Supplementation of Enriched Polyunsaturated Fatty Acids and CLA Cheese on High Fat Diet: Effects on Lipid Metabolism and Fat Profile

Epidemiological studies have demonstrated a positive relationship between dietary fat intake and the onset of several metabolic diseases. This association is particularly evident in a diet rich in saturated fatty acids, typical of animal foods, such as dairy products. However, these foods are the main source of fatty acids with a proven nutraceutical effect, such as the ω-3 fatty acid α-linolenic acid (ALA) and the conjugated linoleic acid (CLA), which have demonstrated important roles in the prevention of various diseases. In the present study, the effect of a supplementation with cheese enriched with ω-3 fatty acids and CLA on the metabolism and lipid profiles of C57bl/6 mice was evaluated. In particular, the analyses were conducted on different tissues, such as liver, muscle, adipose tissue and brain, known for their susceptibility to the effects of dietary fats. Supplementing cheese enriched in CLA and ω-3 fats reduced the level of saturated fat and increased the content of CLA and ALA in all tissues considered, except for the brain. Furthermore, the consumption of this cheese resulted in a tissue-specific response in the expression levels of genes involved in lipid and mitochondrial metabolism. As regards genes involved in the inflammatory response, the consumption of enriched cheese resulted in a reduction in the expression of inflammatory genes in all tissues analyzed. Considering the effects that chronic inflammation associated with a high-calorie and high-fat diet (meta-inflammation) or aging (inflammaging) has on the onset of chronic degenerative diseases, these data could be of great interest as they indicate the feasibility of modulating inflammation (thus avoiding/delaying these pathologies) with a nutritional and non-pharmacological intervention.


Introduction
Human studies have shown that high-fat diets (HFDs) (>30% of energy from fat) can easily induce overweight/obesity, which is considered one of the most risk factors for chronic diseases, such as coronary heart disease (CHD), hypertension, type 2 diabetes (T2D) and cancers [1]. These effects are mainly related to the high level of saturated fatty acids (SFA), which are more abundant in animal products, such as milk and dairy products.

Experimental Design
Twenty-one male C57bl/6 mice six months of age with an initial weight of 27.0 ± 0.72 g were randomly assigned to three groups (seven animals per group) and fed for eight weeks with different diets: (i) control diet (CON), based on the Teklan global 14% protein maintenance diet (ENVIGO, Indianapolis, IN, USA) formulation containing 4% total fat, which provided 13% of the total energy; (ii) control diet + "Pecorino" cheese (CHE) (40%:60%); (iii) control diet + ALA and CLA-enriched "Pecorino" cheese (40%:60%) (ENR). Diets CHE and ENR were characterized by a high fat level, which provide 70% of total energy. The three diets were chosen for the following reasons: CON represents the diet usually used for mice, CHE constitutes the most appropriate control, while ENR represents the effective CLA-and ALA-enriched diet. The chemical composition of the three diets is presented in Table 1. Protein, lipid, carbohydrates and ash contents were determined with official AOAC (Association of Official Agricultural Chemists) methods, while the fatty acid profile was determined by gas-chromatographic analysis, as described later. Diets CHE and ENR were characterized by a high content of lipids and a lower level of carbohydrates. Moreover, CHE and ENR diets showed a higher amount of C16:0, C18:0, C18:1c9 and CLA (C18:2c9t11). The comparison between the two diets with cheese shows that the ENR diet is very rich in CLA and ALA, the levels of which exceed the thresholds (0.3 and 0.7%, respectively) of consumer well-being, as indicated by Mele et al. [4]. The CHE diet, on the other hand, is unable to guarantee these CLA and ALA levels, therefore it does not express the nutraceutical effects of these fatty acids. The dietary approaches proposed in this study were preferred to a simple addition of ω-3 mixed in the standard pellet diet as the added cheese is a dairy product already marketed and this approach should have allowed the mimicry, at least in part, of the consumption of cheese within a diet as varied as that of the consumer. The Pecorino cheese used in the CHE diet was produced according to the specification provided by the PDO consortium of "Pecorino Toscano", while the cheese used in the ENR diet was obtained as described by Mele et al. [4]. Animals were housed in cages with food and water ad libitum, and they were exposed to a 12 h dark/light cycle. During the experimental period, the body weight of mice was recorded weekly. Before sacrifice, mice were fasted for 12 h and their blood was collected, through the caudal vein, to perform a blood glucose test (Glucocard G meter, Menarini Diagnostics, Florence, Italy). Mice were euthanized by urethane injection (100 µg/kg of body weight) and cervical dislocation. Immediately after death, skeletal muscle (SM), brain (B), liver (L) and visceral adipose tissue (VAT) were taken and stored at −80 • C. After the mice died, intracardiac blood was collected in a blood collection tube with EDTA (BD Vacutaine). The blood was used for the measurement of the lipid panel (cholesterol, HDL, LDL and triglycerides) using the Cobas b101 instrument (Roche Diagnostic, Milan, Italy).
All experiments were performed according to the guidelines and protocols approved by the European Union (EU Council 750/2013; D.L. 26/2014) and by the Animal Research Ethics Committee of the University of Pisa, Italy. The authorization number from the Italian Ethical Committee was 12/2019-PR.

Lipid Analysis
Total lipids (TL) of whole samples (52 mg SM, 200 mg L, 150 mg B and 300 mg VAT) were extracted, according to the method described by Rodriguez-Estrada et al. [16], with some modifications. Briefly, samples were resuspended with a chloroform/methanol solution (2:1, v/v) and homogenized by ULTRATURRAX (IKA ® Werke GmbH & Co. KG, Staufen, Germany) for 60 s. After 2 h of incubation at 40 • C, samples were filtered to remove solid phase and then 2 mL of KCl 1M aqueous solution was added. The lower phase (apolar phase) was recovered and placed in a fresh tube. Lipid fractions were separated by thin layer chromatography (TLC) using a Silica Gel 60F254 10 × 20 cm with a mix of hexane and diethyl ether (70:30, v:v) as mobile phase solution. The spots corresponding to the phospholipids (PLs), triglycerides (TGs), free fatty acids (FFAs) and free and esterified cholesterol (FC and EC, respectively) were identified by comparing them with a mix of commercial standards (Sigma-Aldrich: code 17,810 for TGs, 690050C for PLs, MAK044 for FFAs, C4951 for FC and C9253 for EC). Each spot was scraped and recovered separately in a fresh tube with 1 mL of diethyl ether.
For TL and TG, FA methyl esters (FAMEs) were prepared using a methanolic NaOH basic solution (10%), according to Christie [17], with some modifications. Briefly, 1 mg of C19:0 as an internal standard was added to 20 mg of lipids, 1 mL of methanolic NaOH solution (10%) and 1 mL of hexane. Samples were incubated at RT for 1 min and subsequently FAMEs were extracted with 1 mL of hexane.
FAMEs of PLs and FFAs were prepared using a methanolic hydrochloric acid solution (10%), according to Christie [17], with some modifications. Briefly, 1 mg of C19:0 as an internal standard was added to 20 mg of lipids and 2 mL of methanolic hydrochloric acid solution (10%). Samples were incubated at 50 • C and subsequently FAMEs were extracted with 2 mL of hexane and 2 mL of water. After centrifugation at 3000× g for 5 min, the upper phase was transferred into a fresh tube, while the lower phase was washed twice with 2 mL of hexane. FAMEs were separated and identified using a GC-FID (GC 2000 plus, Shimadzu, Columbia, MD, United States), according to Mele et al. [4].

Gene Expression Analysis
Total RNA was extracted using different kits depending on the tissue: mirVanaTM miRNA Isolation Kit (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) for SM, and RNeasy Lipid Tissue (QUIAGEN, Hilden, Germany) for B, L and VAT, according to the manufacturer's instructions. RNA quantification was carried out with a NanoDrop 1000 Spectrophotometer (THERMO ® Scientific). RNA was treated with DNase by TURBO DNA-freeTM kit (Ambion, Austin, TX, USA), following the manufacturer's instructions, to completely remove genomic DNA contamination. Finally, cDNA was synthesized using iScriptTM cDNA Synthesis Kit (Bio-Rad, Hongkong, China), according to the manufacturer's instructions. Real-time PCR was performed using iTaqTM Universal SYBR Green Supermix (Bio-Rad) and Rotor gene Q 6000 system (QIAGEN GmbH, Hilden, Germany). All data were normalized to GAPDH expression. All oligonucleotide pre-designed primers were from Bio-Rad. All information on these primers is available at the website www.bio-rad.com/PrimePCR (1 June 2017).
For this study, we considered the expression of the key genes involved in lipid metabolism, inflammation and the cellular cycle and mitochondrial metabolism (Table S1).

Statistical Analysis
Multiple unpaired comparison tests were performed with one-way ANOVA followed by a Tukey's post hoc multiple comparison test in order to check the effect of specific dietary lipids on lipid metabolism. Data were analyzed by the following linear model (SAS Institute Inc., Cary, NC, USA, 2010): where y ij is the observed trait (fatty acids and cholesterol); µ is the overall mean; T i is the fixed effect of the i th treatment (CON; CHE; ENR); A j = is the random effect of the j th animal ; and ε ij is the random residual term. Data are expressed as means ± SEM. Differences were declared significantly different at a p-value < 0.05.

Effect of ENR Cheese Supplementation on Lipid Fraction Composition
The effect of the ENR cheese supplementation was evaluated in comparison with a standard chow diet (CON) and a diet supplemented with normal cheese (CHE). The mice in the three groups showed a similar initial weight and daily food intake (3.65 ± 0.35, 3.53 ± 0.43, 3.71 ± 0.43 g for CON, CHE and ENR, respectively), however at the end of the treatment, the body weight of ENR group mice was significantly higher with respect to other two groups ( Table 2). Total cholesterol was significantly increased in CHE, while no differences were observed between CON and ENR. This result was due to a higher level of HDL-cholesterol in both cheese diets. LDL-cholesterol was decreased in both experimental groups with respect to CON. Both cheese supplementations are able to markedly reduce cardiovascular risk. On the contrary, TG levels in the blood did not change after cheese supplementation.

Liver
The amount of fat was about 8.5%, which was mainly characterized by TGs (52%) and FFAs (30%), while PLs were the least represented component (8%) ( Table 3). Cholesterol occurred principally as FC, while the EC was about 25 times lower (Table 3).
Total lipids, TGs and PLs did not show significant differences between the three groups, while FC decreased by 43% in CHE and ENR groups. Esterified cholesterol, on the other hand, increased significantly (more than five times) in the ENR mice only.
Regarding the effect of diet on total FA profile, we found a significant increase of SFAs (+47%) and MUFAs (+31%) in CHE and ENR group mice. Conversely, we observed a significant reduction of PUFAs (−46%) in the same groups ( Table 2). The higher level of SFAs was mainly due to the increase in C16:0 (+21% and 13% for mice in the CHE and ENR groups, respectively) (Table S2). There was also a fairly significant increase of C18:0 (+100%) in mice in both experimental groups. Among the other SFAs, there was a significant increase of medium and short chain fatty acids in the CHE and ENR groups (Table S2). As for the PUFAs, the main changes occurred for linoleic acid, which significantly decreased by 59% and 61% for CHE and ENR, respectively. Importantly, C18:2c9t11 (CLA) increased dramatically (+94-fold) in mice of the ENR group (Table 3). On the contrary, we observed a significant reduction of C20:5n3 (−59% and −55% for CHE and ENR, respectively) and C22:3n3 (−78% and −100% for CHE and ENR, respectively) (Table S2).

Brain
The amount of fat was about 10%, which was mainly characterized by PLs (65%) and TGs (22%), while FFAs were the least represented component (8%) ( Table 4). Cholesterol occurred principally as FC, while EC was not revealed (Table 4). All lipid fractions did not show significant differences between the three groups. Also in this case, regarding the fatty acid profile, no effects due to a high-energy diet with cheese supplementation were observed.

Skeletal Muscle
The amount of fat was about 21%, which was mainly characterized by TGs (85%) and PLs (14%), while FFAs were the least represented component (3%) ( Table 5). Cholesterol occurred principally as FC, while EC was about seven times lower (Table 5). The total lipids fraction was not affected by the two experimental diets (Table 4). In contrast, mice of the ENR group showed a significant increase in PLs (+102%) and a reduction in TGs (−12%).
In both experimental groups, cheese consumption resulted in a significant increase in FFA (+11 times). The two experimental groups also displayed a reduction in FC content (−43%), while no significant differences were found for esterified cholesterol.
The supplementation with cheese showed a significant increase in SFAs, in particular, CHE (+54% and +34% in CHE and ENR, respectively); in contrast, MUFAs decreased by 22% and 31% in CHE and ENR, respectively ( Table 5). The level of PUFAs significantly decreased by 42% in CHE, while no differences were observed in ENR. Notably, PUFAn-6 decreased by 48% and 37% in CHE and ENR, respectively. PUFAn-3, on the other hand, increased principally in ENR (+233%); however, we also observed a lower but significant increase (+23%) in CHE ( Table 5).
The higher level of SFAs in CHE and ENR was mainly due to the increase of short and medium chain FAs deriving from dairy product consumption, as also observed in the liver (Table S4). The most diet-influenced MUFAs were principally C18:1 trans isomers (C18:1t9, C18:1t10 and C18:1t11) and oleic acid (Table S4). Finally, many PUFAs showed a significant change with diets containing cheese. In particular, we observed a reduction of C18:2n-6 (−68%), and an increase of n-6 and n-3 PUFAs, deriving from LA and ALA elongation (Table S4).

Adipose Tissue
The amount of fat in the adipose tissue was about 58%, which was mainly characterized by TGs (95%), while PLs represented only 1% (Table 6). Cholesterol occurred principally as FC (7 mg/100 g of total lipids). Lipid fractions were not affected by the two experimental diets ( Table 6).

Effect of Diet on Gene Expression
The effect of diet on the expression of some genes involved in lipid and mitochondrial metabolism and in the control of stress and inflammation is reported in Table 7. The potential pathway of these genes is summarized in Figure 1.   Table 7. Effect of the diet on the gene expression of liver, brain, skeletal muscle and adipose tissue.

Liver
We observed a significant change in the expression of all perilipin genes (PLIN) in the groups supplemented with cheese, compared to the control group: PLIN2 showed a significant increase (+68%) in ENR and a reduction (−23%) in CHE; PLIN3 expression increased (+12%) in CHE and decreased (−42%) in ENR; PLIN4 is the gene that showed the greatest variability of expression, with a significant increase in the CHE (+1037%) and ENR groups (+1620%); PLIN5 increased (+ 38%) in the ENR group and decreased (−83%) in the CHE group.
Genes involved in the ceramide biosynthesis pathway, such as CERK, CERS6 and the Kennedy pathway, such as ETNK1 and CEPT-1, were not influenced by the cheese diet; significant changes were in fact detected.
Three genes involved in the regulation of inflammatory processes showed a significantly lower expression in groups fed with cheese: PRKCQ −13% in CHE and −46% in ENR; TRP53 −54% in CHE and −78% in ENR; FGF21 −70% in CHE and −52% in ENR.
The effect of cheese was also observed in mitochondrial metabolism, where a significantly higher expression was observed for the following genes: PGC1A +57% in the CHE group and +82% in the ENR group; OPA-1 +57% in CHE and +106% in ENR. On the contrary, expression of the following genes was lower: IL1B −65% for both groups; COX10 −24% in the CHE group and −41% in ENR; UCP2 −56% in CHE and −33% in ENR; ATP5G1 −15% in CHE and −42% in ENR; MFN1 −51% in CHE and −57% in ENR.

Brain
Among the genes encoding for PLINs, only PLIN2 and PLIN3 showed lower expression in the ENR group, while CHE showed a level of expression similar to CON.
We also observed a different expression of some genes involved in lipid synthesis. FASN expression (+129%) was significantly higher in ENR, while no effect was observed in the CHE group; SCD and ELOVL2, on the other hand, were not as highly expressed (−10% and −79%, respectively) in mice of the ENR group. The expression of ACLY and SREBP1 decreased in the two experimental groups with respect to the control. Three genes involved in glycerolipid metabolism decreased significantly in the experimental groups: DGAT1 (−37% and −79% in CHE and ENR, respectively); DGAT2 (−47% and −72% in CHE and ENR, respectively); LPIN1 (−15% and −62% in CHE and ENR, respectively). A significant effect was observed for the following genes involved in the regulation of inflammatory processes: PRKCQ (−21% and −77% in CHE and ENR, respectively) and TRP53 (−47% and −82% in CHE and ENR, respectively).
Regarding the genes of the Kennedy pathway, the gene ETNK1 showed reduced expression (−60%) in both groups. On the contrary, the PCYT2 gene was more highly expressed (+81%) in ENR. PISD expression only decreased in ENR. CERK also decreased its activity in the groups with cheese supplementation (−50% for CHE and −68% for ENR), while CEPT showed a significant decrease (−62%) in ENR.

Adipose Tissue
We observed a significant change in the expression of PLIN4, which showed a significant decrease (−81%) in the CHE and ENR groups.
Three genes involved in the regulation of inflammatory processes showed a significant differential expression in the CHE and ENR groups: TRP53 (−73% in the ENR group) and −78%, FGF21 (−88% in both experimental groups) and p21 (−81% in both experimental groups).
The effect of cheese in a high-energy diet was also observed in mitochondrial metabolism, where significantly lower expression was observed for the following genes: UCP2 (−67% in both experimental groups), NDUFS3 (−77% in the ENR group), UQCR10 (−38% and −95% in CHE and ENR, respectively), COX10 (−53% in the ENR group), and KLOTOB (−89% in both experimental groups).

Discussion
Three types of diets were compared: one was the standard pellet diet for mice (CON); the second consisted of CON with pecorino cheese (CHE), which constituted the most appropriate control; finally, the third consisted of the CON diet with pecorino cheese enriched with ω-3 and CLA (ENR). This approach was preferred to a simple addition of ω-3 mixed in the pellet as the added cheese is a dairy product already marketed and this approach should have allowed the mimicry, at least in part, of the consumption of cheese within a standard varied diet. Furthermore, the two diets supplemented with cheese were characterized by high levels of energy intake, so they could be considered as HFDs. This experimental model has therefore allowed us to verify whether supplementation with ω-3 PUFA and CLA is able to mitigate or suppress the effects of HFDs, which consists of an excessive caloric intake and a high amount of SFAs [20]. A first interesting result was the improvement of the blood lipid panel in the CHE and ENR diets, with a significant reduction in LDL cholesterol levels and an increase in HDL levels, which contributes to a reduction of cardiovascular risk in humans, as previously demonstrated by Pintus et al. [3]. The data show how the inclusion of cheese in a diet rich in fat allows the counteraction of the negative effects of a HFD, which are often associated with the development of metabolic disorders.

Lipid Profile
The lack of changes in the brain is justified by the composition of its lipid profile. It is known that the brain is an organ rich in PLs, which play an important role in regulating membrane fluidity [21]. The alteration of the PL profile (in particular, for the level of unsaturation) can affect this aspect by modifying the interaction with other contiguous cells, as well as enzymes and membrane receptor activity [22]. Therefore, there could be a metabolic regulation system, not yet demonstrated, which may counteract PL homeostasis due to external factors, such as diet. However, further investigations are necessary to deepen this regulation system. This would justify the lack of an effect of the diet on the normal lipid profile of the brain, while in the liver a tendency to reduce the level of TGs was observed in mice belonging to the ENR group, although the results were not significant. This trend agrees with Backes et al. [23], whose report stated that a diet enriched with ω-3 tends to decrease TGs in the liver. The lack of significance may be due to the particularities of hepatic metabolism, where there is a continuous flow of both endogenous and exogenous lipids, which can partly mask the effect reported by Backes et al. [23]; on the contrary, this reduction was seen in the muscle.
In skeletal muscle, we observed a significant reduction of TGs in the ENR group. This is linked to a reduction in lipogenesis activity and an increase in β-oxidation activity, as also noted by Backes et al. [23]. In particular, consumption of the ENR diet led to a significant increase in FFAs in the muscle, associated with a possible increase in lipolytic activity which increases the β-oxidation process in accordance with the results of Ibeagha-Awemu et al. [24] and Yan et al. [25]. Furthermore, we observed a significant decrease in FC in both liver and muscle and a simultaneous increase in EC in the liver of mice fed the ENR diet. The increase in hepatic EC may be associated, although this is not proven by experimental data, with an increase in the activity of the ACAT1 protein. This increase is associated with some positive effects for human health, first of all, the increase in EC, which is referred to a lower availability of FC for the production of LDL cholesterol in plasma [26]. Furthermore, a high level of cholesterol in the liver is also associated with a reduction in liver fibrosis, as demonstrated by Tomita et al. [27].
In all organs, we observed an increase in levels of SCFAs and MCFAs in the cheese-fed mice. These FAs are an indicator of the consumption of dairy products, particularly those resulting from the secretion of the mammary gland of ruminants (C4-C16). These FAs are synthesized exclusively in the mammary gland of ruminants thanks to a particular role of the mammary FASN gene, which is able to stop the synthesis of FAs completely at C16:0 and for which it can accumulate C8:0, C10:0 and C14:0 [2]. Other FAs that are significantly increased in the four organs with cheese consumption are all trans C18:1 isomers. These FAs are intermediate or final products of the rumen biohydrogenation process, and monogastric animals can only obtain them by consuming foods derived from ruminants [4]. The increased accumulation of these isomers in mice of the ENR group is due to the higher level of α-linolenic acid, which is one of the substrates of biohydrogenation. High levels of biohydrogenation led to the accumulation of VA and consequently to a proportional increase of CLA. This aspect is interesting because CLA is not present in CON diets, demonstrating the fact that if products of animal origin are not consumed, this important FA is not present along with all the beneficial activities that it permits. The mice of the ENR group ate, on average, about 0.4 g of CLA per day, an amount close to the values obtained to observe effective health effects. In fact, it has been shown that hamsters who ate 0.5 g of CLA per day for 12 weeks showed a reduction in aortic problems [28] and glucose tolerance [29].
An FA that showed a higher level in the ENR group was gamma-linolenic acid (GLA, C18:3n-6). This FA is an ω-6 found in human milk and several oilseeds [30]. Clinical studies have shown that its presence in the diet leads to an alteration of different inflammatory responses [30]. The increase in this fatty acid was observed only in the muscle (both in the TGs and in the PLs).

Gene Expression
Skeletal muscle, liver, brain and adipose tissue are the main site of whole-body FA and carbohydrate oxidation [31]. The relative contribution of these fuels to the energetic demands of the tissues is subject to complex regulation at multiple levels, including substrate availability, hormonal concentrations and the allosteric regulation of enzyme activities by intracellular metabolic intermediates [32]. In normal physiological conditions, glucose availability and flux exert the dominant effect on the oxidized fuel mix [33]. However, an HFD elevates the contribution of FAs to oxidative metabolism, especially in obese subjects [34]. In addition to their role as important oxidative substrates, dietary FAs regulate the expression of many genes [35]. Moreover, the response of those genes is central to the regulation of fatty acid transport and mitochondrial β-oxidation, whose actions are likely to be pivotal in the increased capacity to oxidize fatty acids. The consequence of oxidative stress is converging in the induction and regulation of inflammation, as described by Vitale et al. [36]. For this reason, we focused the analyses on particular genes, as reported in Table S1, concerning lipid metabolism, the mitochondrial oxidation system and the inflammation pathway.

Skeletal Muscle
An important aspect of SM lipid metabolism is the profile of PLINs, which represent the main proteins involved in the regulation of lipid droplet (LD) homeostasis as well as lipolysis and lipogenesis [37]. Recent findings show that modifications of PLIN expression dysregulate intracellular lipid deposition, lead to the accumulation of diacylglycerols or ceramides, impair cellular function and cause lipotoxicity [37]. Our data suggest that both the ENR and CHE diets do not affect the expression of PLIN2, which is considered the marker of LD and thus of the accumulation of fat in non-adipose tissues. Conversely, we revealed an effect on the expression of other proteins belonging to the PLIN family, such as PLIN3 and PLIN5. It is well known that PLIN3 and PLIN5 are involved in lipid oxidation and play a role in the regulation of lipolysis in skeletal muscle [38,39]. In particular, PLIN5 is a fundamental protein for the mobilization of FA, as it acts as a carrier for LD towards the mitochondria, where FAs undergo β-oxidation [38]. Although knowledge about PLIN3 in skeletal muscle is poorer with respect to PLIN5, recent studies suggest that PLIN3 plays a key role in regulating lipolysis by interacting with lipases, similarly to that found for PLIN5 [39]. PLIN5 expression decreased with both experimental diets, suggesting that an increase in dietary fat inhibits FA mobilization. The expression of PLIN3 is decreased only in CHE but not in ENR. It can therefore be hypothesized that supplementation with PUFA ω-3 and CLA induces the mobilization of FAs and therefore β-oxidation by the activity of PLIN3 to compensate for the failure of PLIN5.
An interesting effect was observed for CEPT-1 expression, which is involved in the Kennedy pathway for the biosynthesis of phosphatidylcholine and phosphatidylethalonamine [40]. The ENR diet reduced the expression of CEPT-1, and this probably can lead to an improvement in insulin sensitivity. Indeed, it has been observed that obese mice fed an HFD show reduced insulin sensitivity and increased expression of the CEPT-1 gene [40].
DGAT1 and DGAT2 are responsible for the synthesis and storage of TGs, and levels of expression decreased with the ENR diet. DGAT1 knockout mice have been shown to develop resistance to diet-induced obesity and have improved insulin sensitivity due to reductions in the expression levels of genes involved in lipid uptake and oxidation, thus preventing lipotoxicity [41]. The limited accumulation of triglycerides has also been shown to protect against fatty acid-induced lipotoxicity [42]. Moreover, we observed a reduced expression of ACACA and SCD genes in the ENR diet, suggesting reduced liponeogenesis. This is likely due to the introduction of essential FAs through the diet which are stored in the muscle instead of activating de novo synthesis. The inhibitory effect of PUFAs on liponeogenesis genes has been demonstrated in both mice and humans [43].
As regards the genes of mitochondrial metabolism, interesting data were observed for the UCP2 and UCP3 genes linked to the decoupling of substrate oxidation from ATP synthesis. Both genes showed increased expression in the ENR group compared to CHE. Decreased mitochondrial decoupling in SM has been shown to lead to increased oxidative stress and to the development and progression of certain diseases, such as T2D [44,45]; therefore, PUFA ω-3 could limit this damage by increasing the expression of genes for Ucp.
Both CHE and ENR groups reduced the expression levels of ceramide kinase (CERK) suggesting a reduction in inflammation levels [46]. CERK is involved in the regulation of ceramide phosphorylation and in the production of ceramide-1-phosphate (C1P), a sphingolipid which is commonly implicated in inflammation. A genetic ablation of CERK in mouse embryonic fibroblasts reduces the production of eicosanoids, well known for their involvement in pro-inflammatory processes [47]. This result is also confirmed by the lower expression of NFATC2 and PRKCQ genes in the ENR group, two genes closely related to inflammation [48]. A similar result was also obtained for the TRP53 gene, which showed lower expression in the ENR than in the CHE group. TRP53 is best known as the guardian of the genome, involved in DNA repair phenomena, apoptosis and, together with its p21 downstream, cell cycle blockade and cell senescence. Studies have shown that mice with high TRP53 activity are resistant to cancer but show signs of accelerated aging and die prematurely [49,50]. Regarding the correlation with inflammation, p53 and p21 are known to be potent inducers of cellular senescence, in turn a potent source of pro-inflammatory mediators (the so-called secretory phenotype associated with senescence, SASP). Furthermore, p53 regulates numerous processes related to energy and lipid metabolism, such as glycolysis, oxidative phosphorylation, lipolysis, lipogenesis, β-oxidation, gluconeogenesis and glycogen synthesis [51]. Its activation following obesity or a high-calorie diet would therefore contribute to the onset of inflammation [52]. The fact that the ENR diet induces a reduction of TRP53 expression compared to CHE suggests that ω-3 and CLA supplementation reduces the levels of inflammation induced by HFD.

Liver
In the liver, as regards the expression of PLINs, most interesting was the very high expression level of PLIN4 in both cheese-enriched diet groups, with a significantly higher expression level in ENR with respect to CHE. The role of PLIN4 in the liver, as well as in other organs, is not still clarified. However, a study performed in hepatocyte-specific PLIN2 knockout mice fed with a high-fat western diet (WTD), in agreement with our results, interestingly revealed increased protein expression levels of PLIN4 [53]. Moreover, another study in which C57BL/6 male mice were fed an HFD and simultaneously treated with matcha green tea for six weeks to prevent the excessive accumulation of hepatic lipids, showed higher levels of PLIN4 gene expression than mice fed only with an HFD, while mice fed with an HFD along with matcha green tea supplementation did not accumulate triglycerides in the liver concomitant to liver downregulation of PLIN4 expression levels [54]. Taken together, these results suggest that PLIN4 could play a critical role in liver lipid accumulation in the case of HFDs. However, at present, few data are available on PLIN4 in the liver and other experiments are necessary to understand the beneficial or detrimental role of PLIN4 in this organ.
Genes related to de novo synthesis of FAs, such as ACACA, ACACB, FASN, SREBP-1 and ACLY, showed reduced expression in ENR. Decreased expression of ACACA or ACACB is associated with liver cancer suppression, as demonstrated in rats [55]. In contrast, Nelson et al. [56] observed a completely opposite effect for ACACA and ACACB in mice with a double knockout. Downregulation of the FASN gene is also associated with positive effects on human health. Indeed, the high level of this gene is associated with several types of cancer, as well as chemoresistance and metastasis [57]. The lower expression of SREBP-1 in ENR was related to dietary PUFA ω-3, with a consequent reduction of liver damage, as demonstrated in a previous work [58]. Therefore, an ENR-associated decrease of SREBP-1 expression levels may be related to a beneficial effect of PUFA ω-3-enriched cheese in the liver. The higher level of PCK1 expression in the ENR group compared to CHE increases the health properties of PUFA ω-3-enriched cheese in the HFD diet. This gene is involved in hepatic gluconeogenesis to ensure that glucose production rates match the body's demands for glucose [59]. Recent studies have demonstrated a tumor suppressor role for PCK1 in hepatocellular carcinoma [50]. In particular, several studies have shown that PCK1 promotes cell death, especially in conditions of reduced glucose levels [60].
Moreover, a positive effect of PUFA ω-3-enriched cheese was the increase in the expression of PGC-1α, which was associated with a reduction in hepatic triacylglycerol levels [61].
Another interesting effect of the ENR diet was the lower expression of GDF15, which is the molecule that most correlates with age [62,63]. As is known, biological age is linked to several processes, such as mitochondrial dysfunction, oxidative stress, protein glycation, inflammation and hormonal changes. Many of these stresses are induced by increased expression of TPR53 or EGR-1, which in turn induce the expression of GDF-15 gene [62]. High levels of GDF-15 are associated with increased mortality linked to cardiovascular diseases, such as heart problems, coronary artery disease, atrial fibrillation, T2D and cognitive impairment [62,63]. Increased GDF-15 expression has been also associated with many cancers affecting the mammary gland, colon, pancreas and prostate. Therefore, the reduction in the expression of this gene in the diet with enriched PUFA ω-3 cheese could suggest a positive effect on the health of the consumer.
The positive effect of enriched cheese on the liver, and likely skeletal muscle, is also indicated by the reduction in expression levels of CERK, NFATC2 and PRKCQ, suggesting a positive role of PUFA ω-3 in counteracting inflammation. It is well known that a decrease of NFATC2 expression is associated with a reduction of the expression of cytokines in T cells, in particular IL-2, IL-3, IL-4 and TNF-alpha; the latter, in particular, is a pro-inflammatory cytokine associated with obesity and insulin resistance [64]. Lack of expression of this gene in mice is associated with insulin resistance even with an HFD [65]. Our data suggest the positive role of enriched PUFA ω-3 and CLA cheese in HFDs as well. At the same time, we also observed a reduction in PCYT2, the key enzyme in the biosynthesis of phosphoethanolamine in the Kennedy pathway. The biosynthesis of membrane phospholipids, including phosphoethanolamine, is of fundamental importance for cell growth. High levels of phosphoethanolamine plasmalogens have been observed to be associated with an increased predisposition to ovarian cancer [66]. A similar result was obtained for the TRP53 gene, whose expression levels were reduced more with the ENR diet compared to the levels shown with the CHE. The fact that the diet of the ENR group induces a reduction in p53 compared to the CHE group reinforces our hypothesis that ω-3 supplementation reduces the levels of inflammation induced by a high-fat diet.

Brain
In the brain we observed a significant reduction of TRP53, PRKCQ and CERK levels in the ENR group. In particular, it has been shown that CERK is involved in neurodegenerative diseases, inducing toxicity in rat astrocytes and neurons [67].
Regarding lipid metabolism, we observed a significant decrease of PLIN2 and PLIN3 expression in the ENR group. Very recently, it was demonstrated that PLIN2 in the brain is the only PLIN family member that is modulated with age and neuroinflammation, suggesting that PLIN2 may be connected with the brain, aging and inflammation [68]. Therefore, the decrease in PLIN2 expression in the ENR group suggests that PUFA ω-3 can prevent lipid accumulation and, consequently, possibly lipotoxicity and thus inflammation. It has been shown that lipotoxicity can also occur in the central nervous system in some neurodegenerative disorders [69]. Similarly, mice fed with the ENR diet showed a decrease in the expression level of SCD. A previous study has shown that high levels of SCD are associated with Alzheimer's disease, resulting in high levels of MUFAs [70]. The expression of the ACLY gene was also reduced in the ENR diet, with a healthy effect. Elevated levels of ACLY have been discovered in patients with brain tumors [71], so PUFA ω-3 could contribute to a reduction in the risk of developing these tumors. FASN, a key enzyme in de novo lipogenesis, is highly active in neural stem cells and its deletion impairs adult neurogenesis [72]. Its levels tended to increase in the diet of the ENR group compared to the CHE group, suggesting the positive effect of enriched cheese. As for mitochondrial metabolism genes, UCP2 showed lower levels of expression with the ENR diet. UCP2 deficiency has been shown to cause an increase in resistance to cerebral ischemia, with an increase in cerebral neuronal antioxidant status [73].

Adipose Tissue
We observed a significant reduction in the expression of the CERK and TRP53 genes in the ENR group, suggesting that PUFA ω-3 helps to counteract inflammation also in VAT. Unlike muscle, however, the expression of the NFATC2 and PRKCQ genes did not change. PLIN1 expression tended to decrease with the ENR diet, suggesting that PUFA ω-3 prevents lipid accumulation despite the HFD. DGAT2 showed a reduction in expression in ENR compared to CHE. Inhibition of triglyceride synthesis can improve health status associated with obesity, since endogenous synthesis of triglycerides leads to an increase in the accumulation of lipids in adipose tissue [74]. Among the genes involved in lipogenesis, ACACB and FASN are the genes that have shown greater variability with HSD. ACACB, which regulates the oxidation of FAs, showed a lower expression in the ENR group. ACACB knockout mice have been shown to be protected from HFD-induced obesity and T2D through increased FA and glucose oxidation activity and increased lipolysis, thus promoting the maintenance of insulin sensitivity [75]. On the contrary, the expression of FASN only increased in ENR. This increase is probably due to the simultaneous increase of SREBP-1 expression, a transcription factor that regulates the synthesis of fatty acids [76]. Overexpression of SREBP-1 has been shown to activate genes linked to the FA biosynthesis pathway, leading to an increase in the accumulation of lipids in adipocytes, resulting in hyperglycemia and hypertrophy that causes an inflammatory reaction [77].

Conclusions
This study highlighted two important phenomena: (i) ω-3and CLA-enriched cheese in an HFD reduces the expression of genes involved in inflammation and (ii) promotes metabolism involved in the prevention of cancer and cardiovascular disease. Moreover, our study shows how a diet supplemented with ω-3 fatty acids-enriched cheese determines changes in gene expression, suggesting an overall nutraceutical effect in all the tissues analyzed. Considering the effects of inflammation as a consequence of high-calorie and high-fat diets (metaflammation) or being associated with aging (inflammaging) on the onset of several chronic degenerative diseases, these data are potentially of great clinical interest, as they suggest the possibility of avoiding/delaying these pathologies (and aging itself) by means of a nutritional and non-pharmacological intervention. Further studies are needed to confirm these results, including longitudinal studies to assess the effects of such dietary supplementation on animal lifespans and disease onset.