Biochemical and Molecular Actions of Nutrients

The Chalcone Xanthohumol Inhibits Triglyceride and Apolipoprotein B Secretion in HepG2 Cells1,2 Adele Casaschi, Geoffrey K. Maiyoh, Brent K. Rubio, Rachel W. Li, Khosrow Adeli,* and Andre G. Theriault3 Division of Medical Technology, John A. Burns School of Medicine, University of Hawaii at Manoa, Honolulu, Hawaii; and *Department of Laboratory Medicine, The Hospital for Sick Children, Toronto, Canada ABSTRACT The present study examined the role of xanthohumol (XN), a plant chalcone, on apolipoprotein B (apoB) and triglyceride (TG) synthesis and secretion, using HepG2 cells as the model system. The data indicated that XN decreased apoB secretion in a dose-dependent manner under both basal and lipid-rich conditions (as much as 43% at 15 ␮mol/L). This decrease was associated with increased cellular apoB degradation. To determine the mechanism underlying this effect, we examined triglyceride availability, a major factor in the regulation of apoB secretion. XN inhibited the synthesis of TG in the microsomal membrane and the transfer of this newly synthesized TG to the microsomal lumen (decreases of 26 and 64%, respectively, under lipid-rich conditions), indicating that TG availability is a determining factor in the regulation of apoB secretion under the experimental conditions. The inhibition of TG synthesis was caused by a reduction in diacylglycerol acyltransferase (DGAT) activity, which corresponded to a decrease in DGAT-1 mRNA expression, but not DGAT-2 expression. Microsomal triglyceride transfer protein (MTP) may also control the rate of TG transfer from the microsomal membrane to the active lumenal pool. XN decreased MTP activity in a dose-dependent manner (as much as 30%). Whether the reduction in TG accumulation in the microsomal lumen is predominantly due to DGAT and/or MTP activity remains unknown. In summary, the data suggest that xanthohumol is a potent inhibitor of apoB secretion. J. Nutr. 134: 1340 –1346, 2004. KEY WORDS: ● bioflavonoid ● diacylglycerol acyltransferase

●

apolipoprotein B

●

HepG2

In recent years, several plant flavonoids were identified as hypolipidemic agents in tissue culture, animals, and humans (1– 4). They inhibit the activity of a number of lipogenic enzymes, including hydroxymethylglutaryl (HMG)4 CoA reductase, acyl CoA:cholesterol acyltransferase (ACAT), microsomal triglyceride transfer protein (MTP), and diacylglycerol acyltransferase (DGAT) (5–7). Flavonoids are subcategorized into various groups (i.e., flavonols, flavones, catechins, flavanones, chalcones, anthocyanidins, and isoflavonoids); little is known of the effects of the chalcone flavonoids on lipid

●

microsomal triglyceride transfer protein

metabolism. Chalcones are the immediate precursors to the biosynthesis of flavonoids, and their structure differs considerably from that of the other groups by the inclusion of an open C-ring. Structural differences in these molecules may therefore account for differing effects on metabolism. Higher antioxidant activity is reported for prenylchalcones than for prenylflavanones in the Cu2⫹-mediated oxidation of LDL, suggesting a relation between structure and function (8). Also, many chalcones suppress tumor promotion more effectively than flavonoids themselves [reviewed in Dimmock et al. (9)]. Xanthohumol (XN) is a commonly studied prenylated chalcone found mainly in hops. Hops (Humulus lupulus L.), specifically the female flowers of the hop plant, are used in the brewing industry to add flavor and bitterness to beer. During the brewing process, hop prenylchalcones are readily converted into their isomeric flavonones, namely isoxanthohumol and 6-prenylnaringenin (10). Therefore, beer contains higher levels of prenylflavanones than of prenylchalcones. Tabata et al. (11) reported in 1997 that XN can inhibit the activity of DGAT, a key enzyme in triglyceride (TG) synthesis. DGAT catalyzes the final step in the glycerol phosphate pathway, considered the major pathway for hepatic TG synthesis (12). Progress in understanding DGAT only began with the cloning of DGAT-1 and DGAT-2 in recent years. Although the study by Tabata et al. (11) was crucial in identi-

1 Presented in part at the 4th Annual Conference on Arteriosclerosis, Thrombosis, and Vascular Biology, May 2003, Washington, DC [Casaschi, A., Sosa, M., Rubio, B. & Theriault, A. (2003) Secretion of triglyceride and apoB is inhibited by the plant chalcone, xanthohumol, via reduced DGAT and MTP activity in HepG2 cells. Arterioscler. Thromb. Vasc. Biol. 23: a-54 (abs.)]. 2 Supported by the American Heart Association of Hawaii, grant 0350528Z; the Robert C. Perry Fund of the Hawaii Community Foundation, grant 20020609; and a fellowship from the National Institutes of Health MARC U* STAR Program, grant GM07684 –23 (B.K.R.). 3 To whom correspondence should be addressed. E-mail: [email protected]. 4 Abbreviations used: apoB, apolipoprotein B; apoB-Lp, apoB-containing lipoprotein; ACAT, acyl CoA:cholesterol acytltransferase; BSA, bovine serum albumin; CE, cholesterol ester; DGAT, diacylglycerol acyltransferase; ER, endoplasmic reticulum; FBS, fetal bovine serum; HMG, hydroxymethylglutaryl; LDH, lactate dehydrogenase; MTP, microsomal triglyceride transfer protein; OA, oleic acid; TCA, trichloroacetic acid; TG, triglyceride; XN, xanthohumol.

0022-3166/04 $8.00 © 2004 American Society for Nutritional Sciences. Manuscript received 26 January 2004. Initial review completed 3 March 2004. Revision accepted 12 March 2004. 1340

XANTHOHUMOL INHIBITS ApoB SECRETION

fying chalcones as potential DGAT inhibitors, the experiments did not provide any insights into how XN and DGAT inhibition might regulate hepatic apolipoprotein B (apoB) secretion. Because TG availability is a major factor in the regulation of apoB secretion (13), the present study examined the effects of XN on apoB and TG synthesis and secretion using a well-established human hepatoma cell-line, HepG2, as the model system. This cell-line has been extensively used to study hepatic apoB secretion (14,15). Because chalcones are important micronutrients (9), the present study may contribute to our understanding of the role of these molecules in the treatment of hypertriglyceridemia. MATERIALS AND METHODS Cell culture. Monolayer HepG2 cell cultures (HB 8065; American Type Culture Collection) were maintained in RPMI-1640 medium with 10% fetal bovine serum (FBS) (InVitrogen Life Technologies) at 37°C with 5% CO2 and subcultured in either 35- or 100-mm dishes (Corning Costar) to ⬃80% confluence. Because HepG2 cells are highly dependent on a high concentration of exogenous fatty acids to maintain an adequate supply of lipids for lipoprotein assembly (16), we investigated the effect of XN on apoB secretion in cells incubated with oleic acid (OA). OA was provided as a complex with bovine serum albumin (BSA). The molar ratio of OA:BSA was 8:1, and the concentrations were 0.81 mmol/L for OA and 0.1 mmol/L for BSA. In experiments without oleate, cells were treated with XN in 1% BSA:RPMI. We chose to add 1% BSA to the experimental culture medium to minimize chemical instability of the compound (17). XN (4,2⬘,4⬘-trihydroxy-6⬘-methoxy-3⬘-prenylchalcone, ⬎98% purity; Alexis Biomedicals; Fig. 1) was prepared in DMSO. An appropriate amount of stock solution was diluted in culture medium to give a final maximum DMSO concentration of 0.25%. Stock solutions were kept at 4°C for up to 4 wk. Control cells were treated with the solvent (i.e., DMSO) only. Apo B ELISA. Secretion of apoB into the medium was measured by a noncompetitive binding ELISA procedure, essentially as described by Macri and Adeli (18). Spectrophotometric readings were taken using a microplate reader (Thermomax; Molecular Devices). Cell proteins were digested in 1 mL of 0.1 mol/L NaOH and measured as described below. Apo B synthesis and secretion. Treated and untreated HepG2 cells were preincubated in methionine- and cysteine-free RPMI with or without XN for 30 min and pulsed with a [35S]protein-labeling medium (100 mCi/L of [35S]protein-labeling mix, 1175 Ci/mmol, Perkin Elmer Life Science Research Products, in methionine- and cysteine-free RPMI) with or without XN for 10 min. Following the pulse, cells were chased in the presence or absence of XN for 15, 60, and 120 min in RPMI supplemented with excess methionine and cysteine. At various chase times, cells were harvested from duplicate 35-mm dishes and lysed in solubilization buffer as described previously (19). The lysates were centrifuged and the supernatants were collected for immunoprecipitation using a polyclonal antihuman apoB antibody (Chemicon International). The medium collected at

1341

each time point was also subjected to immunoprecipitation. Samples were subjected to SDS-PAGE, fluorography, and quantitation as described previously (19). Total protein synthesis and secretion were measured by [35S]protein labeling and trichloroacetic acid (TCA) precipitation as described previously (19). Triglyceride synthesis and secretion. To measure the rate of TG synthesis and secretion, treated and untreated cells were labeled with 10 mCi/L [2-3H]glycerol (200 Ci/mol; Perkin Elmer Life Science Research Products) in the last 6 h of a 24-h treatment period. After labeling, the medium was collected and the cells were washed twice with cold PBS. Cellular and medium lipids were extracted with hexane:isopropanol (3:2, v/v) and separated by TLC as described by Theriault et al. (7). Radioactivity associated with TG was counted by liquid scintillation counting and normalized to cell protein. Cell protein samples in parallel dishes were digested in 1 mL of 0.1 mol/L NaOH and measured as described below. DGAT activity assay. Treated and untreated cells were harvested into a TRIS buffer (175 mmol/L, pH 7.8) and homogenized with a sonicator. Whole-cell esterification of diacylglycerol was measured using [14C]palmitoyl-CoA (40 – 60 Ci/mol; Perkin Elmer Life Science Research Products) and 0.4 g cell protein/L, essentially as described by Grigor and Bell (20). The [14C]TG formed was extracted from the assay reaction tube and subjected to TLC in chloroform: acetic acid (96:4). Measurement of DGAT mRNA abundance. Levels of DGAT-1 and DGAT-2 gene expression were assayed by relative RT-PCR. Total RNA was isolated using RNAzol B (Tel-Test), and RNA concentrations were measured by spectrophotometry. Isolated total RNA was reverse-transcribed, and the DGAT mRNA was amplified in an Eppendorf Mastercycler (Eppendorf Scientific) using 10 pmol each of DGAT1-specific primers (sense 5⬘-GGCCTTCTTCCACGAGTACC-3⬘; antisense 5⬘-GGCCTCATAGTTGAGCACG-3⬘) or DGAT2-specific primers (sense 5⬘-CTCAGACCATAGCCTAAACC-3⬘; antisense 5⬘-CAGCTTAGGGGTGTGACATC-3⬘), and Taq DNA polymerase (Eppendorf Scientific) according to the following thermal cycle: 30 s at 94°C, 30 s at 55°C, and 30 s at 72°C for 31 (DGAT-1) or 34 (DGAT-2) cycles (determined to lie within the linear amplification phase). 18S rRNA was used as an internal standard, coamplified with a 1:9 ratio of primers:competimers (Ambion). Amplified products (216 bp for DGAT-1, 307 bp for DGAT-2, and 488 bp for 18S rRNA) were tested on 2% agarose gels and identified by staining with ethidium bromide. Band intensities were analyzed densitometrically using the Gel Doc Gel Documentation System (Bio-Rad Laboratories), and the DGAT mRNA levels were normalized with respect to 18S rRNA. Other methods. Subcellular fractionation of the cytosol, microsomal membrane, and microsomal lumen was conducted essentially as described by Borradaile et al. (21). The cell protein content was measured according to the method of Bradford (22) (i.e., Bio-Rad), using BSA as the standard. The MTP activity in cell extracts was measured by a fluorescent assay according to the manufacturer’s protocol (Roar Biomedical). The albumin secreted into the medium was measured with an ELISA kit (Bethyl Laboratories). The LDL (density 1.006 –1.063 kg/L) was isolated from the medium as described by Luchoomum and Hussain (23). The activity of lactate dehydrogenase (LDH) released into the medium was measured spectrophotometrically using the CytoTox 96 nonradioactive cytotoxicity assay according to the manufacturer’s protocol (Promega). Cell viability was assayed by the trypan blue exclusion technique according to the protocol provided by Invitrogen. Statistical analysis. Data are presented as means ⫾ SD of 3 independent experiments. Statistical differences were analyzed using t-tests; values of P ⬍ 0.05 were considered significant.

RESULTS

FIGURE 1

Structure of xanthohumol.

Xanthohumol inhibits apoB secretion. XN treatment decreased apoB secretion in a dose-dependent manner after 24 h of incubation with and without oleate treatment, as assayed by ELISA (Fig. 2). Secretion was inhibited 13 ⫾ 0.2% at 5 ␮mol/L, 21 ⫾ 4% at 10 ␮mol/L, and 43 ⫾ 2% at 15 ␮mol/L in cells incubated with 1% BSA, and 17 ⫾ 1% at 10 ␮mol/L

1342

CASASCHI ET AL.

TABLE 1 Effects of XN on intracellular apoB degradation in HepG2 cells1,2 Peak to 60 min Control

XN

Peak to 120 min Control

XN

dpm/mg cell protein Intracellular peak3 Depleted4 Secreted Degraded

FIGURE 2 Effects of XN on apoB and albumin secretion in HepG2 cells. Cells were treated with various concentrations of XN (5–15 ␮mol/L) for 24 h in medium containing 0.81 mmol/L OA (complexed to BSA) or 1% BSA. The medium was collected and apoB/albumin secretion was measured by ELISA. Data are expressed as a percentage of control (set as 100%). Values represent the mean ⫾ SD of 3 independent experiments performed in duplicate (n ⫽ 3). aDifferent from OA control, P ⬍ 0.05. bDifferent from BSA control, P ⬍ 0.05.

and 31 ⫾ 2% at 15 ␮mol/L in cells incubated with oleate (0.81 mmol/L complexed to BSA). As a control, albumin secretion was also measured. XN did not affect albumin secretion in cells incubated with 1% BSA over the range of dosages tested, indicating that its effects on apoB secretion are specific. Because HepG2 cells primarily secrete an LDL-sized particle (24), we measured apoB content in this fraction. At 15 ␮mol/L with oleate treatment, XN treatment decreased LDLapoB concentration by 26 ⫾ 4% (n ⫽ 3, P ⬍ 0.05 vs. control; data not shown). Pulse-chase experiments were conducted to investigate the mechanism by which XN inhibited apoB secretion. At 15 ␮mol/L, XN decreased apoB secretion into the culture medium significantly (75% at 120 min) compared to the untreated control (Table 1). The incorporation of [35S]methionine and [35S]cysteine into intracellular apoB at the 15-min chase time was also measured as an index of apoB synthesis. A slight increase of 8% (P ⬎ 0.05) in the incorporation of labeled methionine and cysteine into immunoprecipitable apoB occurred with XN treatment. ApoB recovery in the cells and medium at 120 min was estimated to calculate the amount of apoB lost by the process of intracellular degradation. Recovery (intracellular peak ⫺ depleted ⫹ secreted) was 55% for control cells and 34% for XN-treated cells. This indicated that XN treatment increased the proportion of apoB that was degraded from 45 to 66%. Most of the degradation took place within 1 h. The effects on apoB synthesis and secretion were specific because protein synthesis and secretion, as assessed by TCA precipitation, were slightly stimulated or unchanged, respectively (data not shown). Xanthohumol decreases TG synthesis and secretion. The decrease in apoB secretion was examined to determine whether it was caused by limited TG availability. Cellular TG synthesis decreased significantly under basal and lipid-rich conditions in the presence of XN (15 ␮mol/L; Table 2). Synthesis decreased 63% in cells incubated with 1% BSA and

9110 ⫾ 642 3823 905 2918

9833 ⫾ 730 6764 280 6484

9110 ⫾ 642 5446 1327 4119

9833 ⫾ 730 6785 332 6453

1 Values represent the mean of duplicate measurements of a representative experiment (reproduced in 2 other experiments in duplicate). 2 HepG2 cells were treated with XN (15 ␮mol/L) for 24 h in medium containing OA, then subjected to pulse-chase. 3 Intracellular peaks were calculated from the apoB content at the beginning of the chase (15 min). 4 Depleted values represent the loss of intracellular apoB content between the beginning of the chase (15 min) and the 60- and 120-min chase.

62% in cells incubated with oleate (all P ⬍ 0.05). The secretion of TG into the culture medium was also analyzed simultaneously in untreated cells and in cells treated with XN. Results were similar to intracellular observations (Table 2). XN decreased TG secretion by 51% in cells incubated with 1% BSA and 26% in cells incubated with oleate (all P ⬍ 0.05). XN did not affect the synthesis or secretion of phospholipids (data not shown). Xanthohumol decreases cellular DGAT activity. Wholecell DGAT activity was measured to investigate the mechanism for the reduction in TG synthesis. XN treatment (15 ␮mol/L) markedly inhibited DGAT activity in cells incubated with and without oleate (Table 3). Activity decreased 36% in cells incubated with 1% BSA and 46% in cells incubated with oleate (all P ⬍ 0.05). Xanthohumol decreases DGAT-1 mRNA expression. The effects of XN treatment (15 and 25 ␮mol/L) on DGAT-1 and DGAT-2 mRNA expression were studied to examine the decrease in DGAT activity. Treatment with 25 ␮mol/L XN did not compromise cell viability, as assessed by the trypan blue exclusion technique and leakage of LDH into the medium TABLE 2 Effects of XN on TG synthesis and secretion in HepG2 cells1,2 Treatment

Synthesis

Secretion

mmol [3H]TG/mg cell protein BSA control BSA ⫹ XN OA control OA ⫹ XN

123 ⫾ 6 46 ⫾ 0.2* 301 ⫾ 6 113 ⫾ 2†

2.0 ⫾ 0.2 1.0 ⫾ 0.09* 3.2 ⫾ 0.5 2.4 ⫾ 0.02†

1 Values represent the mean ⫾ SD of 3 independent experiments performed in duplicate (n ⫽ 3). * Different from BSA control, P ⬍ 0.05. † Different from OA control, P ⬍ 0.05. 2 HepG2 cells were treated with XN (15 ␮mol/L) for 18 h in medium with or without OA, then labeled with [3H]glycerol with or without XN or OA for an additional 6 h.

XANTHOHUMOL INHIBITS ApoB SECRETION

TABLE 3 Effects of XN on DGAT activity in HepG2 cells1,2 Treatment

DGAT activity

␮mol [14C]TG/(mg cell protein 䡠 min) BSA control BSA ⫹ XN OA control OA ⫹ XN

9,821 ⫾ 797 6,285 ⫾ 957* 11,730 ⫾ 945 6,297 ⫾ 458†

1 Values represent the mean ⫾ SD of 3 independent experiments performed in duplicate (n ⫽ 3). * Different from BSA control, P ⬍ 0.05. † Different from OA control, P ⬍ 0.05. 2 HepG2 cells were treated with XN (15 ␮mol/L) for 24 h in medium with or without OA, then DGAT activity was determined.

(data not shown). Figure 3A shows the ethidium bromide– stained gel with the corresponding bar chart (Fig. 3B) of a typical experiment performed in duplicate (reproduced in 2 other experiments in duplicate). XN inhibited DGAT-1 mRNA expression in a dose-dependent manner (39 and 59% at 15 and 25 ␮mol/L, respectively). Conversely, XN had no effect on DGAT-2 mRNA expression (⫹2.5 and ⫺2.5% at 15 and 25 ␮mol/L, respectively). Xanthohumol decreases lumenal TG accumulation. The regulatory TG pool responsible for lipoprotein assembly in HepG2 cells is located within the lumen of the endoplasmic reticulum (ER) and the Golgi (16,25). We examined the reduction in apoB secretion in XN-treated cells to determine

1343

whether it was caused by decreased availability of newly synthesized TG within the lumen of these organelles. XN treatment (15 ␮mol/L) significantly reduced the accumulation of newly synthesized TG in the both the cytosol and the microsomal lumen, and decreased TG synthesis in the microsomal membrane. Accumulation decreased 37 and 64% in the cytosol and the microsomal lumen, respectively, and synthesis in the microsomal membrane decreased 26% (all P ⬍ 0.05; Table 4). Xanthohumol decreases MTP activity. Because XN limited the accumulation of TG in the microsomal lumen, we also examined MTP activity. Treatment with XN in 1% BSARPMI medium for 24 h decreased cell MTP activity in a dose-dependent manner. XN decreased MTP activity by 4 ⫾ 2% at 2 ␮mol/L, 18 ⫾ 3% at 10 ␮mol/L, and 30 ⫾ 1% at 25 ␮mol/L (Fig. 4). Treatment with 15 ␮mol/L XN and oleate also decreased MTP activity by 30 ⫾ 4% (n ⫽ 3, P ⬍ 0.05 vs. control; data not shown). DISCUSSION Several flavonoids dramatically reduce apoB secretion in cell culture (1,6,7,26,27). A number of lipogenic enzymes are involved in the mechanism of action, including HMG CoA reductase, ACAT, MTP, and, as recently reported, DGAT (5–7). However, the enzymes and lipids predominantly responsible for the assembly of apoB-Lp in the presence of flavonoids have not been thoroughly investigated. Borradaile et al. (21) recently made progress in this area by ruling out ACAT activity and cholesterol ester (CE) availability in the regulation of apoB-Lp by nanringenin, a citrus flavonoid, in HepG2 cells. Comparing naringenin to selective HMG CoA reductase, ACAT, and MTP inhibitors, they concluded that naringenin inhibits apoB secretion by limiting the accumulation of TG into the active microsomal lumenal pool via MTP inhibition (21,28). Earlier work suggested that MTP, in addition to catalyzing the transfer of lipids to nascent apoB molecules, facilitates the transfer of newly synthesized TG from the microsomal membrane to the lumen (29). In HepG2 cells, the microsomal lumenal TG pool is thought to be the regulatory pool responsible for lipoprotein assembly (16). Although the authors claimed that naringenin reduces the accumulation of lumenal TG via MTP inhibition, a lack of substrate (i.e., TG) caused by DGAT inhibition may offer another explanation for this effect. Because XN inhibits DGAT activity (11), we investigated the effects of XN on several aspects of apoB and triglyceride synthesis and secretion, as well as MTP and DGAT activity, using HepG2 cells as the model system. We initially examined the effects of XN on apoB secretion. TABLE 4 Effects of XN on the cellular distribution of newly synthesized TG in HepG2 cells1,2 Treatment

Cytosol

Lumen

Membrane

mmol [3H]TG/mg cell protein FIGURE 3 Effects of XN on DGAT-1 and DGAT-2 mRNA levels in HepG2 cells. Cells were incubated with XN (15 and 25 ␮mol/L) for 24 h in 1% BSA-RPMI medium and relative RT-PCR was performed. The figure is a representative ethidium bromide–stained gel showing the signals to DGAT-1, DGAT-2, and 18S rRNA (A) with its corresponding bar chart (B). Values in Fig. 3B represent the mean ⫾ SD of the above experiment (A) performed in duplicate.

OA control OA ⫹ XN

3.0 ⫾ 0.36 1.9 ⫾ 0.13*

1.8 ⫾ 0.04 0.65 ⫾ 0.08*

0.43 ⫾ 0.03 0.32 ⫾ 0.01*

1 Values represent the mean ⫾ SD of 3 independent experiments performed in duplicate (n ⫽ 3). * Different from OA control, P ⬍ 0.05. 2 HepG2 cells were treated with XN (15 ␮mol/L) for 22 h in medium containing OA, then labeled with [3H]glycerol with or without XN or OA for an additional 2 h.

1344

CASASCHI ET AL.

FIGURE 4 Effects of XN on MTP activity in HepG2 cells. Cells were treated with various concentrations of XN (2–25 ␮mol/L) for 24 h in 1% BSA-RPMI medium and MTP activity was determined. Data are expressed as a percentage of control (set as 100%). Values represent the mean ⫾ SD of 3 independent experiments performed in duplicate (n ⫽ 3). aDifferent from BSA control, P ⬍ 0.05.

Factors that control the rate of apoB secretion by the liver are of great importance in view of the role of apoB in atherogenesis. The results showed that XN inhibited apoB secretion in a dose-dependent fashion under both basal and lipid-rich conditions. The effect appeared to be slightly more potent under basal condition than with oleate treatment, and the decrease in total apoB secretion was associated with a decrease in LDL-apoB concentration. The effect was specific, because albumin secretion was essentially unchanged. This is the first report to indicate that a chalcone possesses this ability. Interestingly, among the flavonoids tested for their effect on apoB secretion (5,7,27), XN is nearly 5 times more potent than the others. This suggests that flavonoid compounds with an open C-ring and/or prenylated side group may have a superior ability to lower apoB secretion. However, further investigations on the relation of structure to the activity of these compounds are warranted. To examine the mechanism underlying the reduction in apoB secretion, we conducted pulse-chase experiments on the effects of XN on intracellular apoB degradation. Regulation of hepatic apoB secretion is thought to occur at the cotranslational and posttranslational levels, because nascent apoB molecules are either secreted or degraded intracellularly. An ER-localized ubiquitin-proteosome pathway is primarily implicated in the intracellular degradation of apoB [reviewed in Davis (30)]. The apoB that is efficiently translocated across the ER membrane assembles into a lipoprotein particle and is secreted through the secretion pathway. Conversely, any apoB that is not translocated across the ER membrane fails to assemble into a lipoprotein particle and is diverted into the degradation pathway. In the present study, apoB recovery in the cells and medium was greatest in the control cells, indicating that XN treatment increased the proportion of apoB that was degraded. This finding is in agreement with those reported for naringenin and a similar flavonoid compound, taxifolin (7,21). Borradaile et al. (21) further determined that naringenin treatment affects the degradation of apoB via the

proteasomal pathway. We propose that XN acts through the same mechanism. The availability of lipid is a major factor in the mechanism regulating the assembly and secretion of apoB-Lp [reviewed in Davis (30)]. Both dietary and hormonal factors that influence lipid availability modulate the rate of apoB degradation and secretion in cultured hepatocytes. Oleic acid, for example, stimulates apoB secretion by inhibiting protein degradation (31). This fatty acid protects against apoB degradation by increasing apoB lipidation via increased TG synthesis. We investigated whether XN affects apoB secretion and degradation by limiting the synthesis of TG and thus, lipidation of apoB. Interestingly, there was a significant decrease in cellular TG synthesis under both basal and lipid-rich conditions. Because the regulatory TG pool responsible for lipoprotein assembly in HepG2 cells resides within the lumen of the ER and Golgi (16,25), we next examined the effects of XN on the accumulation of newly synthesized TG in this critical pool. We isolated total microsomes consisting of both ER and Golgi. There was a marked decrease in TG accumulation in the microsomal lumen, indicating that TG availability is indeed a major determining factor in the regulation of apoB secretion under the experimental conditions. Interestingly, the reduction in the influx of TG into the microsomal lumen was coupled with a reduction in TG synthesis in the microsomal membrane. This suggests that lack of substrate (i.e., TG) in the microsomal membrane may be critical in determining TG availability for lipoprotein assembly and secretion under the experimental conditions. This finding contrasts with those of Borradaile et al. (21,28), because naringenin treatment had essentially no effect on membrane TG synthesis, indicating that lack of substrate was not involved. The cause of this discrepancy is unknown, but it suggests that the structural differences among these compounds affect lipid metabolism differently. The open C-ring found in XN may have greater lipid-lowering properties. We examined TG synthesis more closely to determine whether the decrease in microsomal TG synthesis caused by XN was associated with a decrease in cellular DGAT activity. DGAT is a key microsomal enzyme in TG biosynthesis; it acylates diacylglycerol at the sn-3 position using fatty acyl CoAs, resulting in the formation of TG (12). Tabata et al. (11) previously reported that XN inhibits DGAT activity. In their studies, using control rat liver microsomes, XN directly inhibited DGAT activity (IC50 ⫽ 50 ␮mol/L). The present study confirmed that XN decreases cellular DGAT activity under basal and lipid-rich conditions. These results compared well with those of Tabata et al. (11) and with the present results on TG synthesis in the whole cell and the microsomal membrane. It is possible that XN, by inhibiting DGAT activity, may limit the synthesis of TG in the microsomal membrane, and thus the transfer of TG into the microsomal lumen for lipoprotein assembly and secretion. Progress in understanding DGAT emerged only with the cloning of DGAT-1 and DGAT-2 in recent years. The DGAT-1 gene was the first to be sequenced and shown to be related to the ACAT gene family (32). Smith et al. (33), using DGAT-1–/– knockout mice, confirmed the existence of alternative mechanisms for synthesizing TG. In 2001 the group cloned DGAT-2 (34) and suggested that it plays a significant role in triglyceride metabolism, possibly by compensating for conditions leading to the absence of DGAT-1 (35). DGAT-2 is a new gene family that has no homology with DGAT-1 (34). Two members of this gene family, DGAT-2A and -2B, were recently identified (36). The function of DGAT-2 re-

XANTHOHUMOL INHIBITS ApoB SECRETION

mains uncertain, but the gene may have specialized functions in intracellular TG storage and in skin development (37). To determine which form of DGAT might decrease TG synthesis under the experimental conditions, we examined the expression of DGAT-1 and DGAT-2 mRNA using relative RT-PCR. XN inhibited DGAT-1 mRNA expression in a dose-dependent manner, suggesting that XN regulates DGAT-1 partly via transcriptional and/or mRNA stabilization control. Conversely, XN had no effect on DGAT-2 mRNA expression. These results were somewhat surprising, given that DGAT-2, not DGAT-1, is thought to be more relevant to VLDL production (38). However, that issue is not completely resolved. The present data suggested that under the experimental conditions, DGAT-1 was primarily involved in the decrease in TG synthesis and availability and the subsequent secretion of apoB-Lp. The inhibitory effect of XN on DGAT-1 mRNA expression may be of interest because DGAT-1 deficiency in mice induced obesity resistance and improved glucose metabolism by reversing leptin and insulin resistance (33,39). This beneficial effect on obesity and insulin resistance unveiled an exciting area of research in terms of both the molecular biology of DGAT and potential treatment methods. Although there are currently no commercial DGAT inhibitors, strategies for their development may be a worthwhile pursuit. However, much work is needed on DGAT and the compensatory mechanisms involved in TG biosynthesis before therapies directed at DGAT become commercially available. Insufficient MTP activity is another reason for the lack of transfer of newly synthesized TG from the microsomal membrane to the active lumenal pool (21,29). The citrus flavonoid naringenin was recently reported to reduce lumenal accumulation of newly synthesized TG via MTP inhibition (21,28). We therefore investigated the possibility that XN would reduce MTP activity. XN, like other flavonoids (5,6), markedly reduced MTP activity in HepG2 cells. This finding not only is novel in the sense that it extends MTP inhibition to chalcones, but also suggests that XN may limit MTP-mediated accumulation of newly synthesized TG within the microsomal lumen. This finding is in agreement with the studies by Borradaile et al. (21,28) on naringenin, with the notable exception that XN-mediated suppression in apoB secretion may also act through DGAT activity and microsomal TG synthesis. In conclusion, the present study indicated that XN is a potent inhibitor of apoB secretion. XN enhanced apoB degradation due to a lack of lipidation of the lipoprotein particle caused by insufficient transfer of newly synthesized TG from the microsomal membrane to the active lumenal pool via DGAT and MTP activity. Whether this reduction in lumenal TG accumulation is predominantly due to MTP and/or DGAT activity remains to be addressed. Specific DGAT inhibitors would be required, and none are currently available. These findings should provide the groundwork for a more comprehensive understanding of the complex effects of chalcones in the prevention of atherosclerosis. LITERATURE CITED 1. Borradaile, N. M., Carroll, K. K. & Kurowska, E. M. (1999) Regulation of HepG2 cell apolipoprotein B metabolism by the citrus flavanones hesperetin and naringenin. Lipids 34: 591–598. 2. Jahromi, M. A. & Ray, A. B. (1993) Antihyperlipidemic effect of flavonoids from Pterocarpus marsupium. J. Nat. Prod. 56: 989 –994. 3. Arai, Y., Watanabe, S., Kimira, M., Shimoi, K., Mochizuki, R. & Kinae, N. (2000) Dietary intakes of flavonols, flavones and isoflavones by Japanese women and the inverse correlation between quercetin intake and plasma LDL cholesterol concentration. J. Nutr. 130: 2243–2250. 4. Merz-Demlow, B. E., Duncan, A. M., Wangen, K. E., Xu, X., Carr, T. P.,

1345

Phipps, W. R. & Kurzer, M. S. (2000) Soy isoflavones improve plasma lipids in normocholesterolemic, premenopausal women. Am. J. Clin. Nutr. 71: 1462–1469. 5. Wilcox, L. J., Borradaile, N. M., de Dreu, L. E. & Huff, M. W. (2001) Secretion of hepatocyte apoB is inhibited by the flavonoids, naringenin and hesperetin, via reduced activity and expression of ACAT2 and MTP. J. Lipid Res. 42: 725–734. 6. Casaschi, A., Wang, Q., Dang, K., Richards, A. & Theriault, A. (2002) Intestinal apolipoprotein B secretion is inhibited by the flavonoid quercetin: potential role of microsomal triglyceride transfer protein and diacylglycerol acyltransferase. Lipids 37: 647– 652. 7. Theriault, A., Wang, Q., Van Iderstine, S. C., Chen, B., Franke, A. A. & Adeli, K. (2000) Modulation of hepatic lipoprotein synthesis and secretion by taxifolin, a plant flavonoid. J. Lipid Res. 41: 1969 –1979. 8. Miranda, C. L., Stevens, J. F., Ivanov, V., McCall, M., Frei, B., Deinzer, M. L. & Buhler, D. R. (2000) Antioxidant and prooxidant actions of prenylated and nonprenylated chalcones and flavanones in vitro. J. Agric. Food Chem. 48: 3876 –3884. 9. Dimmock, J. R., Elias, D. W., Beazely, M. A. & Kandepu, N. M. (1999) Bioactivities of chalcones. Curr. Med. Chem. 6: 1125–1149. 10. Stevens, J. F., Taylor, A. W., Clawson, J. E. & Deinzer, M. L. (1999) Fate of xanthohumol and related prenylflavonoids from hops to beer. J. Agric. Food Chem. 47: 2421–2428. 11. Tabata, N., Ito, M., Tomoda, H. & Omura, S. (1997) Xanthohumols, diacylglycerol acyltransferase inhibitors, from Humulus lupulus. Phytochemistry 46: 683– 687. 12. Coleman, R. A., Lewin, T. M. & Muoio, D. M. (2000) Physiological and nutritional regulation of enzymes of triacylglycerol synthesis. Annu. Rev. Nutr. 20: 77–103. 13. Wu, X., Sakata, N., Lui, E. & Ginsberg, H. N. (1994) Evidence for a lack of regulation of the assembly and secretion of apolipoprotein B-containing lipoprotein from HepG2 cells by cholesteryl ester. J. Biol. Chem. 269: 12375– 12382. 14. Ginsberg, H. N. (1995) Synthesis and secretion of apolipoprotein B from cultured liver cells. Curr. Opin. Lipidol. 6: 275–280. 15. Javitt, N. B. (1990) Hep G2 cells as a resource for metabolic studies: lipoprotein, cholesterol, and bile acids. FASEB J. 4: 161–168. 16. Wu, X., Shang, A., Jiang, H. & Ginsberg, H. N. (1996) Low rates of apoB secretion from HepG2 cells result from reduced delivery of newly synthesized triglyceride to a “secretion-coupled” pool. J. Lipid Res. 37: 1198 –1206. 17. Boulton, D. W., Walle, U. K. & Walle, T. (1999) Fate of the flavonoid quercetin in human cell lines: chemical instability and metabolism. J. Pharm. Pharmacol. 51: 353–359. 18. Macri, J. & Adeli, K. (1993) Development of an amplified enzymelinked immunosorbent assay for sensitive measurement of apolipoprotein B in plasma and tissue culture media. Eur. J. Clin. Chem. Clin. Biochem. 31: 441– 446. 19. Theriault, A., Wang, Q., Gapor, A. & Adeli, K. (1999) Effects of gammatocotrienol on ApoB synthesis, degradation, and secretion in HepG2 cells. Arterioscler. Thromb. Vasc. Biol. 19: 704 –712. 20. Grigor, M. R. & Bell, R. M. (1982) Separate monoacylglycerol and diacylglycerol acyltransferases function in intestinal triacylglycerol synthesis. Biochim. Biophys. Acta 712: 464 – 472. 21. Borradaile, N. M., de Dreu, L. E., Barrett, P. H. & Huff, M. W. (2002) Inhibition of hepatocyte apoB secretion by naringenin: enhanced rapid intracellular degradation independent of reduced microsomal cholesteryl esters. J. Lipid Res. 43: 1544 –1554. 22. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248 –254. 23. Luchoomun, J. & Hussain, M. M. (1999) Assembly and secretion of chylomicrons by differentiated Caco-2 cells. Nascent triglycerides and preformed phospholipids are preferentially used for lipoprotein assembly. J. Biol. Chem. 274: 19565–19572. 24. Thrift, R. N., Forte, T. M., Cahoon, B. E. & Shore, V. G. (1986) Characterization of lipoproteins produced by the human liver cell line, Hep G2, under defined conditions. J. Lipid Res. 27: 236 –250. 25. Zhang, J. & Herscovitz, H. (2003) Nascent lipidated apolipoprotein B is transported to the Golgi as an incompletely folded intermediate as probed by its association with network of endoplasmic reticulum molecular chaperones, GRP94, ERp72, BiP, calreticulin, and cyclophilin B. J. Biol. Chem. 278: 7459 – 7468. 26. Yee, W. L., Wang, Q., Agdinaoay, T., Dang, K., Chang, H., Grandinetti, A., Franke, A. A., Theriault, A. (2002) Green tea catechins decrease apolipoprotein B-100 secretion from HepG2 cells. Mol. Cell Biochem. 229: 85–92. 27. Borradaile, N. M., de Dreu, L. E., Wilcox, L. J., Edwards, J. Y. & Huff, M. W. (2002) Soya phytoestrogens, genistein and daidzein, decrease apolipoprotein B secretion from HepG2 cells through multiple mechanisms. Biochem. J. 366: 531–539. 28. Borradaile, N. M., de Dreu, L. E., Barrett, P. H., Behrsin, C. D. & Huff, M. W. (2003) Hepatocyte apoB-containing lipoprotein secretion is decreased by the grapefruit flavonoid, naringenin, via inhibition of MTP-mediated microsomal triglyceride accumulation. Biochemistry 42: 1283–1291. 29. Wang, Y., Tran, K. & Yao, Z. (1999) The activity of microsomal triglyceride transfer protein is essential for accumulation of triglyceride within microsomes in McA-RH7777 cells. A unified model for the assembly of very low density lipoproteins. J. Biol. Chem. 274: 27793–27800. 30. Davis, R. A. (1999) Cell and molecular biology of the assembly and

1346

CASASCHI ET AL.

secretion of apolipoprotein B-containing lipoproteins by the liver. Biochim. Biophys. Acta 1440: 1–31. 31. Dixon, J. L., Furukawa, S. & Ginsberg, H. N. (1991) Oleate stimulates secretion of apolipoprotein B-containing lipoproteins from Hep G2 cells by inhibiting early intracellular degradation of apolipoprotein B. J. Biol. Chem. 266: 5080 –5086. 32. Farese, R. V., Jr., Cases, S. & Smith, S. J. (2000) Triglyceride synthesis: insights from the cloning of diacylglycerol acyltransferase. Curr. Opin. Lipidol. 11: 229 –234. 33. Smith, S. J., Cases, S., Jensen, D. R., Chen, H. C., Sande, E., Tow, B., Sanan, D. A., Raber, J., Eckel, R. H. & Farese, R. V., Jr. (2000) Obesity resistance and multiple mechanisms of triglyceride synthesis in mice lacking DGAT. Nat. Genet. 25: 87–90. 34. Cases, S., Stone, S. J., Zhou, P., Yen, E., Tow, B., Lardizabal, K. D., Voelker, T. & Farese, R. V., Jr. (2001) Cloning of DGAT2, a second mammalian diacylglycerol acyltransferase, and related family members. J. Biol. Chem. 276: 38870 –38876. 35. Buhman, K. K., Smith, S. J., Stone, S. J., Repa, J. J., Wong, J. S., Knapp, F. F., Jr., Burri, B. J., Hamilton, R. L., Abumrad, N. A. & Farese, R. V., Jr. (2002)

DGAT1 is not essential for intestinal triacylglycerol absorption or chylomicron synthesis. J. Biol. Chem. 277: 25474 –25479. 36. Lardizabal, K. D., Mai, J. T., Wagner, N. W., Wyrick, A., Voelker, T. & Hawkins, D. J. (2001) DGAT2 is a new diacylglycerol acyltransferase gene family: purification, cloning, and expression in insect cells of two polypeptides from Mortierella ramanniana with diacylglycerol acyltransferase activity. J. Biol. Chem. 276: 38862–38869. 37. Stone, S. J., Myers, H. M., Watkins, S. M., Brown, B. E., Feingold, K. R., Elias, P. M. & Farese, R. V., Jr. (2004) Lipopenia and skin barrier abnormalities in DGAT2-deficient mice. J. Biol. Chem. 279: 11767–11776. 38. Meegalla, R. L., Billheimer, J. T. & Cheng, D. (2002) Concerted elevation of acyl-coenzyme A:diacylglycerol acyltransferase (DGAT) activity through independent stimulation of mRNA expression of DGAT1 and DGAT2 by carbohydrate and insulin. Biochem. Biophys. Res. Commun. 298: 317–323. 39. Chen, H. C., Smith, S. J., Ladha, Z., Jensen, D. R., Ferreira, L. D., Pulawa, L. K., McGuire, J. G., Pitas, R. E., Eckel, R. H. & Farese, R. V., Jr. (2002) Increased insulin and leptin sensitivity in mice lacking acyl CoA:diacylglycerol acyltransferase 1. J. Clin. Invest. 109: 1049 –1055.