Protective Effects of Eicosapentaenoic Acid Plus Hydroxytyrosol Supplementation Against White Adipose Tissue Abnormalities in Mice Fed a High-Fat Diet

Objective: Obesity induced by high-fat diet (HFD) elicits white adipose tissue dysfunction. In this study, we have hypothesized that the metabolic modulator eicosapentaenoic acid (EPA) combined with the antioxidant hydroxytyrosol (HT) attenuates HFD-induced white adipose tissue (WAT) alterations. Methods: C57BL/6J mice were administered with a HFD (60% fat, 20% protein, 20% carbohydrates) or control diet (CD; 10% fat, 20% protein, 70% carbohydrates), with or without EPA (50 mg/kg/day), HT (5 mg/kg/day), or both for 12 weeks. Determinations in WAT include morphological parameters, EPA and docosahexaenoic acid content in phospholipids (gas chromatography), lipogenesis, oxidative stress (OS) and inflammation markers, and gene expression and activities of transcription factors, such as sterol regulatory element-binding protein-1c (SREBP-1c), peroxisome proliferator-activated receptor-gamma (PPAR-γ), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) (p65 subunit) and nuclear factor erythroid 2-related factor 2 (Nrf2) (quantitative polymerase chain reaction and enzyme linked immunosorbent assay). Results: HFD led to WAT hypertrophy in relation to PPAR-γ downregulation. WAT metabolic dysfunction was characterized by upregulation of lipogenic SREBP-1c system, mitochondrial energy metabolism depression, loss of the antioxidant Nrf2 signaling with OS enhancement, n-3 long-chain polyunsaturated fatty acids depletion and activation of the pro-inflammatory NF-κB system. EPA and HT co-supplementation diminished HFD-dependent effects additively, reaching values close or similar to controls. Conclusion: Data presented strengthen the importance of combined protocols such as EPA plus HT to attenuate metabolic-inflammatory states triggered by obesity.

assay (Cayman Chemical Company, Ann Arbor, MI, USA) after adjusting the total protein concentration to 7.5 mg/mL per sample. Reduced glutathione (GSH) and glutathione disulfide (GSSG) contents in EWAT were assessed with an enzymatic recycling method described by Rahman et al. [26].

Activities of Antioxidant and Lipogenic Enzymes
Samples of EWAT were homogenized in three volumes of phosphate buffer (30 mM), pH 7.4 containing EDTA (1 mM) and sucrose (250 mM) for determination of antioxidant enzyme activities. The homogenate was subjected to a first centrifugation (1200× g, 10 min, 4 • C) and one supernatant aliquot was used to evaluate superoxide dismutase (SOD) (Cayman Chemical Co., Ann Arbor, MI, USA) and catalase (CAT) activities [27]. Subsequently, the remaining supernatants were centrifuged at 100,000× g for 60 min at 4 • C to perform glutathione peroxidase (GPX) and glutathione reductase (GR) assays [27]. On the other hand, EWAT samples were homogenized in three volumes of buffer, pH 7, containing KH 2 PO 4 (9 mM), K 2 HPO 4 (85 mM), dithiothreitol (DTT) (1 mM) and KHCO 3 (70 mM) and centrifuged (1800× g, 10 min, 4 • C). The fat of the upper layers were discarded, the supernatants were again centrifuged at 100,000× g for 60 min at 4 • C and the aqueous fractions were used to study Acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS) and the glucose-6-phosphate dehydrogenase (G6PD) activities as previously described [12,14]. To assay malic enzyme (ME) activity, EWAT homogenates were centrifuged for 10 min at 40,000× g at 4 • C, then the ME activity was evaluated in the resulting supernatant following the method performed by Wise and Ball [28]. In addition, the activity of lipoprotein lipase (LPL) (enzyme associated with the tissue incorporation of fatty acids) was measured in an epididymal fat pad according to Llobera et al. [29]. The protein content in enzyme extracts was determined by the Bradford assay (Bio-Rad reagent, California, USA).

Quantitative Real-Time PCR
Total RNA was isolated from EWAT samples using Trizol (Invitrogen, Carlsbad, CA, USA), according to the supplier's protocols. Purified RNA (2 µg) was then treated with DNAase I (DNA free kit; Ambion, Austin, TX, USA) and used to generate first-strand cDNA with moloney murine leukemia virus reverse transcriptase (Invitrogen, Carlsbad, CA, USA), utilizing random hexamers (Invitrogen, Carlsbad, CA, USA) and deoxynucleotide triphosphate (dNTP) mix (Bioline, London, United Kingdom), according to the manufacturer's protocol. The resultant cDNA was amplified with specific primers for mice in a total volume of 20 µL. Gene-specific primer sequences used are shown in Table 1. Primers were optimized to yield 95%-100% of reaction efficiency with PCR products by development in agarose gel to verify the correct amplification length. The quality of the DNA extracted was checked by agarose gel (1%) electrophoresis according to Lee et al. [30]. Briefly, samples were mixed with Gel loading dye (0.25% bromophenol blue, 0.25% xylene cyanol and 30% glycerol). Electrophoresis was run at 75 Volt for 20 min. DNA was stained using GelRed ® Nucleic Acid Gel Stain (Biotum, Fremont, CA, USA). Real-Time PCR was performed in a Stratagen Mx3000P System using Brilliant II SYBR ® Green quantitative real time PCR Master Mix (Agilent Technologies, La Jolla, CA, USA) following the manufacturer's recommendations (Applied Biosystems, Foster City, CA, USA). All the expression levels of target genes under study were normalized by the expression of β-actin as an internal control (Applied Biosystems, Foster City, CA, USA). Fold changes between groups were calculated by the 2-(∆∆Ct) method, as established by Pfaffl [31] and Livak and Schmittgen [32]. Sequences are listed in the 5 → 3 direction. Nuclear factor erythroid 2-related factor 2 (Nrf2), glutathione S-transferase (GST), glutamate-cysteine ligase (GCL), sterol regulatory element-binding protein-1c (SREBP-1c), fatty acid synthase (FAS), acetyl-CoA carboxylase (ACC), peroxisome proliferator-activated receptor-gamma (PPAR-γ), Fatty acid transport protein 1 (FATP1), lipoprotein lipase (LPL), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), beta actin (β-actin).

Statistical Analysis
Statistical analysis was performed with GraphPad Prism 6.0 software (GraphPad Prism, Inc., San Diego, USA). The values shown represent the mean ± standard deviation (SD) for 8-10 animals per each experimental group. Evaluation of normality data distribution was performed using the Shapiro-Wilk test. Assessment of the statistical significance of differences between mean values was performed by two-way analysis of variance (ANOVA) and the Tukey's test. A p < 0.05 was considered significant.

HFD-Induced Obesity, Adiposity and WAT Hypertrophy are Decreased by EPA and HT Co-Administration
The body weight (BW) of mice at the beginning of the study was similar. After 12 weeks, all groups fed with CD, with or without supplementation, exhibited comparable final BW. HFD led to an increase in BW gain of 65% (p < 0.05), compared with CD groups, which is related with to augmented adiposity (Table 2A, group e). In fact, all fat pads evaluated, namely, subcutaneous, perirenal and epididymal, in mice fed the HFD showed significant weight increases (Table 2B), therefore the total mass of WAT increased considerably (198%; p < 0.05) over the CD group. Dietary intake was comparable in all experimental groups, but energy consumption was higher in mice subjected to the HFD with or without supplementation over CD groups (Data not shown). EPA and EPA + HT supplementation diminished BW gain produced by the HFD by 30% and 37% respectively (p < 0.05), accompanied with a decrease in fat pads and total WAT of 34% and 32% (p < 0.05), respectively (Table 2B). HT administration was not able to prevent the increased BW observed in HDF-fed mice; however, it reduced the localized and total adiposity (total WAT) by 28% (Table 2B). Specific and morphological parameters were evaluated in the EWAT (Table 2C). The expansion of epididymal fat pad of HFD-fed mice exhibited a lower number of cells per gram of WAT, adipocytes of increased size with augmented TAG content compared with the control group, indicating hypertrophy. The supplementation with EPA, HT or both significantly decreased the hypertrophy observed in WAT of the HFD-fed mice, improving all the parameters evaluated without observing differences between treatments and without reaching control values (Table 2C). Levels of serum adipokines are shown in the Table 2D. Serum adiponectin and leptin were not modified by supplementation when mice were fed the CD. However, the HFD significantly decreased serum adiponectin levels (39%; p < 0.05) while increasing those of leptin (136%; p < 0.05). Supplementation with EPA, HT or EPA + HT improved the profile of serum adipokines in HFD-fed mice, showing a similar effect independently of the treatment, although the values did not reach those observed in CD-fed mice.

HFD-Induced Changes in Fatty Acid Composition in EWAT Phospholipids are Improved by EPA and HT Co-Supplementation
Comparable content of total saturated fatty acids (SFA), monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA) were found in EWAT phospholipids of all groups receiving CD. HFD-fed mice exhibit an increase of total SFA and decrease of total PUFA (Table 3). Regarding the EWAT palmitate levels, comparable values were observed in all groups fed with CD, a significant increment in HFD saline group, and a decrease caused by the administration of EPA or HT alone, with the total recovery towards CD values when supplemented with EPA+HT (Table 3). Moreover, HFD led to a severe reduction in EPA content (81%; p < 0.05) with consequent decrease in DHA content (72%; p < 0.05) in EWAT phospholipids compared to CD-fed animals ( Table 3). The high content of EPA in the adipose tissue of the animals that received the supplementation evidence the effectiveness of the treatment, increasing by 2.86-fold and 3-fold EPA levels in CD groups supplemented with EPA and EPA+HT, respectively. Moreover, the supplementation with EPA to HFD-mice (EPA and EPA+HT groups) restored the depleted levels of EPA in EWAT phospholipids reaching control levels. The incorporation of EPA into EWAT also improved the content of DHA in this tissue, increasing DHA levels in CD animals, with normalization in HFD mice achieving levels observed in CD group (Table 3). It should be noted that in the HFD-fed group that received HT exclusively, the levels of EPA and DHA were also increased in EWAT phospholipids, confirming previous results [18,20]. Surprisingly, the effect of the supplements given together, improved the tissue content of EPA and DHA even more than when each administered individually in animals fed HFD. Furthermore, EPA reduced the n-6/n-3 ratio EWAT of CD animals, and ameliorate this ratio in both groups of mice fed HFD that received EPA supplementation. Also, HT given alone diminished the ratio n-6/n-3 in EWAT of HFD-mice, an effect that was more pronounce after EPA and HT co-supplementation (Table 3). Table 2. General and metabolic parameters in control diet (CD) and high-fat diet (HFD)-fed mice subjected to eicosapentaenoic acid (EPA), hydroxytyrosol (HT) or EPA + HT supplementation.  Values are expressed as mean ± SD; eight to ten mice were included in each experimental group. Significant differences between the groups are indicated by the letter identifying each Values are expressed as mean ± standard deviation (SD). Eight to ten mice were included in each experimental group. Significant differences between the groups are indicated by the letter identifying each group, namely, (a Monounsaturated fatty acids (MUFAs) correspond to C14:1n-7, C16:1n-7 and C18:1n-9. Polyunsaturated fatty acids (PUFAs) correspond to C18:2n-6, C18:3n-3, C20:4n-6, C20:5n-3, C22:5n-3 and C22:6n-3. n-6 LCPUFAs are C20:4n-6 and C22:5n-3, n-3 LCPUFAs are C20:5n-3, C22:5n-3 and C22:6n-3, n-6/n-3 LCPUFAs ratio: C20:4n-6/(C20:5n-3 + C22:5n-3 + C22:6n-3).

HFD-Induced EWAT Oxidative Stress Enhancement is Normalized by EPA and HT Co-Administration
The treatments showed no effect on any of the parameters associated with oxidative stress evaluated in EWAT of CD-fed mice ( Figure 1). HFD feeding induced oxidative stress in EWAT, as evidenced by the increase in markers of oxidative damage TBARS ( Figure 1A), F2-isoprostanes ( Figure 1B) and protein carbonyls ( Figure 1C) accompanied by alteration of the glutathione status ( Figure 1D-G). Supplementation with EPA alone only reduced levels of protein carbonyls ( Figure 1C), without improving the rest of the parameters evaluated in EWAT of HFD animals ( Figure 1A,B,D-G). When HT was given alone, TBARS and F2-isoprostanes levels were reduced ( Figure 1A,B), and the content of protein carbonyls ( Figure 1C) and glutathione levels ( Figure 1D-G) were normalized, reaching control values. Notably, the administration of both supplements, EPA + HT, prevented oxidative damage in all parameters evaluated and deterioration of glutathione status in EWAT of mice receiving the HFD ( Figure 1A-D).

HFD-Induced EWAT Oxidative Stress Enhancement is Normalized by EPA and HT Co-Administration
The treatments showed no effect on any of the parameters associated with oxidative stress evaluated in EWAT of CD-fed mice (Figure 1). HFD feeding induced oxidative stress in EWAT, as evidenced by the increase in markers of oxidative damage TBARS ( Figure 1A), F2-isoprostanes ( Figure 1B) and protein carbonyls ( Figure 1C) accompanied by alteration of the glutathione status ( Figure 1D-G). Supplementation with EPA alone only reduced levels of protein carbonyls ( Figure  1C), without improving the rest of the parameters evaluated in EWAT of HFD animals ( Figure  1A,B,D-G). When HT was given alone, TBARS and F2-isoprostanes levels were reduced ( Figure 1A and B), and the content of protein carbonyls ( Figure 1C) and glutathione levels ( Figure 1D-G) were normalized, reaching control values. Notably, the administration of both supplements, EPA + HT, prevented oxidative damage in all parameters evaluated and deterioration of glutathione status in EWAT of mice receiving the HFD ( Figure 1A-D).  The fall in the mRNA levels and the DNA binding capacity impaired of the cytoprotective transcription factor Nrf2 in EWAT of HFD-fed mice were partially recovered when animals received EPA or HT supplementation (Figure 2A,B). Likewise, the HFD-induced diminution in the expression levels of its downstream genes glutamate-cysteine ligase (GCL) ( Figure 2C) and glutathione-S-transferase (GST) ( Figure 2D) involved in glutathione metabolism were equally improved by EPA or HT. The supplementation with EPA plus HT in HFD mice showed an even greater effect, restoring the expression levels of Nrf2, GCL and GST to control values, and normalizing the DNA binding activity of Nrf2 (Figure 2A-D). A similar behavior was observed in the activity of antioxidant enzymes CAT ( Figure 2E), SOD ( Figure 2F), GR ( Figure 2G) and GPX ( Figure 2H) in EWAT of HFD-mice that received EPA, HT or both supplements, which were partially enhanced by EPA or HT after reduction by HFD and totally recovered by combined EPA and HT ( Figure 2D-H).
Molecules 2020, 25, x FOR PEER REVIEW 10 of 20 The fall in the mRNA levels and the DNA binding capacity impaired of the cytoprotective transcription factor Nrf2 in EWAT of HFD-fed mice were partially recovered when animals received EPA or HT supplementation (Figure 2A,B). Likewise, the HFD-induced diminution in the expression levels of its downstream genes glutamate-cysteine ligase (GCL) ( Figure 2C) and glutathione-S-transferase (GST) ( Figure 2D) involved in glutathione metabolism were equally improved by EPA or HT. The supplementation with EPA plus HT in HFD mice showed an even greater effect, restoring the expression levels of Nrf2, GCL and GST to control values, and normalizing the DNA binding activity of Nrf2 (Figure 2A-D). A similar behavior was observed in the activity of antioxidant enzymes CAT ( Figure 2E), SOD ( Figure 2F), GR ( Figure 2G) and GPX ( Figure 2H) in EWAT of HFD-mice that received EPA, HT or both supplements, which were partially enhanced by EPA or HT after reduction by HFD and totally recovered by combined EPA and HT ( Figure 2D-H).

HFD-Induced Upregulation of SREBP-1c System and Downregulation of PPAR-γ Pathway in EWAT are Partially Recovered by the Combined EPA and HT Protocol
The EWAT of mice subjected to HFD exhibit a significant increase in DNA-binding activity ( Figure 3A) and mRNA levels of SREBP-1c ( Figure 3B). The activation of SREBP-1c was accompanied by higher gene expression and activities of lipogenic enzymes ACC ( Figure 3E,F) and FAS ( Figure 3C,D) and higher activities of G6PD ( Figure 3G) and ME ( Figure 3H) compared to CD values. In addition, mRNA levels of enzyme LPL were decreased, whereas their activity increased in EWAT of HFD-fed mice ( Figure 4D,E). The administration of EPA or HT supplied alone to HFD mice led to a partial recovery of all the lipogenic parameters evaluated in EWAT (Figure 3) including the LPL enzyme ( Figure 4D,E), showing comparable results with both supplements, except for the FAS activity ( Figure 3G). The effect of EPA and HT given together led to a significantly greater recovery of SREBP-1c ( Figure 3A,B) and its target genes FAS, ACC, G6PD, ME ( Figure 3C-H) and LPL ( Figure 4D,E), although without achieving values observed in EWAT of CD mice.

HFD-Induced Upregulation of SREBP-1c System and Downregulation of PPAR-γ Pathway in EWAT are Partially Recovered by the Combined EPA and HT Protocol
The EWAT of mice subjected to HFD exhibit a significant increase in DNA-binding activity ( Figure 3A) and mRNA levels of SREBP-1c ( Figure 3B). The activation of SREBP-1c was accompanied by higher gene expression and activities of lipogenic enzymes ACC ( Figure 3E,F) and FAS ( Figure 3C,D) and higher activities of G6PD ( Figure 3G) and ME ( Figure 3H) compared to CD values. In addition, mRNA levels of enzyme LPL were decreased, whereas their activity increased in EWAT of HFD-fed mice ( Figure 4D,E). The administration of EPA or HT supplied alone to HFD mice led to a partial recovery of all the lipogenic parameters evaluated in EWAT (Figure 3) including the LPL enzyme ( Figure 4D,E), showing comparable results with both supplements, except for the FAS activity ( Figure 3G). The effect of EPA and HT given together led to a significantly greater recovery of SREBP-1c ( Figure 3A,B) and its target genes FAS, ACC, G6PD, ME ( Figure   transport protein 1 (FATP1) ( Figure 4C) were decreased in EWAT of HFD-fed mice compared to the control group. Supplementation whit EPA or HT alone similarly improved the activity and mRNA levels of PPAR-γ and gene expression of FATP1. Moreover, EPA plus HT supplementation led to a significant increase in the binding activity and gene expression of PPAR-γ and in the gene expression of the FATP1 transporter compared to groups that received EPA or HT, without reaching CD values ( Figure 4A-C). Mice subjected to CD with or without supplementation showed comparable results in all parameters evaluated in EWAT mentioned in Figures 3 and 4.

HFD-Induced Activation of Transcription Factor NF-κΒ and Expression of Target Genes Associated with Inflammation in EWAT are Diminished to Values over CD by the EPA Plus HT Protocol
The DNA-binding activity and the gene expression of the transcription factor NF-κΒ were significantly increased in EWAT of mice fed with the HFD compared to those that received the CD ( Figure 5A,B). In parallel, the mRNA levels of the pro-inflammatory cytokine tumor necrosis factor alpha (TNF-α) and interleukin 6 (IL-6) were increased in EWAT of HFD-fed mice ( Figure 5C,D). In the HFD-fed mice that received EPA or HT, the DNA-binding activity of NF-κΒ and the mRNA levels of NF-κΒ, TNF-α and IL-6 were moderately improved, with comparable results between both groups, respectively. The group that received EPA plus HT showed a significantly greater decrease in these values in comparison to the groups supplemented with EPA or HT alone, without reaching those of the control group ( Figure 5). The DNA-binding activity ( Figure 4A) and mRNA levels of PPAR-γ ( Figure 4B) and fatty acid transport protein 1 (FATP1) ( Figure 4C) were decreased in EWAT of HFD-fed mice compared to the control group. Supplementation whit EPA or HT alone similarly improved the activity and mRNA levels of PPAR-γ and gene expression of FATP1. Moreover, EPA plus HT supplementation led to a significant increase in the binding activity and gene expression of PPAR-γ and in the gene expression of the FATP1 transporter compared to groups that received EPA or HT, without reaching CD values ( Figure 4A-C). Mice subjected to CD with or without supplementation showed comparable results in all parameters evaluated in EWAT mentioned in Figures 3 and 4.

HFD-Induced Activation of Transcription Factor NF-κB and Expression of Target Genes Associated with Inflammation in EWAT are Diminished to Values over CD by the EPA Plus HT Protocol
The DNA-binding activity and the gene expression of the transcription factor NF-κB were significantly increased in EWAT of mice fed with the HFD compared to those that received the CD ( Figure 5A,B). In parallel, the mRNA levels of the pro-inflammatory cytokine tumor necrosis factor alpha (TNF-α) and interleukin 6 (IL-6) were increased in EWAT of HFD-fed mice ( Figure 5C,D). In the HFD-fed mice that received EPA or HT, the DNA-binding activity of NF-κB and the mRNA levels of NF-κB, TNF-α and IL-6 were moderately improved, with comparable results between both groups, respectively. The group that received EPA plus HT showed a significantly greater decrease in these values in comparison to the groups supplemented with EPA or HT alone, without reaching those of the control group ( Figure 5).

Discussion
Diet-induced obesity (DIO) in mice represents a model of choice for preclinical nutritional studies [33][34][35]. HFD feeding comprising 60% of the total calories as fat for 12 weeks achieved significant BW gain with concomitant liver and WAT dysfunction. In the case of the liver, HFD induced steatosis in association with n-3 LCPUFAs depletion, oxidative stress with loss of antioxidant defenses [20], a pro-inflammatory state without histological inflammation [17] and a decline in mitochondrial energy metabolism [18]. The drastic metabolic alterations induced by HFD are evidenced by the deranged levels of (i) serum-free fatty acids, the adipokines adiponectin and leptin, and the cytokines TNF-α and IL-6, which are associated with IR development [13], (ii) the development of adipocyte hypertrophy and enhancement in total tissue mass, with higher cell size and TAG content and lower cell number per g of tissue over control values and (iii) the derangement of metabolic parameters in WAT favoring oxidative stress, lipogenesis, mitochondrial dysfunction and inflammation, features that agree with a previous report [13] and resemble those elicited in the liver [17,18,20].
Metabolic alterations can be prevented or attenuated by the combination of active substances, including n-3 LCPUFAs, antioxidants and insulin sensitizers, as reported for DHA, EPA, HT, rosiglitazone and astaxantin [17,18,36,37]. In the present study, EPA and HT co-administration diminished BW gain in HFD-fed mice over CD animals, an effect that was accompanied by a partial

Discussion
Diet-induced obesity (DIO) in mice represents a model of choice for preclinical nutritional studies [33][34][35]. HFD feeding comprising 60% of the total calories as fat for 12 weeks achieved significant BW gain with concomitant liver and WAT dysfunction. In the case of the liver, HFD induced steatosis in association with n-3 LCPUFAs depletion, oxidative stress with loss of antioxidant defenses [20], a pro-inflammatory state without histological inflammation [17] and a decline in mitochondrial energy metabolism [18]. The drastic metabolic alterations induced by HFD are evidenced by the deranged levels of (i) serum-free fatty acids, the adipokines adiponectin and leptin, and the cytokines TNF-α and IL-6, which are associated with IR development [13], (ii) the development of adipocyte hypertrophy and enhancement in total tissue mass, with higher cell size and TAG content and lower cell number per g of tissue over control values and (iii) the derangement of metabolic parameters in WAT favoring oxidative stress, lipogenesis, mitochondrial dysfunction and inflammation, features that agree with a previous report [13] and resemble those elicited in the liver [17,18,20].
Metabolic alterations can be prevented or attenuated by the combination of active substances, including n-3 LCPUFAs, antioxidants and insulin sensitizers, as reported for DHA, EPA, HT, rosiglitazone and astaxantin [17,18,36,37]. In the present study, EPA and HT co-administration diminished BW gain in HFD-fed mice over CD animals, an effect that was accompanied by a partial WAT reduction and EWAT cellularity preservation similar to those observed in the separate supplementations. In agreement with these findings, replacement of dietary lipids in a HFD by 1% to 12% (wt/wt) of EPA plus DHA (DHA > EPA) limited the BW gain and EWAT accumulation in C57BL/6J HFD-fed mice, with similar results being found by replacing a HFD rich in ALA or linoleic acid (C18:2n-6, LA) by mixtures of EPA and DHA at different ratios [38]. The effect of EPA on obesity may be partly mediated by binding to G protein-coupled receptor 120 (GPR120) present in adipocytes and adipocyte tissue macrophages (ATMs) [39], thus regulating adipogenesis, glucose uptake, inflammation and insulin sensitivity [40]. EPA binding and activation of GPR120 promotes several effects, including (i) the interaction with PPAR-γ that favors the differentiation of pre-adipocytes into white adipocytes [40], (ii) the binding to β-arrestin2 and their interaction with transforming growth factor-β-activated kinase binding protein 1 (TAB1) that blocks NF-κB and c Jun N-terminal kinase (JNK) functioning, with inhibition of the low-grade inflammation due to TNF-α and IL-6 production by ATMs triggered by obesity [41] and (iii) the release of fibroblast growth factor 21 (FGF21), resulting in the promotion of brown adipose tissue (BAT) activity and WAT browning [42], with stimulation of mitochondrial energy metabolism depressed by HFD. It is important to note that EPA supplementation in HFD-fed mice upregulates the expression of PPAR-α and the micro-RNAs miR-455 and miR-129-5p in BAT [43]. These mediators enhance the expression of thermogenic markers that stimulate mitochondrial FAO and energy expenditure, pointing to BAT as a crucial anti-obesity target [43]. This contention is further supported by data showing that HT prevents visceral adipogenesis and decreased metabolic activity in BAT and WAT, induced by the exposure to fine particulate matter in mice [44]. The contribution of EPA to the induction of a healthy adipocyte phenotype may also involve its direct binding to PPAR-γ, whose expression and activity are diminished by HFD and partially rescued by EPA. Activation of PPAR-γ induces adipogenesis as an important mechanism of WAT remodeling, which is accomplished through downregulation of SREBP-1c functioning [9,39]. The HT component of the combined protocol elicited diminution of WAT lipid accumulation by regulating genes related to lipogenesis and FAO, but mainly due to the improvement of the cellular antioxidant status that preserves n-3 LCPUFAs loss by HFD [13,45]. The latter mechanism of HT action is exerted directly as a free-radical scavenger or by recovering the Nrf2 system that induces antioxidant enzymes (SOD, CAT, GPX) and those associated with glutathione repletion (GCL, GR) or utilization (GST). At the doses used in our study, the combined EPA and HT administration did not exhibit more potentiated effects on WAT morphology parameters and lipid content than those of the individual compounds. However, the parameters related to WAT metabolic dysfunction are diminished in an additive manner, considering that the specific effects exerted by EPA or HT exhibit convergence in the combined protocol.
In agreement with previous data, WAT expansion in HFD-fed mice led to a significant decrease of serum levels of adiponectin and an increase in those of leptin [13], which are correlated with IR [46] and are partially restored by EPA and HT single or combined supplementations. In this context, the administration of EPA is known to enhance the gene expression and release of adiponectin from adipocytes, either in EWAT from HFD-fed mice [47] or in lean and overweight rats [48], thus evidencing the insulin-sensitizing properties of EPA [17,47,48]. Interestingly, the administration of EPA to obese subjects elevated adiponectin levels without altering BW, suggesting direct involvement of EPA on adiponectin secretion [49]. This conclusion is also supported by EPA supplementation to HFD-fed mice showing maintenance of normal adiponectin levels without diminishing adiposity, pointing to adiponectin as a major factor in EPA-mediated prevention and reversal of IR [50]. Regarding the molecular mechanisms underlying these effects, EPA increase the expression of adiponectin through binding to PPAR-γ, promoting the expression of FGF21 and the activation of AMP-activated protein kinase (AMPK), the energy sensor that favors FAO over lipogenesis [15,42,47,51]. The EPA-induced AMPK activation by phosphorylation in 3T3-L1 adipocytes [51] is in agreement with in vivo studies showing that n-3 LCPUFAs increase the content of WAT AMPK α1 subunit and its phosphorylation status, in association with an enhancement in adiponectin levels [52,53]. In the case of leptin, the enhancement in its serum levels by HFD feeding in mice and their diminution by EPA (Table 2D) agree with studies in humans and rodents [54], suggesting that improved leptin sensitivity concomitantly with decreased production would be secondary to the anti-inflammatory effect of n-3 LCPUFAs [55]. Regarding HT, several preclinical studies reported higher circulating levels of adiponectin with lower leptin values than controls achieved by this polyphenol [13,24,45,56].
Improved expression and activity of the PPAR-γ gene by co-administration of EPA and HT also leads to improved expression of the target genes FATP1 and LPL, an effect that is additive over the individual supplementations after HFD feeding. In particular, upregulation of WAT FATP1 may contribute to the smaller healthy phenotype of adipocytes due to its FA transport function across cell membranes and its acyl-CoA synthase activity to activate FAs for further utilization [57]. LPL, however, exhibited opposite changes regarding its mRNA expression and activity, which may be related to different mechanisms regulating LPL at transcriptional and post-translational levels [13]. The small healthy adipocyte phenotype is also associated with downregulation of key lipogenic genes in EPA-HT-HFD-treated mice. These include genes coding for transcription factor SREBP-1c that controls the expression of the lipogenic enzymes ACC and FAS and the reduced nicotinamide adenine dinucleotide phosphate (NADPH) generating systems G6PD and ME that are required for FA synthesis. The anti-lipogenic effect of the EPA plus HT protocol is related to the repletion of n-3 LCPUFAs due to lower oxidative deterioration by the normoxic state achieved, which downregulates lipogenic SREBP-1c functioning [18]. This contention is supported by previous studies showing (i) diminution in the lipogenic enzyme activities in WAT of sucrose-rich diet-fed rats subjected to fish oil administration [58], (ii) the decrease in the expression of the lipogenic enzyme glycerol-3-phosphate dehydrogenase in EPA-treated 3T3-L1 adipocytes [50] and (iii) the diminished SREBP-1c functioning with prevention of DIO in adipocytes from HFD-fed mice supplemented with EPA [59]. Additionally, HFD-induced prolipogenic state in EWAT with n-3 LCPUFAs depletion are attenuated by HT supplementation, which may preserve the activity of delta-5 and delta-6 desaturases that contribute to n-3 LCPUFAs synthesis, as previously reported in the liver [24], downregulate SREBP-1c or reduce oxidative stress [13]. A major hallmark in the expansion of WAT in obesity is the enhancement in the adipocyte oxidative stress status, which underlies significant mitochondrial dysfunction. This is evidenced by the decreases in the activities of the carboxylic acid cycle regulatory enzyme citrate synthase and the respiratory chain complexes I and II [60], a condition that favors energy deprivation and mitochondrial reactive oxygen species (ROS) production, thus contributing to the attainment of a hypertrophic adipocyte phenotype [61,62]. Data presented indicate that HFD coupled to EPA plus HT supplementation totally avoided oxidative damage to biomolecules and the depletion of antioxidant systems in WAT. This finding mainly relies on the HT component of the combined protocol and may effectively contribute to the hypertrophic to the small healthy adipocyte phenotype transition observed. The reduction of the enlarged WAT in obesity achieved by EPA plus HT is in agreement with studies on the antioxidant behavior exerted by EPA, DHA and the carotenoid astaxantin, alone or in combinations in HepG2-C8 cells [37], EPA and DHA in 3T3-L1 adipocytes and Schwann cells [63,64] or in human macrophages [65], with responses exhibiting synergism. The EPA plus HT protocol also improved the pro-inflammatory status of WAT in HFD-fed mice in an additive manner. Possible mechanisms involved include: (i) disruption of the SFA inflammatory signal in Toll-like receptor 4 that activate NF-κB, (ii) PPAR-γ activation that interferes with the translocation of NF-κB to the nucleus, (iii) binding to GPR120 and prevention of NF-κB activation, (iv) decreased pro-inflammatory adipokine expression and macrophage accumulation, promoting repolarization towards the M2 anti-inflammatory phenotype by SPMs [9,66,67] and (v) preservation of WAT n-3 LCPUFAs and downregulation of NF-κB by HT [13,44].

Conclusions
HFD feeding in mice drastically impacts WAT functioning through the induction of hypertrophy and metabolic dysfunction. WAT hypertrophy involves higher cell size with TAG accumulation and lower cell number, which is related to EPA depletion that blunts PPAR-γ operation. Metabolic dysfunction includes (i) upregulation of the lipogenic SREBP-1c system, (ii) loss of mitochondrial energy metabolism and (iii) downregulation of the antioxidant Nrf2 system, leading to oxidative stress and the secondary n-3 LCPUFAs depletion and activation of the redox-sensitive pro-inflammatory NF-κB system. These effects of HFD were significantly diminished by the co-supplementation with EPA and HT additively, with resulting values being close or similar to those observed for the CD-fed animals. These observations strengthen the contention that EPA and HT co-administration is an important strategy to attenuate obesity-driven metabolic inflammation states. The data reported concerning PPAR-γ upregulation by EPA and HT co-administration may also be of importance as a possible adjuvant therapy to diminish the severity of SARS-CoV-2 infection [68], considering the anti-inflammatory response involving PPAR-γ binding to GPR120 and/or SPMs production from EPA and its product DHA, and the anti-fibrotic effect of PPAR-γ [69].
Author Contributions: L.A.V. and R.V. designed the study. P.I., L.A.V. and R.V. analyzed and interpreted the data. A.E., F.E., S.S.-A., C.C., A.R., and R.V. performed the experiments. P.I., L.A.V., R.V. and T.M. structured and wrote the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by Initiation FONDECYT to Rodrigo Valenzuela grant number [11140174] and Regular FONDECYT to Alejandra Espinosa grant number [1181774].