Apoptosis

, Volume 12, Issue 11, pp 1953–1963

Effect of xanthohumol and isoxanthohumol on 3T3-L1 cell apoptosis and adipogenesis

Authors

  • Jeong-Yeh Yang
    • Department of Animal and Dairy Science, 444 Edgar L. Rhodes Center for Animal and Dairy ScienceUniversity of Georgia
  • Mary Anne Della-Fera
    • Department of Animal and Dairy Science, 444 Edgar L. Rhodes Center for Animal and Dairy ScienceUniversity of Georgia
  • Srujana Rayalam
    • Department of Animal and Dairy Science, 444 Edgar L. Rhodes Center for Animal and Dairy ScienceUniversity of Georgia
    • Department of Animal and Dairy Science, 444 Edgar L. Rhodes Center for Animal and Dairy ScienceUniversity of Georgia
    • Department of Foods and NutritionUniversity of Georgia
Original Paper

DOI: 10.1007/s10495-007-0130-4

Cite this article as:
Yang, J., Della-Fera, M.A., Rayalam, S. et al. Apoptosis (2007) 12: 1953. doi:10.1007/s10495-007-0130-4

Abstract

Xanthohumol (XN), the chalcone from beer hops has several biological activities. XN has been shown to induce apoptosis in cancer cells and also has been reported to be involved in lipid metabolism. Based on these studies and our previous work with natural compounds, we hypothesized that XN and its isomeric flavanone, isoxanthohumol (IXN), would induce apoptosis in adipocytes through the mitochondrial pathway and would inhibit maturation of preadipocytes. Adipocytes were treated with various concentrations of XN or IXN. In mature adipocytes both XN and IXN decreased viability, increased apoptosis and increased ROS production, XN being more effective. Furthermore, the antioxidants ascorbic acid and 2-mercaptoethanol prevented XN and IXN-induced ROS generation and apoptosis. Immunoblotting analysis showed an increase in the levels of cytoplasmic cytochrome c and cleaved poly (ADP-ribose) polymerase (PARP) by XN and IXN. Concomitantly, we observed activation of the effectors caspase-3/7. In maturing preadipocytes both XN and IXN were effective in reducing lipid content, XN being more potent. Moreover, the major adipocyte marker proteins such as PPARγ, C/EBPα, and aP2 decreased after treatment with XN during the maturation period and that of DGAT1 decreased after treatment with XN and IXN. Taken together, our data indicate that both XN and IXN inhibit differentiation of preadipocytes, and induce apoptosis in mature adipocytes, but XN is more potent.

Keywords

Reactive oxygen speciesCaspase-3/7Mitochondrial membrane potentialPeroxisome proliferator-activated receptor γ (PPARγ)CCAAT/enhancer binding protein α (C/EBPα)

Introduction

Obesity is a complex disorder with multiple causes including both genetic and environmental factors. Therefore, the prevention and treatment of obesity are critical to curtail the rising incidence of morbidity and mortality. Potential therapeutic agents, especially from natural products, that have the ability to inhibit adipogenesis or increase cell death by apoptosis could be important tools in preventing obesity. One group of compounds with potential anti-obesity activities are flavonoids, which are either synthetic or natural constituents of foods or drink (tea) [1, 2]. Xanthohumol (XN) is the principal flavonoid found in the hop plant, Humulus lupulus L. (Cannabaceae). XN is largely converted into its isomeric flavanone, isoxanthohumol (IXN) during the wort boiling [3]. XN is the most abundant prenylated flavonoid in hops whereas IXN is the most abundant flavonoid in all types of beer tested [4]. More recently alternative uses for hop compounds and their effects on biological processes have become an area of interest. In particular, XN has been shown to have cancer-inhibiting properties [5, 6]. Its many biological activities include inhibition of PGE2 or NO production [6], anti-tumor activity in hypoxic tumor cells [7], inhibition of diacylglycerol acyltransferase-1 (DGAT1) activity and expression in Raji cells [8], and the induction of apoptosis in various cell types [9, 10]. In addition, XN has also been reported to inhibit TG synthesis in hepatocytes [11]. XN has also been shown to act as a natural ligand for the farnesoid X receptor (FXR), a member of the nuclear hormone receptor superfamily. XN has been shown to activate FXR in vitro and it modulated genes involved in lipid or glucose metabolism [12]. Based on these studies and our previous experience in studying flavonoids, we hypothesized that the prenylated flavonoids from hops and beer would inhibit adipogenesis and induce apoptosis in adipocytes, making them potentially useful as antiobesity agents.

Adipocyte number increases as a result of increased proliferation and differentiation of preadipocytes [13]. A decrease in adipose tissue mass involves the loss of lipids through lipolysis and may also involve the loss of mature and immature fat cells through apoptosis [1416]. Adipocyte differentiation is mediated by a series of programmed changes in gene expression [17]. A cascade of transcription factors, in particular peroxisome proliferators-activated receptor (PPAR) families and CCAAT enhancer binding protein (C/EBP), control the process of adipocyte differentiation. Thereafter the expression of adipocyte genes, including adipocyte lipid binding protein (aP2) and lipid-metabolizing enzymes, dramatically increase [18].

Inducers of apoptosis include both intra- and extracellular stimuli, such as DNA damage, disruption of the cell cycle, detachment of cells from their surrounding tissue, and loss of trophic signaling [19]. Apoptosis occurs primarily through two well-recognized pathways in cells [20]. Both effector mechanisms of apoptosis are associated with caspase activation and include the intrinsic, or mitochondrial-mediated, pathways and the extrinsic, or death receptor-mediated, pathways [21]. The intrinsic pathway of apoptosis relies primarily on the permeabilization of mitochondrial membranes, with associated release of apoptotic mitochondrial proteins, leading to activation of caspase 9 and downstream cleavage of caspase 3, 6, or 7 [22]. The objective of this study was to examine the biochemical mechanism by which XN and IXN induce apoptosis and inhibit adipogenesis in 3T3-L1 cells. Our findings indicate that XN was more potent than IXN in inducing apoptosis in mature adipocytes and in inhibiting adipogenesis in maturing preadipocytes.

Materials and methods

Cell culture

3T3-L1 mouse embryo fibroblasts were obtained from American Type Culture Collection (Manassas, VA) and cultured as described elsewhere [23]. Briefly, cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (GIBCO, Grand Island, NY) containing 10% bovine calf serum (BCS) until confluent. Two days after confluency (D0), the cells were stimulated to differentiate with DMEM containing 10% fetal bovine serum (FBS), 167 nM insulin, 0.5 μM IBMX, and 1 μM dexamethasone for 2 days (D2). Cells were then maintained in 10% FBS/DMEM medium with 167 nM insulin for another 2 days (D4), followed by culturing with 10% FBS/DMEM medium for an additional 4 days (D8), at which time more than 90% of cells were mature adipocytes with accumulated fat droplets. All media contained 100 U/ml of penicillin, 100 μg/ml of streptomycin, and of 292 μg/ml glutamine (Invitrogen, Carlsbad, CA). Cells were maintained at 37°C in a humidified 5% CO2 atmosphere.

Reagents and antibodies

Phosphate-buffered saline (PBS) and DMEM medium were purchased from GIBCO (BRL Life Technologies, Grand Island, NY). XN and IXN were purchased from ALEXIS (San Diego, CA). The viability assay kit (CellTiter 96 Aqueous One Solution Cell Proliferation Assay; containing 3-(4,5-dimethythizol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay reagent (MTS) and Caspase-GloTM 3/7 assay kit were purchased from Promega (Madison, WI). Ascorbic acid and 2-mercaptoethanol were purchased from Sigma (St. Louis, MO, USA). Antibodies specific for polyclonal peroxisome proliferator-activated receptor (PPAR)γ, CCAAT/enhancer binding proteins (C/EBPs), aP2, DGAT1, poly(ADP-ribose) polymerase (PARP), β-Actin, and cytochrome c were from Santa Cruz Biotechnology (Santa Cruz, CA).

MTS cell viability assay

Adipocytes were incubated with XN and IXN. Prior to measuring viability, treatment media were removed and replaced with 100 μl fresh 10% FBS/DMEM medium and 20 μl MTS solution. Cells were then returned to the incubator for an additional 2 h before 25 μl of 10% SDS was added to stop the reaction. The absorbance was measured at 490 nm in a plate reader (μQuantTM Bio-Tek Instruments, Inc. Winooski, VT) to determine the formazan concentration, which is proportional to the number of live cells.

Apoptosis assays

For the assessment of apoptosis, the ApoStrandTM ELISA Apoptosis Detection Kit (Biomol, Plymouth Meeting, PA) was used. This kit detects single stranded DNA, which occurs in apoptotic cells but not in necrotic cells or in cells with DNA breaks in the absence of apoptosis [24, 25]. Adipocytes were incubated with XN and IXN for the times and at the concentrations indicated in the Results and Figure legends. Thereafter, treatment media was removed and the cells were fixed for 30 min and assayed according to the manufacturer’s instructions.

Caspase-3/7 activity assay

Adipocytes were incubated with XN and IXN (times and concentrations indicated in Results and figure legends). Thereafter, 100 μl of caspase-Glo 3/7 (Promega, Madison, WI) reagent was added to each sample and the cells were incubated for 1 h and assayed according to the manufacturer’s instructions.

Measurement of intracellular ROS generation

The determination of ROS generation was based on the oxidation of the nonfluorescent 2,7-dichlorodihydroflourescein diacetate (DCHF) into a fluorescent dye, 2,7-dichloroflourescein (DCF) by peroxide. Control cells and cell treated with XN and IXN (times and concentrations indicated in Results) were analyzed for changes in fluorescence. Following exposure to treatment, cells were washed twice with PBS and then incubated for 30 min at 37°C in the dark with the oxidation sensitive probe, DCHF (Molecular Probes, Eugene, OR) at 2.5 μM. Production of ROS was measured by the change in fluorescence at an excitation wavelength of 495 nm and an emission wavelength of 525 nm.

Measurement of mitochondrial membrane potential

Changes in the mitochondrial trans-membrane potential during apoptosis were measured using 3,3′-dihexyloxacarbocyanine (DiOC6; Molecular Probes, Eugene, OR), which is a cationic dye. Adipocytes exposed to XN and IXN for up to 90 min were incubated with 100 nM DiOC6 for 30 min at 37°C. The cells were then washed twice in PBS and detached from the plates by trypsinzation. Fluorescence was measured with excitation wavelength of 488 nm and emission wavelength of 530 nm.

Quantification of lipid content and oil red O staining

Lipid content was measured using a commercially available kit (AdipoRed Assay Reagent; Cambrex Bio Science Walkersville, Inc.). In brief, XN and IXN along with 0.01% DMSO control were added with the induction medium for days 0–6 of adipogenesis. Medium was changed every 2 days. On day 6, intracellular lipid content was measured by AdipoRed Assay. Cells were washed with PBS (pH 7.4) and 200 μl of PBS was added to the wells. About 5 μl of AdipoRed reagent was added to each well. After 10 min, the plates were placed in the fluorometer and fluorescence was measured with excitation wavelength of 485 nm and emission wavelength of 572 nm. To visualize lipid content, treated cells were stained with oil red O and hematoxylin as described by Suryawan and Hu [26]. After mounting with glycerol gelatin, three images for each dish were captured using ImagePro software (MediaCybernetics, Silver Spring, MD).

Western blot analysis

Whole cell extracts were prepared by washing the cells with PBS and suspending in a lysis buffer (20 mM Tris, pH 7.5, 150 mM sodium chloride, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM sodium vanadate (Na3VO4), 1 μg/ml aprotinin, 1 μg/ml leupeptin, and 100 ug/ml phenylmethylsulfonyl fluoride). After 30 min of rocking at 4°C, the mixtures were centrifuged (10,000g) for 10 min, and the supernatants were collected as whole-cell extracts. To isolate the cytosolic fraction, cells were washed with ice-cold PBS and resuspended in isotonic homogenizing buffer (250 mM sucrose, 10 mM potassium chloride, 1.5 mM magnesium chloride, 1 mM EGTA, 1 mM EDTA, 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonylfluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin, 10 mM HEPES-KOH, pH 7.4). After 30 min incubation on ice, cells were homogenized with a glass Dounce homogenizer (30 strokes) and centrifuged at 700g for 10 min. The supernatant was collected as the cytosolic fraction. The protein concentration was determined by the method of Bradford [27] with bovine serum albumin as the standard. Western blot analysis was performed using the commercial NUPAGE system (Novex/Invitrogen, Carlsbad, CA), where a lithium dodecyl sulfate (LDS) sample buffer (Tris/glycerol buffer, pH 8.5) was mixed with fresh dithiothreitol and added to samples. Samples were then heated to 70°C for 10 min, separated by 12% acrylamide gels and analyzed by immunoblotting as previously described [28]. Immunoblots were developed using ECL kit (Piscataway, NJ, USA). All experiments were repeated at least two times. Representative Western blots are shown along with the graphs of the quantitative data.

Quantitative analysis of Western blot data

Measurement of signal intensity on PVDF membranes after Western blotting with various antibodies was performed using a FluorChemTMdensitomer with AlphaEaseFCTM image processing and analysis software (Alpha Innotech Corporation). For statistical analysis, all data were expressed as integrated density values (IDV). For PPARγ, C/EBPα, aP2, DGAT1, and PARP, IDVs were calculated as the density values of the specific protein bands/β-actin density values and expressed as percentage of the control. For cytochrome c, IDVs are expressed as percentage of 0 h. All figures showing quantitative analysis include data from at least three independent experiments.

Statistical analysis

Data were converted to percent control by dividing raw values of each replicate by the mean control values (separate means were calculated for controls from each time period, when applicable). The ratios were then multiplied by 100. One- or two-way analysis of variance (GLM procedure, Statistica, version 6.1; StatSoft, Inc.) was used to determine significance of treatment effects and interactions. Fisher’s post-hoc least significant difference test was used to determine significance of differences among means. Statistically significant differences are defined at the 95% confidence interval. Data shown are means ± SEM.

Results

XN and IXN reduced cell viability and induced apoptosis

3T3-L1 mature adipocytes were treated with XN or IXN at various doses for 24 or 48 h. After treatment, the number of live cells was determined by the MTS assay. As shown in Fig. 1, both compounds caused a reduction in cell viability in a dose- and time-dependent manner. At the 75 μM concentration XN (51.0 ± 2.1% decrease, P < 0.05) was more effective than IXN (22.1 ± 1.6% decrease, P < 0.05) at 24 h. Half-maximal inhibitory concentrations (IC50) for XN were calculated using the results obtained from viability data. The IC50 values for XN decreased from 75 ± 5 μM after 24 h to 53 ± 3.5 μM after 48 h. However, the IC50 values for IXN were not calculated since maximal inhibition of cell viability was not reached with the concentrations tested.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig1_HTML.gif
Fig. 1

Effect of XN and IXN on viability. Mature 3T3-L1 adipocytes were incubated with XN (A) or IXN (B) at various concentrations (0, 25, 50, 75, 100 μM) for 24 or 48 h. Cell viability was determined by the MTS colorimetric assay. Assays were performed on eight replicates for each treatment. abcd: Within a time period means that are not denoted with a common letter are different, P < 0.05

We next investigated whether the reduction in cell number by both isomers involved apoptosis, using the ssDNA ELISA assay as an indicator of cellular apoptosis. As shown in Fig. 2, exposure of adipocytes to XN or IXN resulted in a dose- and time-dependent induction of apoptosis. At 75 μM, XN increased apoptosis by 115.6 ± 9.7% (P < 0.05) and 229.5 ± 14.4% (P < 0.05) after 24 and 48 h, respectively. IXN also increased apoptosis by 64.1 ± 9.1% (P < 0.05) and 69.2 ± 14.1% (P < 0.05) after 24 and 48 h, respectively. Half maximal effective concentration (EC50) for XN decreased from 73 ± 12.5 μM after 24 h to 56.5 ± 5 μM after 48 h. The EC50 value for IXN after 48 h was 80 ± 12 μM. Thus, XN was more effective than IXN in inducing apoptosis. These results indicate that the decrease in viability was due, at least in part, to induction of apoptosis by both isomers. The 75 μM concentration of XN and IXN was selected for subsequent experiments.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig2_HTML.gif
Fig. 2

Effect of XN and IXN on apoptosis. Mature 3T3-L1 adipocytes were incubated with XN (A) or IXN (B) at various concentrations (0, 25, 50, 75, 100 μM) for 24 or 48 h. Cell apoptosis was evaluated by ssDNA ELISA. Assays were performed on eight replicates for each treatment and repeated twice. abcd: Within a time period means that are not denoted with a common letter are different, P < 0.05

XN and IXN triggered apoptosis by oxidative stress

Several reports suggest an involvement of ROS in the signal transduction pathway leading to apoptosis [29]. To determine the involvement of ROS in XN and IXN-induced apoptosis, ROS levels were measured in adipocytes treated with the two compounds. As shown in Fig. 3, the intracellular hydrogen peroxide levels increased dramatically after cells were treated with XN. In contrast, IXN induced a slow and smaller increase in intracellular hydrogen peroxide. After 20 min incubation of mature adipocytes with XN (Fig. 3A), ROS was 1509.0 ± 125.1% greater than control while the increase after treatment with IXN (Fig. 3B) was only 36.9 ± 7.3%. Furthermore, when adipocytes were pretreated with ascorbic acid (AA) and 2-mercaptoethanol (2-ME) for 3 h before incubation with XN or IXN, ROS production was effectively reduced (Fig. 3C). We next investigated whether the enhanced apoptosis by both compounds involved changes in ROS production. AA or 2-ME pretreatment blocked the increase in apoptosis induced by both compounds (Fig. 3D). This indicates that ROS generation is a critical component in apoptosis induced by both XN and IXN.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig3_HTML.gif
Fig. 3

Effect of XN and IXN on intracellular hydrogen peroxide production. 3T3-L1 adipocytes were incubated with 75 μM XN (A) or IXN (B) and the time-course on ROS production was determined by the change in fluorescence of the oxidized probe as indicated in Materials and Methods. (C) Reduction of XN and IXN-induced ROS generation in cells pretreated with ascorbic acid (AA) or 2-mercaptoethanol (2-ME). Peroxide levels measured as described in Materials and Methods are shown for untreated cells (control); cells exposed to 75 μM XN or IXN for 20 min; cells preloaded with 1 mM AA or 2-ME for 3 h, and cells preloaded with 1 mM AA or 2-ME in the absence or presence of 75 μM XN or IXN for 20 min. (D) Reduction of XN or IXN-induced apoptosis in cells pretreated with AA or 2-ME. 3T3-L1 adipocytes pretreated with 1 mM AA or 2-ME for 3 h prior to exposure to 75 μM XN or IXN for 24 h. Cellular apoptosis was evaluated by ssDNA ELISA. Assays were performed in eight replicates for each treatment and ROS production experiments were repeated twice. abc: Means that are not denoted with a common letter are different, P < 0.05

XN and IXN decreased mitochondrial membrane potential and increased cytochrome c release

To evaluate the changes in mitochondrial membrane potential (Δψm) during the apoptotic process, a strong cationic dye, DiOC6(3), was used. 3T3-L1 mature adipocytes were treated with 75 μM concentration of XN or IXN at the indicated times. Treatment with IXN for 90 min had no effect on the Δψm but XN caused dissipation of the Δψm in a time-dependent manner (Fig. 4A), as indicated by changes in DiOC6(3) up-take. As the dissipation of Δψm is associated with the release of cytochrome c from mitochondria, Western blot analyses for cytochrome c were performed. Treatment of cells with either XN or IXN increased cytoplasmic cytochrome c levels in a time-dependent manner (Fig. 4B). XN (335.1 ± 45.5% increase, P < 0.05) was more effective than IXN (136.4 ± 21.3% increase, P < 0.05) at 48 h.
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig4_HTML.gif
Fig. 4

Effect of XN and IXN on mitochondrial membrane potential and cytoplasmic cytochrome c levels. (A) Time-course of the effect of XN (A) or IXN (B) on MMP. 3T3-L1 adipocytes were incubated with 75 μM XN or IXN for 10, 30, or 90 min. Cells were treated with DiOC6(3) and fluorescence intensity was measured as indicated in Materials and Methods. Assays were performed in eight replicates for each treatment. abcd: Within a time period means that are not denoted with a common letter are different, P < 0.05. (B) 3T3-L1 adipocytes were incubated with 75 μM XN or IXN for 6, 12, 24, or 48 h. Equal amounts of protein from cytosolic fraction were analyzed by Western blotting using specific antibodies. Actin was used as an equal loading control. Densitometric quantitation of the autoradiograms for cytochrome c was performed. Integrated density values were calculated and expressed as % 0 h. All experiments were repeated in triplicate. abc: Means that are not denoted with a common letter are different, P < 0.05

XN and IXN induced activation of caspase-3/7

To determine whether the observed cytochrome c release was accompanied by activation of caspase 3/7, we measured caspase 3/7 activity in mature 3T3-L1 adipocytes. Caspase 3/7 activity increased in a time-dependent manner during treatment with either XN or IXN. After 12 h incubation of mature adipocytes with 100 μM XN (Fig. 5A), caspase 3/7 was 75.0 ± 1.8% greater than control (P < 0.05), while the increase after treatment with IXN (Fig. 5B) was 38.9 ± 4.2% greater than control (P < 0.05).
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig5_HTML.gif
Fig. 5

Effect of XN and IXN on caspase-3/7 activation. 3T3-L1 adipocytes were treated with XN (A) or IXN (B) at various concentrations (0, 25, 50, 75, 100 μM) for 3, 6, 12, or 24 h. Caspase-3 activity was determined by the Caspase-Glo 3/7 luminescent assay. Assays were performed on eight replicates for each treatment. abc: Within a time period means that are not denoted with a common letter are different, P < 0.05

XN and IXN induced PARP cleavage

Activation of caspase-3 leads to the cleavage of a number of proteins, one of which is PARP. We further confirmed the activation of caspase-3 by observing the cleavage of PARP. Treatment of 3T3-L1 mature adipocytes with 75 μM concentration of XN or IXN caused dose-and time-dependent proteolytic cleavages of PARP (116 kDa), with disappearance of the full-length protein and accumulation of 85 kDa fragments (Fig. 6).
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig6_HTML.gif
Fig. 6

Effect of XN and IXN on PARP cleavage. 3T3-L1 adipocytes were treated with 75 μM XN or IXN for 6, 12, or 24 h. Cell lysates were analyzed by Western blotting as described in Materials and Methods, and PARP protein bands were detected using a specific antibody, with β-Actin used as an equal loading control (A). Densitometric quantitation of the autoradiograms for PARP was performed for XN (B) and IXN (C) treatments, respectively. Integrated density values were calculated and expressed as % control. All experiments were repeated three times. abc: Within analytes, means that are not denoted with a common letter are different, P < 0.05

XN and IXN inhibited lipid accumulation

To determine the effect of XN and IXN on lipid accumulation 2-day post confluent preadipocytes were treated with XN or IXN for 6 days as described in Materials and Methods. Results from the AdipoRed assays showed that both flavanones reduced lipid content in a dose dependent manner (Fig. 7A and B). At 25 μM, XN (97.9 ± 0.1% reduction, P < 0.05) was more potent than IXN (17.9 ± 3.1% reduction, P < 0.05). Oil red O staining indicated that 25 μM XN treatment for days 0–6 completely inhibited 3T3-L1 lipid accumulation (Fig. 7C) and cells retained the fibroblastic characteristics. Cell viability assays were performed in parallel with AdipoRed assays to determine if the decrease in lipid accumulation was due to decrease in cell numbers. The data indicated that neither XN or IXN at the tested concentrations had a significant effect on cell viability (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig7_HTML.gif
Fig. 7

Effect of XN and IXN on triacylglycerol accumulation. The differentiation of 2-day post-confluent 3T3-L1 preadipocytes was induced by standard adipogenic medium to initiate adipogenesis as described in Materials and Methods in the absence or presence of XN (A) or IXN (B) at various concentrations (0, 3.125, 6.25, 12.5, 25 μM) during the days 0–6. On day 6 lipid content was measured by AdipoRed assay. All assays were performed on eight replicates for each treatment. abcde: Means that are not denoted with a common letter are different, P < 0.05. (C) On day 6, cellular triglyceride was stained with oil red O dye and cells were photographed at 200× magnification. The experiment was performed on three replicates for each treatment. Representative images are shown

XN and IXN decreased the expression of PPARγ and C/EBPα as well as aP2 and DGAT1

To determine whether the decrease in lipid accumulation with XN or IXN was related to adipocyte marker protein expression, 3T3-L1 cells were treated with XN or IXN from 0–6 days of differentiation. Quantitative analysis revealed that IXN treatment did not significantly alter the expression levels of PPARγ, C/EBPα, and aP2 but XN decreased the expression levels (Fig. 8). Interestingly, the expression of DGAT1 was significantly decreased by both XN (70.9 ± 3.0% decrease, P < 0.05) and IXN (23.6 ± 8.4% decrease, P < 0.05).
https://static-content.springer.com/image/art%3A10.1007%2Fs10495-007-0130-4/MediaObjects/10495_2007_130_Fig8_HTML.gif
Fig. 8

Effect of XN and IXN on expression of adipocyte marker proteins. The differentiation of 2-day post-confluent 3T3-L1 preadipocytes was induced by standard adipogenic medium to initiate adipogenesis as described in Materials and Methods in the absence or presence of 25 μM XN or IXN during the days 0–6. On day 6, cells were lysed, separated on 10% NUPAGE and subjected to immunoblotting to examine the expression levels of PPARγ, C/EBPα, aP2 and DGAT1. Actin was used as an internal reference for sample loading control. Densitometric quantitation of the autoradiograms for PPARγ, C/EBPα, aP2 and DGAT1 was performed. Integrated density values were calculated and expressed as % control. All experiments were repeated in triplicate. abc: Means that are not denoted with a common letter are different, P < 0.05

Discussion

This study elucidates the biological effect of XN and its isomer, IXN, compounds present in extracts from hops (Humulus lupulus), on the adipocyte life cycle in 3T3-L1 adipocytes. Our study showed that treatment of 3T3-L1 mature adipocytes with XN or IXN induces apoptosis, inhibits adipogenesis in a dose-dependent manner, with XN being the more effective compound. Chemically, XN is a prenylated chalcone (2′,4,4′-trihydroxy-3′-prenyl-6′-methoxychalcone) with an open C-ring and IXN (5-O-methyl-8-prenylnaringenin), an isomer of XN, lacks the open C-ring. The biological effects of chalcones depends on their chemical structure [30]. We speculate that the structural differences may be one reason for the difference in the effects of XN and IXN. However, further studies are needed to demonstrate the reason for the difference in the potency between these compounds.

Our finding that XN induced apoptosis in adipocytes is consistent with previous reports that XN and its isomer can induce apoptosis and inhibit cell proliferation in other cell types, including prostate epithelial cells, B-chronic lymphocytic leukemia, and human cancer cell lines [10, 30, 31]. We examined the nature of XN-triggered apoptotic signaling in mature adipocytes, focusing on ROS mediated responses. Oxidative stress has been demonstrated to act as a stimulator of various cell responses, including apoptosis [32, 33]. There is considerable evidence that ROS can induce apoptosis via several pathways, including a mitochondria-dependent pathway in various cells and tissues [34]. The mitochondrion is also a pivotal organelle for the induction of apoptosis via the intrinsic pathway. Increased mitochondrial permeability and dissipation of the electrochemical gradient or membrane potential (MMP) via opening of the mitochondrial permeability transition pore (MTP) triggers cell death by releasing apoptogenic factors from within the mitochondria, with subsequent cytochrome c release, apoptosome formation, and ultimately apoptosis induction [35, 36]. Similarly, numerous reports have demonstrated that ROS generation is an important controller of subsequent apoptotic biochemical changes, including MMP changes and caspase-3 activation [37, 38]. Thus, we assume that the generation of ROS may act on the mitochondria, causing MMP loss and subsequently leading to apoptosis. We found that XN and IXN increased ROS production (Fig. 3A and B) and XN was more potent than IXN. Furthermore, AA or 2-ME pretreatment blocked both the increase in ROS generation and apoptosis induced by both compounds, confirming that apoptosis induced by both compounds is at least partly mediated by ROS. In addition, our data show that XN decreased MMP and increased release of cytochrome c in a time dependent manner. Release of cytochrome c activates apoptotic protease activating factor 1 (Apaf-1), allowing it to assemble the multiprotein caspase-activating complex apoptosome and to bind to and activate procaspase-9 and the downstream effector caspase cascade [39]. Furthermore, activation of caspase-3 leads to the cleavage of a number of proteins, including poly (ADP-ribose) polymerase (PARP). The cleavage of PARP is another hallmark of apoptosis [40]. PARP is a nuclear enzyme that facilitates DNA repair in response to DNA damage [41]. In this regard, Pan et al. showed that XN induced apoptosis by activation of caspase-3, PARP cleavage, and Bcl-2 family protein expression in human colon cancer cells [9]. Consistent with the above results, our results indicate that XN and IXN increased caspase3/7 activity (Fig. 5) and caused time-dependent proteolytic cleavages of PARP (116 kDa), with the accumulation of 85 kDa fragments.

The adipocyte differentiation program is regulated by the sequential expression of transcriptional activators, mainly PPAR families [42]. During adipocyte differentiation, transcriptional factors such as PPARγ and C/EBPs are involved in the sequential expression of adipocyte-specific proteins such as glucose transporter (GLUT) 4 and aP2 [18, 43]. Our study showed that treatment of 3T3-L1 cells with XN or IXN during differentiation suppressed lipid accumulation in maturing preadipocytes in a dose-dependent manner and that XN was the more effective compound in inhibiting adipogenesis. The cell volume of adipocytes is largely dependent on the accumulation of triglyceride (TG). DGAT catalyzes the final step in the glycerol phosphate pathway, considered the major pathway for TG synthesis [44]. Tabata et al. reported that XN inhibited the activity of DGAT [8]. XN also inhibited TG and apolipoprotein B secretion [11]. Similarly, in our study, the expression of DGAT was significantly decreased by XN and IXN.

Farnesoid X receptor expression was shown to be positively correlated with lipid accumulation [45] and XN was shown to act on FXR and regulate downstream gene expression in a manner similar to selective bile acid receptor modulators (SBARM) like guggulsterone and PUFAs [12]. In fact, we have recently shown that cis-guggulsterone also increases apoptosis of mature adipocytes and decreases adipogenesis of maturing preadipocytes [46]. Regulation of adipogenesis by XN might therefore be at least partly mediated by modulating FXR target genes like PPARγ and C/EBPα similar to other SBARMs. Moreover, in our study, the adipocyte-specific proteins PPARγ, C/EBPα, and aP2 were down regulated after treatment with XN whereas IXN treatment did not significantly alter the expression levels of these proteins. These results suggested that the decreased adipogenesis caused by XN was accompanied by a strong inhibition of the adipocyte specific transcription factors.

Conclusion

This report describes a novel discovery that the treatment of 3T3-L1 adipocytes with XN or its isomer, IXN, leads to an enhancement of apoptosis and suppression of adipogenesis. Both XN and IXN caused an increase in intracellular ROS level leading to decreased MMP and activation of caspase-3 and 7, subsequently leading to apoptotic biochemical changes, including PARP cleavage and cytochrome c release, XN being more potent. Furthermore, AA or 2-ME pretreatment significantly blocked both the increase in ROS generation and apoptosis induced by both XN and IXN. Both XN and IXN decreased adipogenesis during the differentiation period, XN being more potent. Moreover, the major adipocyte marker proteins such as PPARγ, C/EBPα, aP2, and DGAT decreased after treatment with XN. Although results from in vitro experiments cannot be directly extrapolated to clinical effects, these studies may help in elucidating the various molecular pathways in adipocytes that are targeted by XN and IXN.

Acknowledgments

This work was supported in part by grants from AptoTec, Inc., and the Georgia Research Alliance and by the Georgia Research Alliance Eminent Scholar endowment held by CAB.

Copyright information

© Springer Science+Business Media, LLC 2007