Effects of glutamine against oxidative stress in the metabolome of rats—new insight

Abstract

Glutamine exerts potential functions against the harmful effects of oxidative stress on animals. However, the systemic metabolic changes related to oxidative stress and glutamine intervention remain largely unknown. Rats were fed a basal diet or a basal diet supplemented with 1% glutamine for 30 days. On day 28, the rats were intraperitoneally injected with either diquat or saline. Oxidative stress alters common systemic metabolic processes, including energy, amino acid, and gut microbiota metabolisms. Compared with the diquat group, the glutamine + diquat group had significantly higher plasma levels of citrate and isobutyrate and urine levels of homogentisate and α-ketoglutarate while lower plasma levels of acetate, creatine, formate, glutamate, leucine, O-acetyl glycoprotein, phenylalanine, pyruvate, α-glucose, and β-glucose and urine levels of benzoate and trigonelline. Glutamine can partially counteract the metabolic effects of oxidative stress. These findings provide new insights into the complex metabolic changes after glutamine supplementation in rats under oxidative stress.

---

Introduction

Oxidative stress, defined as the imbalance between the amount of reactive oxygen species (ROS) and the intra- and extracellular antioxidant systems, can lead to damage of cells, even the whole organism. Several factors such as environmental change, weaning, and infection, result in oxidative stress and can cause growth retardation, disease, and even death for newborn and weaned animals.^1,2

Glutamine is a conditionally essential amino acid that increases cellular adenosine triphosphate levels^3,4 and accelerates the division of cells, such as enterocytes, lymphocytes, and macrophages.⁵ Glutamine also maintains intestinal integrity and function,⁶ modulates intestinal gene expression,⁷ improves nutrient absorption and immune function,^8,9 regulates acid–base balance and cell proliferation, inhibits cell autophagy,¹⁰ and enhances mitochondrial function.¹¹ Moreover, glutamine can improve glutathione production and thus increase the antioxidant capacity in animals.¹² Furthermore, glutamine can improve protein synthesis, inhibit protein breakdown, and enhance the gain : feed ratio of weaned pigs.^7,13 Finally, glutamine supplementation between days 90 and 114 of gestation ameliorates the fetal growth restriction in gilts, increases the survival of suckling piglets, and decreases the preweaning mortality of piglets.¹⁴

Recent studies have revealed significant changes in the plasma metabolite levels of amino acid, fatty acids, and lactate between glutamine and control groups in pigs under non-oxidative stress.¹⁵ However, there is no information about the effects of glutamine on animal or human biological systems under oxidative stress. Metabolomics provides a novel strategy to resolve the changes in metabolic endpoints of physiological regulatory processes of an organism after the administration of specific nutritional interventions. Metabolomics may be applied to understand the effects of glutamine administration on health and disease.

This study is a part of a larger research that involved determining protective effects of glutamine against oxidative stress in rat intestine¹⁶ and the antioxidant decrease of diquat in rat plasma.¹⁷ The main purpose of this study was for the first time to test the hypothesis that glutamine can modulate the global metabolome of rats under oxidative stress. The current study would help define the effects of metabolic modifiers and refine the nutritional requirements of the body to provide strong nutritional support for growth and health.

Materials and methods

Animal experiment and sample collection

The animal experiment was approved by the Animal Care and Use Committee of Sichuan Agricultural University and was performed according to the Guide for the Care and Use of Laboratory Animals of the National Research Council. Thirty eight-week-old female Sprague-Dawley rats weighing 239–272 g were placed in individual metabolic cages and allowed to acclimatize for 2 weeks. The rats were randomly assigned to one of the three isonitrogenous and isocaloric dietary groups (i.e., diquat, glutamine + diquat, and control groups) with 10 rats per group. Each treatment lasted for 30 days. The diquat and control groups were fed the basal diet. The glutamine + diquat group was fed the basal diet containing 1% glutamine (supplied by Beijing Jiakangyuan Technology Development Co., Ltd., Beijing, China). On day 28 (09:00 a.m.), the rats were intraperitoneally injected with diquat (Sigma Chemical Co., St. Louis, MO, USA) at 12 mg kg⁻¹ body weight for the diquat and glutamine + diquat groups or with sterile 0.9% NaCl solution of the same amount for the control group. Urine samples were collected in ice-cooled vessels containing 30 μL of sodium azide solution (1.0% w/v) on days 28 to 29 of the treatment period (24 h). Blood samples were collected (09:00 hours) from the orbital venous plexus after anesthesia with ether into Eppendorf tubes containing sodium heparin 48 h after diquat injection. Whole blood samples were centrifuged at 3500g for 10 min at 4 °C to obtain plasma. All urine and plasma samples were stored at −80 °C until NMR analysis. The dosage selected for this study was based on the results of a previous experiment.^7,15,18 The body weight of each rat was recorded once a week. Rats were allowed free access to food and drinking water. Temperatures between 22 °C and 25 °C, a cycle of 12 h light/12 h dark, and humidity ranging from 50% to 70% were maintained throughout the duration of the study. Clinical observations were carried out during the whole experimental period.

Sample preparation and NMR spectroscopy

Urine samples (550 μL) were added into 55 μL of phosphate buffer (1.5 M NaH₂PO₄/K₂HPO₄, pH 7.4, 100% v/v D₂O) containing 0.1% NaN₃ as bacterial growth inhibitor and 5.0 mM 2,2-dimethyl-2-silapentane-5-sulfonate-d₆ (DSS) as chemical shift reference (δ 0.00 ppm). After mixing with a vortex and centrifugation (4 °C) at 12 000g for 10 min, the supernatant was pipetted into 5 mm NMR tubes for NMR analysis. Plasma samples for NMR analysis were prepared by mixing 200 μL of plasma with 400 μL of saline solution containing 75% D₂O as a field-frequency lock. After vortexing and centrifugation at 12 000g for 10 min at 4 °C, approximately 550 μL of samples were transferred into 5 mm NMR tubes for NMR analysis.

The proton NMR spectra of the urine and plasma samples were obtained at 300 K on a Bruker Avance II 600 MHz spectrometer (Bruker Biospin, Rheinstetten, Germany) operated at a ¹H frequency of 600.13 MHz with a broadband-observe probe. A standard water-suppressed 1D NMR spectrum was derived from urine by employing the first increment of the gradient-selected NOESY pulse sequence (recycle delay–90°–t₁–90°–t_m–90°–acquire data) with a recycle delay of 2 s, a t₁ of 3 μs, a mixing time (t_m) of 100 ms, and a 90° pulse length of 13.70 μs. A total of 128 transients were collected into 49 178 data points at a spectral width of 9590 Hz and an acquisition time of 2.56 s. For the plasma, a water-presaturated Carr–Purcell–Meiboom–Gill pulse sequence [recycle delay–90°–(τ–180°–τ)_n–acquisition] was employed to attenuate the NMR signals from macromolecules. A spin–spin relaxation delay (2nτ) of 76.8 ms and a spin–echo delay τ of 400 μs were employed. Typically, 90° pulse was set to 13.7 μs, and 32 transients were acquired into 49 178 data points for each spectrum with a spectral width of 15 ppm. The other acquisition parameters were the same those as described above. Metabolites were usually assigned by considering the chemical shifts, coupling constants, and relative intensities as in previous reports^19–21 and additional ¹H-¹H correlation spectroscopy and ¹H-¹H total correlation spectroscopy were recorded for selected samples (data not shown).

NMR spectroscopic processes and analysis

An exponential window function with a 1 Hz line-broadening factor was used prior to Fourier transformation. After the manual phase- and baseline-corrections employing Mestrenova 8.1.2 software (Mestrelab Research S.L., Spain), the plasma spectral region spanning δ 0.5–δ 9.0 was segmented into bins with a bucket width of 0.002 ppm. Then, the urinary spectral region spanning δ 0.5–δ 9.5 was bucketed into regions with an equal width of 0.005 ppm by using Mestrenova 8.1.2 software. Plasma and urine chemical shifts were referenced to the peak of the methyl proton of L-lactate at δ 1.33 and the peak of DSS at δ 0.00, respectively. Chemical shifts for urinary citrate were manually corrected because their signals had large inter-sample variations. Ethanol and H₂O signals were excluded from the collected blood to avoid the contributions of these substances to intergroup differentiations and thus obtain only the endogenous metabolite changes induced by the treatment. The discarded regions in the plasma spectra included δ 4.30–δ 5.10 for H₂O, δ 1.16–δ 1.20 and δ 3.60–δ 3.62 for ethanol. The excluded regions in the urine spectra contained δ 4.50–δ 5.30 for H₂O and δ 5.5–δ 6.0 for urea. Afterward, each integral region was normalized to the sum of all integral regions for each spectrum before the pattern recognition analysis.

Multivariate data analysis was conducted on the normalized NMR data sets with the software package SIMCA-P+ (version 11.0, Umetrics, Sweden). Principal component analysis (PCA) of the mean-centered data was performed to show group clustering and to identify possible outliers within the dataset. Results were observed in the form of score and loading plots. The former represented an individual sample, whereas the latter represented an NMR spectral region. Next, the supervised multivariate methods, projection to latent structure-discriminant analysis (PLS-DA) and orthogonal projection to latent structure-discriminant analysis (OPLS-DA) were employed on the data scaled to unit variance as the X-matrix and class information as the Y-matrix.²² The quality of the model was evaluated by the model parameters R²X and Q², which indicate the total explained variation and the model predictability, respectively. The models were validated using a seven-fold cross validation method and a permutation test.^23,24 The loadings from the OPLS-DA were back-transformed by multiplying their respective standard deviations and plotted with signals color-coded with coefficient values (r) in MATLAB (The Mathworks Inc.; Natwick, USA. version 7.1) to reveal significantly altered metabolites.²⁴ In this study, appropriate correlation coefficients were employed as cutoff values (depending on the number of animals used for each group) for statistical significance based on the discrimination significance (P < 0.05). The coefficients were determined using Pearson's product–moment correlation coefficient. The color in the loading plots represents the significance of the metabolite on class discrimination; warm-colored (e.g., red) variables represent higher significance than cold-colored (e.g., blue) variables.

Results

¹H NMR spectra of urine and plasma samples

Fig. 1 and 2 show the typical ¹H NMR spectra of the urine and plasma samples from the control, diquat, and glutamine + diquat groups, respectively. NMR signals were distributed to specific metabolites for ¹H resonance (Table 1). Fifty-one metabolites were assigned to urine. The spectra of the urine samples included resonance from amino acids, organic acids, glucose, allantoin, and choline. Tricarboxylic acid cycle metabolites, such as succinate and citrate, were also detected in the urine samples. The plasma samples mainly included glucose, lactate, lipids, and amino acids.

Fig. 1 Representative one-dimensional ¹H NMR spectra of urine metabolites taken from the control, diquat, and glutamine + diquat groups. The region of δ 6.2–9.5 was magnified 4 times compared with corresponding region of δ 0.5–6.2 for the purpose of clarity. Metabolite keys are given in Table 1.

Fig. 2 Typical 600 MHz ¹H NMR spectra of plasma metabolites taken from the control, diquat, and glutamine + diquat groups. The region of δ 6.0–9.0 was magnified 16 times compared with corresponding region of δ 0.5–6.0 for the purpose of clarity. Metabolite keys are given in Table 1.

Table 1 ¹H NMR data of metabolites in rat urine and plasma^a

Keys	Metabolites	Moieties	δ ^1H (ppm) and multiplicity	Samples^a
a U, urine; P, plasma; * LDL, low density lipoprotein; VLDL, very low density lipoprotein; TMAO, trimethylamine-N-oxide; s, singlet; d, doublet; t, triplet; q, quartet; dd, doublet of doublets; m, multiplet.
1	Bile acids	CH3	0.64 (m), 0.75 (m)	U
2	α-Hydroxy-iso-valerate	δCH3, CH3	0.83 (d), 0.97 (d)	U
3	α-Hydroxybutyrate	CH3	0.89 (t)	U
4	Propionate	CH3	1.06 (t)	U, P
5	Isobutyrate	CH3	1.13 (d)	U, P
6	Ethanol	CH3, CH2	1.19 (t), 3.66 (q)	U, P
7	Methylmalonate	CH3, CH	1.25 (d), 3.75 (m)	U
8	α-Hydroxy-n-valerate	CH3, γCH2	0.89 (t), 1.31 (m)	U
9	Lactate	αCH, βCH3	4.13 (q), 1.33 (d)	U, P
10	Alanine	αCH, βCH3	3.77 (q), 1.47 (d)	U, P
11	Citrulline	γCH2, βCH2	1.56 (m), 1.82 (m)	U
12	Acetate	CH3	1.92 (s)	U, P
13	Acetamide	CH3	1.99 (s)	U, P
14	N-Acetylglutamate	βCH2, γCH2, CH3	2.06 (m), 1.87 (m), 2.03 (s)	U
15	Acetone	CH3	2.24 (s)	U, P
16	Acetoacetate	CH3	2.28 (s)	U
17	Pyruvate	CH3	2.33 (s)	U, P
18	Succinate	CH2	2.40 (s)	U
19	α-Ketoglutarate	βCH2, γCH2	2.45 (t), 3.01 (t)	U
20	Citrate	CH2	2.54 (d), 2.68 (d)	U, P
21	Methylamine	CH3	2.61 (s)	U
22	Dimethylamine	CH3	2.71 (s)	U
23	Methylguanidine	CH3	2.81 (s)	U
24	Trimethylamine	CH3	2.88 (s)	U
25	Dimethylglycine	CH3	2.93 (s)	U
26	Creatine	CH3, CH2	3.04 (s), 3.93 (s)	U, P
27	Creatinine	CH3, CH2	3.04 (s), 4.05 (s)	U, P
28	Ornithine	CH2	3.06 (t)	U
29	Ethanolamine	CH2	3.11 (t)	U
30	Malonate	CH2	3.15 (s)	U
31	Choline	OCH2, NCH2, N(CH3)3	4.07 (t), 3.53 (t), 3.21 (s)	U, P
32	Taurine	–CH2–S, –CH2–NH2	3.27 (t), 3.43 (t)	U
33	TMAO*	CH3	3.27 (s)	U, P
34	Glycine	CH2	3.57 (s)	U
35	Sarcosine	CH2	3.6 (s)	U
36	Phenylacetylglycine	2,6-CH, 3,5-CH, 7-CH, 10-CH	7.30 (t), 7.36 (m), 7.42 (m), 3.67 (s)	U
37	Hippurate	CH2, 3,5-CH, 4-CH, 2,6-CH	3.97 (d), 7.55 (t), 7.63 (t), 7.84 (d)	U
38	N-Methylnicotinamide	CH3, 5-CH, 4-CH, 6-CH, CH2	4.42 (s), 8.21 (d), 8.87 (d), 8.93 (d), 9.24 (s)	U
39	β-Glucose	1-CH, 2-CH, 3-CH, 4-CH, 5-CH, 6-CH	4.47 (d), 3.25 (dd), 3.49 (t), 3.41 (dd), 3.46 (m), 3.73 (dd), 3.90 (dd)	U, P
40	α-Glucose	1-CH, 2-CH, 3-CH, 4-CH, 5-CH, 6-CH	5.24 (d), 3.54 (dd), 3.71 (dd), 3.42 (dd), 3.84 (m), 3.78 (m)	U, P
41	Allantoin	CH	5.39 (s)	U, P
42	Urea	NH2	5.82 (s)	U, P
43	Homogentisate	6-CH, 5-CH	6.67 (d), 6.82 (d),	U
44	p-Hydroxyphenylacetate	6-CH, 2-CH, 3,5-CH	3.6 (s), 6.85 (d), 7.15 (d)	U
45	m-Hydroxyphenylacetate	6-CH, 4-CH, 3-CH	6.92 (m), 7.04 (d), 7.26 (t)	U
46	Indoxyl sulfate	4-CH, 5-CH, 6-CH, 7-CH, CH	7.51 (m), 7.22 (m), 7.28 (m), 7.71 (m), 7.37 (s)	U
47	Nicotinamide	2-CH, 4-CH, 5-CH, 6-CH	8.94 (d), 8.61 (dd), 8.25 (m), 7.5 (dd)	U
48	4-Aminohippurate	CH2, CH	7.6 (d), 6.8 (d), 3.9 (d)	U
49	Benzoate	2,6-CH, 3,5-CH, 4-CH	7.87 (d), 7.49 (dd), 7.56 (t)	U
50	Trigonelline	2-CH, 4-CH, 6-CH, 5-CH, CH3	9.09 (s), 8.85 (m), 8.81 (dd), 8.07 (m), 4.44 (s)	U
51	Formate	CH	8.46 (s)	U
52	LDL*	CH3(CH2)n	0.84 (m)	P
53	VLDL*	CH3CH2CH2C=	0.89 (t)	P
54	Isoleucine	αCH, βCH, βCH3, γCH2, δCH3	3.68 (d), 1.99 (m), 1.01 (d), 1.26 (m), 1.47 (m), 0.94 (t)	P
55	Leucine	αCH, βCH2, γCH, δCH3	3.73 (t), 1.72 (m), 1.72 (m), 0.96 (d), 0.97 (d)	P
56	Valine	αCH, βCH, γCH3	3.62 (d), 2.28 (m), 0.99 (d), 1.04 (d)	P
57	3-Hydroxybutyrate	αCH2, βCH, γCH3	2.28 (dd), 2.41 (dd), 4.16 (m), 1.20 (d)	P
58	Lipids (triglycerides and fatty acids)	CH3(CH2)n, (CH2)n, CH2CH2CO, CH2C=C, CH2–C=O	1.28 (m), 1.58 (m), 2.01 (m), 2.24 (m), 2.76 (m)	P
59	Asparagine	CH2	2.85 (dd), 2.89 (dd)
60	Lysine	αCH, βCH2, γCH2, εCH2	3.76 (t), 1.91 (m), 1.48 (m), 1.72 (m), 3.01 (t)	P
61	N-Acetyl glycoprotein	CH3	2.04 (s)	P
62	O-Acetyl glycoprotein	CH3	2.08 (s)	P
63	Glutamate	αCH, βCH2, γCH2	3.75 (m), 2.12 (m), 2.35 (m)	P
64	Methionine	αCH, βCH2, γCH2, S–CH3	3.87 (t), 2.16 (m), 2.65 (t), 2.14 (s)	P
65	Glutamine	αCH, βCH2, γCH2	3.78 (m), 2.14 (m), 2.45 (m)	P
66	Glycerolphosphocholine	CH3, βCH2, αCH2	3.22 (s), 3.69 (t), 4.33 (t)	P
67	Phosphorylcholine	N(CH3)3, OCH2, NCH2	3.2 (s), 4.21 (t), 3.61 (t)	P
68	myo-Inositol	1,3-CH, 2-CH, 5-CH, 4,6-CH	3.60 (dd), 4.06 (t), 3.30 (t), 3.63 (t)	P
69	Threonine	αCH, βCH, γCH3	3.58 (d), 4.26 (m), 1.32 (d)	P
70	Unsaturated lipids	=CH–CH2C=, –CH=CH–	5.19 (m), 5.30 (m)	P
71	Tyrosine	2,6-CH, 3,5-CH	7.19 (dd), 6.90 (d)	P
72	3,4-Dihydroxymandelate	CH	6.99 (d)	P
73	1-Methylhistidine	4-CH, 2-CH	7.06 (s), 7.79 (s)	P
74	Phenylalanine	2,6-CH, 3,5-CH, 4-CH	7.32 (m), 7.42 (m), 7.37 (m)	P

---

Multivariate data analysis of NMR data

PCA and PLS-DA were initially achieved on the plasma spectral data (Fig. 3A and B). Two principal components were calculated for the treatment groups, with 58.8% and 18.9% of the variables being explained by PC1 and PC2, respectively. PCA results (Fig. 3A) demonstrated that separations in rats from the control, diquat, and glutamine + diquat groups were absent in their metabolic plasma profiles. The plasma metabolic changes in the rats from the three groups were also analyzed employing OPLS-DA. Coefficient analysis showed that glutamine + diquat significantly increased the plasma levels of citrate and isobutyrate and decreased the plasma levels of acetate, creatine, formate, glutamate, leucine, O-acetyl glycoprotein, phenylalanine, pyruvate, α-glucose, and β-glucose compared with diquat (P < 0.05, Fig. 4A and Table 2). Moreover, diquat significantly increased the plasma levels of betaine, citrate, glutamate, low density lipoprotein, lysine, TMAO, α-glucose, and β-glucose and decreased the plasma levels of lipid (triglycerides and fatty acids), unsaturated lipid, valine, and very low density lipoprotein compared with the control treatment (P < 0.05, Fig. 4B and Table 2). PCA and PLS-DA were conducted on the urine spectra of the control, diquat, and glutamine + diquat groups (Fig. 3C and D). The score plots highlighted three clusters that corresponded to the three groups. The metabolic profiles of the control, diquat, and glutamine + diquat groups were compared using OPLS-DA to identify the important urine metabolic changes. The metabolic profile of the glutamine + diquat group was compared with that of the diquat group by using OPLS-DA to observe the effect of glutamine supplementation. The urine levels of homogentisate and α-ketoglutarate were significantly higher in the glutamine + diquat group than in the diquat group (P < 0.05). By contrast, the urine levels of benzoate and trigonelline were lower in the glutamine + diquat group than in the diquat group (P < 0.05, Fig. 5A and Table 3). Multivariate data analysis showed that the urine levels of 4-aminohippurate, acetamide, acetate, alanine, benzoate, citrulline, creatinine, dimethylamine, ethanol, ethanolamine, hippurate, indoxyl sulfate, lactate, methylamine, methylguanidine, methymalonate, m-hydroxyphenylacetate, N-acetylglutamate, nicotinamide, N-methylnicotinamide, phenylacetylglycine, and trigonelline were higher in the diquat group than in the control group (P < 0.05). By contrast, the urine levels of bile acids, citrate, creatine, homogentisate, malonate, and α-ketoglutarate were lower in the diquat group than in the control group (P < 0.05, Fig. 5B and Table 3).

Fig. 3 (A) PCA (R²X = 0.965, Q² = 0.779) and (B) PLS-DA (R²X = 0.468, R²Y = 0.99, Q² = 0.741) score plots based on the ¹H NMR spectra of plasma metabolites obtained from the control (black squares), diquat (green triangles), and glutamine + diquat (red circles). (C) PCA (R²X = 0.405, Q² = 0.171) and (D) PLS-DA score plots (R²X = 0.531, R²Y = 0.977, Q² = 0.595) based on the ¹H NMR spectra of the urine obtained from urinary metabolites obtained from the control (black squares), diquat (green triangles), and glutamine + diquat (red circles).

Fig. 4 OPLS-DA scores plots of plasma metabolites (left panel) and the corresponding coefficient loading plots (right panel) of plasma metabolites taken from the control (black squares), diquat (green triangles), and glutamine + diquat (red circles) ((A), R²X = 0.193, Q² = 0.409; (B), R²X = 0.192, Q² = 0.316; (C), R²X = 0.194, Q² = 0.65). The color map shows the significance of metabolite variations between the two classes. The peaks in the positive direction indicate the metabolites that are more abundant in the groups in the positive direction of the first principal component. The metabolites that are more abundant in the groups in the negative direction of the first primary component are presented as peaks in the negative direction.

Table 2 OPLS-DA coefficients derived from the NMR data of plasma metabolites obtained from the (A) control, (B) diquat, (C) glutamine + diquat groups

Metabolite^a	B (vs. A)^b	C (vs. B)^b	C (vs. A)^b
a Metabolite keys were demonstrated in Table 1.b Correlation coefficients were calculated from OPLS-DA results with positive and negative signs indicating positive and negative correlation in the concentrations, respectively. The correlation coefficient of |r| > 0.602 was used as the cutoff value. ‘‘—’’ means the correlation coefficient |r| is less than 0.602. Analysis of relative integral from metabolites was given in Table S1 (ESI).
3,4-Dihydroxymandelate (72)	—	—	−0.603
3-Hydroxybutyrate (57)	—	—	−0.774
Acetate (12)	—	−0.727	−0.692
Acetoacetate (16)	—	—	−0.742
Acetone (15)	—	—	−0.762
Citrate (20)	0.703	0.635	−0.663
Creatine (26)	—	−0.672	—
Formate (51)	—	−0.692	—
Glutamate (63)	0.736	−0.685	−0.759
Isobutyrate (5)	−0.705	0.604	—
Isoleucine (54)	—	—	−0.680
LDL (52)	0.712	—	−0.630
Leucine (55)	—	−0.733	—
Lipid (triglyceride and fatty acids) (58)	−0.620	—	−0.836
Lysine (60)	0.681	—	−0.613
Methylamine (21)	—	—	−0.658
N-Acetyl glycoprotein (61)	—	—	−0.702
O-Acetyl glycoprotein (62)	—	−0.685	−0.848
Phenylalanine (74)	—	−0.616	−0.759
Pyruvate (17)	—	−0.617	−0.689
TMAO (33)	0.647	—	0.688
Unsaturated lipid (70)	−0.683	—	−0.810
Valine (56)	−0.670	—	−0.661
VLDL (53)	−0.744	—	−0.880
α-Glucose (40)	0.736	−0.765	0.706
β-Glucose (39)	0.687	−0.731	0.777

---

Fig. 5 OPLS-DA scores plots of urinary metabolites (left panel) and the corresponding coefficient loading plots (right panel) of urinary metabolites obtained from the control (black squares), diquat (green triangles), and glutamine + diquat (red circles) ((A), R²X = 0.245, Q² = 0.187; (B), R²X = 0.32, Q² = 0.872; (C), R²X = 0.343, Q² = 0.807). The color map illustrates the significance of metabolite variations between the two classes. The peaks in the positive direction demonstrate the metabolites that are more abundant in the groups in the positive direction of the first principal component. The metabolites that are more abundant in the groups in the negative direction of the first primary component are shown as peaks in the negative direction.

Table 3 OPLS-DA coefficients derived from the NMR data of urine metabolites obtained from the (A) control, (B) diquat, (C) glutamine + diquat groups

Metabolite^a	B (vs. A)^b	C (vs. B)^b	C (vs. A)^b
a Metabolite keys were demonstrated in Table 1.b Correlation coefficients were calculated from OPLS-DA results with positive and negative signs indicating positive and negative correlation in the concentrations, respectively. The correlation coefficient of |r| > 0.602 was used as the cutoff value. ‘‘—’’ means the correlation coefficient |r| is less than 0.602. Analysis of relative integral from metabolites was given in Table S2 (ESI).
4-Aminohippurate (48)	0.726	—	0.78
Acetamide (13)	0.626	—	—
Acetate (12)	0.614	—	0.736
Acetoacetate (16)	—	—	0.672
Alanine (10)	0.711	—	0.686
Allantoin (41)	—	—	−0.714
Benzoate (49)	0.881	−0.602	0.743
Bile acids (1)	−0.955	—	−0.776
Citrate (20)	−0.813	—	−0.642
Citrulline (11)	0.818	—	0.815
Creatine (26)	−0.779	—	−0.723
Creatinine (27)	0.643	—	0.616
Dimethylamine (22)	0.904	—	0.868
Ethanol (6)	0.672	—	—
Ethanolamine (29)	0.738	—	0.748
Hippurate (37)	0.93	—	0.84
Homogentisate (43)	−0.802	0.613	0.773
Indoxyl sulfate (46)	0.898	—	0.879
Lactate (9)	0.612	—	—
Malonate (30)	−0.715	—	−0.737
Methylamine (21)	0.848	—	0.803
Methylguanidine (23)	0.844	—	0.868
Methylmalonate (7)	0.795	—	−0.731
m-Hydroxyphenylacetate (45)	0.837	—	0.844
N-Acetylglutamate (14)	0.81	—	0.745
Nicotinamide (47)	0.801	—	0.737
N-Methylnicotinamide (38)	0.815	—	0.772
Phenylacetylglycine (36)	0.809	—	0.712
p-Hydroxyphenylacetate (44)	—	—	0.647
Propionate (4)	−0.744	—	−0.624
Succinate (18)	−0.689	—	—
TMAO (33)	—	—	−0.620
Trigonelline (50)	0.927	−0.689	0.905
α-Hydroxy-iso-valerate (2)	−0.682	—	—
α-Hydroxy-n-valerate (8)	0.805	—	0.658
α-Ketoglutarate (19)	−0.821	0.603	−0.692

---

Discussion

Effects of diquat injection

Diquat can induce oxidative stress responses. Diquat has been used to induce oxidative stress in different animal models.¹⁸ In this study, diquat increased the urine levels of N-methylnicotinamide and nicotinamide. N-Methylnicotinamide is the methylated metabolite of nicotinamide after the S-adenosylmethionine to S-adenosylhomocysteine conversion in the biosynthesis of cysteine, an essential amino acid in glutathione synthesis.²⁵ Moreover, diquat increased the urine level of indoxyl sulfate, a circulating uremic toxin that increases glomerular sclerosis and interstitial fibrosis. Indoxyl sulfate is a well-known substance of protein-bound uremic retention solutes. Previous studies associated indoxyl sulfate with oxidative stress.^26,27 Therefore, high levels of indoxyl sulfate indicate elevated ROS production in rats. Diquat significantly reduced plasma glutathione, antisuperoxide anion, catalase, and antihydroxy radical levels while increase malondialdehyde level.¹⁷ Thus, diquat supplementation decreased the antioxidant status in rats. Furthermore, two enzymes (ALT and AST) are normally localized in liver cytoplasm and are released into circulation in the presence of liver cell damages. AST and AST/ALT indicate liver function. Significant increases in the plasma levels of AST and AST/ALT (data not shown) suggest hepatic dysfunction caused by oxidative stress.

Diquat can alter bile acid and lipid metabolisms. Bile acids are formed from cholesterol in the liver, stored in the gall bladder, and secreted via the bile into the intestine, in which these acids assist the formation of micelles. Such formation enhances the processing of dietary fat. As part of their enterohepatic circulation, most bile acids (>90%) are reabsorbed in the ileum and returned through the portal vein to the liver.^28,29 The part of the remaining bile acids were excreted into urine, where they were actual quantified. In the present study, diquat decreased the urine levels of bile acids. Diquat supplementation can also affect lipid oxidation. Ketone bodies, such as acetone, 3-hydroxybutyrate, and acetoacetate, are produced through the β-oxidation of fatty acids in the mitochondria. In the present study, the levels of acetone, 3-hydroxybutyrate, and acetoacetate were higher in the diquat group than in the control group, suggesting changes in lipid metabolism. Moreover, 4-aminohippurate is an acyl glycine, a minor metabolite of fatty acids. Previous studies showed 4-aminohippurate is related with renal function and inhibited by bile salts.^30,31 In the present study, 4-aminohippurate levels increased. This is in agreement with the current results: bile acids were decreased by diquat. Diquat supplementation also affected LDL and lipid levels in rats. Collectively, diquat can alter bile acid and lipid metabolism in rats.

Diquat can change energy metabolism. Diquat can increase urinary lactate level in rats. Increased lactate level is linked to increased anaerobic glycolysis, indicating changes in carbohydrate and energy metabolisms. In the present study, the diquat group had higher urinary alanine and plasma glucose levels than the control group. This finding suggests that diquat can alter the glucose–alanine cycle. Furthermore, diquat decreased the levels of succinate, citrate, and α-ketoglutarate, all of which are intermediates of the tricarboxylic acid (TCA) cycle. These results suggest that diquat supplementation can downregulate the TCA cycle. Overall, diquat can affect energy metabolism in rats.

Diquat can also alter amino acid metabolism. In the present study, total protein level increased after diquat injection, indicating that diquat may affect protein synthesis. The above results agree with a previous report that diquat can reduce protein synthesis in cells. As a result, proteins may be decomposed into more amino acids, causing increased levels of amino acids (e.g. glutamate, lysine) in plasma of the current study. Moreover, here, the plasma levels of branched-chain amino acids (e.g. valine) were decreased by diquat possibly because oxidative stress-induced energy expenditure can cause elevated consumptions of branched-chain amino acids to provide energy. In addition, diquat increased the urine levels of citrulline and N-acetylglutamate. Citrulline is an amino acid produced from ornithine and carbamoyl phosphate in the urea cycle. This amino acid is derived from arginine as a by-product of the reaction catalyzed by nitric oxide synthase. In this reaction, arginine is first oxidized into N-hydroxyl-arginine and then oxidized further to citrulline in conjunction with the release of nitric oxide.^18,32 Urea functions in the metabolism of nitrogen-containing compounds. N-Acetylglutamate is crucial for the normal function of the urea cycle; thus, changes in N-acetylglutamate concentration affect the production rates of urea and other substrates.^18,33 In the present study, diquat increased the plasma level of creatinine, suggesting that diquat can affect amino acid metabolism in rats.

Diquat injection can also regulate gut microbiota metabolism by influencing urinary ethanol, short-chain fatty acids (e.g., isobutyrates, propionate, and acetate), and nitrogenous products (urinary methylamine, dimethylamine, and plasma TMAO). These compounds are microbial metabolites of carbohydrates and amino acids, which are produced in the lumen of the small and large intestines.^34,35 In the present study, plasma isobutyrate and urinary acetate levels increased. However, urine propionate decreased in the diquat group, which can be attributed to the fact that gut microbiota can either manufacture or utilize these products. Moreover, the plasma level of microbiotic metabolites such as m-hydroxyphenylacetate significantly increased. Furthermore, diquat increased the urinary excretion of hippurate, which is produced via the renal and hepatic syntheses of glycine and benzoate. Hippurate is the degradation product of flavonols acted upon by intestinal microorganisms.³⁶ This finding corroborates with the increase in benzoate. As a result, a change in the excretion of this compound suggests a corresponding change in the functional metabolism of the microbiota. Variations in urinary hippurate concentration have also been associated with changes in the distribution of intestinal microbial colonies.³⁷ Changes in gut microbial co-metabolites such as phenylacetylglycine upon diquat exposure verified the association of the disturbance to gut microbiota. Through the action of gut microbiota, phenylacetate was transformed from phenylalanine and then conjugated with glycine to produce phenylacetylglycine.³⁷ Mammalian metabolism is significantly affected by the complex gut microbiota. The introduction of diquat into the mammalian system may displace the baseline mammalian-to-microbial behavior, disrupt microbial populations, and eventually affect metabolism. In the present study, urinary acetamide levels significantly increased. Acetamide shows anti-microbial, anti-inflammatory, anti-arthritic, and antibiotic functions.^38,39 Changes in these metabolites can be attributed to alterations in the number and/or activity of intestinal microorganisms. Gut microbiota can significantly affect the development and structure of the intestinal epithelium, the digestive and absorptive capabilities of the intestine, and the host immune system. Therefore, diquat-induced disturbances in the gut microbiota can affect gut health status.

Effects of glutamine under oxidative stress

Glutamine supplementation can reduce oxidative stress. O-Acetylated carbohydrate-bound protein resonance is found in rat blood plasma and can be considered as an alternative “acute-phase” glycoprotein in rats under zearalenone mycotoxin-induced oxidative stress²⁷ and in animal models of human inflammatory conditions.⁴⁰ “Acute phase” acetyl glycoproteins are predominantly synthesized in liver parenchymal cells in response to cytokines.⁴¹ In the current study, the signal intensity of O-acetyl glycoprotein was lower in the plasma from the treatment group than in that from the control group. Reduced levels of O-acetyl glycoproteins in blood plasma can imply decreased oxidative stress. This is in agreement with our study: glutamine + diquat can increase glutathione, anti-superoxide anion, anti-hydroxy radical concentrate, and decrease malondialdehyde content compared with diquat group (data not shown). Moreover, glutamine can alter energy metabolism in rats. In this study, glutamine can decrease plasma glycolytic metabolite (pyruvate) and glucose levels in rats under oxidative stress. Glucose is a major substrate that provides energy for animal growth and development. Glucose was possibly degraded to meet the energy requirement under oxidative stress conditions. Glutamine can also increase urinary α-ketoglutarate and plasma citrate levels under oxidative stress. Considering this result, the tricarboxylic acid cycle was altered in rats. However, the creatine levels were lower in the glutamine + diquat group than in the diquat group. Creatine supplies energy to the muscles in vertebrates in the form of stored creatine phosphate. The creatine levels in the animals are synthesized de novo in the liver by using amino acids, such as arginine, glycine, and methionine. Therefore, glutamine can affect energy metabolism in rats under oxidative stress.

Furthermore, glutamine can change the amino acid metabolism in rats. Glutamine activates signaling pathways to promote protein synthesis and eventually animal growth and development.¹³ Consequently, protein synthesis decreases the amount of amino acids in the plasma. Previous observation of arginine supplementation in growing pigs revealed a similar trend.⁴² In the present study, the levels of glutamate, phenylalanine, and leucine decreased in the plasma of rats. Glutamate is the preferential source of mucosal glutathione synthesis in animals. Moreover, glutamine supplementation decreased the levels of branched-chain amino acids under oxidative stress. These amino acids are key metabolites associated with protein synthesis and cell growth. These results agree with previous findings that glutamine supplementation can significantly reduce serum phenylalanine and leucine concentration in DSS-induced colitis rats on day 7.⁴³

The exposure to glutamine can modify gut microbiota metabolism under oxidative stress. SCFAs (e.g., formate, isobutyrates, and acetate) produced by bacteria in the colon through the fermentation of unabsorbed dietary fiber provide energy for metabolism in the colon. In the present study, isobutyrates were higher and formate and acetate were lower in the glutamine + diquat group than in the diquat group. This result can be attributed to the fact that gut microbiota can either manufacture or utilize these products.

Conclusion

Oxidative stress and glutamine supplementation alter common systemic metabolic processes, including energy, amino acid, and gut microbiota metabolisms. Moreover, glutamine can partially counteract the metabolic consequences associated with oxidative stress, such as amino acid metabolism. This study revealed the metabolic consequences related to oxidative stress and glutamine supplementation. These results have important implications in nutritional research in humans and animals. To the best of our knowledge, this study is the first to systematically identify the metabolome from glutamine supplemented against oxidative stress. However, the mechanism by which glutamine exerts protective effects against oxidative stress in animal tissue intermediary metabolism requires further investigation.

Acknowledgements

This research is funded in part by the National Natural Science Foundation of China (project number is 31301986) and Specific Research Supporting Program for Discipline Construction in Sichuan Agricultural University. We are also grateful to all the study participants for their ongoing assistance.

References

1. J. Yin, W. Ren, G. Liu, J. Duan, G. Yang, L. Wu, T. Li and Y. Yin, Free Radical Res., 2013, 47, 1027–1035.
2. J. Yin, M. M. Wu, H. Xiao, W. K. Ren, J. L. Duan, G. Yang, T. J. Li and Y. L. Yin, J. Anim. Sci., 2013, 92, 612–619.
3. N. Barnabé and M. Butler, Cytotechnology, 2000, 34, 47–57.
4. S. Ahmad, C. W. White, L. Y. Chang, B. K. Schneider and C. B. Allen, Am. J. Physiol.: Lung Cell. Mol. Physiol., 2001, 280, 779–791.
5. J. M. Rhoads, R. A. Argenzio, W. Chen, R. A. Rippe, J. K. Westwick, A. D. Cox, H. M. Berschneider and D. A. Brenner, Am. J. Physiol.: Gastrointest. Liver Physiol., 1997, 272, G943–G953.
6. G. Wu, J. Nutr., 1998, 128, 1249–1252.
7. J. Wang, L. Chen, P. Li, X. Li, H. Zhou, F. Wang, D. Li, Y. Yin and G. Wu, J. Nutr., 2008, 138, 1025–1032.
8. K. Ban and R. A. Kozar, Am. J. Physiol.: Gastrointest. Liver Physiol., 2010, 299, G1344–G1353.
9. S. S. Yoo, C. J. Field and M. I. McBurney, J. Nutr., 1997, 127, 2253–2259.
10. Y. Zhu, G. Lin, Z. Dai, T. Zhou, T. Li, T. Yuan, Z. Wu, G. Wu and J. Wang, Amino Acids, 2015, 47, 2185–2197.
11. P. Groening, Z. Huang, E. F. La Gamma and R. J. Levy, JPEN, J. Parenter. Enteral Nutr., 2011, 35, 249–254.
12. B. Humbert, P. Nguyen, L. Martin, H. Dumon, G. Vallette, P. Maugère and D. Darmaun, J. Nutr. Biochem., 2007, 18, 10–16.
13. P. Xi, Z. Jiang, Z. Dai, X. Li, K. Yao, C. Zheng, Y. Lin, J. Wang and G. Wu, J. Nutr. Biochem., 2012, 23, 1012–1017.
14. G. Wu, F. W. Bazer, G. A. Johnson, D. A. Knabe, R. C. Burghardt, T. E. Spencer, X. Li and J. Wang, J. Anim. Sci., 2011, 89, 2017–2030.
15. Y. Xiao, T. Wu, J. Sun, L. Yang, Q. Hong, A. Chen and C. Yang, J. Anim. Sci., 2012, 90, 4421–4430.
16. L. Xiao, W. Cao, G. Liu, T. Fang, X. Wu, G. Jia, X. Chen, H. Zhao, J. Wang, C. Wu and J. Cai, Anim. Nutr., 2016 DOI:10.1016/j.aninu.2016.04.005.
17. W. Cao, L. Xiao, G. Liu, T. Fang, X. Wu, G. Jia, H. Zhao, X. Chen, C. Wu, J. Cai and J. Wang, Food Funct., 2016, 7, 2303–2311.
18. G. Liu, T. Fang, T. Yan, G. Jia, H. Zhao, X. Chen, C. Wu and J. Wang, RSC Adv., 2014, 4, 56766–56778.
19. J. K. Nicholson and P. J. D. Foxall, Anal. Chem., 1995, 67, 793–811.
20. T. W. M. Fan, Prog. Nucl. Magn. Reson. Spectrosc., 1996, 28, 161–219.
21. D. S. Wishart, T. Jewison, A. C. Guo, M. Wilson, C. Knox, Y. Liu, Y. Djoumbou, R. Mandal, F. Aziat, E. Dong, S. Bouatra, I. Sinelnikov, D. Arndt, J. Xia, P. Liu, F. Yallou, T. Bjorndahl, R. Perez-Pineiro, R. Eisner, F. Allen, V. Neveu, R. Greiner and A. Scalbert, Nucleic Acids Res., 2013, 41, D801–D807.
22. J. Trygg and S. Wold, J. Chemom., 2002, 16, 119–128.
23. O. Cloarec, M. E. Dumas, J. Trygg, A. Crai, R. H. Barton, J. C. Lindon, J. K. Nicholson and E. Holmes, Anal. Chem., 2005, 77, 517–526.
24. F. Lindgren, B. Hansen, W. Karcher, M. SjÖstrÖm and L. Eriks, J. Chemom., 1996, 10, 521–532.
25. C. Huang, H. Lei, X. Zhao, H. Tang and Y. Wang, J. Proteome Res., 2013, 12, 537–545.
26. L. Dou, N. Jourde-Chiche, V. Faure, C. Cerini, Y. Berland, F. Diqnat-George and P. Brunet, J. Thromb. Haemostasis, 2007, 5, 1302–1308.
27. G. Liu, T. Yan, J. Wang, Z. Huang, X. Chen, G. Jia, C. Wu, H. Zhao, B. Xue, L. Xiao and J. Tang, J. Agric. Food Chem., 2013, 61, 11212–11221.
28. J. Y. Chiang, J. Hepatol., 2004, 40, 539–551.
29. D. W. Russell, Annu. Rev. Biochem., 2003, 72, 137–174.
30. J. Sun, L. K. Schnackenberg, D. K. Hansen and R. D. Beger, Bioanalysis, 2010, 2, 207–216.
31. A. Despopoulos, J. Pharmacol. Exp. Ther., 1971, 176, 273–283.
32. N. N. Huynh and J. Chin-Dusting, Clin. Exp. Pharmacol. Physiol., 2006, 33, 1–8.
33. A. J. Meijer and A. J. Verhoeven, Biochem. J., 1984, 56, 559–560.
34. S. Rezzi, Z. Ramadan, L. B. Fay and S. Kochhar, J. Proteome Res., 2007, 6, 513–525.
35. G. W. Tannock, Appl. Environ. Microbiol., 2004, 70, 3189–3194.
36. A. R. Rechner, G. Kuhnle, H. Hu, A. Roedig-Penman, M. H. van den Braak, K. P. Moore and C. A. Rice-Evans, Free Radical Res., 2002, 36, 1229–1241.
37. E. Bohus, M. Coen, H. C. Keun, T. M. Ebbels, O. Beckonert, J. C. Lindon, E. Holmes, B. Noszál and J. K. Nicholson, J. Proteome Res., 2008, 7, 4435–4445.
38. H. Jawed, S. U. Shah, S. Jamall and S. U. Simjee, Int. Immunopharmacol., 2010, 10, 900–905.
39. E. M. Muri and J. S. Williamson, Mini-Rev. Med. Chem., 2004, 4, 201–206.
40. M. Grootveld, A. W. D. Claxson, C. L. Cander, P. Haycock, D. R. Blake and G. E. Hawkes, FEBS Lett., 1993, 322, 266–276.
41. Y. Wang, E. Holmes, H. Tang, J. C. Lindon, N. Sprenger, M. E. Turini, G. Bergonzelli, L. B. Fay, S. Kochhar and J. K. Nicholson, J. Proteome Res., 2006, 5, 1535–1542.
42. Q. He, X. Kong, G. Wu, P. Ren, H. Tang, F. Hao, R. Huang, T. Li, B. Tan, P. Li, Z. Tang, Y. Yin and Y. Wu, Amino Acids, 2009, 37, 199–208.
43. W. Ren, J. Yin, M. Wu, G. Liu, G. Yang, Y. Xion, D. Su, L. Wu, T. Li, S. Chen, J. Duan, Y. Yin and G. Wu, PLoS One, 2014, 9, e88335.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra14469a

---