Obesity is a major risk factor for a number of chronic diseases, including type 2 diabetes, hypertension and CVD(Reference Wolongevicz, Zhu and Pencina1). Increased levels of inflammatory and oxidative biomarkers are the main features of obesity, which is the pathological and physiological feature of many conditions(Reference Trayhurn, Wang and Wood2). Obesity could lead to metabolic disorders of blood glucose and lipids(Reference Franssen, Monajemi and Stroes3, Reference Becker, Dossus and Kaaks4). And numerous studies(Reference Aikawa, Sugiyama and Hill5–Reference Rytter, Vessby and Asgard8) have confirmed that the metabolic disorders of blood glucose and lipids are closely related to inflammation and oxidative stress. However, it is still unclear if amino acid metabolism is altered and if there are links between serum amino acids and inflammation and oxidative stress in obese subjects.
As carbohydrate, protein and fat are the main macronutrients, whose metabolisms are closely related with each other, we hypothesised that imbalanced amino acid metabolism might also exist in obese subjects. The aims of this study were to compare the serum amino acid profiles between obese and non-obese women, and evaluate the associations between amino acid metabolism and inflammation and oxidative stress in a human case–control study.
Methods
A total of 235 middle-aged obese women and 217 non-obese controls were recruited into the present study based on the criteria: (1) BMI of 18–24 kg/m2 (non-obese control group); BMI ≥ 28 kg/m2 (obese group)(Reference Zhou9); (2) fasting blood glucose < 6·1 mmol/l and 2-h postload blood glucose concentration < 7·8 mmol/l; (3) BMI fluctuation < 0·5 kg/m2 in the past 6 months; (4) no disease that could influence the metabolism of nutrients, such as kidney, liver and gastrointestinal diseases; (5) no history of endocrine metabolic disorders, acute infection, injuries or chronic inflammatory diseases; (6) not being pregnant. This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Human Research Ethics Committee of the Harbin Medical University. Written informed consent was obtained from all subjects.
Information on age, weight, height, alcohol use, cigarette smoking, menstrual status, and physical activity at work and at leisure was obtained from questionnaires. Dietary intakes were measured with a validated semi-quantitative FFQ based on frequency(Reference Li, Wang and Zhu10). Energy and nutrient intakes were calculated with the Food Nutrition Calculator (V1·60; Chinese CDC, Beijing, China).
Using a standard protocol, trained examiners measured body weight ( ± 0·1 kg), height ( ± 0·1 cm) and the blood pressure of all participants. BMI was calculated as body weight (kg) divided by the square of height (m). Blood pressure was measured with a standard mercury sphygmomanometer on the right arm after resting for at least 10 min. The mean of the three closest readings was used in the analysis.
Subjects were instructed to fast for 12 h and blood samples were collected in the morning. After the fasting blood sample collection, each subject was given an oral glucose tolerance test and blood samples were collected in 2 h after the 75 g glucose was taken. Antecubital venous blood was collected to obtain serum and plasma, and then stored at − 80°C.
Samples of serum (100 μl) were spiked with acetonitrile (100 μl), vortex-mixed for 30 s, immediately centrifuged at 11 000 g for 10 min at 4°C and the supernatant was recovered. An Alliance 2695 separation module and a 2487 UV detector from Waters were used for the HPLC analysis (CV%, 4·2–5·3 %). Derivatisation and gradient elution of serum amino acids were followed according to the manufacturer's instructions (Waters Corporation, Milford, MA, USA).
Fasting blood glucose, 2-h postload blood glucose, total cholesterol, HDL-cholesterol and TAG were determined using commercial kits (Biosino Biotechnology, Beijing, China), standard enzymatic colorimetric techniques and an auto-analyser (MOL-300, Shanghai, China).
C-reactive protein (CRP) and IL-6 were assayed using ELISA with commercial kits (CRP, Biocheck, Inc., Foster City, CA, USA; IL-6, R&D Systems Europe, Abingdon, UK) according to the manufacturers' instructions. Superoxide dismutase (SOD), malondialdehyde (MDA) and glutathione peroxidase were measured with commercial kits using enzymatic methods (Jiancheng Technology, Nanjing, China).
Statistical analysis
Data were presented as means and standard deviations. Comparisons between obese and non-obese women in the population study were made by using Student's t test for continuous variables and χ2 test for categorical variables. ANCOVA was used to compare serum amino acid levels between obese women and non-obese controls, adjusting for potential covariates (age, smoking, alcohol use, menopause, and physical activity at work and at leisure). Multiple linear regression (adjusted for the aforementioned variables) between serum amino acids and inflammatory and oxidative biomarkers was also calculated in overall, non-obese controls and obese women, respectively. Statistical analyses were carried out with SAS software (version 9.1; SAS Institute, Inc., Cary, NC, USA).
Results
Clinical parameters
As shown in Table 1, BMI, systolic blood pressure, diastolic blood pressure, 2-h postload blood glucose concentration, total cholesterol, TAG, fat and energy intakes (data not shown), CRP, IL-6 and MDA were significantly higher, while HDL-cholesterol, SOD and glutathione peroxidase of obese women were significantly lower than that of non-obese controls. There were no significant differences in the other variables (smoking, alcohol drinking, menopause and physical activity; data not shown).
(Mean values and standard deviations)
SBP, systolic blood pressure, DBP, diastolic blood pressure; TC, total cholesterol; HDL-C, HDL-cholesterol; FBG, fasting blood glucose; 2 h-PG, 2-h postload blood glucose concentration; CRP, C-reactive protein; SOD, superoxide dismutase; MDA, malondialdehyde; GSH-PX, glutathione peroxidase.
* Mean values were significantly different from those of non-obese controls (P < 0·001; Student's t test).
† Mean values were significantly different (P < 0·001; analysis made by ANCOVA), adjusted for age, smoking, alcohol use, menopause and physical activity at work and at leisure.
Amino acid profile
The serum concentrations of amino acids in obese women and non-obese controls are presented in Table 1. Accounting for age, smoking, alcohol use, menopause, physical activity at work and at leisure, serum histidine, arginine, threonine, glycine, lysine and serine were found to be significantly lower in obese women in comparison to non-obese controls. The difference was the greatest for histidine.
Multiple linear regression between serum amino acids and inflammatory and oxidative biomarkers
As shown in Table 2, among the serum histidine, arginine, threonine, glycine, lysine and serine, both histidine and arginine were negatively associated with IL-6, CRP and MDA, and positively associated with SOD; histidine was positively associated with glutathione peroxidase in the overall subjects. The results were consistent as the confounders were included step by step in each model.
(Standardised coefficients, β)†
CRP, C-reactive protein; SOD, superoxide dismutase; MDA, malondialdehyde; GSH-PH, glutathione peroxidase.Values were significantly different: *P < 0·005, **P < 0·001.
† Multiple linear regression: potential confounders including age, smoking, alcohol use, menopause and physical activity at work and at leisure, and BMI were included step by step in Models 1–6; threonine, glycine, lysine and serine were not shown in the table, for their standardised coefficients were not significant.
In addition, as shown in Table S1 (the supplementary material for this article can be found at http://www.journals.cambridge.org/bjn), both histidine and arginine were negatively associated with IL-6, CRP and MDA, and positively associated with SOD in obese women. Histidine was positively associated with SOD and negatively associated with MDA in non-obese controls.
Discussion
In the present study, we found that the serum amino acid profiles could separate between obese and non-obese women. Among the thirteen kinds of amino acids determined, six serum amino acids (histidine, arginine, threonine, glycine, lysine and serine) were significantly lower in obese women than in the non-obese controls, although obese and non-obese subjects had similar dietary protein intake (Table 1).
The difference of serum histidine (mean histidine level in obese women was lower than in the controls by 14·5 %) was the greatest of the amino acids between the two groups, which indicated that histidine might be used as a biomarker of human obesity. Our study reported a lower serum histidine in obese women for the first time. However, subjects with some other conditions were also reported to have lower histidine concentrations. Watanabe et al. (Reference Watanabe, Suliman and Qureshi11) reported that plasma histidine concentrations were significantly lower in chronic kidney disease patients with a history of CVD, protein-energy wasting and inflammation. Wijekoon et al. (Reference Wijekoon, Skinner and Brosnan12) found that fat Zucker diabetic rats had significantly lower plasma histidine than that of lean Zucker diabetic rats. Lee et al. (Reference Lee, Hsu and Lin13) showed that diabetic Balb/cA mice had lower plasma histidine compared with non-diabetic mice. It is obvious that subjects in the previous human and animal studies shared the same feature: inflammation and oxidative stress. According to our results and previous evidence, inflammation and oxidative stress in obese subjects might have a close relationship with a low histidine level.
Further, we found that histidine was significantly negatively associated with inflammation (IL-6: β = − 0·210 and CRP: β = − 0·180, P < 0·05) and oxidative stress (SOD: β = 0·327 and MDA: β = − 0·282, P < 0·05) in obese and overall subjects (Table 2). Findings of previous studies might explain these relationships. First, inflammatory status might lead to the low histidine levels. Several studies have found that histidine supplementation could reduce the levels of IL-6, TNF-α and CRP in animal models with liver injury or diabetes(Reference Lee, Hsu and Lin13, Reference Yan, Wu and Yin14). Son et al. (Reference Son, Satsu and Shimizu15), in their in vitro study, reported that histidine inhibited the H2O2- and TNF-α-induced IL-8 secretion at the transcriptional level in intestinal epithelial cells. Second, because histidine was a free-radicals scavenger and could chelate divalent metal ions in animal and in vitro studies(Reference Babizhayev, Seguin and Gueyne16, Reference Lee, Miyawaki and Bobst17), oxidative stress in obese subjects might be another reason for this negative relationship. However, it was still unclear regarding the causation of lower histidine level in obese women. Further animal studies should be designed to determine if patients-induced obesity could lead to the low histidine level.
Arginine has been well explored in its nutritional and therapeutic roles in treating and preventing human and animal metabolic disorders. Its low level in obese subjects might relate to its bio-functions. Arginine is the substrate for the synthesis of NO and has numerous useful physiological properties and immunomodulatory effects. In diabetic animal models, arginine supplementation was found to reduce white fat mass and endothelium-dependent relaxation, and also to increase the brown adipose tissue(Reference Fu, Haynes and Kohli18, Reference Wu, Collins and Perkins-Veazie19). One clinical trial investigated the supplementation of arginine for 21 d in human subjects and found that arginine increased the antioxidant capacity in the obese humans(Reference Lucotti, Setola and Monti20). Arginine treatment was reported to have decreased inflammatory cytokine levels in chronic renal failure and diabetic rats(Reference Korish21, Reference Pai, Huang and Wu22).
There were six kinds of amino acids found lower in obese subjects with a large sample size. Among them, histidine, threonine and lysine are essential amino acids; arginine, glycine and serine are non-essential amino acids. There could be a higher level of catabolism or utilisation for the essential amino acids; and higher level of utilisation or lower for de novo synthesis of non-essential amino acids in obese subjects. However, as to why the lower circulating levels of these amino acids were found in obese subjects, was still unclear. Whether there are any functional interactions with these amino acids still needs more investigation. As we all know, amino acids were not only the common nutrients, but also cell signalling molecules, regulators of gene expression and the protein phosphorylation cascade(Reference Wu24). Our findings suggest that more attention should be paid to the imbalanced amino acid metabolism in obese subjects.
In the present study, we found that obese subjects had the most significant histidine level and that histidine was negatively associated with inflammation and oxidative stress in obese women. In addition, there have been reports indicating that histidine could suppress appetite and influence body weight through its conversion into neuronal histamine in the hypothalamus of female subjects(Reference Yoshimatsu, Chiba and Tajima25, Reference Kasaoka, Tsuboyama-Kasaoka and Kawahara26). Arginine also has the effects of anti-inflammation, anti-oxidative stress and regulating of the vascular functions. Thus, it is interesting to uncover if obese subjects could benefit from histidine or arginine supplementation in their daily diet.
There are still some limitations in the present study. First, this study only included middle-aged female subjects as they have higher obesity prevalence. The findings from the sample of females alone in this case–control study may not be generalised to the total population. Second, this case–control study may not provide fully convincing evidence of the causal effects of histidine and arginine on chronic inflammation and oxidative stress in obese human subjects; hence, further animal and clinical studies are necessary to clarify these associations.
In conclusion, our study indicated that amino acid metabolic disorder might exist in obese women. Overall, six kinds of serum amino acids were significantly lower in obese women; serum amino acid profiles could discriminate between obese and non-obese women; and histidine and arginine have close relationships with inflammation and oxidative stress in obese women.
Acknowledgements
The authors appreciate the efforts and collaboration of colleagues from the Harbin Center for Disease Control and Prevention in the present study. This study was supported by the National Natural Science Fund of China (nos. 81172649, 81102112 and 81130049). The contributions of the authors to this study are as follows: Y. L. and C.-H. S. were responsible for the study concept and design; R.-N. F., Y.-C. N. and Z. K. performed the experiment and collection of data; K. L. and J. W. carried out the analysis and interpretation of the data; R.-N. F., Y.-C. N. and Y. H. were responsible for the drafting of the manuscript; Y. L. and C.-H. S. contributed also to the critical revision of the manuscript for its intellectual content; K. L. and Y. H. performed the statistical analysis; and Y. L. and C.-H. S. were responsible for the obtaining of funding and supervision. None of the authors had a conflict of interest.
