A role for high density lipoproteins in hepatic phosphatidylcholine homeostasis

Biochimica et Biophysica Acta 1771 (2007) 893 – 900 www.elsevier.com/locate/bbalip

A role for high density lipoproteins in hepatic phosphatidylcholine homeostasis Zhaoyu Li, Luis B. Agellon, Dennis E. Vance ⁎ Department of Biochemistry and Canadian Institutes of Health Research Group on the Molecular and Cell Biology of Lipids, University of Alberta, Edmonton, Alberta, Canada T6G 2S2 Received 2 March 2007; received in revised form 10 April 2007; accepted 16 April 2007 Available online 21 April 2007

Abstract Choline is (95%) found largely in the biosphere as a component of phosphatidylcholine (PC) which is made from choline via the CDP-choline pathway. Animals obtain choline from both the diet and via endogenous biosynthesis that involves the conversion of phosphatidylethanolamine into PC by phosphatidylethanolamine N-methyltransferase (PEMT), followed by PC catabolism. We have uncovered a striking gender-specific conservation of choline in female mice that does not occur in male mice. Female Pemt−/− mice maintained hepatic PC/total choline levels during the first day of choline deprivation and escaped liver damage whereas male Pemt−/− mice did not. Plasma PC levels in high-density lipoproteins (HDLs) were higher in male Pemt−/− mice than those in females before choline deprivation. Interestingly, after choline deprivation for 1 day, female, but not male, Pemt−/− mice increased HDL-PC levels. Glybenclamide, an inhibitor of PC efflux mediated by ABC transporters, eliminated this response to choline deprivation in females. These data suggest that (i) increased PC efflux from extra-hepatic tissues to HDLs in the circulation provided sufficient choline for the liver and compensated for loss of hepatic PC during the initial stages of choline deprivation in female, but not male, Pemt−/− mice, and (ii) plasma HDL in female mice has an important function in maintenance of hepatic PC as an acute response to severe choline deprivation. © 2007 Elsevier B.V. All rights reserved. Keywords: Choline; Phosphatidylcholine; Phosphatidylethanolamine N-methyltransferase; High density lipoprotein; Choline deficiency; Glybenclamide

1. Introduction Choline is an important nutrient for animals [1,2]. Phosphatidylcholine (PC) accounts for N 95% of total choline-containing metabolites with the remainder found in sphingomyelin, glycerophosphocholine, phosphocholine, CDP-choline and choline [3]. Animals obtain choline from two pathways: the diet and endogenous biosynthesis. The only endogenous biosynthesis pathway for choline biosynthesis in mammals involves the conversion of phosphatidylethanolamine (PE) into PC by PE N-methyltransferase (PEMT), followed by PC Abbreviations: PC, phosphatidylcholine; ALT, alanine aminotransferases; AST, aspartate aminotransferaase; CD, choline-deficient; CS, choline-supplemented; PE, phosphatidylethanolamine; PEMT, PE N-methyltransferase; SRB1, scavenger receptor B1; ABCA1, ATPase-binding cassette transporter A1; VLDL, very-low density lipoprotein; HDL, high density lipoprotein ⁎ Corresponding author. Tel.: +1 780 492 8286; fax: +1 780 492 3383. E-mail address: [email protected] (D.E. Vance). 1388-1981/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.bbalip.2007.04.009

catabolism [3]. However, PC provides only a short-term source of choline since complete choline deprivation, achieved by feeding Pemt−/− mice a choline-deficient (CD) diet, is rapidly lethal [2,4]. Nevertheless, mice can survive with basal levels of total choline-containing metabolites (∼ 50% of original level) in the liver during choline deprivation when choline is recycled and redistributed [2,5]. CD-Pemt−/− mice die of liver failure within 4–5 days [4] indicating that Pemt−/− mice are unable to cope with complete choline deprivation. Nevertheless, we explored whether or not CD-Pemt−/− mice initiate a response during the early stages of choline deprivation. PC in plasma high-density lipoproteins (HDLs) is the source of ∼ 40% of biliary PC secreted by the liver, indicating that HDL-PC is an important source for hepatic PC [6]. We found that Pemt−/− mice responded to choline deprivation by mobilizing choline stored in extrahepatic tissues through HDL-PC. However, the appearance of deleterious consequences of choline deficiency was delayed by 1 day in

894

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

female Pemt−/− mice. This gender-specific effect may represent an inherent capacity of female mice to withstand sustained periods of choline insufficiency. 2. Materials and methods 2.1. Animals Pemt−/− mice (C57BL/6; 129/J background) [7] were fed a choline-deficient (CD) diet [a semi-synthetic diet without choline (ICN, Cat #0290138710)] or a choline-supplemented (CS) diet [a CD diet containing 0.4% (w/w) choline chloride]. At the age of 10 to 12 weeks, Pemt−/− mice were fed the CS diet for 24 h (0 day) and then fed the CD diet for 1, 2 or 3 days. Mice were fasted for 12 h before sacrifice. Six to eight mice of each gender were used for each time point in all experiments and assays were performed in duplicate. All data are means ± S.D.

2.2. Assay of liver damage Plasma samples were collected from Pemt−/− mice fed the cholinesupplemented (CS) diet for 24 h (0 day). The mice were subsequently fed the choline-deficient (CD) diet for 3 days. Plasma alanine/aspartate aminotransferase (ALT/AST) activities were measured with a GPT/GOT Kit (Sigma, catalog #P505) as indicators of liver damage. Plasma ALT/AST activities were measured at days 0, 1, 2 and 3.

2.3. Lipid analysis Blood was collected by cardiac puncture with instruments pre-treated with EDTA. Plasma was separated by centrifugation at 2,000 rpm for 20 min in a refrigerated bench-top centrifuge. Bile was collected from intact gall bladders. All samples were stored at − 70 °C before use. Livers were frozen in liquid N2 after dissection. Livers were homogenized with a Polytron in 5 volumes of 10 mM Tris–HCl, pH 7.2, 150 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1:100 protease inhibitor cocktail (Sigma, Cat#P8340). Homogenates were centrifuged for 5 min at 600×g after sonication and supernatants collected. Protein was quantified [8] and total lipids were extracted from liver homogenates [9]. Phospholipids were separated by highperformance liquid chromatography and quantified with an electron-light scattering detector [10]. Phosphatidyldimethylethanolamine was used as an internal standard for quantification.

2.4. Profiles of choline-containing lipids in plasma lipoproteins Fresh plasma samples were combined from 3 to 4 mice and lipoprotein classes were separated by fast protein liquid chromatography. The distribution of choline-containing compounds in lipoprotein classes was assessed with the Phospholipids B kit (Wako).

2.5. Treatment of mice with glybenclamide Female Pemt−/− mice were fed the choline-supplemented (CS) diet containing glybenclamide (Sigma,163 μg/g) for 1 day, and then fed the CD diet containing glybenclamide (163 μg/g diet) for 1 day. Glybenclamide (also called glyburide) is a known inhibitor of ATP-binding cassette (ABC) transporters such as ABCA1 and the cystic fibrosis transmembrane conductance regulator [11–14]. We used glybenclamide to reduce HDL formation via inhibiting ABCA1-mediated PC efflux to apo A1/small HDLs. Control female Pemt−/− mice were fed the CS or CD diet without glybenclamide. Wild-type mice were fed the CS or CD diet containing 163 μg/g glybenclamide.

2.6. Pre-β HDL and apo AI secretion from primary cultured hepatocytes Male and female Pemt−/− mice were fed the CS diet for 1 day and then fed the CD diet for 1 day. Primary hepatocytes were isolated [2,15] and cultured in

CS or CD medium [15]. After the hepatocytes had adhered to the dish (∼ 2 h) the medium was changed to medium without serum. After 1 h the medium was changed and the hepatocytes were incubated in medium without serum for 2 h. The medium was collected and concentrated through a centrifugation filter with a pore size of 5000 kDa. The concentrated medium was loaded on to a 6% nondenaturing polyacrylamide gel for separation of HDL and pre-β HDL particles. Proteins were transferred to polyvinylidene fluoride membranes and blotted with an anti-apo A1 antibody. Proteins in concentrated medium were also separated by electrophoresis on 10%polyacrylamide gels containing 0.1% SDS followed by immunoblotting with a rabbit anti-mouse apoA1 polyclonal antibody, (Biodesign) to measure the amount of apo AI protein by densitometric scanning of the gels. Sandwich enzyme-linked immunosorbent assays (ELISA) were used for analyzing the amount of PC associated with apo AI. A 96-well plate was first incubated with anti-apo AI antibody (dilution 1:400) for 24 h at 4 °C. Concentrated medium was applied to the plate for 30 min, and then discarded (phosphate-buffered saline was used as a negative control). An anti-PC antibody (from Dr. Masato Umeda, Kyoto, Japan; dilution 1:100) was applied to the plate (to measure the amount of PC associated with apo AI) followed by a secondary antibody conjugated to fluorescein isothiocyanate (FITC conjugated goat AntiMouse IgM Antibody, 1:100) [16]. Fluorescence intensity was measured with a fluorimeter and was calculated relative to the fluorescence intensity of male CS mice.

2.7. Clearance and hepatic uptake of plasma HDL-PC HDLs were isolated from mouse plasma by gradient ultracentrifugation followed by dialysis against 0.9% saline [17]. BODIPY-PC (D3771, Invitrogen) dissolved in 95% ethanol was added to purified HDLs and incubated at 37 °C for 12 h. BODIPY-labeled HDLs (25 μg protein) in 100 μl saline were injected into mice via the tail-vein. Blood samples were collected at 10, 30, 60 and 120 min after injection. At 120 min after injection, the liver was removed, perfused with Liver Perfusion Medium (Gibco) then frozen at − 70 °C until use. Fluorescence intensity of PC in plasma and liver homogenates was observed at 503 nm and 512 nm (excitation and emission, respectively). The amount of HDL-PC indicated by BODIPY fluorescence in a sample was calculated using the formula: fluorescence intensity/(injected total fluorescence intensity/plasma HDL-PC) The rate of plasma HDL-PC clearance was calculated from the amount of HDL-PC in plasma at different times. Results were normalized to data from female CS-Pemt−/− mice. Each group included 3 or 4 mice.

2.8. Immunoblotting for hepatic SR-B1 and ABCA1 Liver homogenate proteins were separated by electrophoresis on 8% polyacrylamide gels containing 0.1% SDS and then transferred to polyvinylidene difluoride membranes for immunoblotting with a rabbit anti-mouse SR-B1 antibody (dilution 1:1000, Novus) or a rabbit anti-mouse ABCA1 antibody (dilution 1:1,000, Novus). Protein disulfide isomerase (PDI) was immunoblotted with a rabbit anti-mouse PDI antibody (1:4,000, Novus) as a loading control. Intensity of bands was determined by densitometric scanning of the blots. The amounts of the proteins were normalized to protein disulfide isomerase.

3. Results 3.1. Gender-dependent response to choline deprivation in livers of Pemt−/− mice Analysis of markers of liver damage, plasma ALT/AST activities, showed that when male Pemt−/− mice were fed the CD diet, liver damage was detected within 1 day and plasma ALT/AST activities increased during 3 days of choline deprivation (Fig. 1). However, female Pemt−/− mice did not experience any liver damage during the first day of being fed the CD diet. Liver damage was evident only after the first day of choline deprivation (Fig. 1). Thus, liver damage was delayed by

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

895

3.3. Gender-dependent PC homeostasis in the liver

Fig. 1. Liver damage assays. Plasma was collected from Pemt−/− mice fed the choline-supplemented (CS) diet for 24 h (0 dy) then transferred to the cholinedeficient (CD) diet for 1 to 3 days. Plasma alanine and aspartate aminotransferase (ALT/AST) activities were measured at days 0, 1, 2 and 3 as indicators of liver damage. (A) Plasma ALT activity; (B) plasma AST activity. Data are averages ± S.D. from at least 6 mice of each gender. In some cases error bars are too small to see. *P b 0.05, comparison between CS and CD mice.

1 day in female Pemt−/− mice that were deprived of choline compared to male mice (Fig. 1). 3.2. Gender-dependent difference in PC/PE ratios and PC levels in the liver The ratio of PC to PE was recently suggested to be a key regulator of membrane integrity because liver damage is induced when the hepatic PC/PE ratio is decreased [15]. We provide additional evidence to support this theory by studies on the gender-dependent status of liver damage in Pemt−/− mice during choline deprivation. After 1 day of choline deprivation, the PC/PE ratio in livers of female Pemt−/− mice did not decrease whereas in male Pemt−/− mice this ratio decreased from 2.0 to 1.2 (Fig. 2A) (P b 0.05). Since no liver damage was induced in female Pemt−/− mice after 1 day, but the plasma levels of AST and ALT increased markedly in males (Fig. 1), a correlation existed between the hepatic PC/PE ratio and liver damage in Pemt−/− mice. In our earlier study we observed that a decrease in the PC/PE ratio in CD-Pemt−/− mice resulted from a decrease in the level of PC and an unchanged level of PE [15]. Three days of choline deprivation did not cause any significant change in the level of PE in livers of either male or female mice (Fig. 2B). Thus, the gender difference in the alteration of PC/PE ratio during the first day of choline distribution is due to the differential change of PC levels in male and female Pemt−/− livers.

PC homeostasis in the liver reflects PC anabolism and catabolism, as well as PC secretion and uptake [2,3]. PC concentrations in the gall bladder mirrored hepatic PC levels (Fig. 2B, C). Female, but not male, Pemt−/− mice were able to maintain normal hepatic PC levels for 1 day of choline deprivation (Fig. 2B). Similarly, PC in the gall bladders from female Pemt−/− mice was not decreased during this time (Fig. 2C). A possible explanation for the conservation of hepatic PC in female Pemt−/− mice was that the female mice had decreased their output of PC into bile. However, this does not appear to be the case because the PC content of the gall bladder reflected the PC content of the liver. Interestingly, when fed the CS diet, the female Pemt−/− mice had markedly lower levels of plasma PC compared with males (Fig. 2D). However, after 1 day of the CD diet, the amount of PC in plasma of the female mice approximately doubled, then declined to levels similar to those in the male mice. Since most PC in mouse plasma is associated with HDLs [18], the alterations in plasma PC level completely reflected the HDLPC levels in both female and male Pemt−/− mice (Fig. 3A, B). Thus, when Pemt−/− mice were fed the CD diet, the female, but not the male, mice increased their plasma PC by increasing HDL-PC levels (Figs. 2D, 3A, B). 3.4. Glybenclamide lowers HDL-PC and induces liver damage in female Pemt−/− mice The efflux of cholesterol and PC to HDLs is mediated by ABCA1 and is inhibited by glybenclamide [12,19]. Wild-type mice fed glybenclamide (163 μg/g diet) did not show any liver damage. After 1 day, the presence of glybenclamide (163 μg/g diet) in the diet of female CD-Pemt−/− mice markedly reduced HDL-PC but not in female Pemt−/− mice fed the CS diet (Fig. 4A). Hepatic PC levels decreased by ∼ 30% (P b 0.05) and the PC/PE ratio decreased by ~ 40% as a result of feeding glybenclamide (Fig. 4B, C). Glybenclamide induced liver damage in female Pemt−/− mice within 1 day of choline deprivation (Fig. 4D) but did not induce liver damage after 1 day in female Pemt−/− mice fed the CS diet (Fig. 4D) or in wild-type mice fed either the CS or CD diet for up to 2 days (plasma ALT activities in wild-type mice fed glybenclamide for 2 days: 7.33 ± 0.89 IU/l). Thus, glybenclamide lowers HDL-PC and induces liver damage in female Pemt−/− mice fed a CD diet. 3.5. Metabolism of HDL-PC The efflux of hepatic PC to apo AI from primary hepatocytes isolated from male and female Pemt−/− mice fed the CS or CD diet for 1 day was measured. Fig. 5A shows that hepatocytes from male and female CS mice produced pre-β HDLs but not mature HDLs. Dietary choline deprivation markedly decreased the formation of pre-β HDLs in both male and female Pemt−/− mice (Fig. 5A) although the total amount of apo AI secreted from hepatocytes was not changed significantly (Fig. 5B). The amount of PC associated with apo AI was significantly reduced

896

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

Fig. 2. Phospholipid composition in liver, bile and plasma. Pemt−/− mice were fed the choline-supplemented (CS) diet for 24 h and then fed a choline-deficient (CD) diet for 1 to 3 days. (A) Molar ratio of PC to PE in liver. (B) Amounts of PC and PE in liver. (C) Concentration of biliary PC. (D) Concentration of plasma PC. Data are averages ± S.D. from at least 6 mice of each genotype *P b 0.05 for comparisons between CS and CD mice.

by choline deprivation in both male and female Pemt−/− mice (Fig. 5C). Thus, poor-lipidated pre-β HDLs were secreted by hepatocytes in both male and female Pemt−/− mice during choline deprivation. We next compared the uptake of HDL-PC by livers of male and female Pemt−/− mice fed the CS and CD diet for 1 day. BODIPY-PC labeled HDLs was injected into mice via the tail vein. Blood was collected at 10, 30, 60 and 120 min after injection. After 120 min the livers were perfused with the Liver Perfusion Medium and fluorescence intensity in liver homogenates and plasma was measured with a fluorimeter. The amount of BODIPY-PC in the liver reflects the uptake of HDLPC by the liver. Livers of female Pemt−/− mice fed the CS diet for 1 day contained ∼ 60% less BODIPY-PC than did livers of female mice fed the CD diet or Pemt−/− male mice fed either the CD or CS diet (Fig. 6A), indicating that the hepatic uptake of HDL-PC is significantly upregulated in female compared to male CD- Pemt−/− mice (Fig. 6A). The hepatic fluorescence in male Pemt−/− mice was the same when fed the CS or CD diet. Consistent with these observations on PC uptake by the livers, the clearance of plasma fluorescence was also higher in female CD-Pemt−/− mice than in female mice fed the CS diet. However, in male mice, the level of plasma fluorescence was unaffected by choline deprivation (Fig. 6B). One possible reason for the observed difference in HDL-PC uptake was that the expression of SR-B1, an HDL receptor, was different in the livers. Immunoblotting experiments indicated that hepatic SR-B1 protein levels were the same in female or male mice fed either the CS or CD diet (Fig. 6C). Similarly, the amount of ABCA1, a protein required for PC efflux from the liver, was the same in male and female mice fed the CS or CD diet (Fig. 6C). Thus,

increased HDL-PC levels in plasma mainly resulted from PC efflux from extrahepatic tissues in female Pemt−/− mice during choline deprivation. 4. Discussion Dietary choline deprivation causes a rapid and significant reduction of hepatic PC levels that leads to severe liver damage in Pemt−/− mice. The results of the present experiments uncovered a major difference in the response of male and female Pemt−/− mice during the early stage of total choline deprivation. Whereas male mice experienced a large decline in hepatic PC levels within 1 day after the start of dietary choline deficiency, female mice were able to maintain their hepatic PC level. Subsequently, female Pemt−/− mice exhibited a significant decrease in hepatic PC and eventual liver failure. We found that plasma PC levels (associated with HDL) increased in female Pemt−/− mice during the first day of choline deprivation. The increased hepatic uptake of HDL-PC in female Pemt−/− mice suggests that PC transport via HDL was important for the maintenance of PC homeostasis in the liver. We tested this idea by preventing PC efflux using glybenclamide, an inhibitor of ABCA1 and other ABC transporters [12,19]. Administration of this drug to female CD-Pemt−/− mice abolished the 1-day delay of liver damage. Thus, the data suggest that most of the PC used to maintain the hepatic PC level in female CD-Pemt−/− mice is derived from extra-hepatic sources. Complete choline deprivation is not likely to occur in nature even during periods of starvation because the PEMT pathway provides an endogenous source of choline. However, choline insufficiency might occur and the hepatic need for choline could

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

897

esis in the livers of CD-Pemt−/− mice [2,4]. PC catabolism was also reduced in the livers of CD-Pemt−/− mice [2]. Biliary PC secretion appears to be dependent on the pool size of hepatic PC (Fig. 2B, C), which suggests that down-regulation of biliary PC secretion might not be a mechanism for adaptation to choline deprivation. Secretion of VLDL by Pemt−/− hepatocytes is significantly reduced during choline deprivation [15,20]. Fig. 3 also shows that PC levels in VLDL are decreased in both female and male mice during choline deprivation. PC loading to apo AI from hepatocytes is dramatically decreased in both female and male mice during choline deprivation (Fig. 5). Thus, the major gender difference in PC metabolism in liver appears to be related to HDL-PC metabolism. 4.2. Gender-dependent plasma PC and HDL-PC levels

Fig. 3. Profiles of choline-containing lipids in plasma lipoproteins. Plasma samples were collected from Pemt−/− mice fed the choline-supplemented (CS) diet for 24 h (0 day) then fed the choline-deficient (CD) diet for 1 to 3 days. Plasma lipoproteins were separated on the basis of size and analyzed for content of choline-containing lipids by fast-protein liquid chromatography. Plasma was combined from at least three Pemt−/− mice and assayed in duplicate. Data are from one experiment that is representative of two experiments with similar results. (A) Male Pemt−/− mice; (B) Female Pemt−/− mice.

be provided from extra-hepatic sources via HDL. Our model suggests that the strategy for coping with choline insufficiency is more efficient in females, which might have evolved due to a gender-specific metabolic function, such as lactation. The mobilization of extra-hepatic PC in the CD-Pemt−/− mice was transient since by 48 h after starting the CD diet, the levels of plasma PC in both male and female Pemt−/− mice declined markedly. The results from this study suggest there is much to be learned about whole body choline and PC homeostasis. From which tissues is the HDL-PC derived? What regulates this process and why is it different in male compared to female Pemt−/− mice? Since the amount of SR-B1 was unchanged in CD-Pemt−/− mice, was the delivery of PC to liver determined by the level of HDL-PC? 4.1. Gender-dependent PC homeostasis in liver Hepatic PC homeostasis is determined by PC biosynthesis, PC catabolism, biliary PC secretion, lipoprotein uptake (HDLPC and LDL-PC) and secretion (very-low density lipoprotein (VLDL)-PC and preβ HDL-PC). Lack of choline (both endogenous and exogenous sources) severely impaired PC biosynth-

A striking observation from these experiments was that female Pemt−/− mice had ∼ 60% lower plasma PC levels than did male Pemt−/− mice when fed the CS diet for 24 h. The majority of plasma PC in mice is associated with HDL [18]. When Albers et al. [21] screened 15 different strains of mice fed a chow diet, they found that male mice had a 10–60% higher level of plasma phospholipid (mainly PC) and 10–110% higher level of HDL-phospholipid (mainly HDL-PC) than female mice. Male C57BL/6J mice had 30% higher plasma phospholipid and HDL-phospholipid than female mice [21]. In agreement, Noga and Vance [22] reported that when fed a chow diet, male Pemt−/− mice showed 1.5-fold higher plasma PC level than female Pemt−/− mice. Two differences in experimental protocols between the present experiments and those of Noga and Vance [22] might account for the enhanced difference we observed in plasma PC levels between male and female Pemt−/− mice. First, a CS diet was used in the current experiments instead of a chow diet [22]. Second, the fasting period was 12 h in these experiments but overnight fasting (∼16 h) in the previous study [22]. 4.3. Potential mechanisms of enhanced PC efflux to HDL in response to choline deprivation Our results suggest that female CD-Pemt−/− mice showed a 1-day delay in hepatic choline deficiency compared to male Pemt−/− mice due to different capabilities of PC mobilization from extra-hepatic tissues to HDL particles. Female CS-Pemt−/− mice have a lower level of plasma PC than do male Pemt−/− mice. Once choline deprivation was initiated, PC efflux to HDL and plasma PC levels in male Pemt−/− mice might already have been saturated and could not be up-regulated. In contrast, female CD-Pemt−/− mice might have sufficient capacity to increase HDL-PC to maintain hepatic PC levels and PC/PE ratio in order to prevent liver damage. This acute adaptation might depend on the ability of PC efflux to HDL. However, this response lasted for only 1 day of choline deficiency as the levels of plasma PC declined in CD-Pemt−/− mice of both genders in subsequent days. A decline in plasma PC was expected because extra-hepatic tissues, as sources for PC efflux to HDL, were also choline deprived. Since the clearance of plasma HDL-PC is

898

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

Fig. 4. Glybenclamide lowers HDL-PC and induces liver damage in female Pemt−/− mice fed a choline deficient diet. (A) Profiles of choline-containing lipids in plasma lipoproteins after glybenclamide treatment. Female Pemt−/− mice were fed the choline-supplemented (CS) diet containing glybenclamide (163 μg/g diet) for 1 day (CS + G) and were then fed the choline-deficient (CD) diet containing glybenclamide (163 μg/g) for 1 day (CD + G). Female Pemt−/−mice were fed a CS or CD diet without glybenclamide for 1 day as controls (CS and CD). Lipoproteins in fresh plasma samples were separated by fast-protein liquid chromatography and amounts of choline-containing lipids were measured. Plasma combined from at least 3 female mice was used for each analysis and the assays were performed in duplicate. Data are from one experiment that is representative of two experiments with similar results. (B) Levels of PC and PE in liver homogenates of mice fed glybenclamide in CS (CS1) or CD (CD1) diet for 1 day. Data are averages ± S.D. from 4 independent experiments. *P b 0.05, CS vs. CD. (C) Molar ratio of PC to PE in livers of mice fed the CS (CS1) or CD (CD1) diet for 1 day. Data are averages ± S.D. of 4 mice, *P b 0.05 CD vs. CS. (D) Liver damage after glybenclamide treatment. Plasma alanine aminotransferase (ALT) activities were measured in mice fed the CD or CS diet for 1 day (CS1 and CD1, respectively) as an indicator of liver damage. Data are averages ± S.D. from 4 mice in each group, *P b 0.05 for CS compared to CD.

higher in males than females during choline supplementation and similar during choline deficiency, the mechanism by which the liver handles imported PC might also be different between males and females. An intriguing question is which tissue provides hepatic PC via HDL-PC transport? Our data do not answer this question, since it is still unclear which tissue(s) is the major site of formation of mature HDL [23,24]. A current model is that hepatic PC is originally loaded on to apo AI to form preβ HDL after which, preβ HDL is delivered to extra-hepatic tissues where the particles receive cholesterol and transport it back to liver for disposal [25–27]. The efflux of cholesterol and PC from extrahepatic tissues is linked [26,28]. Thus, preβ HDL receives both cholesterol and PC to form mature HDL. ABCA1 is expressed ubiquitously in all tissues [29], so it is hard to identify which tissue is the major one for HDL formation or HDL-PC efflux. From our experiments in choline redistribution [5], kidney and lung are candidate tissues for increased PC efflux to HDL. Two potential mechanisms are believed to enhance HDL-PC efflux [25–28]. One is that lipid-poor preβ HDL particles are able to receive more lipids from extra-hepatic tissues [25,27]. The other is that increased ABCA1 expression in extra-hepatic tissues can enhance HDL-PC efflux [28]. Our results showed

that choline deprivation decreased the formation of preβ HDL by hepatocytes of both male and female Pemt−/− mice (Fig. 5A). Secretion of poorly lipidated apoA1 by hepatocytes (Fig. 5A) could enhance PC and cholesterol efflux from extra-hepatic tissues. Therefore, male Pemt−/− mice could also maintain high plasma HDL-PC levels in the first day of choline deprivation and female Pemt−/− mice were able to up-regulate their HDLPC at the first day of choline deprivation (Figs. 2D, 3A, B). Recent studies suggest that liver is the major source of PC in HDL whereas extra-hepatic tissues contributes much of the cholesterol to nascent HDL [30,31]. However, in our models, hepatocytes faced PC deficiency because of choline deprivation. Thus, the increased HDL-PC (Fig. 3) suggests that poorlylipidated apo AI secreted from CD hepatocytes (Fig. 5) enhances PC efflux from extra-hepatic tissues of both males and females. Moreover, the amount of hepatic ABCA1 protein was not different between genders or dietary conditions (Fig. 6C). In addition, increased hepatic uptake of HDL-PC by female Pemt−/− mice during the first day of choline deprivation (Fig. 6A), combined with unchanged SR-B1 protein levels (Fig. 6C), suggest that hepatic uptake of HDL-PC is dependent on plasma HDL-PC levels, not on SR-B1 levels, even though hepatic HDL uptake is mainly mediated by SR-B1 [32].

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

899

maintaining plasma HDL-PC levels during the first day of choline deficiency. However, a higher demand of HDL-PC for hepatic PC homeostasis in male Pemt−/− mice caused liver damage in the first day of choline deficiency. A lower demand of HDL-PC for hepatic PC homeostasis in female Pemt−/− mice allowed them to up-regulate plasma HDL-PC that would increase delivery of PC to the liver to maintain hepatic PC homeostasis. During evolution, there might have been pressure

Fig. 5. HDL and apo AI secretion from primary cultured hepatocytes. Male and female Pemt−/− mice were fed the CS diet for 1 day and then fed the CD diet for 1 day. Primary hepatocytes were isolated and cultured in CS or CD medium. (A) HDLs and pre-β HDLs in the medium were separated by electrophoresis on 6% non-denaturing polyacrylamide gels followed by immunoblotting with an antiapo AI antibody. The experiment was repeated with similar results. Std = mature mouse HDL. (B) Apo AI in the medium was separated by 12% SDS-PAGE and analyzed by immunoblotting with an anti-apo A1 antibody. (C) Sandwich ELISA was used for analyzing the amount of PC associated with apo AI proteins. Assay was performed in triplicate and the experiment repeated once, *P b 0.05 CD vs. CS. Fluorescence intensity is relative to fluorescence intensity of male CS HDLs (set as 1.0).

4.4. Why is PC efflux to HDL in Pemt−/− mice dependent on gender? We speculate that male Pemt−/− mice might have a higher requirement for HDL-PC to maintain hepatic PC homeostasis. When choline supply is limited, male Pemt−/− mice might not be able to further increase PC efflux from extra-hepatic tissues to HDL. Nevertheless, male Pemt−/− mice probably do respond to choline deprivation, since male Pemt−/− mice maintain their high HDL-PC levels in plasma during the first day of choline deprivation (Fig. 3A), even though hepatic PC significantly decreased during that time (Fig. 2B). In contrast, female Pemt−/− mice increased HDL-PC to the same level as in male mice during the first day of choline deprivation. Therefore, both male and female Pemt−/− mice responded to choline deprivation by

Fig. 6. Enhanced uptake of HDL-PC by livers of female Pemt−/− mice. Male and female Pemt−/− mice were fed the CS diet for 1 day and then fed the CD diet for 1 day. BODIPY-PC-labeled HDLs were injected into mice via tail-vein and blood was collected after 10, 30, 60 and 120 min. The livers were then perfused with liver perfusion medium and stored at −70 °C until analysis. Fluorescence intensity was measured in plasma and liver homogenates. The amount of fluorescent HDL-PC = sample fluorescence intensity/(total injected fluorescence / plasma HDL-PC fluorescence). The clearance of plasma HDL-PC was calculated from the amount of HDL-PC in plasma at the various time points. (A) Relative amount of fluorescent HDL-PC in the liver. (B) Relative clearance rates of plasma HDL-PC. Results were normalized to data from female CS-Pemt−/− mice. Data are averages ± S.D. from at least 3 mice. P b 0.05. (C) Immunoblotting of hepatic SR-B1 and ABCA1 proteins. Immunoblotting of protein disulfide isomerase (PDI) was used as a loading control. Results are from 2 mice in each group. The experiment was repeated with similar results.

900

Z. Li et al. / Biochimica et Biophysica Acta 1771 (2007) 893–900

for female species to be resistant to choline deficiency that might occur during pregnancy and lactation.

[16]

Acknowledgments [17]

We thank Sandra Ungarian, Susanne Lingrell, Audric Moses, Priscilla Gao and Ted Chan for excellent technical assistance. We thank Dr. Masato Umeda, Kyoto University, for providing anti-PC antibodies and Dr. Jean Vance for helpful comments on the manuscript. This research was supported by a grant from the Canadian Institutes of Health Research (MOP 62935). Z.L was the recipient of a CIHR/HSFC Strategic Training Fellow in Stroke, Cardiovascular, Obesity, Lipids, Atherosclerosis Research (SCOLAR) supported by a grant from AstraZeneca. D.E.V. is holder of the Canada Research Chair in Molecular and Cell Biology of Lipids and Heritage Scientist of the Alberta Heritage Foundation for Medical Research.

References [1] S.H. Zeisel, Choline: an essential nutrient for humans, Nutrition 16 (2000) 669–671. [2] Z. Li, L.B. Agellon, D.E. Vance, Phosphatidylcholine homeostasis and liver failure, J. Biol. Chem. 280 (2005) 37798–37802. [3] D.E. Vance, Phospholipid biosynthesis in eukaryotes, in: D.E. Vance, J.E. Vance (Eds.), Biochemistry of Lipids, Lipoproteins and Membranes, Elsevier, New York, 2002, pp. 205–232. [4] C.J. Walkey, L. Yu, L.B. Agellon, D.E. Vance, Biochemical and evolutionary significance of phospholipid methylation, J. Biol. Chem. 273 (1998) 27043–27046. [5] Z. Li, L.B. Agellon, D.E. Vance, Choline redistribution during adaptation to choline deprivation, J. Biol. Chem. 282 (2007) 10283–10289. [6] I. Portal, T. Clerc, V. Sbarra, H. Portugal, A.M. Pauli, H. Lafont, B. Tuchweber, I. Yousef, F. Chanussot, Importance of high-density lipoprotein-phosphatidylcholine in secretion of phospholipid and cholesterol in bile, Am. J. Physiol. 264 (1993) G1052–G1056. [7] C.J. Walkey, L.R. Donohue, R. Bronson, L.B. Agellon, D.E. Vance, Disruption of the murine gene encoding phosphatidylethanolamine Nmethyltransferase, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 12880–12885. [8] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–543. [9] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911–917. [10] R.H. McCluer, M.D. Ullman, F.B. Jungalwala, Adv. Chromatogr., vol. 25, 1986, pp. 309–353. [11] B.D. Schultz, A.D. DeRoos, C.J. Venglarik, A.K. Singh, R.A. Frizzell, R.J. Bridges, Glibenclamide blockade of CFTR chloride channels, Am. J. Physiol. 271 (1996) L192–L200. [12] N. Wang, D.L. Silver, C. Thiele, A.R. Tall, ATP-binding cassette transporter A1 (ABCA1) functions as a cholesterol efflux regulatory protein, J. Biol. Chem. 276 (2001) 23742–23747. [13] G. Hasko, E.A. Deitch, Z.H. Nemeth, D.G. Kuhel, C. Szabo, Inhibitors of ATP-binding cassette transporters suppress interleukin-12 p40 production and major histocompatibility complex II up-regulation in macrophages, J. Pharmacol. Exp. Ther. 301 (2002) 103–110. [14] P.G. Frank, M.W. Cheung, S. Pavlides, G. Llaverias, D.S. Park, M.P. Lisanti, Caveolin-1 and regulation of cellular cholesterol homeostasis, Am. J. Physiol.: Heart Circ. Physiol. 291 (2006) H677–H686. [15] Z. Li, L.B. Agellon, T.M. Allen, M. Umeda, L. Jewell, A. Mason, D.E. Vance, The ratio of phosphatidylcholine to phosphatidylethanolamine

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

influences membrane integrity and steatohepatitis, Cell Metab. 3 (2006) 321–331. K. Fujimoto, M. Umeda, T. Fujimoto, Transmembrane phospholipid distribution revealed by freeze-fracture replica labeling, J. Cell. Sci. 109 (Pt. 10) (1996) 2453–2460. B.H. Chung, T. Wilkinson, J.C. Geer, J.P. Segrest, Preparative and quantitative isolation of plasma lipoproteins: rapid, single discontinuous density gradient ultracentrifugation in a vertical rotor, J. Lipid Res. 21 (1980) 284–291. A. Jonas, Lipoprotein structure, in: D.E. Vance, J.E. Vance (Eds.), Biochemistry of Lipids, Lipoproteins and Membranes, Elsevier, New York, 2002, pp. 483–504. T.J. Nieland, A. Chroni, M.L. Fitzgerald, Z. Maliga, V.I. Zannis, T. Kirchhausen, M. Krieger, Cross-inhibition of SR-BI- and ABCA1mediated cholesterol transport by the small molecules BLT-4 and glyburide, J. Lipid Res. 45 (2004) 1256–1265. A.A. Noga, Y. Zhao, D.E. Vance, An unexpected requirement for phosphatidylethanolamine N-methyltransferase in the secretion of very low density lipoproteins, J. Biol. Chem. 277 (2002) 42358–42365. J.J. Albers, W. Pitman, G. Wolfbauer, M.C. Cheung, H. Kennedy, A.Y. Tu, S.M. Marcovina, B. Paigen, Relationship between phospholipid transfer protein activity and HDL level and size among inbred mouse strains, J. Lipid Res. 40 (1999) 295–301. A.A. Noga, D.E. Vance, A gender-specific role for phosphatidylethanolamine N-methyltransferase-derived phosphatidylcholine in the regulation of plasma high density and very low density lipoproteins in mice, J. Biol. Chem. 278 (2003) 21851–21859. R.C. Pittman, D. Steinberg, Sites and mechanisms of uptake and degradation of high density and low density lipoproteins, J. Lipid Res. 25 (1984) 1577–1585. C. Glass, R.C. Pittman, M. Civen, D. Steinberg, Uptake of high-density lipoprotein-associated apoprotein A-I and cholesterol esters by 16 tissues of the rat in vivo and by adrenal cells and hepatocytes in vitro, J. Biol. Chem. 260 (1985) 744–750. H.Q. Liang, K.A. Rye, P.J. Barter, Cycling of apolipoprotein A-I between lipid-associated and lipid-free pools, Biochim. Biophys. Acta 1257 (1995) 31–37. P.E. Fielding, C.J. Fielding, Dynamics of lipoprotein transport in the human circulatory system, in: D.E. Vance, J.E. Vance (Eds.), Biochemistry of Lipids, Lipoproteins and Membranes, Elsevier, New York, 2002, pp. 527–552. K.A. Rye, P.J. Barter, Formation and metabolism of prebeta-migrating, lipid-poor apolipoprotein A-I, Arterioscler. Thromb. Vasc. Biol. 24 (2004) 421–428. J.F. Oram, HDL apolipoproteins and ABCA1: partners in the removal of excess cellular cholesterol, Arterioscler. Thromb. Vasc. Biol. 23 (2003) 720–727. C.L. Wellington, E.K. Walker, A. Suarez, A. Kwok, N. Bissada, R. Singaraja, Y.Z. Yang, L.H. Zhang, E. James, J.E. Wilson, O. Francone, B.M. McManus, M.R. Hayden, ABCA1 mRNA and protein distribution patterns predict multiple different roles and levels of regulation, Lab. Invest. 82 (2002) 273–283. J.M. Timmins, J.Y. Lee, E. Boudyguina, K.D. Kluckman, L.R. Brunham, A. Mulya, A.K. Gebre, J.M. Coutinho, P.L. Colvin, T.L. Smith, M.R. Hayden, N. Maeda, J.S. Parks, Targeted inactivation of hepatic Abca1 causes profound hypoalphalipoproteinemia and kidney hypercatabolism of apoA-I, J. Clin. Invest. 115 (2005) 1333–1342. R.R. Singaraja, M. Van Eck, N. Bissada, F. Zimetti, H.L. Collins, R.B. Hildebrand, A. Hayden, L.R. Brunham, M.H. Kang, J.C. Fruchart, T.J. Van Berkel, J.S. Parks, B. Staels, G.H. Rothblat, C. Fievet, M.R. Hayden, Both hepatic and extrahepatic ABCA1 have discrete and essential functions in the maintenance of plasma high-density lipoprotein cholesterol levels in vivo, Circulation 114 (2006) 1301–1309. S. Acton, A. Rigotti, K.T. Landschulz, S. Xu, H.H. Hobbs, M. Krieger, Identification of scavenger receptor SR-BI as a high density lipoprotein receptor, Science 271 (1996) 518–520.