Chrysin 7‐O‐β‐d‐glucopyranoside increases hepatic low‐density lipoprotein receptor expression through AMP‐activated protein kinase activation

Elevated plasma low‐density lipoprotein (LDL) cholesterol level is a risk factor for developing atherosclerosis. Increased LDL receptor (LDLR) expression is expected to reduce the risk of atherosclerotic disease since hepatic LDLR is essential for clearing plasma LDL cholesterol. Here, we screened human LDLR promoter effectors and observed that extracts from peduncles of sweet cherry (Prunus avium) ‘Sato‐Nishiki’ induce LDLR gene promoter activity. We used several analytical and chemical methods to show that chrysin 7‐O‐β‐d‐glucopyranoside (chrysin‐7G) is one of the compounds that stimulate LDLR gene promoter activity in cherry peduncle extracts. Furthermore, synthetic chrysin‐7G increased the expression and activity of LDLR. The chrysin‐7G–mediated increase in LDLR expression and activity was completely abolished by treatment with an AMP‐activated protein kinase (AMPK) inhibitor, compound C. These results indicate that chrysin‐7G increases LDLR expression through AMPK activation and may be a useful compound that can be recycled from waste parts of agricultural products.

Elevated plasma low-density lipoprotein (LDL) cholesterol level is a risk factor for developing atherosclerosis. Increased LDL receptor (LDLR) expression is expected to reduce the risk of atherosclerotic disease since hepatic LDLR is essential for clearing plasma LDL cholesterol. Here, we screened human LDLR promoter effectors and observed that extracts from peduncles of sweet cherry (Prunus avium) 'Sato-Nishiki' induce LDLR gene promoter activity. We used several analytical and chemical methods to show that chrysin 7-O-β-D-glucopyranoside (chrysin-7G) is one of the compounds that stimulate LDLR gene promoter activity in cherry peduncle extracts. Furthermore, synthetic chrysin-7G increased the expression and activity of LDLR. The chrysin-7G-mediated increase in LDLR expression and activity was completely abolished by treatment with an AMP-activated protein kinase (AMPK) inhibitor, compound C. These results indicate that chrysin-7G increases LDLR expression through AMPK activation and may be a useful compound that can be recycled from waste parts of agricultural products.
Cardiovascular disorders are the leading causes of death worldwide. Elevated low-density lipoprotein (LDL) cholesterol level in plasma is a potent risk factor for atherosclerosis and coronary heart disease [1]. The LDL receptor (LDLR) is primarily expressed in the liver, and the liver metabolizes plasma LDL through LDLRmediated endocytosis [2]. High LDLR expression in the liver suppresses plasma LDL-cholesterol levels and is believed to be effective for preventing atherosclerotic diseases.
Chrysin is a flavonoid, classified as a flavone, and is found in honey, propolis, and various plants [14]. Chrysin exhibits several pharmacologic activities against cancer, diabetes, cardiovascular diseases, and neurodegenerative diseases [15,16]. Furthermore, chrysin is found in plants as glycosides; however, whether it is effective in glycoside form is not fully understood [14].
We have previously reported that small compounds of food ingredients increase LDLR expression using Huh-7 cells that stably express a luciferase reporter and are driven by the LDLR gene promoter [17,18]. In the present study, we used this cell line to identify compounds that stimulate the LDLR gene promoter activity from waste parts of agricultural products.

Results
Cherry peduncles contain compounds that stimulate the LDLR gene promoter activity We have previously established a stable cell line that expresses the luciferase reporter gene under the control of the LDLR promoter region from À595 to +36 [17]. Using this stable cell line, we aimed to identify compounds that increase LDLR expression from waste parts of agricultural products, including cherry seeds, cherry peduncles, grape skins, and grape stalks. 50% ethanol extracts of cherry peduncles and grape stalks stimulated the LDLR gene promoter activity (Fig. 1A). Since grape stalks contain high levels of resveratrol, which we have previously reported to increase LDLR expression [19], our later experiments focused on cherry peduncles. Cherry peduncle extracts stimulated the LDLR gene promoter activity in a dosedependent manner (Fig. 1B). Next, the cherry peduncle extracts were divided into ethyl acetate (EtOAc) and  water fractions. Both fractions stimulated the promoter activity of the LDLR gene (Fig. 1C). Although the activating capacity of both fractions was attenuated compared with the extracts before fractionation (whole), the EtOAc fraction stimulated it more than the water fraction, suggesting that the responsible compounds are contained more abundantly in the EtOAc fraction. Therefore, using HPLC we further fractionated the EtOAc fraction into 100 fractions. Several fractions with activation potential are noted, as shown in Fig. 1D. These results indicate that cherry peduncles contain multiple compounds that stimulate the LDLR gene promoter activity.
Chrysin-7G is one of the compounds that stimulate the LDLR gene promoter activity in cherry peduncle extracts Adjacent active fractions were collected and subjected to HPLC analysis to determine the compounds that stimulate the LDLR gene promoter activity. Because fractions 84-86 were mainly occupied by a single compound, this peak was used as an indicator to isolate the active compound. NMR analysis of this active compound revealed that chrysin-7G is the candidate compound ( Fig. 2A). Chrysin-7G, purified at 86% purity from cherry peduncles, stimulated the LDLR gene promoter activity (Fig. 2B, left). To further confirm whether chrysin-7G (1) stimulates the LDLR gene promoter activity, 1 was prepared by Schmidt glycosylation employing imidate 4 and commercially available chrysin (Fig. 3). Synthetic chrysin-7G also stimulated the LDLR gene promoter activity, which was stronger than the purified chrysin-7G (Fig. 2B). These results indicate that chrysin-7G is the compound that stimulates the LDLR gene promoter activity. The synthetic chrysin-7G was used in later experiments. Next, we investigated the cytotoxic effects of chrysin-7G on Huh-7 cells using lactate dehydrogenase (LDH) assay. Treatment with 30-μM chrysin-7G for 24 h did not affect the LDH release from Huh-7 cells (Fig. 2C).

Chrysin-7G increases the LDLR expression and activity
Huh-7 cells were treated with 30-μM chrysin-7G for 24 h to determine the effects of chrysin-7G on the endogenous LDLR expression. Our real-time quantitative PCR and immunoblotting analyses demonstrated that LDLR mRNA and protein levels were increased by the treatment with 30-μM chrysin-7G (Fig. 4A,B). Therefore, Huh-7 cells were treated with 30-μM chrysin-7G for 24 h and subsequently incubated with suggesting that the chrysin-7G-mediated increase of LDLR mRNA leads to increased LDL uptake.
Chrysin-7G stimulates ERK1/2 phosphorylation; however, ERK pathway activation is not involved in the LDLR gene expression stimulation by chrysin-7G Next, we examined whether chrysin-7G treatment stimulates ERK1/2 phosphorylation. As shown in Fig. 6A, 24-h treatment with chrysin-7G increases ERK1/2 phosphorylation in Huh-7 cells. However, chrysin-7G increased LDLR mRNA levels even after the addition of U0126, a specific MEK inhibitor of the ERK pathway (Fig. 6B). These results indicate that the ERK pathway is not involved in the LDLR gene expression stimulation by chrysin-7G.

AMP-activated protein kinase is required for chrysin-7G-mediated stimulation of LDLR expression
To explore the signaling pathway by which chrysin-7G stimulates LDLR gene expression, experiments were conducted using inhibitors of several signaling pathways. We observed that treatment with compound C, an AMP-activated protein kinase (AMPK) inhibitor, completely abolished the upregulation of the LDLR gene promoter activity (Fig. 7A) and LDLR mRNA and protein levels ( Fig. 7B,C) by chrysin-7G. Treatment with compound C reduced the activity of AMPK, as judged by the reduced phosphorylation levels of AMPK (Fig. 7C). In addition, treatment with compound C completely abolished the upregulation of chrysin-7Gmediated LDL uptake (Fig. 7D). The 24-h treatment with chrysin-7G significantly increased the phosphorylation levels of AMPK and its substrate ACC (Fig. 8A). Moreover, chrysin-7G treatment increased the AMPK phosphorylation levels in 1-to 6-h treatment (Fig. 8B). Next, the effect of AMPK activation on the LDLR gene promoter activity was examined. The LDLR promoter activity was increased under cholesterol-depleted conditions that stimulate the SREBP activity, although it was unchanged when treated with AMPK activators, A769662, metformin, and AICAR ( Fig. 9).

Chrysin-7G does not increase the expression levels of Egr1 and C/EBPβ in Huh-7 cells
Because Egr1 and C/EBPβ are involved in the upregulation of LDLR expression by oncostatin M [10] and that AMPK activation increases their expression levels [20,21], we examined the effect of chrysin-7G on the expression levels of Egr1 and C/EBPβ in Huh-7 cells. Treatment with chrysin-7G for 24 h did not stimulate the expression levels of Egr1 and C/EBPβ in Huh-7 cells (Fig. 10

Discussion
In this study, we showed that chrysin-7G, derived from cherry peduncles, activates LDLR by stimulating the LDLR gene promoter activity and that this activation is mediated by the AMPK pathway upregulation.
Agricultural products are significant not only as a source of energy but also as a source of various compounds that affect our health. Our analysis showed that each gram of dry-weight of cherry peduncles contains 0.35-mg chrysin-7G (data not shown). Identification of chrysin-7G, which increases the LDLR activity, from the disposal parts of cherries, shows a way to effectively use the waste part of cherries and increases its value. In addition to chrysin-7G, there may be multiple active compounds in the cherry extract that increase the LDLR gene promoter activity (Fig. 1C,D). In future, to further increase the added value of cherries, we plan to identify these compounds from cherry peduncles.  In the present study, we showed that the chrysin-7G aglycone, chrysin, stimulated the LDLR gene promoter activity at low concentrations (5-20 μM), whereas high chrysin concentrations (60-100 μM) decreased it (Fig. 5A). We have previously reported that 50-100-μM chrysin reduced the FAS gene expression by suppressing the SREBP activity [22]. Given that the LDLR gene promoter is also under the control of SREBP, the reduction of the LDLR gene promoter activity at high chrysin concentrations is possibly because of SREBP activity suppression. On the other hand, low chrysin concentrations conversely stimulated the LDLR gene promoter activity and expression; however, no change in FAS gene expression was noted, suggesting that chrysin acts differently depending on its concentration and that low chrysin concentrations did not suppress the SREBP activity. Considering that low chrysin-7G and chrysin concentrations have stimulatory effects on LDLR expression, it is likely that these compounds exert their effects from outside the cell at low concentrations since flavonoid glycosides are not readily absorbed into the cell.
It is widely known that plant-derived compounds, including flavonoids, regulate signaling pathways in mammalian cells. We have previously reported that the flavonoid kaempferol activates the ERK pathway, and this activation leads to the increase in LDLR expression by activating Sp1, whereas other flavonoids, including luteolin and apigenin, also activate the ERK pathway although do not increase LDLR expression [18]. In this study, we demonstrated that chrysin-7G activates the ERK pathway; however, its activation is not involved in the increase in LDLR gene expression. This evidence suggests that ERK pathway activation alone does not upregulate LDLR gene expression.
Chrysin, but not chrysin-7G, has been reported to activate the AMPK pathway in several cell lines [23][24][25]. Chrysin-7G-containing moringa extracts have been reported to inhibit 3T3-L1 cell differentiation through AMPK activation; however, whether chrysin-7G is involved in AMPK activation is unclear [26]. In the present study, we showed that synthetic chrysin-7G activates the AMPK pathway in Huh-7 cells. Thus, chrysin-7G may be involved in the AMPK pathway activation by moringa extracts. Moreover, we showed that the AMPK pathway activation by chrysin-7G  leads to increased promoter activity and protein levels of the LDLR gene. AMPK activation has been reported to increase LDLR protein levels. AMPK activation downregulates PCSK9 mRNA levels by suppressing the SREBP activity, thereby stabilizing the LDLR protein [27,28]. To our knowledge, no reports on AMPK activation increasing the LDLR gene promoter activity are available. Considering that several AMPK activators did not stimulate the LDLR gene promoter activity (Fig. 7), it appears that chrysin-7G cannot increase the LDLR gene promoter activity by AMPK activation alone. Alternatively, since the strength and persistence of AMPK activation are expected to vary with chrysin-7G and each AMPK activator, this difference may be responsible for the differential effects of these AMPK activators on LDLR gene promoter activity. Among the transcription factors that activate the LDLR gene promoter, it has been reported that AMPK activation suppresses the SREBP-2 and Sp1 activity [29,30]; however, it stimulates the Egr1 and C/EBPβ activity by increasing their expression levels in hepatoma cells [20,21]. Given that chrysin-7G treatment for 24 h increased LDLR expression and activated the AMPK pathway (Figs 4A and 8A) but did not increase the levels of Egr1 and C/ EBPβ expression (Fig. 10), Egr1 and C/EBPβ are unlikely involved in the upregulation of chrysin-7Gmediated LDLR expression. Further studies are required to identify transcription factors to increase LDLR mRNA levels by chrysin-7G.  purchased from Tokyo Chemical Industry (Tokyo, Japan). Compound C (dorsomorphin) was purchased from LC Laboratories (Woburn, MA, USA). DiI-labeled LDL was purchased from Molecular Probes (Eugene, OR, USA). Chrysin, chrysin 7-O-β-D-glucopyranoside (chrysin-7G, common name: Chrysin 7-glucoside, CAS No. 31025-53-3), AICAR, A769662, U0126, and compound C were dissolved in dimethylsulfoxide (DMSO), and metformin was dissolved in water. Sweet cherries and grapes were purchased from a local supermarket.

Cell culture
Huh-7/LDLR-Luc cells were previously established [17] and maintained in medium A containing 2-μgÁmL À1 Blasticidin S. Huh-7 cells were maintained in medium A. All cell cultures were incubated at 37°C with 5% CO 2 atmosphere.

Sample preparation
Dried cherry seeds, cherry peduncles, grape skins, and grape stalks were pulverized in a food mill. Subsequently, each powder was extracted with 10 times the amount of 50% ethanol under reflux conditions. After the filtration, the solvent was removed under reduced pressure to yield a 50% ethanol extract. The extract was dissolved in H 2 O and extracted with EtOAc, and the organic layer was concentrated to yield an EtOAc residue.

Fraction assay
The EtOAc residue (30 mg) was subjected to HPLC (column: YMC-Pack ODS-AQ; 20 mm I.D. × 250 mm; YMC Co., Ltd., Kyoto, Japan) with linear gradient elution from 30% B to 80% B in 65 min (solvent A: 0.1% TFA in H 2 O; solvent B: CH 3 OH) to yield 100 fractions. Each fraction was dissolved in DMSO after drying and subsequently used for the assay.

Isolation of chrysin-7G from cherry peduncles
The EtOAc residue (60 mg) was purified by HPLC (column: YMC-Pack ODS-AQ; 20 mm I.D. × 250 mm; YMC Co., Ltd.) with a similar method as described above to yield a pale yellow powder (6.3 mg). This powder was Luciferase assays in Huh-7/LDLR-Luc cells Huh-7/LDLR-Luc cells were plated on 24-well plates at a density of 1 × 10 5 cells/well before culturing in medium A for 24 h; thereafter, these cells were incubated for 24 h, with and without indicated concentrations of extracts of agricultural products, chrysin, or chrysin-7G. The luciferase activity was measured as previously described [31]. After the addition of acetic anhydride (157 mL, 1.67 mol) at 0°C, the reaction mixture was stirred overnight at room temperature. The reaction was quenched by adding saturated sodium bicarbonate solution, and the mixture was extracted with EtOAc. The combined organic layer was washed with saturated copper(II) sulfate solution and brine successively, dried over anhydrous sodium sulfate, and concentrated in vacuo to give crude penta-O-acetyl-Dglucopyranose (2, 62.0 g) as a colorless solid, which was used in the next reaction without further purification.
To a solution of hydrazine acetate (1.42 g, 15.4 mmol) in DMF (50 mL) was added pentaacetate 2 (5.0 g, 12.8 mmol) with stirring under Ar atmosphere, and the solution was heated to 50°C for 1 h. After cooling down to room temperature, the reaction mixture was diluted with water and extracted with EtOAc. The combined organic layer was washed with water and brine successively, dried over anhydrous sodium sulfate, and concentrated in vacuo to give crude tetra-O-acetyl-D-glucopyranose (3, 3.01 g) as a colorless viscous oil, which was used in the next reaction without further purification. To a solution of tetraacetate 3 (960 mg) in CH 2 Cl 2 (55 mL) were added trichloroacetonitrile (1.38 mL, 6.66 mmol) and 1,8-diazabicyclo [5.4.0]undec-7-ene (83 μL, 546 μmol) at 0°C with stirring under Ar atmosphere. After stirring for 2 h at 0°C and 2 h at room temperature, the reaction mixture was filtered through silica gel pad. Eluent was concentrated in vacuo and the residue was subjected to silica gel column chromatography (Hex/EtOAc = 1 : 1) to give imidate 4 (1.22 g, 58% in 3 steps) as a colorless viscous oil. To a solution of the imidate 4 (1.22 g, ca. 2.47 mmol) and chrysin (147 mg, 496 μmol) in CH 2 Cl 2 (55 mL) was added trimethylsilyl trifluoromethanesulfonate (18 μL, 66 μmol) at 0°C with stirring under Ar atmosphere. The reaction mixture was stirred for 2 days and kept stirring for further 1 day with additional trimethylsilyl trifluoromethanesulfonate (18 μL, 66 μmol). The reaction was quenched by adding saturated sodium bicarbonate solution, and the separated organic layer was washed with water and brine successively, dried over anhydrous sodium sulfate, and concentrated in vacuo. The residue was subjected to silica gel column chromatography (Hex/EtOAc = 5 : 1-1 : 1) to give crude glucoside 5 (141 mg, 49%) as a colorless needle and recover unreacted chrysin (73 mg, 50%). To a solution of compound 5 (2.44 g, 4.17 mmol) in dry methanol (100 mL) was added sodium methoxide (2.55 g, 41.7 mmol) under Ar atmosphere. After stirring for 2 h at room temperature, the reaction was quenched by adding Amberlite IR120B(H)-HG, and the mixture was filtered and concentrated in vacuo. The residue was recrystallized from methanol/water (2 : 1) to give chrysin-7G (1, 623 mg, 36%) as a pale yellow crystal. 14 (m, 3H). 13

LDL uptake assays
Huh-7 cells were incubated with 10-μgÁmL À1 DiI-labeled LDL for 2.5 h. The cells were subsequently washed with phosphate-buffered saline and fixed with 4% paraformaldehyde. Intracellular fluorescent staining was visualized using a fluorescence microscope. The fluorescence intensity was quantified by IMAGE J software (Rasband, WS, National Institutes of Health, Bethesda, MD, USA).

Plasma membrane damage determination
The plasma membrane damage of Huh-7 cells was determined by LDH assay (Dojindo, Kumamoto, Japan). Huh-7 cells were plated in 96-well plates at a density of 2.0 × 10 4 cells/well and were cultured in medium A for 24 h. After incubation for another 24 h in the absence or presence of 30-μM chrysin-7G, LDH assay was performed according to the manufacturer's instructions.

Immunoblotting
Cells were lysed in RIPA buffer containing 50-mM Tris-HCl (pH 7.6), 150-mM sodium chloride, 0.1% (w/v) SDS, 0.5% (w/v) deoxycholate, 1% (v/v) Triton X-100, and protease inhibitors. Lysates were subjected to SDS/PAGE; proteins were transferred onto a PVDF membrane and were probed with the indicated antibodies. Immunoreactive proteins were visualized using Western blotting detection reagents, including ECL (Cytiva, Marlborough, MA, USA). Signals on membranes were detected and quantified using AE-6981 light capture (ATTO, Tokyo, Japan). The signal intensity was quantified by IMAGE J software.

Statistical analysis
All data were presented as mean AE standard error of the mean. Pairwise comparisons of treatments were made using the Student's t-test. Multiple comparisons were made using one-way analysis of variance followed by the Tukey's posthoc test. P < 0.05 were considered statistically significant.