The Journal of Nutrition Biochemical, Molecular, and Genetic Mechanisms

Reduction of Cholesterol Absorption by Dietary Plant Sterols and Stanols in Mice Is Independent of the Abcg5/8 Transporter1,2

Torsten Plo¨ sch,3,4* Janine K. Kruit,3,4 Vincent W. Bloks,4 Nicolette C. A. Huijkman,4 Rick Havinga,4 Guus S. M. J. E. Duchateau,5 Yuguang Lin,5 and Folkert Kuipers4

4Center for Liver, Digestive and Metabolic Diseases, Laboratory of Pediatrics, University Medical Center Groningen, Groningen, The Netherlands and 5Unilever Food and Health Research Institute, Vlaardingen, The Netherlands Downloaded from https://academic.oup.com/jn/article/136/8/2135/4664773 by guest on 25 September 2021

Abstract Dietary supplementation with plant sterols, stanols, and their esters reduces intestinal cholesterol absorption, thus lowering plasma LDL cholesterol concentration in humans. It was suggested that these beneficial effects are attributable in part to induction of genes involved in intestinal cholesterol transport, e.g., Abcg5 and Abcg8, via the liver X receptor (LXR), but direct proof is lacking. Male C57BL/6J mice were fed a purified diet (control), diets containing cholesterol (0.12 g/100 g) only, or in combination with either plant sterols or stanols (0.5 g/100 g) for 4 wk. Plant sterols and stanols dramatically increased neutral fecal sterol excretion (2.2 and 1.4-fold, respectively, compared with cholesterol-fed mice; P , 0.05). Cholesterol and cholesterol ester concentrations were higher in livers of mice fed cholesterol compared with controls (1135% and 1925%; P , 0.05). Plant sterols and stanols completely prevented cholesterol accumulation as well as induction of LXR target genes in liver. Feeding plant sterols and stanols did not alter intestinal expression of Abcg5, Abcg8, or other LXR target genes nor of Npc1l1. Fractional cholesterol absorption in Abcg52/2 mice was reduced to the same extent by dietary plant sterols (49%) as in wild-type littermates (44%). Plant sterol and stanol-induced reduction of cholesterol absorption in mice is not associated with upregulation of intestinal LXR target genes nor is it influenced by Abcg5-deficiency. Our data indicate that dietary plant sterols and stanols inhibit cholesterol absorption within the intestinal lumen independently of LXR. J. Nutr. 136: 2135–2140, 2006.

Introduction taken up by enterocytes via NPC1L1 (7). In the enterocyte, Dietary plant sterols and plant stanols have been recognized as cholesterol is readily esterified by the action of acyl-CoA:cho- efficient modulators of plasma LDL cholesterol concentrations lesterol acyltransferase 2 (ACAT2) and released into lymph in in humans for decades (1,2). In contrast to cholesterol, these association with chylomicrons. In contrast, plant sterols and compounds are only poorly absorbed from the intestine. Based stanols are effectively excreted back to the intestinal lumen by on their structural similarity to cholesterol, it was postulated heterodimers of the ATP-binding cassette transporters, ABCG5 that plant sterols and plant stanols physically interfere with and ABCG8, present at the apical membrane of the enterocyte intestinal cholesterol absorption, e.g., by interference with (8,9). ABCA1 is also highly expressed in the intestine, mainly at cholesterol micellization in the intestinal lumen (3–5). the basolateral domain of the cell membrane, but its role in Recently, proteins involved in cholesterol absorption as well sterol absorption is unclear (10–12). Expression of ABCA1, as in intestinal sterol excretion were identified. Niemann-Pick ABCG5, and ABCG8, but not of NPC1L1, is under control of a C1-Like protein 1 (NPC1L1)6 is considered the target of the nuclear receptor, the liver-X-receptor (LXR), activated by cholesterol absorption inhibitor ezetimibe and was shown to be oxysterols (13,14). Activation of LXR by synthetic ligands in responsible for the majority of cholesterol uptake into entero- mice was shown to increase fecal neutral sterol loss and to cytes (6,7). Recent findings indicate that plant sterols are also reduce fractional cholesterol absorption, as a consequence of increased Abcg5/Abcg8 expression, leading to enhanced choles- 1 Supported in part by Unilever Food and Health Research Institute, Vlaardingen, terol excretion back into the intestinal lumen (14,15). The Netherlands and the Dutch Heart Foundation, grant 2004T048 to T.P. and The discovery that cholesterol absorption can be actively 2001B043 to J.K.K. 2 Supplemental Tables 1 and 2 are available with the online posting of this regulated at the level of the enterocyte opens the possibility that paper at jn.nutrition.org. plant sterols and plant stanols, in addition to their postulated 3 These authors contributed equally. physicochemical effects (3–5), also influence cholesterol absorp- 6 Abbreviations used: Abca1, Abcg5, Abcg8: ATP-binding cassette transporter tion by regulating expression of transport proteins. Specifically, a1, g5, g8; Acat2: acyl-CoA:cholesterol acyltransferase 2; LXR: liver-X-receptor; plant sterols and stanols may act as LXR agonists either directly Npc1l1: Niemann-Pick C1-Like 1 protein. * To whom correspondence should be addressed. E-mail: t.plosch@med. or after their conversion into oxyphytosterols or oxyphytostanols, umcg.nl. respectively. Plant sterols and stanols were shown to act as LXR

0022-3166/06 $8.00 ª 2006 American Society for Nutrition. 2135 Manuscript received 27 February 2006. Initial review completed 24 April 2006. Revision accepted 11 May 2006. activators in in vitro experiments (16,17) and for very specific TABLE 1 Composition of purified diets1–3 sterols, also in vivo in mice (18). Cell lines frequently lack ABCG5/ABCG8 expression and therefore may accumulate Control Cholesterol Plant sterol Plant stanol T0901317 higher plant sterol and stanol levels than enterocytes in vivo. g/kg To test whether plant sterols and stanols present in functional food items are able to induce LXR-activated gene expression Cholesterol - 1.20 1.20 1.20 - in vivo, we fed C57BL/6 mice a purified diet virtually free of Plant sterol esters - - 8.33 - - cholesterol and the same diet enriched with cholesterol (0.12 g/ Plant stanol esters - - - 8.33 - 100 g), cholesterol and plant sterol fatty acid esters (0.12 and T0901317 - - - - 0.150 0.83 g/100g), or cholesterol and plant stanol fatty acid esters 1 All diets contained (g/kg): calcium caseinate, 161.40; vitamin mix, 11.30; mineral (0.12 and 0.83 g/100 g) for 4 wk. The plant sterol and stanol mix, 39.70; Arbocel BC-200, 56.70; fat (soybean oil), 126.10; carbohydrates (corn), fatty acid ester doses were equivalent to 0.5 g/100 g sterol or 599.90; L-cysteine HCl, 2.00; choline bitartrate, 2.80. stanol, respectively. A purified diet with the established LXR 2 The purified diets were provided by Unilever (Vlaardingen, The Netherlands). agonist T0901317 (0.015 g/100 g) was used as a positive control Analysis of sterol composition revealed that the control diet contained 0.03% (wt/wt) sterols, mainly b-sitosterol (53.8% of all sterols). Plant sterol fatty acid for gene expression studies (19). Furthermore, we measured esters were mainly b-sitosterol, campesterol, and stigmasterol (36.5, 20.4 and Downloaded from https://academic.oup.com/jn/article/136/8/2135/4664773 by guest on 25 September 2021 intestinal gene expression profiles and fractional cholesterol 14.9%, respectively). Plant stanol esters were mainly b-sitostanol and campestanol absorption in Abcg5 knock-out mice and their wild-type (50.3 and 24.6%). The plant sterol or stanol ester dose is equivalent to 5 g of sterol littermates fed both the cholesterol diet and the cholesterol or stanol/kg diet. 3 diet enriched with plant sterols. Enterocytes in Abcg5 knockout Vitamin mix (g/kg mix): nicotinic acid, 3; Ca pantothenate, 1.6; pyridoxine-HCl, 0.7; thiamine-HCl, 0.6; riboflavin, 0.6; folic acid, 0.2; biotin, 0.02; cyanocobalamin, 5; mice are not protected against plant sterol accumulation; as a all-rac-a-tocopheryl acetate (50%), 15; all-trans-retinyl palmitate, 0.8; cholecalciferol, consequence, they should be more susceptible to LXR-mediated 1; phylloquinone, 0.1; cornstarch, 971.38. The AIN-93G mineral mix contained (g/kg activation of gene expression by plant sterols. mix): calcium carbonate, 236.91; potassium hydrogen phosphate, 196; sodium The aim of this study was to determine whether the well- chloride, 74; magnesium oxide, 24; potassium citrate, 70.78; potassium sulfate, 46.6; AIN mineral mix, 91.71; cornstarch, 260 (38). characterized inhibitory action of plant sterols and plant stanols on intestinal cholesterol absorption is dependent on the activa- tion of LXR in different mouse models. Life Science) dissolved in Intralipid (20%; Fresenius Kabi) and an oral dose of 1.2 mCi of 14C-cholesterol (Amersham Bioscience) dissolved in medium-chain triglyceride oil. 14C and 3H activity was measured by Materials and Methods liquid scintillation counting. Blood samples obtained by retroorbital puncture 48 h after administration were used for the calculation of Animal experiments. Male, 3-mo-old C57BL/6J mice were purchased cholesterol absorption. from Harlan. Mice were housed in temperature-controlled rooms (21C) with 12-h light cycling and consumed a purified diet and water ad RNA isolation and PCR procedures. Total RNA was extracted from libitum. The composition of the purified diet (prepared by Unilever) is frozen tissues with TriReagent (Sigma) and quantified photometrically. given in Table 1. All mice were fed the purified control diet for 2 wk (run-in cDNA synthesis was performed using recombinant M-MLV reverse period); then, they were assigned to 1 of 5 treatment groups (n ¼ 6) based transcriptase (10 U/mL), the appropriate buffer, dNTPs (500 mmol/L), on their body weights and were fed the specific diets (control, cholesterol random nonamers (1 mmol/L), RNAse inhibitor (2 U/mL; all from Sigma) diet, plant sterol diet, plant stanol diet, T0901317 diet) for 4 wk. and total RNA (50 ng/mL). The reaction mix was incubated for 10 min at T0901317 was purchased from Cayman Chemicals. 25C for primer annealing, 60 min at 37C for synthesis, and 5 min at At the end of the experiment, feces were collected for 24 h. Mice were 94C to denature the RT enzyme. Real-time quantitative PCR was anaesthetized by i.p. injection with Hypnorm (fentanyl/fluanisone, 1 mL/ performed as previously described (20). Primers (Invitrogen) and kg) and diazepam (10 mg/kg). Bile and tissues were collected as described fluorogenic probes (Eurogentec) used in these studies were described (20). All experimental procedures were approved by the local Ethical elsewhere (Srebp1a, Srebp1c, Srebp2, Srb1, Acat2, Hmgcr, Cyp7a, Committee for Animal Experiments of the University of Groningen. Abca1, Abcg5, Abcg8, 18S rRNA (20); Ldlr (28); Npc1l1 (29). All data were subsequently normalized to 18S rRNA, which was analyzed in Analytical procedures. Biliary bile salt concentrations were measured separate runs. enzymatically (21). Phospholipids and cholesterol in bile were deter- mined as described by Bo¨ ttcher et al. (22) and Gamble et al. (23), Statistics. Statistical analyses were performed using SPSS 10.1 for respectively, after extraction according to Bligh and Dyer (24). The same Windows. Differences among the dietary groups (control, cholesterol, extraction method was applied for hepatic lipids, after which commer- plant sterol, plant stanols) were evaluated using Kruskal-Wallis analysis cially available kits were used for the determination of unesterified followed by the Mann-Whitney-U-test. The positive control (T0901317 cholesterol (Free cholesterol C, 279–47106, Wako), and for total diet) was not included in the statistical analysis. Data presented are cholesterol and triglycerides (Cholesterol CHOD-PAP, 11489232–216, means 6 SD. Differences with P , 0.05 were considered significant. and Tri/GB, 12146029–216, respectively; Roche). Plasma lipids were measured using the same kits. Cholesterol ester concentrations were calculated from total and unesterified cholesterol concentrations. Results Feces were weighed and homogenized to a powder. Aliquots of fecal powder were used for analysis of total neutral sterol and bile salt content Fecal neutral sterol excretion. Growth rates of male C57BL/ according to Arca et al. (25) and Setchell et al. (26), respectively. 6J mice fed the purified diet with or without cholesterol,

2/2 cholesterol and plant sterols, or cholesterol and plant stanols for Fractional cholesterol absorption in Abcg5 mice. Fractional 4 wk did not differ (data not shown), indicative of similar food cholesterol absorption in Abcg52/2 mice and wild-type littermates was intake in all groups. As anticipated, fecal neutral sterol excretion measured using the plasma dual-isotope ratio method as previously described (27). Male Abcg52/2 mice and littermate controls were housed in mice fed the cholesterol diet was increased by 56% and that of as described above. All mice were fed the purified diet for 2 wk (run-in bile salts by 131%; the latter is indicative of increased bile salt period). Then, mice were fed the purified diet enriched in cholesterol or synthesis. The addition of plant sterols and stanols induced a in cholesterol and plant sterols (Table 1) for 1 wk. Subsequently, mice massive further increase in fecal neutral sterol output compared were administered an i.v. injection of 2.4 mCi of 3H-cholesterol (NEN with mice receiving the cholesterol diet only (2.2- and 1.4-fold

2136 Plo¨ sch et al. for plant sterols and plant stanols, respectively; Fig. 1). Plant sterols reduced bile salt excretion by 31%. For comparison, treatment of mice fed the control diet with the LXR agonist T0901317 stimulated fecal neutral sterol and bile salt loss 3.6- and 1.7-fold, respectively, compared with mice fed the control diet only (data not shown).

Plasma and hepatic lipid composition. Plasma cholesterol, triglyceride, and phospholipid concentrations did not differ among the 4 groups (Supplemental Table 1). Nevertheless, mice fed the cholesterol diet had significantly greater concentrations of both unesterified cholesterol (1135%) and cholesterol ester (1925%) in their livers than those fed the control diet (Fig. 2). Together, this resulted in a hepatic cholesterol concentration that was 4-fold that of controls. Concurrently, the hepatic triglyceride Figure 2 Hepatic triglyceride (A) and cholesterol (B) concentrations in C57BL/ concentration was 1.8-times greater in cholesterol-fed mice than 6 mice fed the control diet alone or supplemented with cholesterol with or Downloaded from https://academic.oup.com/jn/article/136/8/2135/4664773 by guest on 25 September 2021 in mice fed the control diet. In mice fed plant sterols or plant without plant sterols or stanols for 4 wk. Values are means 6 SD, n ¼ 6. stanols, cholesterol and triglyceride concentrations were signifi- aDifferent from control; bdifferent from cholesterol, P , 0.05. cantly lower than in cholesterol-fed mice and did not differ from those in mice fed the control diet. T0901317-treated mice had synthesis. Concomitantly, the expression of the LXR target elevated hepatic triglyceride (5.2-fold) and total cholesterol gene Srebp1c was increased by cholesterol feeding alone concentrations (0.6-fold) compared with mice fed the control diet. (187%) and in combination with plant sterols and stanols Biliary cholesterol excretion is a major contributor to hepatic (189% and 187%, respectively). All other genes analyzed did cholesterol turnover and an important determinant of fractional not differ among groups. cholesterol absorption. Bile flow (Supplemental Table 2) and Gene expression was measured in 3 segments along the biliary bile salt output (not shown) did not differ among the lengths of the small intestine. The groups did not differ in the groups. Hepatobiliary sterol output rates did not differ signif- expression of Abcg5, Abcg8, Npc1l1 (Fig. 3), Srb1, Acat2,or icantly between groups fed cholesterol with or without plant Hmgcr (data not shown), indicating that treatment did not sterols and stanols because of large variations between individ- influence intracellular cholesterol homeostasis at the level of ual mice. gene expressions. However, large interindividual differences were noticed. The expression of Abca1 was significantly greater Hepatic and intestinal gene expression. To identify potential than in controls in the medial section of the small intestine of differences in LXR-mediated effects exerted by cholesterol and/ mice fed the cholesterol or plant sterol diet (Fig. 3). Feeding the or plant sterol and stanol feeding, gene expression patterns in T0901317 diet induced 2- to 9-fold increases in the expression of liver tissue were determined (Table 2). Mice fed the control diet Abcg5, Abcg8, and Abca1 along the whole axis of the small supplied with the synthetic LXR agonist T0901317 were used as intestine compared with the control diet. T0901317 did not a positive control. Feeding the cholesterol diet significantly affect the expression of Npc1l1, Srb1, Acat2,orHmgcr. upregulated the LXR target genes Abcg5 and Abcg8 (1194% and 1143%, respectively) encoding the major regulator of Fractional cholesterol absorption in Abcg52/2 and wild- hepatobiliary cholesterol excretion. In mice fed plant sterols or type mice. The Abcg5/Abcg8 heterodimer prevents accumula- stanols, this upregulation was significantly less pronounced tion of plant sterols in enterocytes and is thought to contribute to (187% for Abcg5 and 194% for Abcg8 with sterols and 163% regulation of cholesterol absorption. Mice lacking Abcg5 and 161% for Abcg5 and Abcg8, with stanols, respectively). (Abcg52/2) should therefore, in principle, be prone to LXR Expression of Abcg5 and Abcg8 in the mice treated with activation upon consumption of a plant sterol diet. We measured T0901317 was increased by .300% compared with mice fed intestinal gene expression levels and fractional cholesterol the control diet. A similar induction by cholesterol feeding, in absorption in wild-type and Abcg52/2 mice fed cholesterol parallel with reduction by plant sterols or stanols, occurred for only or cholesterol and plant sterol esters (Fig. 4). Unexpectedly, Cyp7a1, the gene encoding cholesterol-7a-hydroxylase. Expres- plant sterols did not differentially affect fractional cholesterol sion of Cyp7a1 is considered representative for bile salt absorption in wild-type and Abcg52/2 mice: plant sterol supplementation led to comparable significant reductions in fractional cholesterol absorption of 44% in wild-type mice and 49% in Abcg52/2 mice. Expression of typical LXR target genes did not differ between wild-type and Abcg52/2 mice (data not shown).

Discussion Since the early 1950s, plant sterols and stanols were known to interfere with intestinal cholesterol absorption and to lower plasma cholesterol concentrations in humans (1,2). Several Figure 1 Fecal sterol output of C57BL/6 mice fed the control diet alone or supplemented with cholesterol with or without plant sterols or stanols for 4 wk. concepts were proposed to explain this effect (30). First, plant Neutral sterols do not include plant sterols and stanols. Values are means 6 SD, sterols and stanols may replace cholesterol from micelles in the n ¼ 6. aDifferent from control; bdifferent from cholesterol, P , 0.05. intestinal lumen and therefore may lower the amount of

Plant sterols do not influence Abcg5 expression 2137 TABLE 2 Hepatic mRNA expression levels in male C57BL/6 mice fed a purified diet supplemented with cholesterol, plant sterols, or plant stanols for 4 wk1,2

Control Cholesterol Plant sterol Plant stanol T09013173

Abcg5 1.00 6 0.31 2.94 6 0.59a 1.87 6 0.61ab 1.63 6 0.45ab 3.59 6 1.37 Abcg8 1.00 6 0.29 2.43 6 0.31a 1.94 6 0.46ab 1.61 6 0.36ab 3.40 6 1.99 Abca1 1.00 6 0.15 1.38 6 0.15 1.23 6 0.28 1.10 6 0.36 1.39 6 0.29 Hmgcr 1.00 6 0.37 0.55 6 0.22 0.67 6 0.24 0.77 6 0.47 0.89 6 0.34 Srebp1a 1.00 6 0.14 1.09 6 0.18 1.12 6 0.25 1.03 6 0.33 1.16 6 0.29 Srebp1c 1.00 6 0.38 1.87 6 0.41a 1.89 6 0.64a 1.87 6 0.91a 2.42 6 1.11 Srebp2 1.00 6 0.16 0.70 6 0.18 0.79 6 0.17 0.69 6 0.27 0.98 6 0.32 Ldlr 1.00 6 0.37 0.99 6 0.23 1.02 6 0.32 0.91 6 0.38 1.33 6 0.42 Cyp7a1 1.00 6 0.33 3.30 6 1.59a 1.72 6 0.45a 1.95 6 1.06 2.36 6 0.51 Cyp8b1 1.00 6 0.53 1.09 6 0.30 0.80 6 0.29 0.68 6 0.20 1.39 6 0.23

1 Values are means 6 SD, n ¼ 6. aDifferent from control; bdifferent from cholesterol, P , 0.05. Downloaded from https://academic.oup.com/jn/article/136/8/2135/4664773 by guest on 25 September 2021 2 Gene expression was normalized to 18S rRNA. Expression of the control diet group was set to 1.00. 3 The T0901317-treated group was not included in the statistical analysis. cholesterol present in an absorbable form (5). Second, plant To determine whether this reduction was associated with sterols and stanols may block uptake of cholesterol into the changes in the expression of LXR target genes in the intestine or enterocyte or interfere with intracellular routing. This route may the liver, gene expression profiles of those tissues were analyzed. include the recently discovered ezetimibe-sensitive pathway Along the axis of the small intestine, no significant changes in involving NPC1L1 (6) and possibly aminopeptidase N (CD13) gene expression were observed, with the exception of Abca1, (31). Third, it was suggested that interference with the esteri- which was upregulated by cholesterol feeding in the medial fication machinery inside the enterocyte and inhibition of section of the small intestine. Plant sterol and stanol supple- chylomicron formation might be responsible for the observed mentation did not show any additional effect. As anticipated, plasma cholesterol-lowering effects of plant sterols, as proposed treatment of mice with the synthetic LXR agonist T0901317 almost 50 y ago (32). induced the expression of the LXR targets Abcg5, Abcg8, and Very recently, it was proposed that induction of intestinal Abca1 (20). Npc1l1, involved in intestinal cholesterol absorp- genes involved in the control of cholesterol absorption may be tion, was not influenced by plant sterol and stanol supplemen- involved (16–18, 30). During the last few years, it was shown tation nor by T0901317 treatment. Our observations are in line that enterocytes actively excrete sterols back to the intestinal with the very recent findings of Calpe-Berdiel et al. (35), which lumen by the action of the Abcg5/Abcg8 heterodimer. This were published while this work was in progress. These authors process is regulated at the level of gene expression by the nuclear fed different mouse strains a Western-type diet with and without transcription factor LXR upon binding of oxysterols. Because phytosterols and did not find differences in the intestinal plant sterols are able to activate LXR either directly or after expression of LXR target genes. We conclude that the observed oxidation to oxyphytosterols in in vitro studies (16–18), it is effects of plant sterols and stanols on cholesterol absorption are possible that the lowering of cholesterol absorption by plant not caused by LXR-mediated induction of intestinal gene sterols and stanols may also be due to LXR activation and expression in mice but must be the result of upstream events. subsequent increased expression of Abcg5/Abcg8 in enterocytes. This conclusion is further supported by gene expression Stigmasterol, a plant sterol present in many commercially avail- profiles in liver: feeding cholesterol alone did induce the able plant sterol formulations, was shown to activate LXR in the expression of Abcg5 and Abcg8. Based on the assumption that adrenal gland of mice lacking both Abcg5 and Abcg8 (33). LXR activation by plant sterol- and stanol-derived metabolites To test the hypothesis that the decrease in cholesterol occurs, we expected an additional increase in Abcg5/g8 expres- absorption by plant sterols and stanols is related to induction sion as a result of plant sterol or stanol supplementation. of Abcg5/g8 expression mediated by LXR, we fed C57BL/6 mice However, the opposite occurred: the addition of plant sterols or a diet virtually free of sterols, enriched with cholesterol, or stanols partially prevented the effects usually associated with enriched with cholesterol and either plant sterols or stanols for cholesterol feeding, i.e., it lowered the hepatic expression of 4 wk. Fecal neutral sterol excretion increased with the addition Abcg5/g8 compared with mice fed cholesterol only. of plant sterols or stanols to the cholesterol diet, indicating the Our data contrast with those of recently published in vitro expected inhibitory effect on intestinal cholesterol absorption studies demonstrating LXR activation by plant sterols in cell potentially enforced by an enhanced hepatobiliary clearance of lines (16–18). Cell lines frequently differ from their source cholesterol. The latter possibility, however, was excluded tissues in gene expression profiles. The commonly used CACO-2, because hepatobiliary lipid excretion rates in all groups of HepG2 or HEK293 cell lines, for example, express only limited mice were not affected by plant sterol or stanol feeding. levels of Abcg5 and Abcg8, which makes them prone to the In contrast to the situation described in humans, mice did not accumulation of plant sterols in vitro. This limits the conclusions have lower plasma cholesterol concentrations due to plant sterol that can be drawn from experiments in which cells were loaded and stanol supplementation but did have the expected reduction with plant sterols and stanols (36). We speculated that in vivo, in cholesteryl ester concentration in the liver. Hepatic cholesteryl enterocytes do not react to plant sterols and stanols via LXR ester concentrations were shown to reflect intestinal cholesterol because they are largely protected from their accumulation by uptake in mice (34). We therefore conclude that supplementa- the action of Abcg5/g8. We therefore measured fractional tion of the cholesterol-containing diet with either plant sterols or cholesterol absorption in mice lacking Abcg5 which, as a stanols indeed reduces intestinal cholesterol absorption in mice. consequence, hyperabsorb plant sterols and accumulate them in

2138 Plo¨ sch et al. Figure 4 Fractional cholesterol absorption in Abcg52/2 and wild-type mice fed a diet supplemented with cholesterol or cholesterol and plant sterols for 1 wk. Values are means 6 SD, n ¼ 5–6. aDifferent from cholesterol alone, P , 0.05. Downloaded from https://academic.oup.com/jn/article/136/8/2135/4664773 by guest on 25 September 2021

additional evidence that the cholesterol-lowering effects of plant sterols and stanols are independent of LXR. Based on our findings, we speculate that a combination of dietary plant sterol and stanol supplementation with ezetimibe would have cumulative inhibitory effects on intestinal choles- terol absorption in humans because plant sterols and stanols would putatively lower the relative amount of cholesterol available for absorption, whereas inhibition of NPC1L1 by ezetimibe would decrease total sterol uptake. Very recent data (37) demonstrate that a combination of dietary plant sterols and ezetimibe has no additional effect on plasma LDL cholesterol concentrations in human hypercholesterolemic patients com- pared with the monotherapies. However, plasma lathosterol concentrations were significantly higher upon treatment with plant sterols combined with ezetimibe compared with single treatments, which may indicate a greater reduction in the absorption of cholesterol, which is compensated for by a more pronounced increase in (hepatic) cholesterol synthesis. Data presented in this study demonstrate that induction of Abcg5/Abcg8 transporter activity by LXR activation does not play a role in plant sterol- and stanol-induced reduction of intestinal cholesterol absorption in mice.

Acknowledgments Figure 3 Intestinal gene expression in C57BL/6 mice fed the control diet alone or supplemented with cholesterol with or without plant sterols or Elke Trautwein is kindly acknowledged for the useful discus- stanols for 4 wk. Expression was measured along 3 points of the small intestine sions on the design and manuscript of the study. Wim Kloots (labeled proximal, mid, and distal) and corrected for that of 18S rRNA. T091317 was very instrumental in preparing the experimental diets. We (T09) was used as a positive control for LXR activation (not included in the are grateful to Renze Boverhof for fecal sterol analysis and to statistical analysis). Values are means 6 SD, n ¼ 6. aDifferent from control, P , 0.05. Paula Jansen and Dieter Lu¨ tjohann for plant sterol analyses and helpful discussions. plasma and various tissues, including the intestinal mucosa (15). Moreover, mice lacking both Abcg5 and Abcg8 were shown to Literature Cited be prone to LXR activation by certain sterols, namely, stigmas- terol, in selected tissues (33). 1. Pollak OJ. Reduction of blood cholesterol in man. Circulation. 1953;7:702–6. When Abcg52/2 mice and their wild-type littermates where 2. Best MM, Duncan CH, van Loon EJ, Wathen JD. Lowering of serum fed cholesterol-rich diets, they did not differ in fractional cholesterol by the administration of a plant sterol. Circulation. cholesterol absorption, in line with previous studies (15). 1954;10:201–6. Moreover, the addition of plant sterols to the cholesterol diet 3. Ikeda I, Tanaka K, Sugano M, Vahouny GV, Gallo LL. Inhibition of reduced fractional cholesterol absorption to the same extent in cholesterol absorption in rats by plant sterols. J Lipid Res. Abcg52/2 mice and their wild-type littermates. This clearly 1988;29:1573–82. demonstrates that the decline in cholesterol absorbed from the 4. Ikeda I, Tanaka K, Sugano M, Vahouny GV, Gallo LL. Discrimination between cholesterol and sitosterol for absorption in rats. J Lipid Res. small intestine upon plant sterol supplementation is independent 1988;29:1583–91. of the action of the Abcg5/Abcg8 heterodimer. Moreover, we did 5. Ikeda I, Tanabe Y, Sugano M. Effects of sitosterol and sitostanol on not detect differences in intestinal gene expression responses to micellar solubility of cholesterol. J Nutr Sci Vitaminol (Tokyo). 2 2 plant sterols between Abcg5 / and wild-type mice, providing 1989;35:361–9.

Plant sterols do not influence Abcg5 expression 2139 6. Altmann SW, Davis HR Jr, Zhu LJ, Yao X, Hoos LM, Tetzloff G, Iyer 21. Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum SP, Maguire M, Golovko A, et al. Niemann-Pick C1 like 1 protein is bile acids. Clin Chim Acta. 1976;70:79–86. critical for intestinal cholesterol absorption. Science. 2004;303:1201–4. 22. Bo¨ttcher CFJ, van Gent CM, Pries C. A rapid and sensitive sub-micro 7. Davis HR, Zhu LJ, Hoos LM, Tetzloff G, Maguire M, Liu J, Yao X, Iyer phosphorus determination. Anal Chim Acta. 1961;24:203–4. SP, Lam MH, et al. Niemann-Pick C1 Like 1 (NPC1L1) is the intestinal 23. Gamble W, Vaughan M, Kruth HS, Avigan J. Procedure for determi- phytosterol and cholesterol transporter and a key modulator of whole nation of free and total cholesterol in micro- or nanogram amounts body cholesterol homeostasis. J Biol Chem. 2004;279:33586–92. suitable for studies with cultured cells. J Lipid Res. 1978;19:1068–70. 8. Lee MH, Lu K, Hazard S, Yu H, Shulenin S, Hidaka H, Kojima H, 24. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and Allikmets R, Sakuma N, et al. Identification of a gene, ABCG5, purification. Can J Biochem Biophys. 1959;37:911–7. important in the regulation of dietary cholesterol absorption. Nat 25. Arca M, Montali A, Ciocca S, Angelico F, Cantafora A. An improved Genet. 2001;27:79–83. gas-liquid chromatographic method for the determination of fecal 9. Berge KE, Tian H, Graf GA, Yu L, Grishin NV, Schultz J, Kwiterovich P, neutral sterols. J Lipid Res. 1983;24:332–5. Shan B, Barnes R, Hobbs HH. Accumulation of dietary cholesterol in 26. Setchell KD, Lawson AM, Tanida N, Sjovall J. General methods for the sitosterolemia caused by mutations in adjacent ABC transporters. analysis of metabolic profiles of bile acids and related compounds in Science. 2000;290:1771–5. feces. J Lipid Res. 1983;24:1085–100. 10. McNeish J, Aiello RJ, Guyot D, Turi T, Gabel C, Aldinger C, Hoppe 27. van der Veen JN, Kruit JK, Havinga R, Baller JF, Chimini G, Lestavel S, KL, Roach ML, Royer LJ, et al. High density lipoprotein deficiency and Staels B, Groot PH, Groen AK, Kuipers F. Reduced cholesterol

foam cell accumulation in mice with targeted disruption of ATP-binding Downloaded from https://academic.oup.com/jn/article/136/8/2135/4664773 by guest on 25 September 2021 absorption upon PPARdelta activation coincides with decreased intes- cassette transporter-1. Proc Natl Acad Sci U S A. 2000;97:4245–50. tinal expression of NPC1L1. J Lipid Res. 2005;46:526–34. 11. Drobnik W, Lindenthal B, Lieser B, Ritter M, Christiansen WT, Liebisch 28. Grefhorst A, Elzinga BM, Voshol PJ, Plo¨ sch T, Kok T, Bloks VW, van G, Giesa U, Igel M, Borsukova H, et al. ATP-binding cassette der Sluijs FH, Havekes LM, Romijn JA, et al. Stimulation of lipogenesis transporter A1 (ABCA1) affects total body sterol metabolism. Gastro- by pharmacological activation of the liver X receptor leads to enterology. 2001;120:1203–11. production of large, triglyceride-rich very low density lipoprotein 12. Mulligan JD, Flowers MT, Tebon A, Bitgood JJ, Wellington C, Hayden particles. J Biol Chem. 2002;277:34182–90. MR, Attie AD. ABCA1 is essential for efficient basolateral cholesterol efflux during the absorption of dietary cholesterol in chickens. J Biol 29. Kruit JK, Plo¨ sch T, Havinga R, Boverhof R, Groot PH, Groen AK, Chem. 2003;278:13356–66. Kuipers F. Increased fecal neutral sterol loss upon liver X receptor activation is independent of biliary sterol secretion in mice. Gastroen- 13. Schwartz K, Lawn RM, Wade DP. ABC1 gene expression and apoA-I- terology. 2005;128:147–56. mediated cholesterol efflux are regulated by LXR. Biochem Biophys Res Commun. 2000;274:794–802. 30. Ostlund RE Jr. Phytosterols and cholesterol metabolism. Curr Opin Lipidol. 2004;15:37–41. 14. Yu L, York J, von Bergmann K, Lu¨tjohann D, Cohen JC, Hobbs HH. Stimulation of cholesterol excretion by LXR agonist requires 31. Kramer W, Girbig F, Corsiero D, Pfenninger A, Frick W, Jahne G, Rhein ATP-binding cassette transporters G5 and G8. J Biol Chem. 2003;278: M, Wendler W, Lottspeich F, et al. Aminopeptidase N (CD13) is a 15565–70. molecular target of the cholesterol absorption inhibitor ezetimibe in the enterocyte brush border membrane. J Biol Chem. 2005;280:1306–20. 15. Plo¨ sch T, Bloks V, Terasawa Y, Berdy S, Siegler K, van der Sluijs F, Kema I, Groen A, Shan B, et al. Sitosterolemia in ABC-Transporter G5- 32. Swell L, Trout EC Jr, Field H Jr, Treadwell CR. Intestinal metabolism of deficient mice is aggravated on activation of the liver-X receptor. C14-phytosterols. J Biol Chem. 1959;234:2286–9. Gastroenterology. 2004;126:290–300. 33. Yang C, Yu L, Li W, Xu F, Cohen JC, Hobbs HH. Disruption 16. Plat J, Mensink RP. Increased intestinal ABCA1 expression contributes of cholesterol homeostasis by plant sterols. J Clin Invest. 2004;114: to the decrease in cholesterol absorption after plant stanol consumption. 813–22. FASEB J. 2002;16:1248–53. 34. Schwarz M, Davis DL, Vick BR, Russell DW. Genetic analysis of 17. Plat J, Nichols JA, Mensink RP. Plant sterols and stanols: effects on intestinal cholesterol absorption in inbred mice. J Lipid Res. mixed micellar composition and LXR (target gene) activation. J Lipid 2001;42:1801–11. Res. 2005;46:2468–76. 35. Calpe-Berdiel L, Escola-Gil JC, Ribas V, Navarro-Sastre A, Garces- 18. Kaneko E, Matsuda M, Yamada Y, Tachibana Y, Shimomura I, Garces J, Blanco-Vaca F. Changes in intestinal and liver global gene Makishima M. Induction of intestinal ATP-binding cassette transporters expression in response to a phytosterol-enriched diet. Atherosclerosis. by a phytosterol-derived liver X receptor agonist. J Biol Chem. 2005;181:75–85. 2003;278:36091–8. 36. Ho SS, Pal S. Margarine phytosterols decrease the secretion of 19. Repa JJ, Turley SD, Lobaccaro JA, Medina J, Li L, Lustig K, Shan B, atherogenic lipoproteins from HepG2 liver and Caco2 intestinal cells. Heyman RA, Dietschy JM, Mangelsdorf DJ. Regulation of absorption Atherosclerosis. 2005;182:29–36. and ABC1-mediated efflux of cholesterol by RXR heterodimers. 37. Jakulj L, Trip MD, Sudhop T, von Bergmann K, Kastelein JJ, Vissers Science. 2000;289:1524–9. MN. Inhibition of cholesterol absorption by the combination of dietary 20. Plo¨ sch T, Kok T, Bloks VW, Smit MJ, Havinga R, Chimini G, Groen plant sterols and ezetimibe: effects on plasma lipid levels. J Lipid Res. AK, Kuipers F. Increased hepatobiliary and fecal cholesterol excretion 2005;46:2692–8. upon activation of the liver X receptor is independent of ABCA1. J Biol 38. Reeves PG. Components of the AIN-93 diets as improvements in the Chem. 2002;277:33870–7. AIN-76A diet. J Nutr. 1997;127:838S–41.

2140 Plo¨ sch et al.