Obesity and its associated disorders are becoming a major health issue in many countries. The resulting low-grade inflammation not only affects the periphery but also the central nervous system. We set out to study, in a time-dependent manner, the effects of a high-fat diet on different regions of the central nervous system with regard to the inflammatory tone. We used a diet-induced obesity model and compared at several time-points (1, 2, 4, 6, 8, and 16 weeks) a group of mice fed a high-fat diet with its respective control group fed a standard diet. We also performed a large-scale analysis of lipids in the central nervous system using HPLC-MS, and we then tested the lipids of interest on a primary co-culture of astrocytes and microglial cells. We measured an increase in the inflammatory tone in the cerebellum at the different time-points. However, at week 16, we evidenced that the inflammatory tone displayed significant differences in two different regions of the central nervous system, specifically an increase in the cerebellum and no modification in the cortex for high-fat diet mice when compared with chow-fed mice. Our results clearly suggest region-dependent as well as time-dependent adaptations of the central nervous system to the high-fat diet. The differences in inflammatory tone between the two regions considered seem to involve astrocytes but not microglial cells. Furthermore, a large-scale lipid screening coupled to ex vivo testing enabled us to identify three classes of lipids—phosphatidylinositols, phosphatidylethanolamines, and lysophosphatidylcholines—as well as palmitoylethanolamide, as potentially responsible for the difference in inflammatory tone. This study demonstrates that the inflammatory tone induced by a high-fat diet does not similarly affect distinct regions of the central nervous system. Moreover, the lipids identified and tested ex vivo showed interesting anti-inflammatory properties and could be further studied to better characterize their activity and their role in controlling inflammation in the central nervous system.

Oxysterols are bioactive lipids derived from cholesterol that are linked to inflammatory processes. Because obesity and metabolic syndrome are characterized by inflammation and altered cholesterol metabolism, we sought to investigate the variations of oxysterol levels and their metabolic pathways induced by obesity in the liver, hypothalamus, adipose tissue and plasma. To this end, we used diet-induced and genetic (ob/ob and db/db) models of obesity. Among the oxysterols measured, we found that 4β-oxysterol levels were consistently decreased in the high-fat diet study, at different time-points and in the ob/ob model. Overall, we did not find any correlation between cytochromes mRNA expression and variations of oxysterol levels. We also measured the levels of hepatic primary bile acids, in these three models and found similar profiles between HFD and ob/ob mice. However, although they are downstream metabolites of oxysterols, the variations in bile acid levels did not reflect the variations of their precursors. Our data show that, when considering oxysterol metabolism, the high-fat diet and ob/ob models are more closely related when compared to the db/db model. However, we were able to discriminate between lean and obese phenotypes based on liver oxysterol (4β-hydroxycholesterol, 27- hydroxycholesterol, 7-hydroxycholestenone) levels and enzyme (CYP3A11, CYP27A1, CYP7A1) expression.

Oxysterols are cholesterol derivatives that have been suggested to play a role in inflammatory diseases such as obesity, atherosclerosis, or neuroinflammatory diseases. However, the effect of neuroinflammation on oxysterol levels has only been partially studied so far. We used an HPLC-MS method to quantify over ten oxysterols both in in vitro and in vivo models of neuroinflammation. In the same models, we used RT-qPCR to analyze the expression of the enzymes responsible for oxysterol metabolism. Using the BV2 microglial cell line, we explored the effect of lipopolysaccharide (LPS)-induced (M1-type) and IL-4-induced (M2-type) cell activation on oxysterol levels. We also used LPS-activated co-cultures of mouse primary microglia and astrocytes. In vivo, we induced a neuroinflammation by administering LPS to mice. Finally, we used a mouse model of multiple sclerosis, namely the experimental autoimmune encephalomyelitis (EAE) model, that is characterized by demyelination and neuroinflammation. In vitro, we found that LPS activation induces profound alterations in oxysterol levels. Interestingly, we could discriminate between control and LPS-activated cells based on the changes in oxysterol levels both in BV2 cells and in the primary co-culture of glial cells. In vivo, the changes in oxysterol levels were less marked than in vitro. However, we found in both models increased levels of the GPR183 agonist 7α,25-dihydroxycholesterol. Furthermore, we studied in vitro the effect of 14 oxysterols on the mRNA expression of inflammatory markers in LPS-activated co-culture of microglia and astrocytes. We found that several oxysterols decreased the LPS-induced expression of pro-inflammatory markers. These data demonstrate that inflammation profoundly affects oxysterol levels and that oxysterols can modulate glial cell activation. This further supports the interest of a large screening of oxysterol levels when studying the interplay between neuroinflammation and bioactive lipids.