Association Between Parameters of Cortisol Metabolism, Biomarkers of Minerals (Zinc, Selenium, and Magnesium), and Insulin Resistance and Oxidative Stress in Women with Obesity

This is a cross-sectional study with women divided into a group of those with obesity (n = 80) and a control group (n = 94). Statistical analysis was conducted using the SPSS program. There were high values of GPx and TBARS and reduced values of SOD in women with obesity compared to the control group. Obese women showed increased concentrations of cortisol in serum and urine as well as hypozincemia, hyposelenemia, and hypomagnesemia and increased urinary excretion of these minerals. There was a negative correlation between the cortisol/cortisone ratio and erythrocyte zinc and selenium concentrations and a significant positive correlation between GPx and SOD activity and erythrocyte and plasma concentrations of zinc and selenium. The results of the study suggest the influence of adiposity on the increase in cortisol concentrations and the role of this hormone in the compartmentalization of the minerals zinc, selenium, and magnesium. However, the association study does not allow identifying the impact of such action on the antioxidant defense system and insulin sensitivity.


Introduction
Obesity is characterized by an excessive accumulation of body fat resulting from the imbalance between energy intake and energy expenditure, with a complex and multifactorial etiology involving genetic and environmental factors that can compromise health and increase morbidity and mortality [1].This disease contributes to an increased risk of developing other important chronic diseases such as type 2 diabetes mellitus, cardiovascular disease, and some types of cancer [2].
Adipose tissue dysfunction is the pathophysiological basis of obesity and is characterized by adipocyte hypertrophy, reduced angiogenesis, local hypoxia, fibrosis, chronic low-grade inflammation, and important hormonal disorders such as changes in the metabolism of cortisol, a glucocorticoid with compromised secretion and sensitivity in individuals with obesity [3][4][5].
In this regard, literature has demonstrated the relevance of deregulation of the hypothalamic-pituitary-adrenal (HPA) axis in the presence of adipose tissue dysfunction as a contributing factor to accentuate the synthesis and secretion of cortisol, which consequently favors the increase of the risk for the development of metabolic disorders associated with obesity [6].It is noteworthy that cortisol acts as an important regulator of endocrine function, metabolism, and adipocyte differentiation, contributing to adipogenesis and increased visceral fat stores [7].Furthermore, this hormone is also involved in the central response to stress and indirectly influences neuroprotection mechanisms [8].
Moreover, studies have demonstrated the influence of cortisol on the metabolism of micronutrients, in particular the minerals zinc, selenium, and magnesium, with some mechanisms already substantiated.Molecular-based research reveals the participation of cortisol in increasing the expression of genes coding for metallothionein and the zinc transporter protein ZIP-14, proteins that accumulate zinc in specific tissues such as the liver and adipose tissues, favoring the manifestation of hypozincemia in individuals with obesity [5,9].
Cortisol also influences the homeostasis of magnesium and selenium by favoring the manifestation of oxidative stress as well as by altering the activity and expression of enzymes of the antioxidant defense system.Oxidative stress, exacerbated by the presence of cortisol, in turn, favors the depletion of components of the non-enzymatic antioxidant defense system, such as selenium and magnesium.Thus, the reduction in the concentrations of these nutrients in the serum of obese individuals may compromise their functions in lipid metabolism, insulin action, and antioxidant activity [8,[10][11][12][13][14].
In this perspective, although the influence of adipose tissue dysfunction on cortisol metabolism has already been demonstrated, with an important impact on the homeostasis of magnesium, zinc, and selenium, there is still a gap in the literature with robust data that can substantiate the effect of such alterations in manifestation of insulin resistance and oxidative stress in obesity.Thus, this study aimed to verify the relationship between parameters of cortisol metabolism, mineral biomarkers (zinc, selenium, and magnesium), and insulin resistance and oxidative stress in women with obesity.

Characterization of the Study and Experimental Protocol
This is a cross-sectional study involving 174 adult women aged between 20 and 50 years who were divided into two distinct groups according to their body mass index (BMI): obesity group (n = 80) consisting of women with BMI ≥ 35.00 kg/m 2 and the eutrophic group (n = 94) formed by women with BMI between 18.50 and 24.99 kg/m 2 .The recruitment of study participants took place from 2018 to 2020 according to the demand for care in health institutions and through informed consent.
The selection of volunteers was based on the following eligibility criteria: • Present eutrophic or diagnosis of grade II or III obesity.
• Not being a smoker or ex-smoker.

Assessment of Anthropometric Parameters
The anthropometric assessment was performed by measuring body weight, height, and waist and hip circumferences according to established technical procedures [15,16].To assess the nutritional status of study participants, BMI was calculated from the division between the participant's weight and height squared [17].In addition, waist/hip was calculated [16].

Assessment of Food Consumption
The amount of energy, macronutrients, and magnesium, zinc, and selenium consumed by the study participants was obtained by analyzing the 3-day food record using the Nutwin program (version 1.5) based on data from the Brazilian Food Composition Table [18].The analysis of the amounts of minerals magnesium, zinc, and selenium was conducted based on the insertion of data from the study by Ferreira [19], the Food Composition The reference values used for minerals were: 255 mg of magnesium/day for women aged 19 to 30 years and 265 mg of magnesium/day for those over 30 years [22], 45 μg of selenium/day [23], and 6.8 mg of zinc/day [24].Intake of energy, magnesium, zinc, and selenium macronutrients was corrected for intra-and interpersonal variations.In addition, dietary intake of these minerals was also adjusted for energy intake.
The values of energy, macronutrients, and minerals intake were entered into the multiple source method (MSM) online platform version 1.0 for intrapersonal and interpersonal variability adjustments, corrected by statistical modeling techniques, as well as for estimating the usual dietary intake of energy and these nutrients through logistic regression analysis.This program can be accessed through the MSM website (https:// nugo.dife.de/ msm/) [25][26][27][28].
The dietary values of macronutrients and minerals were also adjusted for energy using the residual method, avoiding distortions generated by differences in energy intake.After verifying the normality of data distribution, intake values were adjusted in relation to energy by calculating the nutrient [29][30][31].

Collection and Processing of Biological Material
Blood collection was performed in the morning between 7 and 9 am, with the participants fasting for at least 12 h, and 20 mL of blood was collected, which was distributed in different tubes: (1) vacuum tube containing ethylenediaminetetraacetic acid (EDTA) for analysis of selenium, markers of oxidative stress, and glycated hemoglobin (8 mL), (2) vacuum tube with clot activator for analysis of parameters of glycemic control and lipid profile (4 mL), (3) dry tube for analysis of cortisol metabolism markers (4 mL), and (4) tube containing citrate for analysis of zinc and magnesium.

Separation of Blood Components
To separate the plasma and erythrocyte mass, the tubes containing EDTA and citrate were centrifuged at 1764 × g for 15 min at 4 °C and then stored at -20 °C.The erythrocyte mass resulting from the centrifuge process of the tubes was washed with 3.5 mL of isotonic saline solution (0.9% NaCl), being carefully homogenized by inversion, and subsequently centrifuged at 2401 × g for 10 min.This procedure was repeated three times to remove contaminants from the erythrocytes (platelets and leukocytes).After the last centrifugation, the saline solution was aspirated and discarded, and the erythrocyte mass was carefully extracted with the aid of an automatic pipette and transferred to demineralized microtubes for analysis of erythrocyte magnesium, zinc, selenium, and hemoglobin.The dried tubes with clot activator were centrifuged at 1764 × g for 15 min at 4 °C to obtain the serum, which was stored at -80 °C for analysis of the other parameters.
The 24-h urine was collected in a previously demineralized amber plastic bottle.After homogenizing the total urine content, a 4-mL aliquot was taken and transferred to properly demineralized microtubes and kept in a freezer at − 20 °C for later analysis of selenium and zinc.A 4-mL aliquot was transferred to microtubes and stored in a freezer at − 80 °C for later analysis of urinary markers of cortisol metabolism.The remaining volume was acidified with 3 mol/L hydrochloric acid and homogenized to avoid magnesium precipitation and to better preserve the mineral in the urine [32].Subsequently, 4 mL of urine was removed and distributed into two microtubes, which were kept in a freezer at − 20 °C.
To calculate the urinary volume, we considered the density (1.015 g/mL) and the mass of the urine (difference between the weight of the bottle before and after the 24-h urine collection, measured on a semi-analytical scale).

Evaluation of the Biochemical Parameters of Magnesium, Selenium, and Zinc
Analyses of plasma, erythrocyte, and urinary concentrations of minerals were carried out at the Laboratory of Atomic Emission Spectrometry-Embrapa (National Center for Research on Maize and Sorghum, SeteLagoas-MG).The elemental analysis of minerals was carried out in an inductively coupled plasma spectrometer-optical emission spectrometry with an axial view configuration and a V-Groove nebulizer (720 ICP/OES, Varian Inc., CA, USA).The adopted reference values are shown in Table 1.

Determination of Glycemic Control Parameters
Fasting plasma glucose concentrations were determined using the enzymatic colorimetric method, with values equal to or greater than 125 mg/dL being considered indicative of diabetes according to the criteria defined by the American Diabetes Association [41].
Serum insulin concentrations were analyzed using the chemiluminescence method, using values between 6 and 27 μU/mL as a reference standard.Glycated hemoglobin (HbA1C) was analyzed using the ion exchange chromatography method, and values equal to or greater than 6.5% were adopted as the cutoff point for the diagnosis of diabetes [4].
To assess insulin resistance, the updated version of the homeostasis model assessment for insulin resistance (HOMA) was used, calculated from fasting glucose and fasting insulin concentrations.HOMA-IR values greater than 2.71 mean insulin resistance [41,42].

Determination of Antioxidant Enzyme Activity
The activity of the enzymes glutathione peroxidase and superoxide dismutase in erythrocytes was evaluated in an automatic biochemical analyzer (Labmax 240 model; Lagoa Santa, MG, Brazil) using the Ransel 505 kit (Randox Laboratory, Crumlin, UK) and according to the methodology proposed by Paglia and Valentine [43] and the Ransod kit (Randox Laboratory, Crumlin, UK), respectively, according to the methodology recommended by the manufacturer.Hemolysate hemoglobin concentration was determined to express enzyme activity in units of enzyme per gram of hemoglobin (U/gHg).The reference values proposed by Randox for glutathione peroxidase activity (27.5 to 73.6 U/ gHg) and superoxide dismutase (1102 to 1601 U/gHg) were used.
Catalase activity was determined in a Bel Photonics UV/ Vis spectrophotometer (model SF200DM; Osasco, SP, Brazil) at 240 nm according to the method proposed by Aebi [44], which is based on the decay of absorbance resulting from the reduction of H 2 O 2 to water by the catalase present in the sample measured during a time of 30 s.For the reaction medium, a solution of hydrogen peroxide (30 mM) in 50 mM sodium and potassium phosphate buffer (pH 7.0) was used.In a quartz cuvette, 1 mL of H 2 O 2 solution and 2 mL of sample dilution were added.Enzyme activity was expressed in mmol H 2 O 2 consumed per minute and gram per gram of hemoglobin (mmol.min−1 .g−1 /gHb).

Determination of Plasma Concentrations of Substance Reactive to Thiobarbituric Acid
The determination of plasma concentrations of thiobarbituric acid-reactive substances (TBARS) was performed according to the method proposed by Ohkawa, Ohishi, and Yagi [45].The calibration curve was prepared using concentrations of 0.5 to 8.0 nmol/mL of the standard reagent 1,1,3,3-tetraethoxypropran (Sigma-Aldrich®).Absorbance was measured at a wavelength of 532 nm using a Bel Photonics UV/Vis spectrophotometer (model 1102; Osasco, SP, Brazil).

Serum Cortisol
Serum cortisol concentrations were determined using the chemiluminescence method with values from 6.70 to 22.60 μg/dL as a reference standard for the morning period [46].

Serum Adrenocorticotropic Hormone (ACTH)
Serum ACTH concentrations were determined using the chemiluminescence method, with values below 46.0 pg/mL being the reference standard.

Cortisol-Binding Globulin (CBG) Serum
CGB concentrations in serum were determined using the radioimmunoassay method, with values ranging from 40.0 to 154.0 μg/mL as a reference standard.Urinary Cortisol, Cortisone, Tetrahydrocortisol (THF), and Tetrahydrocortisone (THE) Urinary concentrations of cortisol, cortisone, tetrahydrocortisol (THF), and tetrahydrocortisone (THE) were analyzed in 24-h urine samples by isotope dilution liquid chromatography and tandem mass spectrometry, with the following reference values adopted: 3.0 to 43.0 μg/24 h for urinary cortisol, 5 to 122 μg/24 h for urinary cortisone, 0.5 to 1.5 mg/24 h for THF, and 1.2 to 3.5 mg/24 h.

Enzymatic Activity of 11β-Hydroxysteroid Dehydrogenase 2 Enzyme (11β-HSD2)
The activity of the 11β-HSD 2 enzyme was calculated using the ratio of urinary metabolites of cortisol and cortisone (THF/THE), considering that this ratio reflects the enzymatic activity of 11β-HSD type 2 in the kidneys, adopting the reference value 0 0.11 to 0.67 μg/24 h, as shown in Table 2.

Enzymatic Activity of 5β-Reductase
The activity of the 5β-reductase enzyme can be inferred from the THF/cortisol and THE/cortisone ratio, with the proportions shown in Table 2.

Statistical analysis
Considering the number of missing data in the database, 23% in the control group, and 3.6% in the group of women with obesity, it was decided to carry out a data imputation process with the objective of "completing" the bank and enabling the analysis with all individuals in the study.Single imputation was performed using the predictive mean matching (PMM) method, which is indicated for quantitative variables.It was conducted in three stages [47] using the function created by [48]: (1) estimation of a regression model where the variable of interest (to be imputed) was the response variable and the remaining variables collected were the explanatory ones, (2) estimation of the value of the variable of interest for subjects with missing data, and (3) pairing of the value of the predicted variable of interest for subjects with missing data, with the closest adjusted value (based on the calculation of the Euclidean distance).In cases where there was more than one adjusted value with a distance equal to the minimum distance found, the value to be imputed was chosen randomly among those that suffered a tie.The process was carried out in the R program (R Development Core Team, 2019).
The data were organized in Microsoft Excel® spreadsheets to carry out the descriptive analysis of the observed variables and were subsequently exported to the SPSS program (version 25.0) and GraphPad Prism (version 8.0) for statistical analysis of the results.Continuous variables were expressed as mean ± SD.
The assumption of normality of variables was assessed using the Kolmogorov − Smirnov test.Then, for purposes of comparison between groups, Student's t-test was used for continuous variables with normal distribution and the Mann-Whitney test for those with non-parametric distribution.For the study of correlations, Pearson's or Spearman's correlation coefficient was used for variables with parametric and non-parametric distribution, respectively.Tests were considered significant when p-values < 0.05.
Kernel-generalized canonical correlation analysis (KGCCA) was used according to the methodology of Tenenhaus, Philippe and Frouin [49] to better understand the interrelationships between the blocks of variables, being a method that combines non-linearity and multigroup analysis.For this analysis, all variables were standardized (mean zero and variance one) to make the blocks comparable, and a possible strategy is to standardize the variables and then divide each block by the square root of its number of variables [50].
In this study, four domains of variables were analyzed, which were named cortisol metabolism, minerals, glycemic control, and oxidative stress.The present study was not only to define the link between the four domains above but also to reveal the variables that participate most in each domain under the hypothesis that cortisol metabolism influences the deficiency of minerals and that this deficiency can impact in some way on glycemic control and oxidative stress.To apply the KGCCA method, the centroid scheme function and the fully connected design matrix from the cited hypothesis were used.
Kernel canonical correlation analysis (KCCA) was also used to determine the correlation between domains, and the kernel function chosen was the Gaussian one.The nonlinear extension of KGCCA and KCCA makes the model more complex, so it is not possible to formulate test statistics based on the unknown nonlinear function.Thus, we consider a permutation test approach for determining p-values, as performed in Bae et al. (2020).Furthermore, we set the number of samples for the permutation test at 1000.The RGCCA and kernlab packages of the R version 4.2.2 software [51] were used to process these analyses.

Anthropometric Parameters for the Assessment of Nutritional Status
Mean values and SD for age and anthropometric parameters used in the assessment of the nutritional status of the study participants are shown in Table 1.It is observed that there was a statistical difference between the anthropometric parameters evaluated (p < 0.05).The study did not reveal a statistically significant difference (p > 0.05) regarding the parameters of glycemic control between women with obesity and the control group.In addition, there were high levels of GPX and TBARS (p < 0.05) and reduced levels of SOD (p < 0.05) in women with obesity when compared to the control group.The results show a statistically significant difference (p < 0.05) in relation to magnesium and selenium intake between the evaluated groups (Table 3).
Table 4 presents the mean values and standard deviations of cortisol metabolism markers in women with obesity and the control group.It was found that there was a statistically significant difference (p < 0.05) between the evaluated parameters.
Obese women had reduced (p < 0.05) plasma and erythrocyte concentrations of magnesium, selenium, and zinc, as well as high values of these minerals in the urine compared to the control group, as shown in Table 5.
The simple linear correlation analysis between the adiposity parameters and markers of cortisol metabolism, serum, and urinary cortisol of the study participants is shown in Fig. 1.The results reveal a significant positive correlation (p < 0.05) between all the anthropometric parameters evaluated and serum and urinary cortisol.
The simple linear correlation analysis between markers of cortisol metabolism and plasma, erythrocyte, and urinary concentrations of minerals is shown in Table 6.There was a positive correlation (p < 0.05) between urinary concentrations of tetrahydrocortisone (THE) with plasma zinc and plasma selenium, and urinary magnesium.The data also showed a negative correlation (p < 0.05) between the cortisol/cortisone ratio and erythrocyte concentrations of zinc and selenium.In the control group, there was a negative correlation (p < 0.05) between urinary concentrations of cortisol and erythrocyte selenium, a negative correlation (p < 0.05) between urinary concentrations of cortisol and magnesium in the urine, as well as a negative correlation (p < 0.05) between urinary concentrations of THF and zinc and erythrocyte selenium.Figure 2 presents the significant results of the simple linear correlation analysis between magnesium, zinc, and selenium, and oxidative stress markers of the study participants.The results reveal a significant positive correlation (p < 0.05) between the activity of the enzyme glutathione peroxidase, selenium, and zinc in the erythrocytes, a positive correlation (p < 0.05) between the activity of the enzyme superoxide dismutase and the plasmatic and erythrocyte zinc concentrations, as well as a negative correlation (p < 0.05) between the concentrations of TBARS and the concentrations of zinc and selenium in the erythrocytes of women with obesity.
Figure 3 presents the significant results of the simple linear correlation analysis between magnesium, zinc, and selenium and markers of glycemic control of the study participants.The results demonstrate a positive correlation (p < 0.05) between serum insulin concentrations, HOMA-IR, and urinary zinc concentrations in obese women.The data also reveal the existence of a negative correlation (p < 0.05) between serum concentrations of insulin, HOMA-IR, and urinary concentrations of selenium in participants in the control group, as well as a positive correlation (p < 0.05) between urinary concentrations of zinc and fasting glucose in these participants.
Figure 4 shows the correlation between blocks to define the link between the four domains: cortisol metabolism, minerals, glycemic control, and oxidative stress and to reveal the variables that most participate in each domain under the hypothesis that cortisol metabolism influences the deficiency of minerals and that this deficiency can impact in some way form in glycemic control and oxidative stress, being therefore important for a better understanding of the relationship between such variables.
Correlations between blocks of variables obtained through canonical analysis did not reveal a significant (p > 0.05) relationship between cortisol metabolism and markers of nutritional status related to minerals, and even between parameters of minerals and oxidative stress.
Although the correlation between markers of cortisol metabolism and mineral nutritional status was also not significant (p > 0.05), it was found that tetrahydrocortisone is the variable in the group that significantly contributes to a possible relationship between cortisol metabolism and minerals.However, there was a significant correlation between mineral parameters and glycemic control (r = 0.949, p = 0.024), highlighting that magnesium in the diet significantly contributes to the relationship between the mineral group and the other groups.

Discussion
In the present study, markers of cortisol metabolism, concentrations of minerals magnesium, zinc, and selenium were evaluated, as well as the association between these parameters and the index of insulin resistance and markers of oxidative stress in obese women.
The evaluation of cortisol metabolism parameters revealed a statistically significant difference between all parameters between the evaluated groups except for ACTH.The results show high concentrations of cortisol in serum and urine, as well as reveals a reduction in urinary cortisone values; this fact demonstrates the occurrence of alterations in the HPA axis in women with obesity probably due to Fig. 1 Simple linear correlation analysis between the adiposity parameters and serum and urinary cortisol in all participants.*Pearson or Spearman linear correlation coefficient (p < 0.05).BMI, body mass index; HC, hip circumference; WC, waist circumference the inefficiency of serum cortisol in inhibiting the releasing hormone of corticotropin (CRH) and adrenocorticotropin (ACTH), substances that induce cortisol secretion [3,52].
Associated with this, this result can also be ratified by the fact that the study revealed an increase in the THF/THE and cortisol/cortisone and cortisol/THF ratios in women with obesity, which reflect a reduction in the activity of the 11β-HSD type 2 enzyme, demonstrating impairment in the inactivation of cortisol by the action of this enzyme, which reinforces a possible compromise in the negative feedback mechanism of the HPA and, consequently, changes in cortisol metabolism [53].
Data from this study demonstrate a positive correlation between adiposity parameters and markers of cortisol metabolism, which may suggest the influence of adipose tissue dysfunction, characterized by the presence of chronic low-grade inflammation and oxidative stress, important contributing factors to the manifestation of changes in cortisol metabolism.
With a view to evaluating the probable influence of cortisol on the homeostasis of the minerals zinc, magnesium, and selenium in the study participants, a correlation analysis was carried out which revealed a negative result between the concentrations of cortisol in the urine and the values of zinc found in the plasma and erythrocytes.Furthermore, a negative correlation was found between the cortisol/cortisone ratio and erythrocyte zinc concentrations.
In this regard, the fact that high concentrations of cortisol favor the expression of genes coding for metallothionein and zinc transporter protein ZIP-14 is probably the factor that contributed to the manifestation of hypozincemia observed in obese women evaluated in this study.
The present study also demonstrated a negative correlation between serum cortisol concentrations and the cortisol/ cortisone ratio and selenium values in erythrocytes, as well as a positive correlation between urinary concentrations of cortisol and selenium in the urine.This result reinforces the influence of cortisol on the distribution of selenium in the evaluated biological compartments, which is due to the action of cortisol as a chronic stressor in the body, which favors an increase in the demand for antioxidant nutrients such as selenium.
The data from the correlation analysis showed a positive result between serum and urinary cortisol and the concentration of magnesium in the urine, suggesting the role of cortisol, though indirectly, in the excretion of this mineral in obese individuals.In this regard, it is noteworthy that higher cortisol concentrations, as occurs in the presence of adipose tissue dysfunction, may constitute an important contributing factor to increase urinary magnesium excretion since this hormone favors the excessive production of reactive species of oxygen, contributing to accentuate the condition of hypomagnesemia in women with obesity evaluated in this study.
The literature demonstrates that these minerals can also influence cortisol metabolism.The hypomagnesemia, for example, contributes to increase in the concentrations of cortisol and its metabolites in the body, and this fact may contribute to justify the high values of these hormones in participants with obesity.It is to intensify that in situations of magnesium concentration, this mineral favors the reduction and release of ACTH, although indirectly, which consequently reduces the reduction of cortisol [12].
Regarding zinc, studies show that both high and low concentrations of this mineral in serum promote changes in adrenal secretion [54].In the study by Chen et al. [55], it was found that this mineral inhibits the binding of glucocorticoids to its receptor as there is a zinc binding region in the receptor for these hormones, reducing the effects of cortisol in the body.On the other hand, it has been shown that zinc deficiency in the diet may be a contributing factor to increased cortisol secretion since the reduction in the Fig. 2 Simple linear correlation analysis between magnesium, zinc, and selenium and oxidative stress markers of study participants.*Pearson or Spearman linear correlation coefficient (p < 0.05).GPX, glutathione peroxidase; SOD, superoxide dismutase; TBARS, substances reactive to thiobarbituric acid; Zn, zinc; Se, selenium; Mg, magnesium serum concentration of this nutrient increases the activity of the hypothalamic-pituitary-adrenal axis, followed by an increase in secretion of this glucocorticoid from the adrenal cortex [56][57][58].
Selenium deficiency, in turn, can contribute to changes in the adrenal gland, which limits the synthesis of cortisol.This fact occurs as result of the deficiency of this mineral, compromising GPx activity, accentuating oxidative stress in the adrenal gland [59].
In this discussion, it is highlighted that the concentrations of minerals magnesium, selenium, and zinc in blood components were reduced in women with obesity, which possibly favored the excessive production of reactive species, as verified in the data obtained in this study.The results of this study reveal the existence of a positive correlation between erythrocyte selenium concentrations and the activity of the antioxidant enzyme superoxide dismutase.In addition, a positive correlation between selenium concentrations in plasma and the activity of the enzyme glutathione peroxidase was also verified, which demonstrates the important role of selenium as an active center of the enzyme glutathione peroxidase.Thus, selenium deficiency alters the activity of these important antioxidant enzymes, a fact confirmed in the associations identified in the present study.The positive correlation between plasma and erythrocyte zinc concentrations and the activity of the antioxidant enzymes superoxide dismutase and glutathione peroxidase demonstrates the importance of adequate values of this micronutrient in maintaining the antioxidant defense system.Zinc is part of the superoxide dismutase structure and regulates the activation of erythroid nuclear factor associated with factor 2 (Nrf2), which in turn stimulates the expression of enzymes of the antioxidant defense system, justifying the possible influence of this nutrient on antioxidants enzymes evaluated [60].
Deficiency in zinc in obese individuals seems to influence glycemic control parameters.In this study, a significant positive correlation was found between urinary zinc excretion and fasting insulin and HOMA-IR values in women with obesity.On this data, the reduction of zinc in the blood components due to the higher urinary excretion may have limited the glycemic control of the participants since this mineral participates in the secretion of insulin and maintenance of the glycemic homeostasis [3][4][5][60][61][62][63]. Zinc deficiency contributes to insulin resistance by reducing insulin receptor β-subunit phosphorylation and activation of PI3K and Akt proteins.Thus, the deficiency in the mineral reduces the transport of glucose into the cells, favoring hyperglycemia [7,64,65].
Although the association study between magnesium, selenium, and insulin resistance did not reveal a significant result in the evaluated obese women, it is important to highlight that magnesium deficiency also reduces the autophosphorylation of the β subunit of the insulin receptor and of the AKT and PI3K enzymes, compromising the translocation of GLUT-4 to the plasma membrane, favoring hyperglycemia [9,66].In addition, selenium is a mineral with an important role in the expression of the glucagon-like peptide receptor in β cells, favoring the action of this peptide in stimulating proinsulin gene expression in response to carbohydrate Fig. 4 Diagram of the canonical correlation between markers of cortisol metabolism, minerals, and markers of glycemic control and oxidative stress of study participants.ACTH, adrenocorticotropic hormone; CBG, cortisol-binding globulin; CAT, catalase; GPX, glutathione peroxidase; HOMA-IR, homeostasis model assessment insu-lin resistance; HbA1C, glycated hemoglobin; Mg, magnesium; Se, selenium; SOD, superoxide dismutase; TBARS, substances reactive to thiobarbituric acid; THF, tetrahydrocortisol; THE, tetrahydrocortisone; Zn, zinc ingestion and in the regulation of phosphorylation of substrates of the insulin signaling cascade [10,67].Thus, selenium deficiency can also favor insulin resistance.
It is noteworthy that although obese women consume high amounts of magnesium and selenium when compared to the control group, the data from this study show that this fact does not seem to have influenced the blood concentrations of these minerals since these were reduced.Another interesting aspect deals with the fact that the study did not reveal the possible influence of such reduced values of these minerals on insulin resistance markers since no significant association was evidenced.
The results of this study make it possible to identify the influence of adiposity on cortisol metabolism, the impact of this hormone on zinc, selenium, and magnesium homeostasis.However, data from this research do not suggest that the influence of cortisol on the distribution of analyzed minerals is capable of potentiating insulin resistance and oxidative stress present in obesity.In this discussion, the contribution of important factors that may have influenced such results, such as the adequate glycemic control of women with obesity, is highlighted.In addition, other factors besides mineral deficiency are also important for the manifestation of oxidative stress in this population group.
It should be noted that the assessment of the 24-h urine cortisol concentration and excretion of cortisone and its metabolites, in particular the determination of the enzymatic activity of the intracellular 11β-HSD enzymes and A-ring reductases, which allows the assessment of underlying mechanisms, constitutes relevant markers in the analysis of cortisol metabolism, allowing to obtain important information about its intracellular regulation.
Some limitations of this study can be highlighted, for example, the cross-sectional nature of the study, the lack of data on selenoprotein P levels, and concentrations of inflammatory parameters, as well as molecular markers that allowed a better discussion of the results found in this study.Additionally, the use of other parameters to evaluate oxidative stress in this study could have contributed to a better understanding of the action of minerals in the reduction of reactive oxygen species present in obesity.

Conclusions
The obese women evaluated in this study have disorders in the metabolism of cortisol, characterized by increased serum and urinary concentrations and reduced urinary cortisone and 11β-HSD 2 enzyme activity.The study demonstrates the presence of alterations in the homeostasis of the evaluated minerals, such as hypozincemia, hyposelenemia, and hypomagnesemia.
In addition, obese women also have elevated concentrations of TBARS, increased GPX activity, and reduced SOD, suggesting the presence of oxidative stress.
The results of the study suggest the influence of adiposity on the increase in cortisol concentrations, the possible role of this hormone in concentrations of the minerals zinc, selenium, and magnesium in obese woman.However, the association study does not allow identifying the impact of such action on the antioxidant defense system and insulin sensitivity.

Fig. 3
Fig.3 Simple linear correlation analysis between minerals zinc and selenium and markers of glycemic control of study participants.*Pearson or Spearman linear correlation coefficient (p < 0.05).HOMA-IR, homeostasis model assessment insulin resistance; HbA1C, glycated hemoglobin; Se, selenium; Zn, zinc Table of the Brazilian Institute of Geography and Statistics, and the National Nutrient Database for Standard Reference of the US Department of Agriculture in the Nutwin program (version 1.5) [20, 21].

Table 1
Reference values adopted for the minerals magnesium, zinc, and selenium

Table 2
Indices of the activity of enzymes that participate in the metabolism of cortisol 11β-HSD, 11β-hydroxysteroid dehydrogenase; THE, tetrahydrocortisone; THF, tetrahydrocortisol.

Table 4
Mean values and standard deviations of metabolic markers in women with obesity and control group *Significantly different values between obese patients and the control group, Mann-Whitney test (p < 0.05).ACTH, adrenocorticotropic hormone; CBG, cortisol-binding globulin; THF, tetrahydrocortisol; THE, tetrahydrocortisone.

Table 5
Mean values and standard deviations of plasma, erythrocyte, and urinary concentrations of zinc, selenium, and magnesium in obese women and control group *Significantly different values between obese patients and control group, Student's t-test or Mann-Whitney test (p < 0.05).Zn, zinc; Se, selenium; Mg, magnesium.