Effects of Different Vegetable Oils on the Nonalcoholic Fatty Liver Disease in C57/BL Mice

Background Nonalcoholic fatty liver disease (NAFLD) is the most common hepatic disorder, affecting 22–28% of the adult population and more than 50% of obese people all over the world. Modulation of the fatty acids in diet as a means of prevention against nonalcoholic fatty liver disease in animal models (NAFLD) remains unclear. The treatment of NAFLD has not been described in specific guidelines so far. Thus, the justification for the study is to check modifications in macronutrients composition, fatty acids, in particular, play a significant role in the treatment of NAFLD regardless of weight loss. Aim To investigate different vegetable oils in prevention and progression of NAFLD in animal models. Methods For the experiment were used fifty C57BL/6J mice male fed with high fat and fructose diet (HFD) to induce the NAFLD status and they received different commercial vegetable oils for 16 weeks to prevent steatosis. Liver steatosis and oxidative stress parameters were analyzed using biochemical and histological methods. Fatty acids profile in the oils and in the liver samples was obtained. Results The high fat and fructose diet led to obesity and the vegetable oils offered were effective in maintaining body weight similar to the control group. At the end of the experiment (16 weeks), the HFHFr group had a greater body weight compared to control and treated groups (HFHFr: 44.20 ± 2.34 g/animal vs. control: 34.80 ± 3.45 g/animal; p < 0.001; HFHFr/OL: 35.40 ± 4.19 g/animal; HFHFr/C: 36.10 ± 3.92 g/animal; HFHFr/S: 36.25 ± 5.70 g/animal; p < 0.01). Furthermore, the HFD diet has caused an increase in total liver fat compared to control (p < 0.01). Among the treated groups, the animals receiving canola oil showed a reduction of hepatic and retroperitoneal fat (p < 0.05). These biochemical levels were positively correlated with the hepatic histology findings. Hepatic levels of omega-3 decreased in the olive oil and high fat diet groups compared to the control group, whereas these levels increased in the groups receiving canola and soybean oil compared to control and the high fat groups. Conclusion In conclusion, the commercial vegetable oils either contributed to the prevention or reduction of induced nonalcoholic fatty liver with high fat and fructose diet, especially canola oil.


Introduction
Nonalcoholic fatty liver disease (NAFLD) is the most common hepatic disorder, afecting 22-28% of the adult population and more than 50% of obese people all over the world [1,2]. NAFLD is characterized by abnormal fat accumulation in the liver (simple steatosis), which can be reversed by changes in lifestyle. Te progression of NAFLD to nonalcoholic steatohepatitis (NASH) covers a broad spectrum ranging from NAFLD with development of infammatory changes to possible cirrhosis and cellular hepatocarcinoma. NAFLD/NASH is strongly associated with metabolic abnormalities such as obesity, insulin resistance, and dyslipidemia [3][4][5][6][7]. Te mechanisms leading to disease progression and the treatment required have not been fully clarifed.
although it may be stated that the cause of these conditions is polygenic and multifactorial. Tere is an association between genes and the participation of the environment, alongside the diet and sedentarism, whose importance has been demonstrated [8,9].
Te currently most accepted mechanism is the classical two-hit theory. Te frst hit is characterized by steatosis resulting from an imbalance in the formation and turnover of triacylglycerides, and it is believed to be afected by insulin resistance. Te second hit involves oxidative stress and infammatory cytokines production leading to hepatic injury [5,10]. Te role of diet in physiopathology and in the progression of NAFLD is not clear. A high-calorie diet, excess of fatty acids (especially saturated ones), and high simple sugars intake, with greater importance on fructose and saccharose, have seemed to be implicated in the disease development. Fructose is used as a food preservative, mainly present in sugar-sweetened beverages, and it is known for stimulating de novo lipogenesis and increasing liver oxidation/infammation [11][12][13]. In addition to fructose/saccharose, the consumption of saturated fatty acids is implicated in the pathogenesis of NAFLD [13,14]. Despite obtaining contradictory results, specifc fatty acids have been proposed to be involved in hepatic lipogenesis, possibly acting on the induction or reversal of the signs and symptoms of NAFLD [15][16][17][18].
Te treatment of NAFLD has not been described in specifc guidelines so far. Lifestyle changes are the primary disease treatment option and weight loss is strongly recommended for NAFLD patients. However, recent studies have demonstrated that modifcations in macronutrient composition and fatty acids, in particular, play a signifcant role in the treatment of NAFLD regardless of weight loss [15][16][17][18]. An increase in saturated fat intake is usually associated with increased insulin resistance, which may induce the progression of NAFLD, whereas monounsaturated (MUFA) and polyunsaturated (PUFA) fatty acids may prevent it by inhibiting genes involved in the mechanisms of de novo lipogenesis [13]. Using diferent n-3, n-7, n-6, and n-9-rich lipid formulations, Siddiqui et al. [13] demonstrated NAFLD reduction.
Te aim of this present study was settled in order to investigate the efect of diferent vegetable oils on the prevention of NAFLD.
Te vegetable oils, extravirgin olive oil ( ® Gallo), soybean oil, and canola oil ( ® Liza) were purchased at the local market. Te fatty acid profles of lard and of respective oils are listed in Table 1, as well as peroxidability index [20]. Vitamin, mineral mix, choline, and L-cystine were purchased from Rhoster (Araçoiaba da Serra, Brazil).
Food intake and weight were determined per cage (2 animals/cage) over a period of 16 weeks, and are reported as mean food intake and weight in g/day. At the end of experiment, animals were starved for 12 hours and then anesthetized with ketamine and xylazine diluted in saline at the proportion of 1 : 1:2 ml. It was administrated dose applications of 10 µl/g weight each. Blood was immediately collected by cardiac puncture, left to rest at room temperature for 30 minutes, and centrifuged at 3500 rpm at 4°C for serum separation afterward. Serum was stored at −80°C for later analysis. Liver, epididymal adipose tissue, and retroperitoneal adipose tissue were weighed and frozen in aluminum parts for further analysis.

Hepatic
Histology. Liver fragments were fxed in 10% bufered formalin for 24 hours and embedded in parafn. Histological preparations containing 5 µm thick sections were stained with hematoxylin and eosin (H&E) for semiquantitative assessment of steatosis and/or steatohepatitis on 10 representative microscope felds at 40× magnifcation by a pathologist, who was blind to the samples. All divergent results were analyzed and discussed using a system of coobserver microscopic analysis [21].

Biochemical and Hepatic
Analyses. Serum triacylglycerols (TAG), total cholesterol (TC), and cholesterol present in high-density lipoprotein (HDL) were determined using commercial kits (Labtest Diagnóstica S.A., Vista Alegre, Brazil), and very low-density lipoprotein (VLDL) cholesterol was then calculated. Te same kits were used for the hepatic determinations as well. Total hepatic fat was extracted by adapted Bligh and Dyer method [22], and total TAG and total hepatic cholesterol levels were determined using commercial kits.
A direct transesterifcation method was applied to determine the hepatic fatty acid profle [23]. Fatty acid methylated esters were separated by gas chromatography 2 Evidence-Based Complementary and Alternative Medicine (Shimadzu Europe, Duisburg, Germany) using an instrument equipped with an AOC-20i self-injector (Shimadzu Europe, Duisburg, Germany) and a SP-2560 fused silica column (100 m, 0.25 mm I.D, flm thickness 0.20 μm). Helium was used as the carrier gas. Synthetic air was used as the fame ionization detector at 250°C. Injections were performed in split mode and the time of fatty acid retention was determined by comparison with an external standard (Supelco 37 component FAME Mix).

Analysis of Glycemia and Insulin
Resistance. Animals glycemia was determined at the end of experiment using obtained samples from animal's tail and the freestyle lite Abbot ® A glucometer. Te triglyceride/HDL-cholesterol ratio was calculated as a predictor of insulin resistance [24].

Analysis of Lipid Peroxidation and Antioxidant
Parameters. Hepatic malondialdehyde MDA was determined by Gerard-Monnier et al. method [25] with some modifcations, as it is thoroughly described in S1 text section. Hepatic reduced glutathione (GSH) was determined from hepatic tissue by Sedlak and Lindsay method [26] with adaptations, and vitamin A (α-tocopherol) was determined by adapted Arnaud et al. method [27]. Complete methodology is detailed and described in S1 text section.

Statistical Analysis.
One-way analysis of variance (ANOVA) was applied to the data of the various groups, followed by the Tukey post-test, using the GraphPad Prism software, version 5.00 for Windows (GraphPad Software, San Diego, CA, USA), with the level of signifcance set up at p < 0.05. Data are reported as mean ± standard deviation.

Efects of MUFAs and PUFAs on Body Weight Gain, Food
Intake, and Energy Intake. HFHFr animals' body weight started to increase signifcantly compared to control from 5th week on (p < 0.05). From 9th week, HFHFr animals' body weight increased signifcantly compared to control (35.00 ± 4.30 g/animal vs. 29.80 ± 2.48 g/animal, p < 0.01) and to treated groups (HFHFr/OL: 30.00 ± 1.89 g/animal; HFHFr/ CN: 31.20 ± 2.34 g/animal; HFHFr/S: 30.63 ± 2.66 g/animal; p < 0.05). At the end of the experiment (16 weeks), HFHFr group had a greater body weight compared to control and treated groups (HFHFr: 44.20 ± 2.34 g/animal vs. Control: 34.80 ± 3.45 g/animal; p < 0.001; HFHFr/OL: 35.40 ± 4.19 g/ animal; HFHFr/C: 36.10 ± 3.92 g/animal; HFHFr/S: 36.25 ± 5.70 g/animal; p < 0.01), suggesting that the oils were efective in maintaining body weight. Tere was some variation in mean food intake throughout the experimental period due to control animals ingesting more food and HFHFr animals ingesting less food. Among the treated groups, HFHFr/OL group ingested more food than any of them. Energy intake data were positively correlated with food intake data, whereas control and HFHFr groups showed lower energy intake throughout the experiment, the experimental groups, particularly HFHFr/OL group, showed greater energy intake. However, these variations were not expressed by the efcacy of diet in favoring weight gain as determined by the feed efciency rate (FER), which only difered between control and HFHFr animals. Te results are shown in Table 2.

Liver Weight and Epididymal and Retroperitoneal Adipose Tissue Weight and Teir Relationship with Body
Weight. Hepatic tissue weight was increased in the HFHFr group compared to control. Among treated groups, the PUFA-rich group (soybean oil) showed a signifcant reduction of hepatic tissue compared to the HFHFr group (p < 0.05), even though no signifcant reduction was observed in relation to body weight throughout the experiment. Te HFHFr/CN group showed a reduction of retroperitoneal adipose tissue compared to the HFHFr group (p < 0.05), although there was no diference in relation to body weight. Te sum of epididymal + retroperitoneal weight was higher for the HFHFr group compared to control (p < 0.05), but it did not difer from that of treated groups. No diference in epididymal adipose tissue or its relation to body weight was observed over 16 weeks of the experiment. Te results are shown in Table 3.  Evidence-Based Complementary and Alternative Medicine group were signifcantly higher than those of HFHFr group (0.291 ± 0.049 vs.0.126 ± 0.027), whereas those of HFHFr/S group were lower than control (0.168 ± 0.037 vs. 0.258 ± 0.060) (Figure 1(a)). Te same results were observed for VLDL values (Figure 1(b)). HFHFr/S group showed a higher total cholesterol level than control (4.208 ± 1.102 vs.

Efects of Vegetable
2.818 ± 0.599) (Figure 1(c)). HDL-C levels were higher for HFHFr/OL and HFHFr groups compared to control (2,175 ± 0.624 vs. 2.175 ± 0.420 vs. 1.457 ± 0.321), while they were lower in soybean oil group compared to HFHFr group ( Figure 1(d)). Glycemia and the marker of insulin resistance (triacylglycerol/HDL-C ratio) did not difer among groups (Figures 1(e) and 1(f )). About treated groups, the HFHFr/ CN group was the only one that did not show changes in serum parameters.

Te MUFAs and PUFAs of Vegetable Oils
Reduced the Accumulation of Hepatic Fat. As expected, the western diet (HFHFr) increased the accumulation of total hepatic fat compared to control (p < 0.01). Among treated groups, the group receiving canola oil, mainly rich in MUFAs and PUFAs, showed a reduction in total hepatic fat accumulation compared to HFHFr group (p < 0.01) (Figure 2(a)). Terefore, these data were positively correlated with the hepatic histology fndings. HFHFr group exhibited moderate and severe difuse microvesicular hepatic steatosis (Figure 2(b)). Groups that received a diet rich in MUFAs and PUFAs showed reduced steatosis compared to HFHFr group. Whereas HFHFr/OL group exhibited mild to moderate macro and microvesicular steatosis, HFHFr/CN and HFHFr/S groups exhibited mild steatosis. No group showed fbrosis or infammation (Figure 2(b)), as it can be seen in Figure 3(c) based on the NAFLD score. Hepatic triacylglycerol levels were higher in the groups receiving a western diet compared to control (38.48 ± 16.86 vs. 17.28 ± 5.4, p < 0.01). Although the groups receiving diets rich in MUFAs and PUFAs showed reduced triacylglycerols, the diference was nonsignifcant (Figure 2(d)). Hepatic cholesterol did not show changes among groups.

Efects of Vegetable Oils on Oxidative Stress Parameters.
Oxidative stress, fnal resulted product of lipid peroxidation, was assessed on the basis of the levels of MDA.  (Figure 3(a)). Antioxidant protection was assessed on the basis of hepatic GSH levels, which were found to be higher in groups HFHFr/CN and HFHFr/S, preventing oxidative stress compared to control and to the group receiving western diet (control: 0.856 ± 0.340 vs. HFHFr: 1.121 ± 0.275; vs. HFHFr/CN: 3.808 ± 1.984 vs. HFHFr/S: 4.254 ± 1.623) (Figure 3(b)). Vitamin E levels were signifcantly reduced in all treated groups compared to control (p < 0.05) (Figure 3).

Efect of the Diets on the Composition of Hepatic Fatty
Acids. Te diet rich in omega-9 (olive oil group) and the diet rich in omega-6 (soybean oil group) increased the levels of palmitic acid (C16 : 0) compared to the group receiving western diet (HFHFr) (p < 0.05), with no changes in C18 : 0 levels or in the total SFA sum. C16 : 1 levels were higher in  the control group than in any other group (p < 0.001), while there was no diference with regard to C18 : 1 or oleic acid (omega-9) levels or the total sum of fatty acids among all groups. Whereas an olive oil-rich diet reduced the hepatic levels of omega-6 (C18 : 1 n-6) compared to control (14.41 ± 1.43 vs. 21.90 ± 3.92), soybean oil-rich diet resulted in increased hepatic levels of C18 : 1 n-6 compared to control and to HFHFr group (HFHFr/S: 29.63 ± 1.44 vs. HFHFr: 18.19 ± 1.55 vs. control: 21.90 ± 3.92, p < 0.05). Omega-3 hepatic incorporation was reduced in both HFHFr/OL and HFHFr groups compared to control (p < 0.05), whereas it was increased in groups receiving canola oil (HFHFr/CN) and soybean oil (HFHFr/S) compared to control and to the HFHFr group (p < 0.05). Arachidonic acid levels were lower in the treated groups than in control (p < 0.05). Despite EPA (C20 : 5 n-3) was signifcantly increased in the groups receiving the canola oil and soybean oil diets compared to control (p < 0.05), it was only increased in the canola oil group compared to the HFHFr group (p < 0.001). Although DHA (C22 : 6 n-3) levels were higher in canola oil group, they did not difer signifcantly. Te group receiving soybean oil had higher amounts of PUFAs compared to control (p < 0.05). Te results are shown in Table 4.

Discussion
Even though it is well known that the MUFAS and PUFAS, in special long chain n-3 PUFAs (EPA and DHA), can afect hepatic metabolic processes, their precise efects across diferent physiological situations are not fully understood. Tus, not only is the quantity of ingested fats but also the fatty acids composition is of pivotal importance for human health.
Meanwhile, the three vegetable oils used in the present study yielded positive results regarding the metabolic syndrome-related parameters. Te HFHFr group showed an increase in weight compared to control and the remaining groups. Te animals treated with the respective oils maintained a gradual and homogeneous weight gain similar to control. Food intake varied throughout the experiment and it was positively correlated with energy intake. However, diet intake afected weight gain only in the group receiving Evidence-Based Complementary and Alternative Medicine western diet, whereas this did not occur in the groups receiving PUFAs and MUFAs, as determined by the diet effciency rate [35]. Diferent types of fatty acids have diferent oxidation and deposition rates that may contribute to fat accumulation and weight gain. Reports have suggested positive efects of monounsaturated fatty acids (MUFAs) on weight control by increasing postprandial fat oxidation and diet-induced thermogenesis compared to SFAs or n-3 polyunsaturated fatty acids (PUFAs; faxseed oil-rich in linolenic acid) [36]. Fatty acids regulate lipogenesis through various transcription factors and nuclear receptors, including PPARs, hepatocyte nuclear factor (HNF)-4, liver X receptor (LXR), and sterol response element-binding proteins (SREBPs) [37]. Moreover, a previous study showed that oil with a high P/S ratio was relevant to lower body fat accumulation, and high-MUFA oil with a high P/S ratio (HMHR; consisting of 60% MUFAs from the total fatty acids with a ratio of 5) may prevent HFD-induced increased in body weight and body fat. Terefore, quantity of ingested fats, as well as composition of fatty acids, feature a pivotal role in human health [36].
PUFAs afect essential fatty acids intake and adipose tissue lipolysis. Te present fndings for the groups receiving diferent oils demonstrate that there was no change in epididymal adipose tissue. Even though retroperitoneal adipose tissue was lower in the group receiving the canola oil diet, animals receiving western diet developed a greater volume of retroperitoneal adipose tissue and sum of adipose tissues. According to the chromatography fndings in this study, canola oil is rich in n-3, especially in regard to the major incorporation of EPA and DHA. Such fatty acids regulate lipogenesis through various transcription factors and nuclear receptors, including PPARs. As it is known, PPAR is highly expressed in white adipose tissue, and its activation plays a key role in adipocyte development and diferentiation resulting in lipolysis as consequence [36].
Several studies have demonstrated that omega-6 and omega-3 fatty acids suppress lipogenesis and NAFLD development by inhibiting SREBP1c and transcriptional genes involved in lipogenesis [13,[38][39][40][41][42]. In the present study, it was the HFHFr group that showed greater liver weight as well as greater accumulation of total hepatic fat and of hepatic triacylglycerols. It was the HFHFr/S group that otherwise featured lower accumulation of total hepatic fat and also lower steatosis score. Nevertheless, these data were positively correlated with histological analysis, as in accordance with macro and micro vesicular steatosis the present steatosis was classifed as mild in HFHFr/S and HFHFr/CN groups and as moderate in HFHFr/OL group. Te therapeutic efects of low dietary omega 6 : 3 ratio have been observed in animal models and in clinical trials with NAFLD/NASH patients. Te signifcance of balanced omega 6 : 3 ratio was more evident in recent trials in which treatment of NAFLD patients with omega-3 fatty acids, such as eicosapentaenoic acid (EPA) or DHA, reduced steatosis [37].
According to Hijona et al. [43], the biochemical quantitation of hepatic fat shows a good correlation with the histological classifcation of hepatic steatosis proposed by Kleiner and Brunt [44]. Te HFHFr/OL group showed higher serum levels of triacylglycerols and VLDL compared to the HFHFr groups, whereas the soybean oil group showed lower levels. Studies conducted on rodents have shown that diets enriched with extravirgin olive oil increase the activity of both acetyl-CoA carboxylase and fatty acid synthase and also reduce the activity of carnitine palmitoyltransferase, thus increasing lipogenesis and reducing β oxidation [45]. Furthermore, the reduced soybean oil values may be associated with the action of PUFAs (n-6), present in greater quantities in this vegetable oil, in terms of either both lipogenesis suppression and regulation of PPRɑ expression, with the consequent degradation of fatty acids [13]. Conversely, the group receiving olive oil displayed greater quantities of HDL-C. In a study carried out with 200 volunteers, Covas et al. [45] demonstrated that the consumption of extravirgin olive oil increased the plasma levels of HDL-C. Fasting glycemia and the insulin predicting factor did not difer among the groups studied. Yan et al. [46] also found no signifcant diference among groups in a study that compared diferent proportions of n-6/n-3. In the present study, canola oil did not show signifcant changes in the biochemical parameters among the treated groups.
Te present data demonstrated a greater quantity of palmitic acid in the HFHFr/OL and HFHFr/S groups. It may be a fact related to the aforementioned described fndings, with greater steatosis in the olive oil group and an increase in serum cholesterol in the soybean oil group. Palmitic acid is known for promoting increased infammation of hepatic cells [47]. Although the diference was not statistically signifcant, oleic acid levels were higher in the group receiving olive oil though. Te olive oil presented lower amount of serum cholesterol. Tese results indicate that the molecular mechanism, through EVOO extracts, promotes a hypocholesterolemic efect through HMGCoAR activity modulation, a crucial enzyme in cholesterol biosynthesis and also the well-known target of statins [48]. Both canola oil and soybean oil induced higher EPA levels compared to control, even though these levels were higher only in the canola oil group than in the HFHFr group, while DHA showed no changes. Morrison et al. demonstrated that EPA was more efective than DHA regarding the parameters of steatosis [47]. Hanke et al. [49] demonstrated an attenuation of NAFLD through use of diferent quantities of canola oil and a greater hepatic quantity of EPA and DHA, alongside a reduction of infammatory parameters. Te present fndings show that all the oils used for various groups' treatment increased lipid peroxidation, as demonstrated by the analysis of hepatic MDA. However, in relation to antioxidant protection based on GSH determination, an increase in hepatic GSH levels was only observed in groups treated with canola oil and soybean oil, while vitamin E levels were lower in the treated groups.

Conclusions
In summary, lipid modulation in a vegetable oil contained diet was able to reserve a number of adverse metabolic efects of high fat diet and high fructose diet, in special the usage of soybean oil and of canola oil. Terefore, canola oil is suggested to be useful to treat metabolic syndrome comorbidities and NAFLD. Tis appears to be an interesting low-cost and easily applicable alternative for the prevention/treatment of steatosis induced by rich western style diet in either or both patients with NAFLD and metabolic syndrome.

Data Availability
Te data used to support this study are available from the corresponding author upon request.

Additional Points
Core tip: It is a fact that there is no guideline or more targeted treatment. Due to the increase in the consumption of ultra-processed foods, rich in saturated fat and fruit, the prevalence of disease increased. Tus, studied dietary methods through dietary modulation for the prevention of nonalcoholic liver disease in a wide scenario increased prevalence of nontransmissible chronic diseases.

Ethical Approval
All animal experiments conformed to the internationally accepted principles for the care and use of laboratory animals (Animal use Ethics Committee, Brazil; protocol no. No. 017/2015, Te Institutional Animal Care and Use Committee at Faculdade de Medicina de Ribeirão Preto, SP, Brazil). Arrive guidelines statement: Te authors have read the ARRIVE guidelines, and the manuscript was prepared and revised according to the arrival guidelines.

Conflicts of Interest
Te authors declare that they have no conficts of interest.

Authors' Contributions
M.C.S. conducted the research, wrote the article, performed the statistical analysis, and contributed to the study design and data interpretation. A.L.M.C.S and O.P.P contributed to technical support. R.L.Z contributes to the interpretation of histological data analysis. A.A.J. contributed to the technical support and data interpretation. All authors read and approved the fnal manuscript.