The Dietary Replacement of Soybean Oil by Canola Oil Does Not Prevent Liver Fatty Acid Accumulation and Liver Inflammation in Mice

A high-carbohydrate diet (HCD) is a well-established experimental model of accelerated liver fatty acid (FA) deposition and inflammation. In this study, we evaluated whether canola oil can prevent these physiopathological changes. We evaluated hepatic FA accumulation and inflammation in mice fed with a HCD (72.1% carbohydrates) and either canola oil (C group) or soybean oil (S group) as a lipid source for 0, 7, 14, 28, or 56 days. Liver FA compositions were analyzed by gas chromatography. The mRNA expression of acetyl-CoA carboxylase 1 (ACC1) was measured as an indicator of lipogenesis. The mRNA expression of F4/80, tumor necrosis factor-α (TNF-α), interleukin (IL)-1β, IL-6, and IL-10, as mediators of liver inflammation, were also measured. The C group stored less n-6 polyunsaturated FAs (n-6 PUFAs) and had more intense lipid deposition of monounsaturated FAs (MUFAs), n-3 PUFAs, and total FAs. The C group also showed higher ACC1 expression. Moreover, on day 56, the C group showed higher expressions of the inflammatory genes F4/80, TNF-α, IL-1β, and IL-6, as well as the anti-inflammatory IL-10. In conclusion, a diet containing canola oil as a lipid source does not prevent the fatty acid accumulation and inflammation induced by a HCD.


Introduction
Hepatic lipid accumulation is the hallmark of non-alcoholic fatty liver disease (NAFLD), one of the most common diseases worldwide, which can progress to steatohepatitis, fibrosis, cirrhosis, and liver function failure [1]. NAFLD has also been associated with other disorders, such as obesity and cardiovascular diseases [2], and occurs due to an imbalance between the synthesis and exportation of lipids from the liver. The liver lipid accumulation occurs through increased dietary lipid intake, abnormal repartitioning of triacylglycerol (TAG) from adipose tissue to the liver, and increased de novo fatty acid (FA) synthesis and lipogenesis (DNL) [3].
Animals fed a stipulated diet of macronutrient composition provide experimental alternatives to overcome the limitation of studies in humans, where there is a large variability in the daily diet composition. In this context, we previously demonstrated that the diet for maintenance of laboratory adult rodents proposed by the American Institute of Nutrition (AIN-93-M) is more inflammatory and lipogenic than a high-fat diet in the liver of Swiss mice [4]. The carbohydrate content in the AIN-93 diet is 72.1% [5] versus 50-55% in the rodent chow diet [6]. Thus, the AIN-93-M diet can be considered a high-carbohydrate diet [4,[7][8][9][10]. The n-3 polyunsaturated FAs (n-3 PUFAs) inhibit hepatic lipogenesis and inflammation, preventing NAFLD [11][12][13]. In contrast, higher liver levels of saturated FAs (SFAs), monounsaturated FAs (MUFAs), and elevated n-6:n-3 PUFA ratio have been associated with higher inflammation state and NAFLD occurrence [14][15][16].
In the Western diet, vegetable oil consumption as a source of FAs has been increasing [17][18][19]. One of the most consumed oils in the Western diet is canola oil, a cheaper substitute for olive oil. Canola oil, when compared to soybean oil, contains lower levels of SFAs and a lower n-6:n-3 PUFA ratio [19].
Despite numerous studies, the potential beneficial effects of canola oil still need confirmation. For example, the lifespan has been shown to be shorter in stroke-prone spontaneously hypertensive rats fed canola oil as a sole lipid source than soybean oil [30][31][32][33][34]. One study has reported memory impairments and reduced synaptic integrity in a transgenic mouse model of Alzheimer's disease [35]. Rats fed canola oil have shown insulin resistance [17] and higher blood pressure [36], compared to those fed with soybean oil. In addition, no study has compared the effect of canola oil and soybean oil based lipid intake on liver FAs' deposition and inflammation.
A high-carbohydrate diet is a well-established experimental model of liver FA accumulation and inflammation [4,37,38]. Herein, we evaluate whether canola oil can prevent the liver FA deposition and inflammation induced by a high-carbohydrate diet. For comparative purposes, the reference group received soybean oil as the source of lipids.

Animals and Diets
The experimental protocol was approved by the Animal Ethics Committee of The State University of Maringá (CEUA).
We used male Swiss mice (six weeks old) receiving standard rodent chow (Quintia-Nuvilab ® , Colombo city, Brazil) from weaning. The mice were individually housed and maintained at a controlled temperature (23 ± 1 • C), humidity (55 ± 10%), photoperiod (12 h light/12 h darkness), and had free access to water and food.
We prepared the diets with highly refined ingredients purchased from the Rhoster Company (Araçoiaba da Serra, SP, Brazil). The diet composition was based on purified diets for the maintenance of laboratory adult rodents proposed by the American Institute of Nutrition (AIN-93-M). The protein, carbohydrate, and total fat contents in the diets were 14.0, 72.8, and 4.0 g/100 g, respectively. The main FA compositions of the diets were measured (Table 1).
Carbohydrate composition was represented by cornstarch (46.6%), dextrinized cornstarch (15.5%), and sucrose (10.0%) [5]. As soybean oil is the source of lipids in the AIN-93 diet, the control group was represented by mice fed with the AIN-93 diet (high-carbohydrate diet) with soybean oil as a fat source (S group). The experimental group received the AIN-93 diet (high-carbohydrate diet) with canola oil as a fat source (C group).
After receiving the diets for 0 (before starting the diets), 7, 14, 28, or 56 days, the mice were fasted from 5:00 p.m. to 8:00 a.m. the following day and then euthanized. We measured blood glucose, TAG, and cholesterol concentrations, according to the manufacturer's instructions. The livers were removed, weighed, and stored in liquid nitrogen until further analysis. The treatment schedule of the animals is shown in Figure 1. Carbohydrate composition was represented by cornstarch (46.6%), dextrinized cornstarch (15.5%), and sucrose (10.0%) [5]. As soybean oil is the source of lipids in the AIN-93 diet, the control group was represented by mice fed with the AIN-93 diet (high-carbohydrate diet) with soybean oil as a fat source (S group). The experimental group received the AIN-93 diet (high-carbohydrate diet) with canola oil as a fat source (C group).
After receiving the diets for 0 (before starting the diets), 7, 14, 28, or 56 days, the mice were fasted from 5:00 p.m. to 8:00 a.m. the following day and then euthanized. We measured blood glucose, TAG, and cholesterol concentrations, according to the manufacturer's instructions. The livers were removed, weighed, and stored in liquid nitrogen until further analysis. The treatment schedule of the animals is shown in Figure 1. . The high-carbohydrate diet containing canola oil or soybean oil as a lipid source was administered for 0 (before starting the diets), 7, 14, 28, or 56 days.

Analysis of Liver Fatty Acid Composition
We used triturated samples (100 mg) from frozen livers to determine the total lipid content and FAs. We transesterified total lipids utilizing the method of Figueiredo et al. (2016) [39]. Methyl ester tricosanoic acid (23:0me; Sigma, St. Louis, EUA) served as an internal standard. FA methyl esters (FAME) separation was performed by gas chromatography in a Thermo Scientific™ TRACE™ Ultra Gas Chromatographer (Thermo Scientific™, Waltham, MA, USA). The equipment had a flame ionization detector (FID), a split/split-less injector, and a fused silica capillary column CP-7420 . The high-carbohydrate diet containing canola oil or soybean oil as a lipid source was administered for 0 (before starting the diets), 7, 14, 28, or 56 days.

Analysis of Liver Fatty Acid Composition
We used triturated samples (100 mg) from frozen livers to determine the total lipid content and FAs. We transesterified total lipids utilizing the method of Figueiredo et al. (2016) [39]. Methyl ester tricosanoic acid (23:0me; Sigma, St. Louis, EUA) served as an internal standard. FA methyl esters (FAME) separation was performed by gas chromatography in a Thermo Scientific™ TRACE™ Ultra Gas Chromatographer (Thermo Scientific™, Waltham, MA, USA). The equipment had a flame ionization detector (FID), a split/split-less injector, and a fused silica capillary column CP-7420 (Select FAME, 100 m size, 0.25 mm of internal diameter, and 0.25 µm film thickness of the cyanopropyl stationary phase). The operational parameters were: the gas flow rates used were 1.2 mL min −1 for the carrier gas (H 2 ), 30 mL min −1 for the make-up gas (N 2 ), and 30 and 300 mL min −1 for the FID gas H 2 and synthetic air, respectively. The injected sample volume was 1.0 µL, with a split injection ratio of 1:40. The column temperature was maintained at 165 • C for 18 min and then ramped to 235 • C (4 • C min −1 ) for 20 min. The injector and detector temperatures were kept at 230 • C and 250 • C, respectively.
We identified FAMEs by comparison of the retention times of the sample constituents with Sigma FAMEs. Retention times and peak areas were determined using the Chrom-Quest™ software (Thermo Fisher Scientific™, MA, USA). FA levels were calculated according to Visentainer (2012) [40]. FA contents in the diets and livers are expressed as mg/100 mg.

Statistical Analysis
Results are presented as mean ± standard error, as analyzed by ANOVA (one-way) and Tukey's post test. We compared each FA of the S group or C group using unpaired Student's t-test. We performed Statistical analyses using the Graph-Pad Prism software. The 95% level of confidence (p < 0.05) was accepted for all comparisons.
On day 56, the C group exhibited higher (p < 0.05) levels of arachidic acid (20:0) in comparison with the S group. On the other hand, on day 56, the C group showed lower (p < 0.05) levels of heneicosanoic acid (21:0), in comparison with the S group. The contents of lauric acid (12:0), myristic acid (14:0), palmitic acid (16:0), and stearic acid (18:0) did not differ in C vs. S group (Table 3). Table 2. Food intake, body weight, liver weight, relative liver weight, serum glucose, triacylglycerol, and cholesterol, of mice fed with a diet containing soybean oil (S group) or canola oil (C group) as a lipid source, at 0, 7, 14, 28, or 56 days after starting the diets.  Results are expressed as mean ± standard error of three replicates for each group. p < 0.05 as compared with day 0 a , day 7 b , day 14 c , and day 28 d , and S group *. ∆%: percentage change (day 0 vs. day 56).

Analysis of Fatty Acids (FAs) Family Composition and n-6:n-3ratio
The liver lipid deposition (calculated by the sum of all FAs) and liver deposition of SFA, MUFA, n-6 PUFA, and n-3 PUFA were intensified (p < 0.05) during the experimental period (day 0 vs. day 56) for  Table 7). Table 7. Fatty acids family composition (mg/100 g of sample) and n-6:n-3 PUFA ratio in the livers of mice fed with a diet containing soybean oil (S group) or canola oil (C group) as a lipid source, at 0, 7, 14, 28, or 56 days after starting the diets. Results are expressed as mean ± standard error of three replicates for each group. Abbreviations: SFA, total saturated fatty acids; MUFA, total monounsaturated fatty acids; PUFA, total polyunsaturated fatty acids; SUM, the sum of all fatty acids evaluated. p < 0.05 as compared with 0 a , day 7 b , day 14 c , and day 28 d ; and S group *. ∆%: percentage change from day 0.
On day 56, the C group showed a higher (p < 0.05) value for the sum of all fatty acids evaluated. The C group also exhibited higher (p < 0.05) levels of MUFA and n-3 PUFA. In contrast, the C group showed lower (p < 0.05) levels of n-6 PUFA, while the SFA levels were similar (C group vs. S group) ( Table 7).
The levels of PUFAs and MUFAs, particularly the levels of α-linolenic acid and oleic acid in the livers of the C group reflected the lipid composition of canola oil in the diet. The C group showed lower liver concentrations of γ-linolenic acid and arachidonic acid (synthesized from linoleic acid), and higher liver levels of dihomo-α-linolenic acid, EPA, and DHA (synthesized from α-linolenic acid). Results are expressed as mean ± standard error. p < 0.05 as compared with day 0 a , day 7 b , day 14 c , day 28 d , and S group *.
The levels of PUFAs and MUFAs, particularly the levels of α-linolenic acid and oleic acid in the livers of the C group reflected the lipid composition of canola oil in the diet. The C group showed lower liver concentrations of γ-linolenic acid and arachidonic acid (synthesized from linoleic acid), and higher liver levels of dihomo-α-linolenic acid, EPA, and DHA (synthesized from α-linolenic acid).
The total amount of FAs between days 0 and 56, in the livers of the S and C groups increased by 229.5% and 272.3%, respectively. The mechanisms by which a high-carbohydrate diet increases lipid deposition involve the intensification of the generation of acetyl-CoA from glucose. Acetyl-CoA activates the transcription factors sterol regulatory element-binding proteins (SREBP-1c) and carbohydrate response element binding protein (ChREBP), which regulate key genes involved in the lipid synthesis, such as ACC1 [41][42][43].
The higher concentrations of n-3 PUFAs in the livers of the C group could prevent liver DNL via downregulation of SREBP-1c and ChREBP gene expression and stimulation of FA oxidation [44,45]. However, in this study, the diet containing canola oil as a lipid source did not prevent liver lipid accumulation.
The higher liver lipid accumulation in livers from the C group could be explained by the greater amount of MUFAs in the C diet. In agreement with this affirmation, Duwaerts et al. [46] reported that an MUFA-enriched diet is more steatogenic than an SFA-enriched diet, particularly when combined with complex carbohydrates such as starch. In addition, mice fed with diets rich in oleic acid have shown high liver lipid deposition [47,48]. Oleic acid, the main MUFA in the C group's diet and liver, promotes liver steatosis, oxidative stress, apoptosis, and the increased production of TNF-α and IL-6 [47][48][49].
Higher levels of n-3 PUFAs, such as EPA (20:5n-3) and DHA (22:6n-3), are expected to prevent liver inflammation, as they are precursors to anti-inflammatory mediators [50]. However, the livers of the C group exhibited higher inflammation, as suggested by the higher gene expressions of F4/80, TNF-α, IL1-β, and IL-6. The cytokines TNF-α, IL1-β, and IL-6 are involved in the inflammatory process by producing other cytokines that promote chronic inflammation [51]. Moreover, F4/80 is a marker of the liver recruitment of macrophages from resident Kupffer cells and circulating monocytes, which play a central role in the progression of NAFLD [52]. Furthermore, the simultaneous increase of gene expression of IL-10 (an anti-inflammatory cytokine) represents a negative feedback mechanism, in an attempt to protect the liver against an exacerbated inflammatory response [53].
In agreement with our results, other reports have also demonstrated that canola oil promotes higher oxidative stress and inflammation than soybean oil, safflower oil, or flax oil [30,54]. It is likely that other components of canola oil, which were not investigated in this study, could influence oxidative stress and inflammation; for example, the production of cyclic FAs monomers and/or the loss of phenolic compounds during the industrial refining of canola oil [55][56][57].
The main limitation of this investigation was the restricted time period of evaluation (56 days). Additional limitations of the study included a reduced number of biomarkers of lipogenesis and inflammation. Despite these limitations, we can conclude that the replacement of soybean oil by canola oil as a lipid source did not prevent the liver FA accumulation and inflammation induced by a high-carbohydrate diet in mice.