.574 kcal/mol, indicating strong binding affinity (Fig. 6I). Next, we validated the stability of the AKG-UCP2 interaction using the CESTA method. The results showed that binding between AKG and UCP2 increased significantly as temperature rose (P < 0.05) (Fig. 6J and K). These findings demonstrate that AKG alleviates oxidative stress in intestinal epithelial cells by modulating UCP2.

Fig. 6.

Role of AKG-mediated uncoupling protein 2 (UCP2) in regulating redox in Caco-2 cells. (A) Average OCR. (B) Glucose consumption in culture. (C) Lactic acid concentration in culture. (D) ATP content. (E) ADP content. (F) Mitochondrial membrane potential was detected by staining with JC-1 dye: red, aggregates; green, monomers. Scale bar is 50 μm. (G) Mitochondrial membrane potential statistical plots (relative to the control). (H) ROS level. (I) Binding region of AKG and UCP2 protein. (J) Immunoblots of relative UCP2 protein expression in cellular thermal shift assay (CETSA) by AKG treatment. (K) Quantification of relative UCP2 protein expression in CETSA assay by AKG treatment. The values are presented as the mean ± SEM (n = 3 - 6). * represents a significant difference between the control group and the AKG group. * indicates differences of P ≤ 0.05. ** and *** indicate differences of P ≤ 0.01 and P ≤ 0.001, respectively.

Discussion

The nutrients support for intestine is closely linked to animal health and growth performance, with the gut exhibiting a preferential utilization of amino acids among various substrates (Cherbuy et al., 1995; He et al., 2022; He and Wu, 2022). Dietary glutamate is almost completely absorbed or degraded by the small intestinal mucosa (Reeds et al., 1996), resulting in less availability in the large intestine. Here, we employed a coat-technology for the controlled delivery of glutamate and its metabolite AKG into the hindgut of broilers. Our results demonstrated that hindgut-targeted delivery of AKG significantly increased final body weight and average daily gain, and decreased F/G. However, the growth promoting effect of coated L-Glu is weaker than that of coated AKG. It suggested that L-Glu might convert into AKG to exert its physiological effect; or AKG might scavenge the hindgut ammonia to save amino acids (Blachier et al., 2022; Mayer and Moran, 2022; Pengpeng and Tan, 2013). Previous studies have reported that direct dietary supplementation with L-Glu or AKG failed to improve growth performance in various animals. For instance, adding 3 % glutamate to growing pig diets showed no significant effect on body weight (Kong et al., 2015), and supplementing 1 % AKG directly into the diets of broilers or Cherry Valley ducks likewise failed to enhance growth performance (Guo et al., 2017; Tomaszewska et al., 2020). Therefore, we speculate that dietary supplementary L-Glu and AKG supposed to be absorbed in the small intestine and enter the bloodstream (Burrin and Stoll, 2009; Lambert et al., 2006), making it difficult for them to reach the hindgut in effective concentrations. In the digestion-metabolism trial, the ileal digestibility of dietary glutamate was approximately 90 %, leaving only 10 % available reached the hindgut (Aguirre et al., 2024; Yoon and Kong, 2025). Considering that the present study employed a coating strategy for targeted delivery and that the delivery efficiency was expected to be high, we estimated based on the ileal digestibility that only 0.1 % supplementation would be required. Together with our previous findings for other nutrients (e.g., coated malic acid at a 0.1 % inclusion level improved broiler growth performance) (Jiang et al., 2025), we ultimately selected a 0.1 % supplementation level for the current study, which is substantially lower than those used in previous reports. Here, our study observed hindgut-targeted delivery of lower concentration of Glu and AKG has beneficial effects on broiler growth performance, indicating that this targeted delivery system is an effective strategy for delivering bioactive nutrients and achieving cost reduction and efficiency enhancement. Though directly supplementation of L-Glu or AKG to livestock and poultry diets has shown limited effectiveness (Guo et al., 2017; Kong et al., 2015; Tomaszewska et al., 2020), it would be better to set the uncoated L-Glu or AKG groups to highlight the effect of coated L-Glu or AKG on broiler growth performance.

Targeted controlled release in the intestine is well established in pharmaceutical applications. For example, delivering nutrients like quercetin, anthocyanins, and resveratrol directly to the colon reduces oxidative stress and improves colitis (Tang et al., 2025). Unlike in medicine, few reports exist on targeted intestinal release in animal production. Available studies suggest this strategy can enhance growth performance in livestock and poultry. For example, targeted release of encapsulated essential oils has been shown to improve growth, enhance digestibility, reduce the F/G, and increase short-chain fatty acid content in broilers (Meligy et al., 2023). Similarly, targeted release of encapsulated Enterococcus faecalis improved daily weight gain and reduced the feed conversion ratio in broilers (Han et al., 2013; Zhang et al., 2024b). Our previous research demonstrated that hindgut-targeted delivery of malate significantly increased body weight gain and improved feed efficiency in broilers compared to direct dietary supplementation (Jiang et al., 2025). These findings suggest that hindgut-targeted delivery technology can be applied to a broader range of nutrients, enabling more precise regulation of hindgut nutrition.

AKG is a key metabolite produced from glutamate breakdown and acts as a bioactive signaling molecule (Macelline et al., 2025). Multiple studies showed that AKG regulates intestinal health in various experimental models. In swine, dietary supplementation with 1 % AKG improved intestinal health under LPS challenge. It increased villus height, decreased crypt depth (Hou et al., 2011), and reduced pro-inflammatory cytokine expression in the jejunum of weaned piglets (He et al., 2017). Additionally, it altered cecal microbiota composition in low-protein diet models (Chen et al., 2018a). In mouse models of colitis, AKG administered through drinking water (0.5 % or 1 %) or diet (1 %) preserves colonic structure, reduces inflammation, and improves microbial diversity (Li et al., 2019; Si et al., 2022; Tian et al., 2021). To investigate how AKG promotes growth, this study used hindgut-targeted delivery. However, the results showed that this treatment did not significantly affect intestinal morphology, inflammatory cytokine mRNA expression, or SIgA content. It only increased the mRNA expression of the tight junction protein Occludin. Additionally, 16S rRNA sequencing revealed limited effects on cecal microbiota, with abundance changes observed only in certain taxa. Several factors may explain these differences. First, our study used healthy broilers without an inflammatory challenge, unlike previous reports (He et al., 2017; Hou et al., 2011; Li et al., 2019; Si et al., 2022; Tian et al., 2021). Second, previous studies typically administered AKG through diet or water at 0.5–1 %, allowing rapid foregut absorption and primary action in the small intestine (He et al., 2017; Hou et al., 2011). Our intervention specifically targeted the hindgut. Third, we used a lower dosage (0.1 %) compared to the 1 % level commonly used in earlier work (Chen et al., 2018a; He et al., 2017; Hou et al., 2011; Li et al., 2019; Si et al., 2022; Tian et al., 2021). These differences may explain why we observed no pronounced effects on intestinal morphology, barrier function, or microbiota.

AKG is a key intermediate in ammonia nitrogen metabolism, rapidly converting ammonia into glutamate and glutamine (Xiao et al., 2016). Low-protein diets can use AKG to improve nitrogen utilization efficiency. Supplementing a low-protein diet with 1 % AKG increased serum free amino acid concentrations in pigs (Chen et al., 2018b). Another study showed that daily oral administration of AKG solution (0.4 g/kg body weight) to turkeys on a normal diet significantly increased blood levels of proline and leucine (Tatara et al., 2005). However, in this study, hindgut-targeted delivery of AKG did not increase blood amino acid levels. Instead, it significantly reduced serum levels of phenylalanine and tryptophan (aromatic amino acids) and valine (a branched-chain amino acid) in broilers. We speculate that when AKG is added directly to the diet, it is rapidly absorbed in the small intestine and enters the liver, where it participates in amino acid metabolism (Wang et al., 2015). In contrast, the targeted delivery system in this study delivered AKG directly to the hindgut. In the anaerobic hindgut environment, amino acids undergo deamination (Blachier, 2023), and AKG reacts with ammonia to form glutamate (Xiao et al., 2016). This mechanism may explain the observed decrease in serum amino acids and indicates that AKG does not promote broiler growth by elevating systemic amino acid levels. Our results suggested that hindgut metabolism has a significant impact on whole-body amino acids metabolism. Further study might focus on the dose-effect of hindgut delivery AKG on serum amino acids and broiler growth.

Hindgut health is essential for livestock and poultry growth and development. This study shows that hindgut-targeted delivery of AKG reduces intestinal mucosal OXPHOS, downregulates OXPHOS-related gene expression, and alleviates oxidative damage. OXPHOS in intestinal epithelial cells provides the energy needed to maintain gut health (Guerbette et al., 2024). However, dysfunction in this mitochondrial process can lead to oxidative damage (Fujiwara et al., 2021). This oxidative damage disrupts the intestinal barrier, increases permeability, and causes intestinal leakage, all of which impair animal growth performance (Oke et al., 2024). Previous studies show that AKG supplementation reduces intestinal oxidative damage. For example, Hou et al. found that adding 1 % AKG to the diet reduced oxidative damage in the intestinal mucosa and reversed the growth-inhibiting effects of LPS in weaned piglets (Hou et al., 2010). Similarly, another study found that adding 0.5 % AKG to drinking water reduced intestinal oxidative stress and improved barrier function in mice with colitis (Si et al., 2022). From the perspective of energy metabolism regulation, this study demonstrates that hindgut-targeted delivery of AKG alleviates oxidative damage via the modulation of mitochondrial OXPHOS function. These findings provide new experimental evidence for improving broiler intestinal health through targeted hindgut metabolism regulation.

AKG modulates mitochondrial OXPHOS differently depending on the cell type. In palmitate-injured HepG2 cells, it enhances OXPHOS efficiency and improves mitochondrial function by activating the AMPK–PGC-1α/Nrf2 pathway (Cheng et al., 2024). In contrast, AKG suppresses OXPHOS in human pluripotent stem cells by inhibiting ATP synthase (TeSlaa et al., 2016). It is generally assumed that increasing ATP availability would elevate intracellular ATP levels and thereby improve intestinal health (Guo et al., 2017). Our results show that AKG reduces OXPHOS in intestinal epithelial cells and alleviates oxidative damage. Specifically, AKG increased oxygen consumption while decreasing ATP content in these cells. Based on these observations, we hypothesize that AKG reduces the P/O ratio and promotes proton leak (Brand et al., 1993). We further propose that AKG may improve the redox state of intestinal cells by inducing uncoupling. Among the five main uncoupling proteins, UCP2 is predominantly expressed in the intestine (Jin et al., 2017; Zhao et al., 2019). Substantial evidence links UCP2 to redox regulation. In vivo studies show that UCP2-knockout mice exhibit higher oxidative stress in colonic epithelial cells than wild-type mice (Derdák et al., 2006). In cellular models, UCP2 overexpression protects intestinal epithelial cells from hydrogen peroxide-induced oxidative damage (Su et al., 2025). UCP2 maintains intestinal epithelial barrier integrity and reduces oxidative stress (Huang et al., 2024; Su et al., 2025). Appropriate UCP2 activity effectively supports animal health and growth (Liu and Lai, 2016). In this study, co-treatment with genipin eliminated AKG's ability to reduce OXPHOS and alleviate redox imbalance. Molecular docking analysis and CETSA validation confirmed that AKG binds favorably to UCP2. This suggests that AKG alleviates redox imbalance in intestinal epithelial cells by specifically regulating UCP2. For the detailed mechanism, earlier work indicated that AKG cannot directly cross the plasma membrane (Baracco et al., 2019). But extracellular AKG increases its intracellular concentration via specific signaling pathway (He et al., 2018; Yuan et al., 2022). These findings support further exploration of UCP2 as a target for developing novel antioxidants in animal production. Other TCA cycle intermediates have also been found to activate UCP2 (Wang et al., 2025b; Yin et al., 2024). Further study may focus on the application of these intermediates to regulate broiler growth.

In conclusion, hindgut-targeted delivery of AKG improved body weight gain and reduced the F/G in broilers. AKG enhances broiler growth by directly modulating UCP2 in intestinal epithelial cells, reducing OXPHOS and alleviating redox imbalance. This study demonstrates that hindgut-delivered AKG playing a novel role in metabolism and broiler growth.

CRediT authorship contribution statement

Xue Yang: Writing – original draft, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Xiaohua Feng: Writing – review & editing, Validation, Methodology. Wei Huang: Writing – review & editing, Validation. Xinyu Gan: Methodology, Formal analysis. Hongfeng Jiang: Data curation. Yunlong Zhu: Methodology. Yujun Wang: Validation. Genghui Li: Resources. Xiaofeng Peng: Methodology. Jing Rong: Validation. Ruifan Wu: Supervision. Canjun Zhu: Supervision. Lina Wang: Supervision. Songbo Wang: Supervision. Qingyan Jiang: Supervision, Project administration, Funding acquisition. Gang Shu: Writing – review & editing, Visualization, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization.

Disclosures

The authors declare no conflict of interest.

Appendix. Supplementary materials

Figure S1. Effect of hindgut delivery AKG on intestinal health of broilers. (A) mRNA expression of inflammatory cytokines in broiler mucosa. (B) Content of secretory immunoglobulin A (SIgA) in the cecal mucosa. The values are presented as the mean ± SEM (n = 6). ^# indicates a significant difference between the control group and the L-Glu group, ^# indicates differences of P ≤ 0.05. ^## and ^### indicate differences of P ≤ 0.01 and P ≤ 0.001, respectively.

Figure S2. Effect of hindgut-delivered AKG on the metabolism of different classes of amino acids in broilers. (A) Alkaline amino acids. (B) Acidic amino acids. (C) Polar neutral amino acids. (D) Other non-essential amino acid. The values are presented as the mean ± SEM (n = 7 - 8). * represents a significant difference between the control group and the AKG group. ** indicates differences of P ≤ 0.01.