Oral Supplementation of Phosphatidylcholine Attenuates the Onset of a Diet-Induced Metabolic Dysfunction–Associated Steatohepatitis in Female C57BL/6J Mice

Background & Aims Changes in phosphatidylcholine levels in the liver have been associated with the development of metabolic dysfunction–associated steatotic liver disease. Here, the effects of supplementing phosphatidylcholine on the development of early signs of metabolic dysfunction–associated steatohepatitis were assessed. Methods Male and female C57BL/6J mice were fed a liquid control or a fructose-, fat-, and/or cholesterol-rich diet for 7 or 8 weeks. The diets of female mice were fortified ± phosphatidylcholine (12.5 mg/g diet). In liver tissue and portal blood, indices of liver damage, inflammation, and bacterial endotoxemia were measured. J774A.1 cells and human monocytes preincubated with phosphatidylcholine (0.38 mmol/L) were challenged with lipopolysaccharide (50–100 ng/mL) ± the peroxisome proliferator-activated receptor γ (PPARγ) activator pioglitazone (10 μmol/L) or ± a liver receptor homolog 1 (LRH-1) antagonist 1-(3′-[1-(2-[4-morpholinyl]ethyl)-1H-pyrazol-3-yl]-3-biphenylyl)ethanon (1–10 μmol/L). Results In fructose-, fat-, and/or cholesterol-rich diet–fed mice the development of fatty liver and the beginning of inflammation were associated with significantly lower hepatic phosphatidylcholine levels when compared with controls. Supplementing phosphatidylcholine significantly attenuated the development of fatty liver and inflammation, being associated with protection against the induction of PPARγ2, and activation of nuclear factor of κ light polypeptide gene enhancer in B-cell inhibitor α whereas Lrh1 expression was unchanged. The protective effects of phosphatidylcholine on the lipopolysaccharide-induced activation of J774A.1 cells and human monocytes were attenuated significantly by the PPARγ activator pioglitazone and the LRH-1 antagonist. Conclusions Our data suggest that phosphatidylcholine levels in the liver are lower in early metabolic dysfunction–associated steatohepatitis in mice and that supplementation of phosphatidylcholine can diminish the development of metabolic dysfunction–associated steatotic liver disease through mechanisms involving LRH-1/PPARγ2/ nuclear factor κ-light-chain enhancer of activated B-cell signaling.

BACKGROUND & AIMS: Changes in phosphatidylcholine levels in the liver have been associated with the development of metabolic dysfunction-associated steatotic liver disease.Here, the effects of supplementing phosphatidylcholine on the development of early signs of metabolic dysfunction-associated steatohepatitis were assessed.

RESULTS:
In fructose-, fat-, and/or cholesterol-rich diet-fed mice the development of fatty liver and the beginning of inflammation were associated with significantly lower hepatic phosphatidylcholine levels when compared with controls.Supplementing phosphatidylcholine significantly attenuated the development of fatty liver and inflammation, being associated with protection against the induction of PPARg2, and activation of nuclear factor of k light polypeptide gene enhancer in B-cell inhibitor a whereas Lrh1 expression was unchanged.The protective effects of phosphatidylcholine on the lipopolysaccharide-induced activation of J774A.1 cells and human monocytes were attenuated significantly by the PPARg activator pioglitazone and the LRH-1 antagonist.
CONCLUSIONS: Our data suggest that phosphatidylcholine levels in the liver are lower in early metabolic dysfunction-associated steatohepatitis in mice and that supplementation of phosphatidylcholine can diminish the development of metabolic dysfunction-associated steatotic liver disease through mechanisms involving LRH-1/PPARg2/ nuclear factor k-light-chain enhancer of activated B-cell signaling.(Cell Mol Gastroenterol Hepatol 2024;17:785-800; https:// doi.org/10.1016/j.jcmgh.2024.01.009)W ith a still-increasing prevalence, metabolic dysfunction-associated steatotic liver disease (MASLD), until recently referred to as nonalcoholic fatty liver disease, 1 by now is the most prevalent liver disease worldwide.Indeed, it is estimated that approximately 30% of the general population worldwide is afflicted with MASLD. 2 Ranging from simple steatosis and steatohepatitis (metabolic dysfunction-associated steatohepatitis [MASH]) to fibrosis and even cirrhosis, MASLD comprises a wide range of disease. 3Besides age, genetic predisposition, and general overnutrition, in more recent years, several dietary factors such as a diet rich in saturated fatty acids and sugars have been associated with the development of MASLD. 4,5The resulting chronic excessive inflow of lipids and enhanced de novo lipogenesis is by now well acknowledged as one factor in the multifactorial development of MASLD (for overview, see Hughey et al 6 ).Furthermore, changes of intestinal microbiota composition and an increased translocation of lipopolysaccharide (LPS) subsequently leading to activation of Toll-like receptor 4 (TLR4) depending signaling cascades have been shown repeatedly to be associated with the development of MASLD, [7][8][9][10][11] and also may interact with hepatic lipid homeostasis. 12Although some progress in understanding has been made, to date the molecular mechanisms associated with the development of MASLD are not fully understood and universally accepted therapies are lacking.
Phosphatidylcholine comprises approximately 40%-50% of total cellular phospholipids and is the most abundant phospholipid in mammalian cells and subcellular organelles (for overview, see van der Veen et al 13 ).6][17] Herein, impairments of very low density lipoprotein (VLDL) are discussed to play a pivotal role 15 ; however, it also has been suggested that a lack of phosphatidylcholine may enhance inflammatory processes through VLDL-independent mechanisms. 16Specifically, it has been suggested that phosphatidylcholine can alter macrophage polarization and reduce the inflammatory response, for example, LPS-induced expression of proinflammatory cytokines such as tumor necrosis factor a (TNF-a) 18 through as yet not fully determined mechanisms.
Starting from this background, the aim of the present study was to assess if an oral supplementation of phosphatidylcholine prevents the onset of an early diet-induced MASH in mice and to determine the molecular mechanisms involved.

Results
Phosphatidylcholine Levels in Livers of Fructose-, Fat-, and Cholesterol-Rich Diet-Fed Female and Fructose-and Fat-Rich Diet-Fed Male Mice As expected, mice fed a MASLD-inducing diet (eg, a fructose-, fat-, and cholesterol-rich [FrFC] or fructose-and fat-rich [FrF] diet), regardless of sex, developed early signs of MASH, as seen by a significantly higher nonalcoholic fatty liver disease activity score (NAS), alanine aminotransferase activity, and numbers of neutrophil granulocytes when compared with control diet-fed mice.Also, lymphocyte antigen 6 complex locus G6D (Ly6G)-positive cells and myeloperoxidase (MPO) activity were significantly higher in livers of male and female mice fed the MASLD-inducing diets when compared with control diet-fed mice (Figures 1B-F and 2B-F, and Tables 1 and 2).The development of early MASH in both male and female mice fed a MASLD-inducing diet was associated with a significantly lower phosphatidylcholine concentration in liver tissue compared with control diet-fed mice (P < .05)(Figures 1A and 2A).

Effects of Supplementing Phosphatidylcholine on Markers of Liver Damage in FrFC-Fed Female Mice
Despite similar caloric intake and body weight (Table 1), NAS and herein especially signs of hepatic inflammation such as the number of neutrophil granulocytes, MPO activity, and Ly6G-positive cells were significantly lower in livers of FrFCfed mice treated concomitantly with phosphatidylcholine when compared with FrFC-fed mice (P < .05for all parameters) (Table 1 and Figure 1C, D, and F).In contrast, steatosis scoring and hepatic triglyceride levels, as well as alanine aminotransferase activity in plasma, were similar between FrFC-fed groups regardless of additional treatment (Table 1 and Figure 1E).However, NAS and liver to body weight ratios of FrFCþphosphatidylcholine-fed mice still were significantly higher than in control diet-fed groups (P < .05)(Figure 1C and Table 1).Liver to body weight ratios also were significantly lower in FrFCþphosphatidylcholine-fed mice when compared with FrFC-fed mice (P < .05)(Table 1).Applying Sirius red staining, no signs of fibrosis were detected in any of the groups (Table 1).Nitric oxide (NO) concentration in liver tissue was significantly higher in livers of FrFC-fed mice compared with all other groups (P < .05)(Figure 1G).NO concentrations in livers of FrFCþphosphatidylcholine-fed mice still were significantly higher than in control diet-fed groups (P < .05)(Figure 1G).Arginase activity, suggested to be the counterplayer of inducible nitric oxide synthase, 19 also Abbreviation used in this paper: ApoB, apolipoprotein B; C, control diet; FrF, fructose-and fat-rich; FrFC, fructose-, fat-, and cholesterolrich; LPS, lipopolysaccharide; LRH-1, liver receptor homolog 1; Ly6G, lymphocyte antigen 6 complex locus G6D; MASLD, metabolic dysfunction-associated steatotic liver disease; MASH, metabolic dysfunction-associated steatohepatitis; MPO, myeloperoxidase; mRNA, messenger RNA; NAS, nonalcoholic fatty liver disease activity score; NFkB, nuclear factor k-light-chain enhancer of activated B cells; NO, nitric oxide; PPARg, peroxisome proliferator-activated receptor g; PC, phosphatidylcholine; IkBa, nuclear factor of k light polypeptide gene enhancer in B-cell inhibitor a; TLR4, Toll-like receptor 4; TNF-a, tumor necrosis factor a; VLDL, very-low-density lipoprotein.
was significantly lower in livers of FrFC-fed mice compared with all other groups (P < .05)(Figure 1H).No differences in arginase activity were found between FrFCþphosphatidylcholine-fed mice and the 2 control groups (Figure 1H).Because mice did not show signs of overweight and only early signs of MASH, no differences in markers of glucose metabolism were found between groups (Table 1).

Effect of Supplementing Phosphatidylcholine on Apolipoprotein B Expression in Liver Tissue of FrFC-Fed Mice
Because it has been suggested that phosphatidylcholine may alter VLDL secretion through apolipoprotein B (ApoB)dependent mechanisms, 15,20 we next determined if the protective effects found for the supplementation of phosphatidylcholine were associated with changes in ApoB expression in liver.Neither mRNA expression of Apob nor protein levels in liver tissue differed between groups (Figure 3).It has been suggested that phosphatidylcholine may improve intestinal barrier function 21,22 and that an increased intestinal translocation of LPS leading to an activation of TLR4 and subsequently an activation of nuclear factor k-light-chain enhancer of activated B cells (NFkB)dependent signaling may be critical in the development of MASLD. 7,8,23Therefore, concentration of TLR4 ligands (LPS) in portal plasma and mRNA expression of Tlr4 and dependent signaling cascades in liver tissue were determined.Concentrations of TLR4 ligands (LPS) were significantly higher in portal plasma of both FrFC-fed groups than in control groups (P < .05)(Figure 4A), and were similar between FrFC-fed groups (Figure 4A).Expression of Tlr4 mRNA in liver tissue did not differ between groups because expression varied considerably within groups (Figure 4B).Phosphorylation of nuclear factor of k light polypeptide gene enhancer in B-cell inhibitor (IkBa) in liver tissue being indicative of NFkB activation 24 also was significantly higher in livers of FrFC-fed mice when compared with control diet-fed group (P < .05)(Figure 4D), while being almost at the level of controls in FrFCþphosphatidylcholine-fed groups (Figure 4D).In line with these findings, TNF-a protein concentration was significantly (control diet [C] vs FrFC and Cþphosphatidylcholine vs FrFC), and by trend, higher in livers of FrFC-fed mice than in FrFCþphosphatidylcholine-fed mice (FrFC vs FrFCþphosphatidylcholine: P ¼ .09)(Figure 4E).Because results of studies suggest that phosphatidylcholine may regulate peroxisome proliferator-activated receptor g (PPARg) and that PPARg may alter NFkB activation, 25,26 we next determined the mRNA expression of the 2 splice variants of PPARg: Pparg1 and 2. Expression of Pparg1 mRNA expression was similar between groups (Figure 4F).In contrast, Pparg2 mRNA expression was significantly higher in livers of FrFC-fed mice than in all other groups (P < .05)(Figure 4G).In line with these findings, PPARG2 protein levels in liver homogenate were significantly higher in livers of FrFC-fed mice compared with control diet-fed and FrFCþphosphatidylcholine-fed animals (Figure 4H).In livers of FrFCþphosphatidylcholine-fed animals PPARG2 protein concentration was almost at the level of controls (Figure 4H).Immunohistochemical staining of liver sections with a specific PPARg2 antibody revealed that this splice variant of the nuclear receptor is not detectable in hepatocytes, but rather in immune cells (eg, Kupffer cells) found in liver tissue (representative pictures of staining are shown in Figure 4I).In contrast, expression of Lrh1 (Nr5a2) mRNA, suggested to be regulated by certain phosphatidylcholines, 27 did not differ between groups (Figure 4J).

Effects of Supplementing Phosphatidylcholine on LPS-Dependent Activation of J774A.1 Cells and Primary Human Monocytes Isolated From Peripheral Blood
To further delineate the mechanisms related to the effects of phosphatidylcholine on the LPS-induced inflammatory response, J774A.1 cells used as a model of Kupffer cells as well as primary monocytes isolated from buffy coats of healthy donors were stimulated with LPS in the presence or absence of phosphatidylcholine.As expected, concentrations of NO in medium and mRNA expression of Tnfa both were significantly higher in J774A.1 cells treated with LPS when compared with control groups (P < .05)(Figure 5A and B).The preincubation with phosphatidylcholine significantly blunted the LPS-induced Tnfa mRNA expression (Figure 5B) and NO formation (P < .05)(Figure 5A); however, NO levels still were higher in cell culture medium of LPSþphosphatidylcholine-treated cells than in both control groups (P < .05)(Figure 5A).In human monocytes, TNF-a protein concentration was significantly higher in LPS-challenged cells compared with control groups and LPSþphosphatidylcholine-treated cells (Figure 5C).Although significantly lower than in LPS-challenged monocytes, TNF-a protein levels still were higher in LPS-phosphatidylcholine-treated cells than in both control groups (Figure 5C).

Effects of Phosphatidylcholine on LPS-Induced Activation of J774A.1 Cells and Human Monocytes Preincubated With a PPARg Agonist and an Antagonist, Respectively
To further delineate the molecular mechanisms underlying the protective effects of phosphatidylcholine, and to delineate if PPARg2 might be critical herein, J774A.1 cells and human primary monocytes, respectively, were either preincubated with phosphatidylcholine, GW9662, an antagonist of PPARg shown to attenuate the induction of PPARg2 in mice with diet-induced MASLD, 10 or the PPARg activator pioglitazone.Both pretreating J774A.1 cells and human monocytes with phosphatidylcholine and GW9662 significantly attenuated the LPS-dependent increase of NO and TNF-a protein levels in cell culture medium (P < .05)(Figure 5D-F).When cells were pretreated with a combination of phosphatidylcholine and GW9662, the effects were similar to the effects found for either compound alone (Figure 5D-F).In contrast, preincubating J774A.1 cells and human monocytes, respectively, with phosphatidylcholine and the PPARg agonist, pioglitazone, abolished the protective effects of phosphatidylcholine on the release of NO and TNF-a protein levels into the cell culture media (Figure 5G-I).

Effects of Phosphatidylcholine on LPS-Induced Activation of J774A.1 Cells or Primary Human Monocytes Preincubated With an Antagonist of Liver Receptor Homolog 1 (NR5A2)
Because activation of liver receptor homolog 1 (LRH-1) has been suggested as a potential target for certain phosphatidylcholines, 28 J774A.1 cells and primary human  monocytes, respectively, were either preincubated with phosphatidylcholine or an antagonist of LRH-1.The antagonist of LRH-1 significantly abolished the protective effects of phosphatidylcholine on the release of NO and TNF-a protein levels into the cell culture media (Figure 6).

Discussion
Results of both human and animal studies have suggested that the development of MASLD is associated with lower levels of phosphatidylcholine. [14][15][16][17]29 In line with these findings and despite being fed sufficient amounts of choline, female and male mice fed MASLD-inducing diets had significantly lower hepatic phosphatidylcholine levels when compared with mice fed a control diet in the present study. Thse lower levels of phosphatidylcholine in liver tissue were associated with the development of hepatic macrovesicular steatosis and signs of liver inflammation.In livers of female mice receiving the FrFC diet enriched with phosphatidylcholine, these alterations were attenuated markedly, with most inflammatory markers being almost at the level of controls.Because mice still were of normal weight and were only fed the FrFC diet for 8 weeks, no alterations in glucose tolerance were observed yet.Indeed, we showed before that when using this model of lean MASLD, the development of alterations in glucose metabolism and insulin resistance might take longer than in models of overweight-associated MASLD.10,30,31 Further suggesting that in the present study phosphatidylcholine predominantly affected inflammatory alterations, NO levels also were almost at the level of controls in livers of female mice fed the FrFC diet supplemented with phosphatidylcholine.NO has been reported before to be indicative of an induction of inducible nitric oxide synthase in settings of MASLD, 32 and also M1 polarization.33 In line with these findings, arginase activity, also discussed to be an indicator of macrophage polarization, 34 was lower in livers of female FrFC-fed mice.Again, these alterations were abolished almost completely in livers of female mice fed the FrFC diet fortified with phosphatidylcholine.These findings are also in line with those of others proposing that a lack of phosphatidylcholine may add to the development of proinflammatory alterations in settings of MASLD.13,35 Interestingly, and somewhat contrasting the findings of the present study suggesting that phosphatidylcholine derived from egg lecithin may especially attenuate the progression of steatosis to MASH, results of Niebergall et al, 15 using liver-specific elimination of phosphocholine cytidyltransferase knock-out mice (LCTa -/-), suggested that normalizing hepatic phosphatidylcholine levels may not attenuate disease progression to inflammatory stages but may only affect the development of liver steatosis.Differences between the present study and that of Niebergall et al 15 may have resulted from differences in feeding protocols, for example, 7 days of a high-fat diet vs 8 weeks of a high-fructose, high-fat, and cholesterol-rich diet, but also suggest that other mechanisms besides being a precursor for the synthesis of betaine may be critical in the beneficial effects of phosphatidylcholine in the setting of MASLD (see later).15 In summary, our results suggest that the development of MASLD is related to a decrease in phosphatidylcholine levels in liver tissues of male and female mice, even in a nonmalnutrition state.Our results also suggest that enriching the diet with phosphatidylcholine may diminish the development of MASLD in female mice and that this is related predominantly to a suppression of inflammatory alterations.However, if similar protective effects also are found in male mice and later stages of the disease, as well as in human beings, remain to be determined.Also, it remains to be determined if phosphatidylcholine from other sources than egg lecithin, having shown to have higher phosphatidylcholine levels (w3 times higher) when compared with soy lecithin 36 , has similar beneficial effects on the development of diet-induced MASLD.

The Protective Effects of Phosphatidylcholine Are Not Related to Changes in ApoB Expression in Liver Tissue
Studies have suggested that phosphatidylcholine may be critical in the regulation of ApoB and VLDL-dependent hepatic lipid export. 37In the present study, protein levels of ApoB did not differ between the 2 FrFC groups.It has been suggested before by the results of others that the effects of phosphatidylcholine on ApoB might be sex-and diet-specific. 38Indeed, Noga et al 39,40 showed that phosphatidylethanolamine N-methyltransferase catalyzes the conversion of phosphatidylethanolamine to phosphatidylcholine and that in female phosphatidylethanolamine N-methyltransferase knock-out mice fed a high-fat/highcholesterol diet, VLDL secretion and plasma ApoB100 lipoprotein levels were maintained normally whereas both were diminished in male mice.In the present study, we only used female mice to assess the effects of fortifying the diet with phosphatidylcholine.It may be that when performing experiments alike in male mice, supplementing the diet with phosphatidylcholine might have altered ApoB levels in the liver.Also, we only determined ApoB levels by Western blot and did not perform functional assays to assess VLDL secretion.

The Protective Effects of Phosphatidylcholine Are Not Related to Protection Against the Translocation of LPS but Protection Against the Activation of NFkB Signaling
It has been shown previously that the development of MASLD is associated with impairments of intestinal barrier function and an increased translocation of LPS, resulting in activation of TLR4-dependent signaling cascades in the liver. 41It also has been shown previously in settings of inflammatory bowel disease that phosphatidylcholine may attenuate inflammatory alterations in the gut and subsequently improve intestinal barrier function. 42,43In the present study, concentrations of TLR4 ligands, for example, LPS being indicative of intestinal barrier function, were significantly higher in portal blood of both FrFC-fed groups regardless of additional treatments.Differences between the results of our study and those of others might have been related to differences in concentrations of phosphatidylcholine used.Indeed, in the present study, mice received 12.5 mg phosphatidylcholine per gram diet while in the trial assessing the effects of phosphatidylcholine in models of inflammatory bowel disease up to 100 mg/kg body weight were used. 43Somewhat in contrast to our findings for TLR4 ligand concentration, Tlr4 mRNA expression in liver tissue did not differ between groups.Indeed, it has been shown before by us and others that Tlr4 expression in liver tissue is not always induced in the presence of its ligands. 9Still, supporting the findings for TLR4 ligand concentration, phosphorylation of IkBa being the inhibitor of NFkB, 44 and TNF-a also were both at the level of controls in FrFC-fed mice concomitantly treated with phosphatidylcholine while being significantly induced in livers of FrFC-fed mice.Also, pretreating J774A.1 cells or human monocytes with phosphatidylcholine attenuated the induction of Tnfa mRNA and formation of NO or TNF-a protein, further supporting the hypothesis that phosphatidylcholine may interfere with TLR4-NFkB-dependent signaling in immune cells.Indeed, it has been suggested before by the results of others that phosphatidylcholine may interfere with the LPS-TLR-NFkB signaling cascade. 45,46However, results of these studies also have suggested that phosphatidylcholine may not directly interfere with NFkB but rather that other mechanisms may be involved.For instance, results of studies in mice have suggested that injecting adipose tissue of mice with phosphatidylcholine may down-regulate PPARg expression in fatty tissue. 47Also, recent studies have reported that both docosahexaenoic acid-phosphatidylcholine and eicosapentanoic acid-phosphatidylcholine suppress NFkB-dependent signaling through enhancing levels of PPARg, 48 further suggesting that PPARg may act as a counterplayer of NFkB signaling, thereby dampening proinflammatory responses. 49n contrast, in a recently published study, it was shown that a high-fat diet-induced MASLD was associated with an induction of PPARg in liver tissue.This induction was dampened when phospholipase D1, an enzyme catalyzing the formation of phosphatidic acid, from phosphatidylcholine was knocked out. 50Furthermore, structural studies have shown that phospholipids may be potential ligands of LRH-1. 51Indeed, it was reported that treatment with the phospholipid dilauroyl phosphatidylcholine protected mice from the development of MASLD and insulin resistance through LRH-1-dependent mechanisms because the protective effects of dilauroyl phosphatidylcholine were lost in LRH-1 -/-mice. 27In the present study, Pparg1 mRNA expression was found to be similar between groups, whereas Pparg2 mRNA expression and PPARG2 protein concentrations were found to be induced in liver of FrFC-fed mice, and, herein, especially in immune cells, while not being detectable in hepatocytes.Furthermore, in livers of mice fed the FrFC diet supplemented with phosphatidylcholine, Pparg2 mRNA and protein levels were almost at the level of controls.Somewhat contrasting the findings of others reporting a down-regulation of LRH-1 in patients with MASLD or MASH, 52 in the present study, expression of Lrh1 mRNA was similar in female and male mice, respectively, fed the MASLD-inducing diets and controls.Also, the addition of phosphatidylcholine had no effects of Lrh1 expression in livers of female FrFC-fed mice.Supporting the hypothesis that phosphatidylcholine may have protected FrFC-fed mice through diminishing the activation of PPARg2, pretreating LPS-stimulated cells with the PPARg inhibitor GW9662 and phosphatidylcholine diminished the induction of NO production and TNF-a expression in J774A.1 cells or human monocytes to an almost similar extent.Also, the concomitant treatment of LPS-challenged cells with phosphatidylcholine and the PPARg agonist pioglitazone abolished the protective effects of phosphatidylcholine on NO formation and TNF-a protein release into cell culture medium.These data contrast the findings of others discussed earlier. 48,49owever, it has been shown before that an induction of PPARg2 is associated with the development of MASLD in various rodent models 10,53 and that targeting PPARg2 with the PPARg antagonist GW9662 may attenuate the development of MASLD. 10,54It also has been shown in these studies that the protective effects of the PPARg inhibitor were associated with a marked protection against inflammatory alterations associated with the development of MASLD, for example, the induction of proinflammatory cytokines and an increase in NO in liver tissue and cell culture supernatants. 10Furthermore, results of the in vitro experiments of the present study also suggest that the beneficial effects of phosphatidylcholine may be related to LRH-1 because the protective effects of phosphatidylcholine on LPS-induced production of NO and proinflammatory cytokines were abolished almost completely when cells were treated concomitantly with an LRH-1 antagonist.
Taken together, our data suggest that although intestinal barrier function was disturbed in FrFC-fed mice, enriching the diet with phosphatidylcholine had no effect on these alterations but rather protected the liver from the development of MASLD through suppressing the LPS-dependent activation of PPARg2 and, subsequently, NFkB in liver tissue.These results by no means preclude that at other concentrations or diseases of other etiologies (eg, inflammatory bowel disease) phosphatidylcholine may alter intestinal barrier dysfunction.Also, our data do not preclude that the formation of phosphatidic acid and a subsequent induction of PPARg may be critical in lipid accumulation in settings of MASLD.Rather, our data suggest that in a setting of nonobese diet-induced MASLD, the beneficial effects of phosphatidylcholine may be related predominantly to a counterregulation of LPS-TLR4-PPARg2-NFkB-dependent proinflammatory responses in the liver.The exact role of LRH-1 and its interplay with PPARg2 and NFkB herein needs to be determined in future studies.

Conclusions
Taken together, results of our study further bolster the hypothesis that alterations in hepatic phosphatidylcholine levels are critical in the development of MASLD and that supplementing phosphatidylcholine may attenuate or at least diminish the development of MASLD.Results of the present study further suggest that the protective effects of phosphatidylcholine are related to a protection of alterations in the LRH-1/PPARg/NFkB signaling cascade in the liver.However, further studies are needed to determine exact molecular mechanisms underlying the interaction of phosphatidylcholine with these nuclear receptors as well as cell types involved.Further studies also are needed to determine if supplementation of phosphatidylcholine also attenuates the progression of MASLD in human beings, and if this also is related to changes in LRH-1/PPARg2/NFkB signaling.

Animals and Treatment
Female and male C57BL/6J mice were obtained from Janvier (Janvier SAS, Le-Genest-Saint-Isle, France) and housed in a specific-pathogen-free barrier facility.All procedures were approved by the local institutional animal care and use committee (Bundesministerium für Wissenschaft, Forschung und Wirtschaft, Referat für Tierversuchswesen und Gentechnik, Vienna, Austria) and animals were handled in accordance with the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes.All experiments were performed under controlled conditions (12 h/12 h light/dark cycle, w24 C, w55% relative humidity) and mice had free access to tap water at all times.Animal trial 1: after an adaptation period to the facility and the intake of a liquid diet, female C57BL/6J mice, aged 6-8 weeks, were assigned randomly to the following groups (n ¼ 6-8/group): C: mice fed a liquid control diet (69E% carbohydrate, 12E% lipids, 19E% protein; Ssniff, Soest, Germany), Cþphosphatidylcholine (PC): mice fed a liquid control diet enriched with PC (12.5 mg/g diet), FrFC: mice fed a fructose-, fat-, and cholesterol-rich diet (60E% carbohydrate, 21E% lipids, 19E% protein with 50% wt/wt fructose and 0.16% wt/wt cholesterol; Ssniff), and FrFCþPC: mice fed a FrFC diet fortified with PC (12.5 mg/g diet) for 8 weeks.The sample size was determined based on previous findings. 30Phosphatidylcholine originating from egg lecithin (purity, 98%; Lipoid, Ludwigshafen, Germany) was used in all experiments.The dose of phosphatidylcholine was based on previous studies by other investigators. 55As detailed previously, 30 the intake of the liquid diet was documented daily, and within dietary groups, mice received the amount of diet that was consumed the previous day by the group with the lowest food intake (pair-group).This pair-group had access to diet ad libitum.The body weight was monitored weekly and the food intake was monitored daily.Moreover, a glucose tolerance test was performed during week 7 of the feeding trial.For that purpose, mice were fasted for 6 hours.Blood glucose levels were determined before and 15, 30, 60, 90, and 120 minutes after the intraperitoneal glucose injection.Animal trial 2: male C57BL/6J mice were assigned randomly to the following groups (n ¼ 7/group): C: mice fed a liquid control diet (69E% carbohydrates, 12E% fat, 19E% protein; Ssniff) or FrF: mice fed a liquid fructose-and fat-rich diet (60E% carbohydrates, 25E% fat, 15E% protein; Ssniff) for 7 weeks.
After 7-8 weeks of feeding, mice were anesthetized (intraperitoneally) with ketamine/xylazine in the morning, blood was taken from the portal vein, and mice then were killed by cervical dislocation.Hepatic tissue samples were collected in neutral-buffered formalin or immediately snapfrozen in liquid nitrogen and were stored in a À80 C freezer for further experiments.Killing and all following measurements were performed in a mixed order of groups.

Evaluation of Liver Damage and Inflammation and Blood Parameter of Liver Damage
Paraffin-embedded liver sections (4 mm) were stained with H&E (Sigma-Aldrich Chemie GmbH) to evaluate liver histology using NAS, as detailed by others. 56To assess hepatic fibrosis, liver sections were stained with Picrosirius red and counterstained with fast green (Sigma-Aldrich Chemie GmbH) as detailed previously. 57Neutrophil granulocytes were stained and counted in liver sections using a commercially available kit (naphthol AS-D chloroacetate kit; Sigma-Aldrich Chemie GmbH) as detailed previously, 58 or were stained with Ly6G antibody (anti-Ly6G, #ab238132; Abcam, Cambridge, UK).In brief, antigen retrieval was performed and endogenous peroxidase activity was quenched by incubating sections with a peroxidase block (Dako Staining Solutions, Agilent Technologies, Vienna, Austria).Sections were blocked with 1% bovine serum albumin in phosphate-buffered saline with Tween 20 and incubated with a monoclonal antibody (anti-Ly6G, #ab238132; Abcam) for 1.5 hours at room temperature before incubating with secondary antibody for 20 minutes at room temperature (anti-rabbit; Dako Staining Solutions, Agilent Technologies).After staining with 3,3 0 -Diaminobenzidine chromogen solution (Dako Staining Solutions, Agilent Technologies) and counterstaining with hematoxylin (Sigma-Aldrich Chemie GmbH), sections were covered with a coverslip.Neutrophils and Ly6G-positive cells were counted and the mean of 8 microscopic fields was determined using a microscope (DFC 450 C; Leica, Wetzlar, Germany).Alanine aminotransferase activity in plasma was measured in a routine laboratory (Veterinary Medical University of Vienna, Vienna, Austria).

Assessment of Arginase and MPO Activity and NO and Phosphatidylcholine Concentrations
Arginase and MPO activity and NO concentration in liver tissue were assessed as described previously. 30,59In brief, to determine arginase activity, liver tissue was homogenized in 50 mmol/L Tris-HCl, pH 7.5, containing 0.1% Triton X-100, 10 mmol/L MnCl 2 , and protease inhibitor cocktail (Sigma-Aldrich Chemie GmbH), and liver tissue was homogenized in phosphate buffer for measuring MPO activity.NO concentration in liver tissue as well as in cell culture supernatant was measured using a commercially available Griess reagent assay (#G2930; Promega, Mannheim, Germany).Phosphatidylcholine concentration in liver tissue was measured using a commercially available assay following the manufacturer's instructions (#MAK049; Sigma-Aldrich Chemie GmbH).

Immunohistochemical Staining of PPARg2
To determine in which cells PPARg2 is expressed predominately in liver, liver sections were stained with a specific monoclonal primary antibody.In brief, after antigen retrieval and peroxidase blocking, liver sections were stained with a monoclonal primary antibody against PPARg2 (#sc-390740; Santa Cruz Biotechnology) overnight at 4 C and biotin-labeled secondary antibody for 30 minutes and enzyme-labeled streptavidin (Dako Staining Solutions, Agilent Technologies) for 30 minutes, respectively.Representative pictures were taken using a microscope with an integrated camera (DM4000 B LED; Leica).

Assessment of TLR4 Ligand Concentration in Plasma
Commercially available reporter gene assays with TLR4transfected HEK293 cells were used to determine the levels of TLR4 ligands in plasma, following the manufacturer's instructions (InvivoGen, Toulouse, as described previously detail. 60zyme-Linked Immunosorbent Assays, RNA Isolation, and Real-Time Polymerase Chain Reaction Using commercially available enzyme-linked immunosorbent assay kit concentrations of TNF-a (DuoSet; R&D Systems, Inc, Minneapolis, MN) were analyzed in cell culture supernatant following the manufacturer's instructions.RNA from liver tissue and cells was extracted using TRItidy G (PanReac AppliChem).After measuring concentration of RNA, complementary DNA was synthetized with a reversetranscription system (Promega GmbH).Using primers listed in Table 3, real-time polymerase chain reaction was performed.In female mice, expression of the respective genes was normalized to 18S and glyceraldehyde-3-phosphate dehydrogenase, of which a geometric mean was calculated as previously described. 31,61In male mice, expression of genes was normalized to the geometric mean of 18S, b-actin, and glyceraldehyde-3-phosphate dehydrogenase.

Statistical Analysis
Using PRISM (version 7.03; GraphPad Software, San Diego, CA), a Grubb's outlier test was performed for all measurements before further statistical analysis.To analyze differences between 2 groups, an unpaired Student t test was used.To determine statistical differences between 3 or more groups, a 1-way analysis of variance and a 2-way

Figure 1 .
Figure 1.Effect of supplementing phosphatidylcholine on markers of liver damage in female C57BL/ 6J mice fed a FrFC diet.(A) PC in liver tissue.(B) Representative pictures of H&E staining in liver tissue (magnification, 200Â and 400Â).(C) NAS of liver sections.(D) Number of neutrophil granulocytes per microscopic field in liver tissue.(E) Alanine aminotransferase (ALT) activity in plasma.(F) MPO activity.(G) NO concentration.(H) Arginase activity in liver tissue.Data are presented as box and whisker plots, n ¼ 6-8 mice per group.a P < .05compared with mice fed the control diet.c P < .05compared with mice fed the controlþphosphatidylcholine diet.d P < .05compared with mice fed the FrFCþphsophatidylcholine diet.NAFLD, nonalcoholic fatty liver disease.

Figure 3 .
Figure 3.Effect of supplementing phosphatidylcholine on apolipoprotein B mRNA and protein levels in liver tissue of female C57BL/6J mice fed a FrFC diet.(A) Expression of apolipoprotein B (Apob) mRNA.(B) Hepatic protein concentration of ApoB normalized to b-actin.Data are presented as box and whisker plots, n ¼ 6-8 mice per group.

Figure 4 .
Figure 4. Effect of supplementing phosphatidylcholine on TLR4 ligands in portal blood, and expression of Tlr4, TNF-a, Pparg1 and 2, as well as activation of IkBa in liver tissue in female C57BL/6J mice fed a FrFC diet.(A) Levels of TLR4 ligands in plasma.(B) mRNA expression of Tlr4 in liver tissue.(C) Representative pictures of Western blots.(D) Hepatic protein concentration of phospho-IkBa.(E) Hepatic protein concentration of TNF-a.Hepatic mRNA expression of (F) Pparg1 and (G) Pparg2.(H) Hepatic protein concentration of PPARG2 normalized to b-actin.(I) Representative pictures of PPARG2 staining (magnification, 400Â and 630Â).(J) Hepatic mRNA expression of Lrh1 (Nr5a2).Data are presented as box and whisker plots, n ¼ 6-8 mice per group.a P < .05compared with mice fed the control diet.c P < .05compared with mice fed the controlþphosphatidylcholine diet.d P < .05compared with mice fed the FrFCþphosphatidylcholine diet.Lrh1 (Nr5a2), liver receptor homolog 1.

Figure 5 .
Figure 5. (See previous page).Effect of phosphatidylcholine on the LPS-induced activation of J774A.1 cells and primary human monocytes isolated from buffy coats of healthy donors.(A) NO concentration in cell culture medium.(B) mRNA expression of Tnfa in stimulated J774A.1 cells treated with phosphatidylcholine. (C) TNF-a protein in supernatant of LPSstimulated human monocytes treated with phosphatidylcholine. (D) NO concentration and (E) TNF-a protein concentration in supernatant of LPS-stimulated J774A.1 cells treated with phosphatidylcholine and GW9662.(F) TNF-a protein concentration in cell culture media of LPS-stimulated human monocytes treated with phosphatidylcholine and GW9662.(G) NO concentration and (H) TNF-a protein concentration in cell culture media of LPS-stimulated J774A.1 cells treated with phosphatidylcholine and pioglitazone.(I) TNF-a protein concentration in LPS-stimulated human monocytes treated with phosphatidylcholine and pioglitazone.Data are presented as box and whisker plots, n ¼ 4-7.(A and B) n ¼ 7. a P < .05compared with naïve J774A.1 cells.b P < .05compared with J774A.1 cells treated with phosphatidylcholine. d P < .05compared with J774A.1 cells treated with LPSþphosphatidylcholine. (C) n ¼ 4. a P < .05. compared with naïve human monocytes.b P < .05compared with human monocytes treated with phosphatidylcholine. d P < .05compared with human monocytes treated with LPSþphosphatidylcholine. (D and E) n ¼ 6. a P < .05compared with J774A.1 cells treated with LPS.(F) n ¼ 4. a P < .05compared with human monocytes treated with LPS.(G and H) n ¼ 6. a P < .05compared with J774A.1 cells treated with LPS.c P < .05compared with J774A.1 cells treated with LPSþphosphatidylcholineþpioglitazone.(I) n ¼ 5. a P < .05compared with human monocytes treated with LPS.c P < .05compared with human monocytes treated with LPSþphosphatidylcholineþpioglitazone.n.d., not detectable; NOx, nitric oxide; Pio, pioglitazone.

Table 1 .
Effect of Supplementing Phosphatidylcholine on Food Intake, Body Weight, Glycemia, and Triglyceride Concentration in Liver Tissue in Female C57BL/6J Mice Fed a FrFC Diet NOTE.Data are presented as means ± SD, n ¼ 6-8 mice per group.AUC, area under the curve.a P < .05compared with mice fed the control diet.c P < .05compared with mice fed the controlþphosphatidylcholine diet.d P < .05compared with mice fed the FrFCþphosphatidylcholine diet.

Table 2 .
Food Intake, Body Weight, and Liver Damage in Male C57BL/6J Mice Fed a FrF Diet a NOTE.Data are presented as means ± SD, n ¼ 7 mice per group.a P < .05.

Table 3 .
Primer Sequences Used for Real-Time Polymerase Chain Reaction