Ghrelin and Leptin and Their Relations with Insulin Resistance in Diabetes Mellitus Type 2 Patients

Ghrelin and leptin are hunger hormones related to type 2 diabetes mellitus (T2DM), and the pathogenesis of T2DM is the abnormality in insulin secretion and insulin resistance (IR). The aim of this study is to evaluate ghrelin and leptin concentrations in blood and to specify the relationship of these hormones as dependent variables with some biochemical and clinical measurements in T2DM patients. In this study, forty one T2DM and forty three non-diabetes mellitus (non-DM) subjects, aged between 40-60 years and with normal weight, were enrolled. Fasting serum ghrelin and leptin were estimated by enzymelinked immunosorbent assay (ELISA). In our results ghrelin was significantly increased, and leptin was significantly decreased, in T2DM patients compared with non-DM subjects. Ghrelin was positively correlated with the fasting blood glucose (FBG) and IR, but inversely related to the insulin sensitivity (IS). Leptin was negatively correlated with mean arterial pressure (MAP), FBG, glycated hemoglobin (HbA1c), IR, low-density lipoprotein cholesterol, nitric oxide (NO), and alanine aminotransferase (ALT), as well as showed a linear correlation with IS and a strong dependence on sex. The area under the curve (AUC) value shows ghrelin and leptin as biomarkers for T2DM. In conclusion ghrelin and leptin hormones have predictive ability to predict T2DM, as they are significantly associated with IR, IS, free radicals, and lipid profile.


Introduction:
Type 2 diabetes mellitus is a chronic disease due to deficiency in insulin secretion and insulin actions 1 . Many studies are presented to explain the association of hunger hormones (ghrelin and leptin) with IR [2][3][4][5] .
Ghrelin is an orexigenic hormone, first isolated from the stomach of humans and rats in 1999 6 . Studies have focused on the relationship between ghrelin and diabetes complications, especially hypertension, atherosclerosis, and obesity 7 . Patients with T2DM has a lower ghrelin level than healthy subjects 4 . Ghrelin has a role in glucose metabolism, and involved in insulin secretion and IS 8, 9 . Leptin is an anorexigenic hormone synthesized in adipose tissue, and affects the hypothalamus to alter food intake behavior and energy expenditure 10 . The role of hyperleptinemia in the pathogenesis of IR and T2DM has been postulated, and a higher level of leptin was found in patients with T2DM than in healthy subjects 11,12 . Additionally, strong associations of leptin with body mass index (BMI), IR, and insulin production in T2DM patients have been found 3 . Leptin may affect cardiovascular functions and increasing blood pressure 13 . Leptin may be a marker of hepatic fibrosis 14 , and thus it may be associated with alanine aminotransferase (ALT), which is a liver injury marker. Becerril revealed the physiological relationship between leptin and nitric oxide (NO, a nitrogen-based free radical secreted by endothelial cells) 15 . Leptin concentrations depend on sex, as female subjects have higher serum leptin concentrations than males 16, 17 . However, no sufficient studies clearly show the association of both orexigenic and anorexigenic hormones with almost all biochemical parameters related to T2DM. The purpose of our study is to evaluate ghrelin and leptin concentrations in blood and to specify the relationship of these hormones as dependent variables with some biochemical and clinical measurements in T2DM patients.

Material and Methods: Subjects
This study included eighty four subjects from Sulaimani Governorate, Kurdistan Region of Iraq. Forty-one participants (16 men and 25 women) with T2DM attended the Diabetic and Endocrine Clinical Center of Sulaimani Governorate. The rest of the subjects (14 men and 29 women) were healthy and selected randomly as the control group. The age of the participants ranged between 40-60 years, they had BMI ≤25 kg/m 2 . The present study excluded those with chronic diseases (hypertension, cancer, cardiovascular, gastrointestinal, liver, kidney, autoimmune, and thyroid diseases), pregnant, smokers, or alcoholic. The blood samples were collected from April to October 2018.
This study was approved by the ethics committee of the College of Medicine at University of Sulaimani-Kurdistan Region of Iraq.

Clinical Parameters
The clinical history of each subject was recorded by questioning them orally on the day of sample collection and the participants provided informed verbal consent. A mechanical scale was used to measure the height and weight of the participants, and the BMI was calculated, by applying BMI formula 18 . Blood pressure was measured with mercury sphygmomanometer from the right hand while the subject was kept at rest, after resting for five minutes. By applying the mean arterial pressure (MAP) equation, MAP was calculated according to Yu, et al 19 .
IR was calculated by applying the homeostasis model assessment-insulin resistance (HOMA-IR) formula described by Matthews, et al 20 . The quantitative insulinsensitivity check index (QUICKI) was used for assessing IS 21 . Laboratory Measurements About 10 mL of fasting blood was withdrawn from the antecubital vein of all participants, while they were seated. We collected 1 mL of blood in a vacuumed ethylenediaminetetraacetic acid (EDTA) tube for assessing the HbA1c test immediately, while the rest of the blood was collected in a vacuumed gel tube to obtain serum. Sera were separated from the blood and stored in small Eppendorf tubes for later analysis.
NO was estimated by the colorimetric method based on the Griess reaction 23 . Malondialdehyde (MDA) was determined by a simple and sensitive spectrophotometric method and Thiobarbituric acid (TBA) reagent from (Merck KGaA, Darmstadt, Germany) used 24 .
Fasting serum ghrelin and leptin hormones were estimated by enzyme-linked immunosorbent assay (ELISA). Human ELISA kits, Product number CEA991Hu (Cloud-Clone Corp., Houston, TX, USA), and BioVendor research and diagnostic products, Cat number RD191001100 (Brno, Czech Republic), were used for ghrelin and leptin estimation, respectively.

Statistical Analysis
The data of ghrelin and leptin hormones were expressed as a median (interquartile range, IQR). A p-value <0.05 was considered statistically significant. By using GraphPad Prism 7 (San Diego, CA, USA), a Mann-Whitney non-parametric t-test was applied to compare all the studied variables between non-DM controls and T2DM patients. The receiver operating characteristic (ROC) curve was created to compare the sensitivity and specificity of variables in both control and patient groups.
All the data were log-transformed. Pearson (r) and non-parametric Spearman (r s ) correlations were used to analyze the correlation coefficient of ghrelin and leptin, respectively. To predict the relationship of ghrelin and leptin with all other variables, stepwise multiple regression was performed. All the analyses were performed with Statistical Package for Social Sciences (SPSS) version 24.0 (IBM, Armonk, NY, USA).

Results
As shown in Table 1, there were significant differences between T2DM and non-DM subjects in serum ghrelin and leptin concentrations. Depending on Mann-Whitney test, ghrelin was significantly higher in T2DM patients compared with non-DM subjects, while leptin was highly significantly lower in T2DM patients compared with non-DM subjects ( Table 1). Serum ghrelin was higher in both men and women with T2DM, but the difference was significant only in men. On the other hand, by applying unpaired t-test, both men and women showed significantly lower serum leptin concentrations in T2DM patients compared with non-DM subjects, respectively (Table 1).  Serum ghrelin showed a significant positive correlation with FBG and HOMA-IR, and a negative correlation with weight and QUICKI. It was not correlated with the rest of the studied parameters according to Spearman correlation coefficient (r s ) analysis. Leptin showed a strong inverse correlation with most of the studied parameters, including height, systolic blood pressure (SBP), MAP, FBG, HbA1c, HOMA-IR, LDL-C, ALT and NO. It showed a positive correlation with sex and QUICKI. The remaining parameters showed no significant relation with serum leptin levels on Pearson correlation (r) analysis (Table 3).

Table 1. Serum ghrelin and leptin concentration (ng/mL) in males, females and all subjects of non-diabetes mellitus (non-DM) and type 2 diabetes mellitus (T2DM) patients.
Stepwise multiple regression was performed to test the relation between ghrelin and leptin with all the studied parameters, and the results are shown in Tables 4 and 5.
As shown in Table 4 the independent predictors for ghrelin in the total studied population group are ALT, FBG, and age as first, second, and third predictors, respectively.
The independent predictors for leptin in the total studied group are illustrated in Table 5. Sex was the major predictor for serum leptin. FBG, weight, and NO were the second, third and fourth entrant predictors for leptin.

Discussion:
This study showed the prevalence of both ghrelin and leptin in T2DM patients and non-DM subjects in Iraqi Kurdistan Region.
We found that ghrelin significantly increased in T2DM patients, in contrast to previous studies 4,7,25 . The increment of ghrelin may be related to hyperglycemia and IR in our population with T2DM, as ghrelin induces insulin resistance 26 or may be related to BMI as our subject were within normal BMI, a previous study reported the peripheral ghrelin is inversely correlated with BMI and they had lower ghrelin level in  27,28 . Another possible reason for significantly higher ghrelin concentration is the antidiabetes medication, metformin, which causes an elevation in plasma ghrelin level and this explanation approved by Nabel M, as they found serum ghrelin level in control diabetic patients was lower than that of control healthy group while it was higher in diabetic patients' using metformin daily 29 .
On the other hand, serum leptin showed a highly significant lower concentration in T2DM compared with non-DM subjects. Previous studies observed a variation in the level of serum leptin, so our result may be partly explained by ethnic differences. For instance, Abdelgadir, et al observed that the Sudanian T2DM patients had lower leptin levels than non-DM subjects with BMI ≤25 30 , whereas in a Polish study, a reverse outcome was obtained 12 . Saudi Arabians with DM had a moderately higher leptin level than non-diabetic subjects 31 . And Iraqi diabetic normal weight men had lower serum leptin than normal weight healthy subjects 32 . As with our ghrelin results, metformin may also explain the lower leptin levels 11 . The other possible reason of decreasing in leptin level of T2DM patients in our study may be related to the BMI, Zamil, et al result confirmed our result as they concluded that obesity had a significant effect on leptin concentration 32 .
In our population, women showed a higher concentration of serum ghrelin in both T2DM patients and non-DM subjects than men. This result is frequently recorded and several studies have explained the interactions between ghrelin and estrogen, as they reported stimulatory effect of plasma estrogen on ghrelin receptors expression 33 . Testosterone also has an inverse relation with ghrelin, overexpression of ghrelin may inhibit production of testosterone in human testes which may be due to the activities of ghrelin receptor 34 . And another possible reason of the sexual dimorphism of ghrelin is the difference in fat mass contents and distribution in female than in male 35 however, body fat content was not estimated in our study. On the other hand, subjects with T2DM showed significantly lower serum leptin levels in our population, both men and women. Women had higher leptin concentrations than men, which accords with previous studies 16,17 . Higher leptin levels in women may be attributed to a higher ratio of adipose tissue and higher leptin production rate per unit mass of adipose tissue 36 and estrogen levels in women, which stimulate leptin mRNA expression 37 .
Our results, established by ROC curve analysis, show that ghrelin is a useful T2DM disease marker for male subjects and this is possibly because of the regulatory effect of steroid hormone on ghrelin level as the ghrelin and ghrelin receptor expression occurs in the Leydig and Sertoli cells of human testes and ghrelin affect testosterone secretion via stimulation of cyclic guanosine monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP) 34 , or due to the stimulatory effect of estrogen hormone on ghrelin, while leptin is a useful biological marker for T2DM regardless of sex which may be explained by many possible reasons. Including the body fat mass distribution, and the stimulatory effect of estrogen and progesterone on serum leptin in women. In men, testosterone inhibits leptin production in adipose tissue. Leptin may have a direct role on spermatogenesis and the function of testes due to the distribution of leptin receptors in the seminiferous tubules and seminal plasma on the testes cells 38 .
According to the nonparametric Spearman's coefficient (r s ) of correlation, ghrelin exhibits a positive correlation with FBG and HOMA-IR. This result is in line with Vestergaard et al, who found that ghrelin administration induced IR 39 . The main reason for that association is the stimulatory effects of ghrelin on glucose synthesis and decreasing the peripheral glucose uptake 40 and in turn, it causes insulin resistance due to defect in insulin action and leads to reduced insulin sensitivity.
By applying stepwise multiple regression analysis, ALT was found to be a significant predictor for ghrelin, along with 39 FBG and age. ALT is the first predictor for ghrelin and this is due to the effect of ghrelin on liver inflammation 41 , and the inhibitory effect of ghrelin O-acyltransferase administration on serum ALT activity 42 . FBG was another independent predictor of ghrelin hormone. This is due to the enhancement effect of ghrelin on glucagon secretion via its effect on the liver and brain, and stimulate hepatic glucose production through its effect on the brain 43 .
Pearson correlation (r) was used to determine the relationship of leptin with other studied biochemical parameters. Serum leptin inversely correlated with FBG, HbA1c, HOMA-IR, LDL-C, MAP and NO concentrations. Leptin deficiency causes a severe IR, which is associated with hyperglycemia 44,45 . The inverse correlation between leptin and HbA1c mechanism may be due the prolonged hyperglycemia which lowers serum leptin level. As previous study found a negative correlation between serum leptin level and HbA1c in T2DM 46 . The negative correlation between leptin and IR may be related to the normal body weight of our population as obesity has an important linear effect on increasing serum leptin and IR 3 . Consistent with our results, Mohiti-Ardekani, et al found that leptin and HOMA-IR had a negative correlation in non-obese T2DM patients 47 . Leptin has a linear relation with QUICKI, which accords with the findings of previous studies as they found a strong significant correlation of leptin with QUICKI 48,49 . This indicates the improvement effects of leptin on hepatic IS, which is due to the effect of leptin on liver through autonomic nervous system 50 .
Our stepwise regression found that the major predictor for leptin in the total studied subjects is sex, which accounts for 41% of the leptin variation. The sex differences in leptin may be due to factors such as subcutaneous fat in women 46 , estrogens, which stimulate leptin mRNA expression 51 , and glucocorticoids, which may play an important role on leptin synthesis. Fat mass is a significant determinant of plasma leptin concentrations, as mentioned by Fujita, et al 52 and Zamile et al, also found a higher serum leptin in overweight diabetic than normal weight T2DM patients 32 . These support our result about the effect of weight on leptin level. A hypoglycemic effect is another physiological function of leptin, and leptin is a key regulator of glucose homeostasis in rodents and humans 53 . Finally, NO is a predictor of leptin, which contributes to NO metabolism in humans 15 . Interestingly, the current results showed ghrelin a linear correlation with higher FBG and IR. Also, it is evident from our results that leptin is associated with lower FBG, LDL-C levels, IR, hypertension, reactive oxygen species, and oxidative stress in DM patients.

Conclusion:
Ghrelin induced IR and hyperglycemia in our subjects, because of the stimulatory effects of orexigenic hormones on glucose and insulin secretion. While, lower level of leptin is associated with higher FBG and IR in T2DM patients. Ghrelin is a biomarker of T2DM in male subjects, but leptin is a useful marker for the disease regardless of sex. Ghrelin and leptin have predictive ability to predict T2DM disease, as they were significantly associated with IR, IS, free radicals, and lipid profile. their help in blood samples collection, performing the experiments and in handling the instruments. The authors have not announced any conflict of interest, and no specific grant from any funding agency in the public, commercial, or not-for-profit sectors has been received.