The Role of High Fat Diets and Liver Peptidase Activity in the Development of Obesity and Insulin Resistance in Wistar Rats

High-fat diets (HFD) have been widely associated with an increased risk of metabolic disorders and overweight. However, a high intake of sources that are rich in monounsaturated fatty acids has been suggested as a dietary agent that is able to positively influence energy metabolism and vascular function. The main objective of this study was to analyze the role of dietary fats on hepatic peptidases activities and metabolic disorders. Three diets: standard (S), HFD supplemented with virgin olive oil (VOO), and HFD supplemented with butter plus cholesterol (Bch), were administered over six months to male Wistar rats. Plasma and liver samples were collected for clinical biochemistry and aminopeptidase activities (AP) analysis. The expression of inducible nitric oxide synthase (iNOS) was also determined by Western blot in liver samples. The diet supplement with VOO did not induce obesity, in contrast to the Bch group. Though the VOO diet increased the time that was needed to return to the basal levels of plasma glucose, the fasting insulin/glucose ratio and HOMA2-%B index (a homeostasis model index of insulin secretion and valuation of β-cell usefulness (% β-cell secretion)) were improved. An increase of hepatic membrane-bound dipeptidyl-peptidase 4 (DPP4) activity was found only in VOO rats, even if no differences in fasting plasma glucagon-like peptide 1 (GLP-1) were obtained. Both HFDs induced changes in hepatic pyroglutamyl-AP in the soluble fraction, but only the Bch diet increased the soluble tyrosyl-AP. Angiotensinase activities that are implicated in the metabolism of angiotensin II (AngII) to AngIV increased in the VOO diet, which was in agreement with the higher activity of insulin-regulated-AP (IRAP) in this group. Otherwise, the diet that was enriched with butter increased soluble gamma-glutamyl transferase (GGT) and Leucyl-AP, iNOS expression in the liver, and plasma NO. In summary, VOO increased the hepatic activity of AP that were related to glucose metabolism (DPP4, angiotensinases, and IRAP). However, the Bch diet increased activities that are implicated in the control of food intake (Tyrosine-AP), the index of hepatic damage (Leucine-AP and GGT), and the expression of hepatic iNOS and plasma NO. Taken together, these results support that the source of fat in the diet affects several peptidases activities in the liver, which could be related to alterations in feeding behavior and glucose metabolism.


Introduction
Obesity is considered a serious chronic disease with adverse consequences due to the excessive accumulation of adipose tissue. Usually, obesity is caused by an imbalance between energy intake and energy expenditure [1]. The prevalence of obesity in the population has increased alarmingly in In order to analyze the effects of HFD on liver peptide metabolism, we administered diets with different amounts and fat sources to rat groups in this study. We demonstrate that monounsaturated HFD is not related to metabolic manifestations because overweight, dyslipidemia, and high cholesterol levels were present in the animals that were fed with a saturated HFD. However, monounsaturated HFD mainly affects parameters that are related to glucose metabolism, probably in agreement with a greater use of fatty acids for energy. At the same time, the analyzed diets had a differential effect on several aminopeptidase activities with important roles in the development of metabolic syndromes.

Animals and Treatments
Adult male Wistar rats were purchased from Harlan Interfauna Ibérica S.A. (Barcelona, Spain). The rats had free access to experimental diets and water during 24 weeks, and they were maintained at a controlled temperature (20-25 °C) and humidity (50 ± 5%) environment in a 12 hours light/dark cycle. At the beginning of the study, the mean body weight and age were 495 g and six months, respectively. Experimental procedures for animal use and care were in accordance with European Communities Council Directive 2010/63/UE and Spanish regulation RD 53/2013, and the study was approved by the Institutional Animal Care and Use Committee of the University of Jaén. Rats were randomly assigned into three groups (n = 5 each): In the standard diet (S) group, rats were fed with a commercial chow for experimental animals. In HFDs, one group of rats was fed with a diet supplemented with 20% of virgin olive oil (VOO), and the second group of rats was fed with a diet supplemented with 20% butter plus cholesterol (0.1%) (Bch) in order to reach the average cholesterol content of the Western diet. The HFD diets were isocaloric. The food composition and nutritive value of different diets are shown in Supplemental Table S1. The food intake for each group was measured, and the animals were weighed once a week. At the end of the experimental period, a glucose tolerance test was performed. Animals were perfused with a saline solution through the left cardiac ventricle under Equithensin anesthesia (2 mL/kg Body Weight), and a sample of blood was collected for GLP-1, insulin, NO and other biochemical parameters. The blood samples were centrifuged for 10 min at 2000 g to obtain the plasma. Liver tissue samples were collected for angiotensinase activities, ENKand TRH-degrading activities, hepatic damage markers, and immunoblot for hepatic iNOS.

Glucose Tolerance Test
A glucose tolerance test (GTT) was performed after overnight fasting. For the GTT, rats were i.p. injected (8 mL/kg BW) with a single dose of glucose that was dissolved in saline (2.0 g/kg BW). Blood glucose was measured by using a glucometer (Roche Accu-Check Inform.). A baseline (fasting) blood glucose measurement was taken before glucose administration, and further measurements were made at regular intervals thereafter (20, 40, 60 and 90 min).

Determination of Blood Parameters
Plasma insulin levels were determined by using an ELISA kit (#10-1113-01) that was purchased from Mercodia Developing Diagnostics (Winston Salem, NC, USA). Plasma GLP-1 concentrations were determined by using an ELISA kit (#107444-51-9) that was purchased from Cayman (Ann Arbor, MI, USA).
Plasma levels of NO were analyzed by the Griess method as a summation of NO2-and NO3with an assay kit (Total Nitric Oxide and Nitrate/Nitrite Parameter#KGE001) that was purchased from R&D Systems (Minneapolis, MN, USA).

Determination of Insulin Resistance
A homeostasis model of insulin resistance (HOMA2-IR) was calculated by using the HOMA Calculator v.2.2.3 software [51]. HOMA-2 is an update and actualization of the HOMA equation, where a lineal regression is established between glucose and insulin by adjusting real physiology. It is possible, with this program, to calculate the HOMA2-%B, an index of insulin secretion and a valuation of β-cell usefulness (% β-cell secretion), and to know HOMA2-%S, an index for estimating sensitivity to insulin (% insulin sensitivity).

Sample Preparation for Aminopeptidase Activities Assay
Plasma samples were used for the peptidase activities assay. Samples from the liver were quickly removed and frozen in dry ice. To obtain a soluble fraction, tissue samples were homogenized in ten volumes of a 10 mM HCl-Tris buffer (pH 7.4) and ultracentrifuged (100,000 g for 30 min at 4 °C). The resulting supernatants were used to measure soluble (sol) enzymatic activity and protein content, assayed in triplicate. To solubilize membrane proteins, pellets were re-homogenized in an HCl-Tris buffer (pH 7.4) plus Triton X-100 (1%). After centrifugation (100,000 g for 30 min at 4 °C), supernatants were used to measure membrane-bound (mb) activity and proteins in triplicate. To ensure the complete recovery of activity, detergent (Triton X-100) was removed from the medium by adding the adsorbent polymeric SM-2 Biobeads (100 mg/mL), which were purchased form Bio-Rad (Richmond, VA, USA), to the samples and the shaking for 2 h at 4 °C [52].

Peptidase Activities Assay
The DPP4 family, GGT, LeuAP, pGluAP, TyrAP and angiotensinase (AspAP, GluAP, AlaAP, and IRAP) activities were determined in a fluorometric assay while using arylamidase as substrate according to the method of Ramírez et al. [53]. The substrate and substrate solution for each activity are listed in Supplemental Table S2. Briefly, 10 µl of each supernatant were incubated for 30 min at 37 °C with 100 µl of the substrate solution [54]. Later, enzymatic reactions were ceased after adding 100 µl of a 0.1 M acetate buffer (pH 4.2). The β-naphthylamine (β-NA) that was released as a product of proteolytic activity was fluorometrically assessed at 412 nm emission and at a 345 nm excitation wavelength. Specific peptidases activities were expressed as pmol of the β-NA that was hydrolyzed per minute and per milligram of protein. Fluorogenic assays were linear with respect to time of hydrolysis and protein concentration. The protein concentration was determined according to Bradford method [55] with BSA (Bovine Serum Albumin) as standard. All chemical products were supplied by Sigma (St. Louis, MO USA).

Protein Extraction and Western Blot Analysis
For the Western blotting, total protein was isolated from the liver samples. The hepatic tissues were homogenized in a cooled protein extraction buffer (Tris-HCl, 0.1 M, pH 7.5; aprotinin, 0.1 mg/mL; sodium pyrophosphate, 0.1 M; sodium fluoride, 0.1 M; EDTA (Ethylene Diamine Tetraacetic Acid), 0.01 M; sodium orthovanadate, 0.01 M; and PMSF (Phenyl Methyl Sulfonyl Fluoride), 0.002 M), mixed with 10% Triton X-100 and centrifuged (12,000g for 30 min at 4 °C). The supernatants were collected, and then the protein contents were analyzed with a micro Lowry reaction (DCTM Protein Assay #500-0116) purchased from Bio-Rad Laboratories (Hercules, CA, USA). A Western blot analysis was performed by using iNOS antibodies (BD Biosciences, CA USA). The same amount of extracted protein from each sample (30~40 µg) was loaded for the sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) immunoblot analysis. The separated proteins were electrophoretically transferred into a nitrocellulose membrane (Amersham), which was then blocked with 5% BSA and 0.1% Tween for 60 min at room temperature. The membranes were incubated with the primary antibody overnight at 4 °C. The regions containing proteins were visualized by using ECL (Amersham ECL prime Western blotting detection reagent # RPN2232) purchased from GE Healthcare Life Sciences (Buckinghamshire, UK). Each band was normalized by the corresponding value of α1-tubulin as an internal control. A densitometric analysis was performed by Image J 1.36b (NIH) Software. The primary antibodies against the proteins are listed in Table S3.

Statistical Analysis
A statistical analysis was performed by one-way ANOVA followed by Tukey's post-hoc test for multiple comparisons. A comparison of parameters between two groups was performed with an unpaired Student's t-test. The relationship between variables were assessed by Pearson's correlation coefficient. The presence of significant difference was estimate with the Sigmaplot v11.0 software (Systat Software, Inc., San Jose, CA, USA), and p values below 0.05 (p < 0.05) were considered statistically significant. All the data are presented as mean ± standard error of the mean (SEM).

Plasma Nitric Oxide (NO) and Hepatic Inducible NO Synthase (iNOS)
In order to investigate the effects of HFDs on oxidative stress, the expression of hepatic iNOS was determined by using a Western blot assay ( Figure 5A,B), and the plasma NO concentrations (NOx) were measured by the conversion of nitrate (NO3-) in to nitrite (NO2-) by nitrate reductase ( Figure 5C). The Bch-fed rats exhibited a significant increase in iNOS expression ( Figure 5B: S, 1.0 ± 0.2; VOO, 1.3 ± 0.1; Bch: 2.3 ± 0.5). Moreover, the concentration NOx was significantly higher in the Bch group compared with the S and VOO groups. Interestingly, the VOO group showed lower NOx levels than the S group ( Figure 5C: S, 1.4 ± 0.1 µmol/l; VOO, 0.9 ± 0.0 µmol/l; Bch, 2.1 ± 0.2 µmol/l).

Discussion
High fat diets have demonstrated the ability to be a good model of obesity and metabolic syndrome in laboratory animals models [6][7][8]56,58,59]. However, the efficiency in the use of energy also influences the development of obesity in rodents. A study by Liu [7] demonstrated that obesity development was influenced by gender and the efficiency in the use of energy. In the current study, the animals were fed with two isoenergetic HFDs but with different degree of saturation in their fatty acids (mainly monounsaturated in the VOO group and saturated in the Bch group) and minor components (polyphenols in the VOO group and cholesterol in the Bch group). While the Bch animals achieved a significantly higher body weight than the control group, the VOO rats did not show significant differences with the control group (data not shown). Nevertheless, the greater gain of body weight in the Bch rats were not associated with a higher food intake. Indeed, daily food intakes (g/100g BW) were lower in the two HFDs at the end of the experimental period.
Regardless of the energetic-density of the diet, animals are usually able to adapt their food intake and get the same amount of energy [60]. The differences in body weight gain between the HFDs may be explained by the fact that saturated fatty acids show higher levels of energetic efficiency and are able to decrease the mitochondrial oxidative capacity in the liver and the skeletal muscle [61]. However, the inclusion of VOO in the diet seems to have beneficial effects on body weight control [62,63], increase diet-induced thermogenesis [64], regulate UCP1 (uncoupling protein 1) expression in the adipose tissue [65], and enhance thermogenesis by increasing the UCP1 content in brown adipose tissue and increasing noradrenaline and adrenaline secretions [66]. These results suggest that the unsaturated dietary fatty acids and the quantity of dietary fat may have a significant effect on the regulation of thermogenic conditions. At the same time, these effects could be related to the polyphenols present in VOO [13,67].
The differences in final body weight do not seem to be associated with altered values of fasting plasma glucose and insulin in the three groups of animals. Nevertheless, our results showed differences in the intraperitoneal glucose tolerance test (GTT). During the GTT, the higher levels of plasma glucose were reached at 20 min after i.p. injection in the S and Bch groups. However, in VOO animals, the maximum values were delayed until 90 min after i.p. injection, and they remained high at 120 min. These results contradict with previous works [7,8,60], but the different index of insulin resistant that was calculated in our animals (HOMA-IR, HOMA-%S and HOMA-%B) were similar in the three diets. Still, mean values of fasting insulin/glucose and the HOMA2-%b were lower in the VOO group.
In the fed (postprandial) state, GLP-1 peptide has the ability to stimulate insulin secretion and the synthesis of glycogen in the liver, thus reducing postprandial hyperglycemia. On the other hand, the physiological inhibitory control of GLP-1 on gastric emptying and its contribution to the regulation of food intake and satiety suggests that impaired secretion and/or activity of GLP-1 may be involved in the pathogenesis of obesity. A previous study indicated that, in morbid obesity, the faster inactivation of circulating GLP-1 could result in lower plasma levels of this peptide and could contribute to eating behavior abnormalities [23]. Plasma DPP4 activity is the main factor that is implicated in the metabolism of GLP-1, as well as its main regulator [23,68]. In the present study, an increase of the hepatic DPP4 family activity was only found in the VOO-fed rats, even if no differences in fasting plasma GLP-1 levels were obtained between the three diets, probably because of the rapid inactivation of this peptide. Nevertheless, the high DPP4 activity in the VOO animals could have been related to high postprandial GLP-1 levels, which in agreement with other previous works [17,69,70]. During the GTT, the stimulation of the intestinal L cell did not take place, explaining the higher glucose values in the VOO group.
In addition to GLP-1, another peptide with important antidiabetic role able to improve insulin sensitivity and body weight control is cyclo-histidine-proline (cHP) [43,71]. Cyclo-histidine-proline can be found in hepatic cells [42,57], and it is a metabolic product of TRH by the activity of pyroglutamate aminopeptidase (pGluAP) [57].
The role of pGluAP and cHP in the obesity is not yet clear, but cHP plus zinc is effective in decreasing blood glucose concentrations in genetically obese (ob/ob), type 2 diabetic mice [72]. Moreover, it has been reported that HFDs raise TRH expression [73] and alter pGluAP activity, effects that seem to depend on the fat source [44]. Our results indicated that HFDs induced changes in hepatic TRH metabolism, increasing the pGluAP activity in the soluble fraction of liver, with the consequence of the overproduction of cHP.
Several orexigenic peptides, such as enkephalins (ENK), are able to modify the intake behavior and the energy metabolism in animals [48,49]. A high fat diet changes the hypothalamic expression of this peptide, increases the propensity to overeat, the gain of body weight, and the excess of white adipose tissue [48]. Enkephalin concentration changes in the hepatic tissue and plasma during liver disorders [47], and the liver cells can express Met-ENK immunoreactivity, suggesting that ENK is an endogenous opioid that is produced by the liver [74]. Ours results suggest an increase in soluble tyrosyl aminopeptidase activity (TyrAP), which was implicated in the degradation of ENK in the livers of the Bch group, which could have been related to the differences in food intake and body weight gain in these animals. In fact, ENK was recently proposed as a potential therapeutic peptide for HFD-induced obesity and metabolic disorders [75].
The renin-angiotensin system (RAS) is an important regulator of blood pressure and water balance. Beside the RAS, other local systems have been described in several tissues, including the liver [24,74,76,77]. Changes in the balance of different angiotensin bioactive peptides (AngII, AngIII, AngIV, Ang2-10, and Ang1-7) have been implicated in different chronic diseases, such as liver fibrosis, portal hypertension, hepatic tumors [37,78,79], and metabolic alterations, whereas the blocking of angiotensin receptors has been shown to improve glycemic control and reduce hepatic triglyceride levels [80]. The relative amounts of angiotensines are regulated by the activity of several aminopeptidases, namely angiotensinases ( Figure 3A). It is known that these activities are modulated by dietary fat [5,29,30,54,81,82]. Previous results have demonstrated that liver angiotensinases are altered by the thyroid status [83] and obesity [84,85]. The present results confirm the effect of a diet that is enriched with virgin olive oil on liver aminopeptidase activity, increasing the activity of alanyl and cystinyl aminopeptidase. Alanyl aminopeptidase (angiotensinase M) is implicated in the metabolism of AngIII to AngIV [86], whereas cystinyl aminopeptidase (IRAP/AT4) has been described as the insulin regulated aminopeptidase (IRAP) and also as the specific receptor to bind AngIV [83,85,87,88]. This result could indicate an increase of local levels of AngIV, the peptide that binds to the AT4 receptor that is colocalized with the glucose transporter GluT4 [27]. These changes were observed in the membrane-bound fraction, which suggest a more specific functional role for these enzymes [87].
Otherwise, the diet that was enriched with butter and cholesterol increased peptidase activities that have been widely proposed as hepatic biomarkers (GGT and LeuAP) [89,90]. Moreover, the activity levels of the GGT in the liver were significant and positively correlated with plasma triglycerides and total cholesterol. It has been reported that HFDs significantly increased GGT activity, whereas antioxidants (vitamin E) and hypocaloric diets decreased GGT levels [91,92]. Published data have shown that HFDs are the major triggers for oxidative stress [92], and both the cytoprotective role of NO in the liver and hepatic iNOS expression could be an adaptive response to minimize inflammatory injury, hepatic tumor, and cirrhosis [60]. Recently, studies that treated nonalcoholic fatty liver disease (NAFLD) showed that the hepatic expression of iNOS was markedly increased in HFD-fed mice [40,93]. In our results, only animals that were fed with the Bch diet significantly increased the plasmatic levels of nitrate, nitrites, and iNOS expression in the liver, probably as a response to the oxidative stress damage. The main effects of HFDs on peptidase activities are summarized in Figure 6. The butter plus cholesterol diet (Bch) increased soluble and/or membrane-bound activities that are implicated in the control of food intake (tyrosine aminopeptidase (TyrAP)), markers of hepatic damage (leucine aminopeptidase (LeuAP) and gamma glutamyl transferase (GGT)), and the expression of hepatic iNOS, which is involved the reactive oxidative stress (ROS) pathway and induces an increase of plasmatic nitrogen oxides (NOx) levels.

Conclusions
In conclusion, these results demonstrate that the metabolic response to a high fat diet that is enriched with different fat sources could be related to changes in hepatic peptidases that are implicated in the regulation of glucose metabolism and oxidative stress. The diet that was supplemented with butter plus cholesterol altered peptidases activities that are associated with the impaired in the control of food intake and hepatic damage, and it increased body weight. However, the diet that was supplemented with virgin olive oil affected peptidases that are involved in glucose homeostasis (DPP4 and angiotensinases), but it did not alter body weight. Taken together, these results support the beneficial effect of virgin olive oil on energy metabolism and body weight control. Further studies should be performed in order to analyze the expression of key enzymes in fatty acids metabolism, and its relation to peptidase activities.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Table S1: Composition of diets; Table S2: Substrates for Aminopeptidases incubation media; Table S3: Antibodies used for Westernblotting.