Bariatric Surgery Normalizes Protein Glycoxidation and Nitrosative Stress in Morbidly Obese Patients

The results of recent studies indicate the key role of nitrosative stress and protein oxidative damage in the development of morbid obesity. Nevertheless, the effect of bariatric surgery on protein oxidation/glycation and nitrosative/nitrative stress is not yet known. This is the first study evaluating protein glycoxidation and protein nitrosative damage in morbidly obese patients before and after (one, three, six and twelve months) laparoscopic sleeve gastrectomy. The study included 50 women with morbid obesity as well as 50 age- and gender-matched healthy controls. We demonstrated significant increases in serum myeloperoxidase, plasma glycooxidative products (dityrosine, kynurenine, N-formyl-kynurenine, amyloid, Amadori products, glycophore), protein oxidative damage (ischemia modified albumin) and nitrosative/nitrative stress (nitric oxide, peroxy-nitrite, S-nitrosothiols and nitro-tyrosine) in morbidly obese subjects as compared to lean controls, whereas plasma tryptophan and total thiols were statistically decreased. Bariatric surgery generally reduces the abnormalities in the glycoxidation of proteins and nitrosative/nitrative stress. Noteworthily, in the patients with metabolic syndrome (MS+), we showed no differences in most redox biomarkers, as compared to morbidly obese patients without MS (MS−). However, two markers: were able to differentiate MS+ and MS− with high specificity and sensitivity: peroxy-nitrite (>70%) and S-nitrosothiols (>60%). Further studies are required to confirm the diagnostic usefulness of such biomarkers.


Introduction
Obesity is the leading problem for public health worldwide. This multifactorial disease increases the risk of hypertension, cardiovascular disease, type 2 diabetes (T2DM) and cancer [1,2]. The patho-mechanism of obesity and its metabolic complications has not been fully understood. We still do not know why some morbidly obese patients develop metabolic syndrome (MS), whereas others do not. Adipose tissue distribution is regulated by the interaction of many proteins and hormones (amlodipine), whereas patients with T2DM took metformin, gliclazide and/or insulin. None of the patients received β blockers with a nitric oxide donative action (e.g., nebivolol), nitrates nor captopril.
The control group included 50 healthy lean women (BMI < 25 kg/m 2 , aged 28 to 56) treated in q Specialist Dental Clinic at the Medical University of Bialystok, Poland. Only subjects with blood count and biochemical blood parameters (Na + , K + , INR, ALT, AST and creatinine) within the reference range were enrolled in the study.
Exclusion criteria for both lean and morbidly obese women were as follows: malignancy, acute inflammatory diseases, infectious diseases (HIV/AIDS, hepatitis A, B, or C), autoimmune diseases (Crohn's and Hashimoto's disease, ulcerative colitis,), cardiovascular diseases (with the exception of arterial hypertension in the OB group), metabolic diseases, such as osteoporosis, type 1 diabetes, mucopolysaccharidosis and gout), digestive, respiratory or genitourinary systems diseases. All subjects were instructed not to take antibiotics, nonsteroidal anti-inflammatory drugs, glucocorticosteroids and antioxidant supplements (including iron preparations) for three months prior to bariatric surgery. Nevertheless, after laparoscopic sleeve gastrectomy, mineral and vitamin supplementation have been implemented in all patients (WLS Optimum, FitForMe©). All patients declared not smoking nor drinking alcohol for three months prior to blood sampling.

Blood Collection and Laboratory Measurements
Blood samples were collected from patients after overnight fasting into serum and ethylenediaminetetraacetic acid (EDTA)-coated tubes (S-Monovette SARSTEDT) and centrifuged at 4 • C and 4000 rpm for 10 min. For twenty-four hours prior to blood sampling, patients had not performed any intense physical activity. Butylated hydroxytoluene (10 µL 0.5 M BHT/1 mL serum/plasma) was added to all samples to protect against oxidation [23]. The serum and plasma samples were stored at −80 • C for further analysis. The Abbott analyzer (Abbott Diagnostics, Wiesbaden, Germany) was used to assay the blood counts and biochemical laboratory parameters. Homeostatic model assessment index (HOMA-IR) was calculated according to Matthews et al. [24].

Redox Assays
All reagents (unless otherwise stated) were purchased from Sigma-Aldrich Nümbrecht, Germany and Sigma-Aldrich Saint Louis, MO, USA. The absorbance and fluorescence were measured using Infinite M200 PRO Multimode Microplate Reader (Tecan Group Ltd., Männedorf, Switzerland). Redox assays were performed in duplicate samples. Due to the high number of determinations, a detailed description of biochemical methods was given in references. The results were standardized to 1 mg of total protein. The total protein concentration was analyzed using a commercial kit (Thermo Scientific PIERCE BCA Protein Assay; Rockford, IL, USA), according to manufacturer's instructions.
The formation of amyloid cross-β structure was analyzed colorimetrically using thioflavin T [27]. The characteristic fluorescence was measured at 435/485 nm.
The formation of Amadori product was analyzed colorimetrically using nitro blue tetrazolium (NBT) assay [28]. The absorbance was measured at 525 nm and extinction coefficient of 12,640 cm −1 mol −1 l for monoformazan was used.

Protein Oxidative Damage
The concentration of total thiols was determined colorimetrically at 420 nm using Ellman's reagent [30]. The concentration of thiol groups was calculated from the calibration curve using reduced glutathione (GSH) as a standard.
Serum ischemia modified albumin (IMA) concentration was measured colorimetrically. The method is based on the measurement of the exogenous cobalt (Co 2+ ) binding facility of the human serum albumin [31]. The absorbance was assessed at 470 nm.
Plasma S-nitrosothiols concentration was measured colorimetrically based on the reaction of the Griess reagent with Cu 2+ ions [33,35]. The absorbance was analyzed at 490 nm.
Plasma nitro-tyrosine concentration was analyzed using an ELISA commercial kit (Immundiagnostik AG; Bensheim, Germany), according to the manufacturer's instructions.

Statistics
GraphPad Prism 8.3.0 (GraphPad Software, Inc., San Diego, CA, USA) was used for statistical data analysis. The distribution of results was assessed using the Shapiro-Wilk test, while the Leven test was used to assess the homogeneity of variance. Due to lack of normal distribution of results, ANOVA Kruskal-Wallis test and Dunn's post hoc test were used for comparison of quantitative variables. Multiplicity adjusted p-value was also calculated. Correlations between redox biomarkers and clinical parameters were assessed using Spearman correlation coefficient. The value of redox biomarkers in predicting the risk of MS development was determined using ROC (Receiver Operating Characteristic) analysis analyzing the area under the curve (AUC). The cut-off point with the highest sensitivity and specificity was determined. The test probability p (materiality limit) was established as 0.05 (two-sided).
We observed increased kynurenine plasma content in OB (+17%, p = 0.0206) before laparoscopic sleeve gastrectomy in comparison with the controls ( Figure 2B).

Protein Oxidative Damage
There were no statistically significant differences in the plasma concentrations of total thiols in morbidly obese women as compared to controls ( Figure 3A). The plasma concentrations of ischemia modified albumin (IMA) were greater in morbidly obese patients before (+35%, p < 0.0001) sleeve gastrectomy than in healthy controls ( Figure 3B).
Additionally, we checked correlations between the analyzed nitrosative/nitrative stress biomarkers and clinical parameters 12 months after bariatric surgery (Table S3) (Table S3).

Discussion
This is the first study evaluating protein glycoxidation, protein oxidative damage and nitrosative/nitrative stress in morbidly obese patients before and after (one, three, six and twelve months) laparoscopic sleeve gastrectomy. We demonstrated significant increases in plasma glyco-oxidative products, protein oxidation and nitrosative/nitrative stress in morbidly obese subjects as compared to lean controls. Noteworthily, in patients with metabolic syndrome (MS+), we showed no differences in most redox biomarkers, as compared to morbidly obese patients without MS (MS−). However, two markers were able to differentiate MS+ and MS− with high specificity and sensitivity: peroxy-nitrite (>70%) and S-nitrosothiols (>60%). After bariatric surgery physiological redox homeostasis was markedly improved.
The main aim of obesity surgery is not to lose weight to achieve an aesthetic effect, but to improve the metabolic state of obese patients, prolong their live and improve its quality. Bariatric treatment maintains long-term weight loss and reduces the risk of several diseases such as hypertension, T2DM or cardiovascular disorders [18,36]. In this study, we showed that bariatric surgery resulted in the reduction of class 3 obesity to class 1, 12 months after procedure. Along with weight loss in morbidly obese patients, we also observed an improvement in carbohydrate and lipid metabolism.
The results of recent studies indicate the key role of nitrosative/nitrative stress in the development of morbid obesity. It has been shown that the progression of metabolic disturbances accompanying obesity is paralleled by increase in myeloperoxidase activity, NO formation as well as protein nitrosative damage [37]. Importantly, nitrosative stress is linked to endothelial dysfunction and thus excessive platelet aggregation or induction of atherogenesis [38,39]. However, it is still unclear whether bariatric surgery reduces/eliminates nitrosative stress to the level observed in lean control. Due to the short half-life, direct analysis of RNS quantities is virtually impossible. Thus, the evaluation of oxidative/nitrosative stress is more often based on compounds formed by the reaction of ROS/RNS with cellular components. Markers of protein oxidative damage are among the best because proteins are present in every cell. The products of their oxidation are relatively stable and easy to analyze [40].
We have shown that glycoxidation rate (↑dityrosine, ↑kynurenine, ↑N-formyl-kynurenine, ↑amyloid, ↑Amadori products and ↑glycophore) as well as protein oxidative damage (↑IMA) were significantly higher in the preoperative plasma of obese patients as compared to the control group. This fact is not surprising, because in morbidly obese patients many metabolic pathways such as sorbitol, polyol, hexosamine and protein kinase C are activated [41,42]. Under these conditions, glucose autoxidation also occurs, resulting in the formation of reactive ketoaldehyde, which further enhances non-enzymatic glycation [10,43]. Glycoxidation products (especially AGE, advanced glycation end products) can also activate the pro-inflammatory pathway NF-KB, which increases the expression of pro-inflammatory cytokines and chemokines, growth factors, adhesion molecules, and nitric oxide synthases (mainly iNOS) [15,44]. In our study, glycation and oxidation of proteins generally normalize after bariatric surgery; nevertheless, the fluorescence of N-formyl-kynurenine is significantly increased, while the tryptophan content is statistically reduced even one year after surgery. This condition undoubtedly results from the improvement of metabolic parameters after the bariatric treatment. Indeed, in patients after surgery, a better lipid and carbohydrate profile is observed. Nevertheless, the decrease in blood glucose level deserves special attention. Although protein glycation occurs very slowly in vivo, it intensifies dramatically under the hyperglycemic conditions [10,45]. Interestingly, the content of dityrosine correlates negatively with weight loss at the end of the experiment.
Until twelve months after laparoscopic sleeve gastrectomy we also observed a diminished MPO activity; however, this decline is not at the same level of the control group. A slight decrease in MPO expression may be due to the fact that, despite enormous weight loss, patients from the study group were still obese. It has been demonstrated that MPO takes part in the peroxidation of LDL cholesterol, which is embedded in atherosclerotic plaques of blood vessels [46]. MPO also initiates/promotes acute and chronic inflammation, as well as atherosclerosis by the production of pro-oxidants such as HOCl and tyrosyl radicals [47].
In the OB group we also found enhanced NO level which dynamically reduced after the bariatric treatment. Interestingly, total NO concentration was also significantly higher in MS 6 + subgroup than MS 6 -. Although the bioavailability of NO has been shown to be reduced both in the animal model of obesity and diabetes [48], as well as in obese and T2DM patients [49,50], it is worth noting that obesity is an inflammatory disease. Indeed, NO generated by iNOS (inducible nitric oxide synthase) plays a crucial role in inflammatory processes and has the highest capacity to produce NO [51]. Therefore, in morbidly obese patients, chronic inflammation may lead to the increased formation of nitric oxide. NO is also a substrate for peroxy-nitrite formation. ONOO − is a strong oxidant with harmful effect on the cellular biomolecules (e.g., amino acids, lipids and nucleic acids) [52]. Nevertheless, the main target of its action are proteins. Indeed, ONOO − oxidizes thiol groups and tyrosine in signaling proteins/enzymes as well as in proteins involved in energy cell metabolism [53]. In our study, we found greater concentration of ONOO − and nitro-tyrosine in morbidly obese patients before laparoscopic gastrectomy as compared to lean controls. Nevertheless, their levels are gradually reduced after bariatric treatment. Before surgery, we also observed an increase in plasma S-nitrosothiols concentration. This may be an adaptive reaction to the overproduction of RNS in obese patients. Indeed, in biological systems, S-nitrosothiols are involved in transport and cellular storage of nitric oxide [54]. S-nitrosothiols also protect cells against the toxic effects of NO 2 , NO + , NOand ONOO - [54]. Therefore, it is not surprising, that along with the normalization of protein nitrosative/nitrative damage, the concentration of S-nitrosothiols returned to the values found in the lean controls.
Obesity is a disease with complex etiology in which redox imbalance seems to play a special role [55]. Although we are unable to determine whether oxidative stress is the cause or effect of metabolic complications in obesity, this process can exacerbate inflammation as well as disturb the lipid and carbohydrate metabolism. However, we did not show any differences between patients with obesity and metabolic syndrome from patients with obesity alone. The lack of change may result from a small study group, but also from the complexity of metabolic disturbances in obese patients. Additionally, it should be remembered that in our study we measured only circulating redox biomarkers. The evaluation of plasma oxidation/glycation of proteins and nitrosative/nitrative stress may not necessarily reflect metabolic changes in target organs such as adipose tissue, liver or skeletal muscles [56]. However, before the surgery, peroxy-nitrite content was significantly higher in the OB + MS subgroup compared to OB. Similar changes have been observed for S-nitrosothiols. Using the ROC analysis, we have demonstrated the diagnostic usefulness of these parameters in differentiating metabolic disturbances in the course of obesity. Indeed, plasma ONOO − and S-nitrosothiols with high sensitivity and specificity distinguish obese patients from those with OB and MS. The peroxy-nitrite content was also positively correlated with insulin, uric acid and CRP levels. Although the correlations observed are relatively poor, they may suggest the contribution of peroxy-nitrite and S-nitrosothiols in the progression of metabolic changes in obesity. Nowadays, redox biomarkers are increasingly used in differential diagnosis/prediction of the severity of many diseases such as hypertension [9], diabetes [57], chronic heart disease [7], chronic kidney disease [58] and stroke [59]. Nevertheless, in order to confirm the diagnostic usefulness of redox biomarkers in morbid obesity patients, long term observations on a larger number of cases are necessary.
In our work, we did not compare the effect of different techniques of bariatric surgery on the oxidative/nitrosative stress of obese patients. Laparoscopic sleeve gastrectomy (LSG) reduces gastric capacity, which leads to lower intake and absorption of food and decreased secretion of ghrelin. Although laparoscopic sleeve gastrectomy and laparoscopic adjustable gastric banding (RYGBP) are the most standard surgical procedures, biliopancreatic diversion (BPD) and mini gastric bypass (mini-GBP) are among the most effective but also have more side effects [60,61]. In future studies, due to the different mechanism of weight loss, it is necessary to compare various bariatric procedures on cellular redox homeostasis of obese patients.
Despite the undoubted advantages, our study also has some limitations. We assessed only selected biomarkers of glycoxidation and nitrosative/nitrative stress, so we cannot fully characterize the effect of bariatric surgery on redox homeostasis in morbidly obese patients. Furthermore, we are not able to determine the source of NO formation and we have not evaluated the expression of proinflammatory cytokines, which are inseparable from the oxidation and glycation of proteins [62]. Finally, we are not able to eliminate the influence of drugs and diet on the assessed redox biomarkers.
In conclusion, we have shown that protein glycoxidation and nitrosative/nitrative stress is markedly deranged in obese patients. Although oxidative/nitrosative stress contributes to the progression of metabolic changes in obesity [63,64], oxidation and glycoxidation of plasma proteins did not differ between MS+ and MS. Within the broad spectrum of studied parameters, only peroxy-nitrite and S-nitrosothiols differentiated morbidly obese patients with metabolic syndrome from those without. Bariatric surgery generally reduces the abnormalities in the oxidation and glycation of proteins as well as nitrosative/nitrative stress. Further studies are required to confirm the diagnostic usefulness of such biomarkers.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-3921/9/11/1087/s1. Table S1: Clinical characteristics of the control (C) morbid obesity without metabolic syndrome (MS−) and morbid obesity with metabolic syndrome (MS+); Table S2: Correlations between the analyzed oxidative/nitrosative stress and clinical parameters in patients with morbid obesity (OB) at the beginning of the study; Table S3: Correlations between the analyzed oxidative/nitrosative stress and clinical parameters in patients with morbid obesity (OB) at the end of the study.

Conflicts of Interest:
The authors declare no conflict of interest.