In the present study, we investigated (i) the effects of long-term consumption of HF diet supplemented with LC n-3 PUFA in the form of PL or TG on adiposity, oxidative stress and inflammation and (ii) whether the PL or TG carrier can affect these parameters. For this purpose, we designed a HF diet containing 20% of fat, rich in saturated fatty acids and with a relatively high n-6/n-3 ratio to mimic the lipid-enriched foods as consumed in typical Western diets. Importantly, all diets contained the same amount of PL, in the form of PL-DHA in HF-ω3PL diet and in the form of lecithin PL-LA in all other groups. This formulation ensured that the possible observed effects of PL-ω3 can be attributed to the location of n-3 PUFA in the diet (on PL rather than TG) and not to the presence of polar lipids that would have biased the interpretation.
We show that altogether, long-term intake of HF diet supplemented with LC n-3 PUFA protects against inflammation and oxidative stress induced by HF diets. This is in agreement with numerous studies documenting the beneficial health effects of LC n-3 PUFA [19, 37]. In this study, the supplementation of HF diet with LC n-3 PUFA in the PL carrier was slightly lower than that in the TG carrier. Therefore, our results indicate that the effects of LC n-3 PUFA from lecithin rich in PL-DHA is more efficient than LC n-3 PUFA from tuna oil (mainly TG-DHA). This supports the results of a previous study reporting that in healthy humans the metabolic effects of krill oil (mainly PL) are similar to those of fish oil but at lower dose of LC n-3 PUFA . Importantly in previous studies, the PL carrier was provided by the bulk marine sources, i.e. krill oil [25–28]. Usually, the percentage of PL in these products is 40%, which means that a part of n-3 PUFA brought by the krill oil was in fact bound to TG carrier in these studies.
In the present study, a dietary intake of LC n-3 PUFA during 8 weeks resulted in an increased incorporation of DHA into plasma, liver and WAT lipids, the FA composition of the organs reflecting the fatty acid profiles of the diets. Regardless of the LC n-3 PUFA carrier, the n-6/n-3 ratio in plasma, liver and WAT of mice fed HF- ω3PL and HF- ω3TG was lower than in the HF diet. The ratio was almost similar in both HF-ω3PL and HF-ω3TG groups, while the HF-ω3TG group showed a significantly higher proportion of DHA than the HF-ω3PL group. A decreased n-6/n-3 ratio in tissues has been reported to reduce atherosclerosis due to the inhibition of systemic and vascular inflammation in apolipoprotein E-deficient mice. The authors attributed these protective effects to the anti-inflammatory properties of n-3 PUFA . In addition, the proportion of EPA in WAT was higher in mice fed the HF diets supplemented with LC n-3 PUFA compared with HF, suggesting that the turnover of DHA in eWAT is high. The increase in EPA concentration paralleled the increased level of DHA in HF-ω3TG more than in HF-ω3PL.
LC n-3 PUFA are known to reduce metabolic inflammation in human and rodents [39–41]. Our data show that the HF diet induced higher IL-6, MCP-1 levels in plasma and in eWAT than LF diet. Interestingly, no activation of these pro-inflammatory markers was observed in HF-ω3PL and in HF-ω3TG groups, even if the EPA + DHA dose in the HF- ω3PL is lower of that in the HF-ω3TG group. This may indicate that LC n-3 PUFA derived from PL provide a better bioavailability and/or bioactivity than those esterified into TG. Our results suggest that LC n-3 PUFA, regardless of their molecular form, could inhibit the low-grade inflammation by directly inhibiting macrophage immigration through the inhibition of MCP-1.
Our data are consistent with previous studies reporting that the inflammatory response in WAT induced by HF diet in obese diabetic animals was prevented by the supplementation HF with n-3 PUFA either in the form of PL or TG [6, 21]. Batetta et al. concluded that such anti-inflammatory effects can be due to observed lower levels of arachidonic acid in membrane phospholipids .
Recent studies revealed the complementary role of metabolic endoxotemia in the low-grade inflammation. Laugerette et al. observed that endotoxin transporter LBP was positively correlated with plasma IL-6 in mice fed a palm oil-based high-fat diet, which was reversed using rapeseed oil . Other works evidenced a link between low-grade inflammation or related metabolic disorders and plasma LBP and sCD14 [43, 44]. In our study, the HF group presented the highest plasma concentration of LBP, consistently with inflammatory markers. A significant decrease in the level of plasma sCD14 was observed in HF-ω3PL and HF-ω3TG groups. Altogether, our results indicate that the supplementation of HF diet with LC n-3 PUFA can lower plasma concentrations of endotoxin transporters. Further mechanisms should be investigated to elucidate the implication of n-3 PUFA sources in the regulation of endotoxemia-induced metabolic inflammation.
Circulating leptin levels are directly associated with the mass of WAT and inflammation . Our results showed that the plasma levels of leptin, mainly produced by adipocytes, were decreased in the two supplemented groups with LC n-3 PUFA in the form of PL or TG vs HF mice. Plasma leptin was also lower in HF-ω3TG group independently of WAT mass, suggesting a relationship between the lower metabolic inflammation and the leptin. Our results are in agreement with a previous study showing that long-term intake of dietary n-3 PUFA by rats resulted in a significant decrease in plasma leptin levels .
Adiponectin plays an important role as insulin-sensitizing adipokine which production is decreased in obesity and in conditions associated with insulin resistance [5, 47]. In our study, we did not observed a significant difference in circulating adiponectin level among groups. In addition, the glucose and insulin tolerances were not affected by the different diets. Our results are in agreement with previous study reporting that the level of adiponectin was not different between HF group (35% of fat) and DHA/EPA in the form of PL or TG supplemented to the HF diets for 8 weeks . Conversely, other studies showed a decrease in the level of plasma adiponectin after feeding mice a HF diet; this level was restored by the supplementation HF diet with DHA and EPA [6, 29]. Altogether, such modifications in plasma adiponectin concentrations are reported with different lipid types (n-6 PUFA-rich corn oil instead of lard) and/or concentrations in the diet (35% w/w instead of 20%) and using larger doses of fish oil (15-40% in total dietary lipids) compared with the present study (5%).
Regarding adiposity, HF diet did not induce significantly higher body weight gain than the low fat control, although a trend towards heavier adipose tissue was observed. This can be due to our choice of preparing LF and HF groups using similar ingredients. Both LF and HF were semi-synthetic and based mainly on corn starch and casein, while many studies in the literature reach weight gain by comparing semi-synthetic high-fat diet with regular chow based on more various ingredients . Among the HF diets, HF-ω3PL diet led to a lower weight gain and a reduced adipose tissue compared with HF group without affecting the mass of muscles. However these effects were not observed in mice fed HF-ω3TG. These results suggest that the PL carrier can decrease body fat deposition more than the TG carrier. This is consistent with a previous study reporting that obese mice fed LC n-3 PUFA carried by PL developed less body weight gain than those fed LC n-3 PUFA carried by TG . We further investigated the effect of HF diet containing LC n-3 PUFA in the form of PL or TG on the cell size distribution of adipose tissue by using the Coulter counter method (Multisizer IV, Beckman Coulter) which allows a more precise cell-size distribution because the number of cell counted is much higher than the one of microscopic methods. The most noticeable difference between groups was in the adipocyte size distribution. Mice fed HF diet showed an increased adipocyte size compared with those fed the LF diet. Differential effects were observed regarding LC n-3 PUFA supplementation; HF-ω3PL and LF diets induced similar adipocyte size whereas HF-ω3TG and HF diets led to larger adipocytes. Our results following a 20% w/w high-fat diet for 8 weeks are consistent with those of Rossmeisl et al. . These authors observed that the PL carrier decreased adipocyte area compared with the control HF diet in obese mice, which was not observed using the TG carrier. Of note, endothelial lipase presents specificity towards DHA-containing PL, thus generating Lyso-PL containing DHA . Moreover, our PL source also contained Lyso-PC-DHA. We may hypothesize that such DHA-containing lipid species might exert specific biological activity, because superior biological effects of LysoPC-DHA have otherwise been demonstrated regarding DHA accretion in the brain and anti-inflammatory activity [50, 51]. However, further research is required to better understand the mechanism of action of PL carrier.
Skurk et al. investigated the secretory capacity of adipocyte fractions from the same individual. They demonstrated that the large adipocytes were implicated in the induction of pro-inflammatory genes such as those coding for IL-6 and MCP-1 . Consistently, HF diet-induced-inflammation can be associated with large adipocytes. Regarding the impact of supplementing with n-3 PUFA, noticeably, the HF-ω3PL diet had a greater anti-adiposity effect than the HF-ω3TG diet associated with smaller adipocytes. Here, the anti-adiposity effect of HF-ω3PL diet could be partly associated with the lower metabolic inflammation in the mice. In contrast, the anti-inflammatory effect of HF-ω3TG was not associated with a decrease in adiposity.
We further investigated the effect of the composition of the dietary high-fat, and more specifically of the presence of LC n-3 PUFA on lipid peroxidation and oxidative stress. LC PUFA are molecules susceptible to oxidation because they contain many double bonds. PUFA oxidation leads to the formation of secondary end-products such as 4-HNE derived from n-6 PUFA and 4-HHE derived from n-3 PUFA . Our previous studies in mice showed that these markers can induce oxidative stress and inflammation, even when consumed in moderately oxidized dietary fat . They also provoke oxidative stress in vitro in Caco-2 cells . In the present study, similar levels of plasma 4-HHE were measured in the three HF mice groups; however the plasma level of 4-HNE was increased in HF mice compared with the other groups. Previously, Esterbauer et al. found that the basal concentrations of 4-HNE in the human serum were in the range 0–700 nM. In our study, plasma 4-HNE in mice remained within this range observed in humans . In liver and in WAT, the levels of 4-hydroxy-2-alkenals were lower than 2 nmol/g, which is in the same order of magnitude than in another study reporting basal concentrations of 4-HNE in the liver of mice . We thus suggest that the dietary high-fat used in this study may slightly enhance the oxidative stress through the induction of 4-HNE in plasma, albeit without increase of 4-HNE in tissues. LC n-3 PUFA supplementation prevented this phenomenon.
We propose that the increase of oxidative stress after HF diet consumption is linked to an alteration of anti-oxidant defenses. In the gastro-intestinal tract, a defense system including GPx2, is able to detoxify lipid peroxidation products and protect against inflammation [30, 56, 57]. Our results show a significant increase of GPx2 mRNA in the duodenum of mice fed HF vs HF-ω3PL diets. Interestingly, a significant difference of GPx2 mRNA in the jejunum was also observed between HF vs HF-ω3PL mice. Meanwhile, we observed a significant difference between HF vs HF-ω3TG groups in the jejunum. This suggests that (i) the HF diet-induced-inflammation could alter the anti-oxidant defense system in the intestine and (ii) the supplementation of HF diet with LC n-3 PUFA in the form of PL is more bioactive in the upper intestine than those in the form of TG.
Regarding the concentration of α-tocopherol, a lipid-soluble antioxidant vitamin, we show that it was significantly increased in the WAT of HF-ω3PL and HF-ω3TG groups as compared to HF mice. Interestingly, tocopherol level in WAT was significantly greater in HF-ω3PL than in HF-ω3TG mice. Of note, in this study we took care to adjust dietary tocopherol to obtain similar concentrations in our diets. Thus, our results can be attributed to the direct effects of PL-bound vs TG-bound LC n-3 PUFA. Consistently, Choi et al. recently reported that high-fat diet enhanced the oxidative stress through the decreased level of tocopherols in the livers of rats . Thus, our findings suggest that dietary supplementation with LC n-3 PUFA is beneficial for decreasing lipid peroxidation in high fat-fed mice, and that the PL carrier may induce a superior bioavailability to tocopherols or lower need for their use to counteract oxidative stress. The enhanced level of α- tocopherol following LC n-3 PUFA in the form of PL can also be an effective defense against oxidative stress and inflammation. Further studies should analyze the amount and distribution of vitamin E among different tissues after the diets. The effect of dietary n-3 PUFA in the form of PL and TG on the activity of anti-oxidant systems should also be clarified by further studies in humans. Altogether, HF diet induced concomitant increase in plasma 4-HNE, GPx-2 activation in the small intestine and lower tocopherol level in WAT; all of these markers being reversed by the supplementation with LC n-3 PUFA.