When cholesterol is not cholesterol: a note on the enzymatic determination of its concentration in model systems containing vegetable extracts
© Jové et al; licensee BioMed Central Ltd. 2010
Received: 10 May 2010
Accepted: 21 June 2010
Published: 21 June 2010
Experimental evidences demonstrate that vegetable derived extracts inhibit cholesterol absorption in the gastrointestinal tract. To further explore the mechanisms behind, we modeled duodenal contents with several vegetable extracts.
By employing a widely used cholesterol quantification method based on a cholesterol oxidase-peroxidase coupled reaction we analyzed the effects on cholesterol partition. Evidenced interferences were analyzed by studying specific and unspecific inhibitors of cholesterol oxidase-peroxidase coupled reaction. Cholesterol was also quantified by LC/MS. We found a significant interference of diverse (cocoa and tea-derived) extracts over this method. The interference was strongly dependent on model matrix: while as in phosphate buffered saline, the development of unspecific fluorescence was inhibitable by catalase (but not by heat denaturation), suggesting vegetable extract derived H2O2 production, in bile-containing model systems, this interference also comprised cholesterol-oxidase inhibition. Several strategies, such as cholesterol standard addition and use of suitable blanks containing vegetable extracts were tested. When those failed, the use of a mass-spectrometry based chromatographic assay allowed quantification of cholesterol in models of duodenal contents in the presence of vegetable extracts.
We propose that the use of cholesterol-oxidase and/or peroxidase based systems for cholesterol analyses in foodstuffs should be accurately monitored, as important interferences in all the components of the enzymatic chain were evident. The use of adequate controls, standard addition and finally, chromatographic analyses solve these issues.
Cholesterol function is essential for membrane physiology, bile acids and steroid hormones biosynthesis. However, an elevated level of cholesterol in plasma is implicated in atherosclerosis and other cardiovascular diseases [1, 2]. Therefore, minimizing dietary cholesterol intake is often recommended as a primary measure for lowering cholesterolemia . In the intestinal tract, dietary lipids are first emulsified in the lumen by bile components (biliary salts and phospholipids) and then encapsulated into micelles. Cholesterol can be then transferred from micelles to gut wall and thereafter to bloodstream . Epidemiological and experimental evidence demonstrate that the consume of vegetable foods allows to a lowering effect on cholesterol plasma levels and diminished risk of atherosclerosis progression [5, 6]. It is known that cholesterol esters, phenol compounds and other vegetable derived nutrients can block the entry of most cholesterol into micelles, partially preventing its absorption . While developing a model of "in vitro" digestion based on published methods  we analyzed mixtures of cholesterol with selected foodstuffs and bile, we detected strong interferences in a widely used method for cholesterol quantification arising from different vegetable foods like cocoa and/or green tea.
This study characterized those potential interferences and presents different solutions to solve them.
Vegetable extracts induce an apparent increase in cholesterol content in "in vitro" experiments using an enzymatic method
Phytoesterol interferences were ruled out as ergosterol and other sterols (data not shown) did not offer fluorescence in the complete system. Metal chelation (EDTA-DTPAC) did not inhibit significantly fluorescence in the complete system in the presence of vegetable extracts except in the case of tea extract (figure 2C). The potential contribution of vegetable derived peroxidases was ruled out by heat denaturation of vegetable extracts, even resulting in the increase of fluorescence (figure 2C). Finally, the inhibitory activity of catalase [126.96.36.199] (leading to fluorescence decreases to 50%) suggests that there is a vegetable extract dependent H2O2 production which may be derived by the previously described interaction of sample antioxidants with horseradish peroxidase . Moreover, in the presence of vegetable extracts, the fluorescence in both cholesterol-dependent and independent reactions offered similar values suggesting the possibility of using the cholesterol independent fluorescence as a blank to account the vegetable extract derived interferences (figures 2A and 2B).
To test this alternative solution, cholesterol standard curves in the presence of vegetable derived extracts were developed (figure 2D and 2E). The slope of the cholesterol concentration-fluorescence linear relationship was inversely related to the concentration of vegetable extract, in a given range (up to 10 mg/ml in cocoa extract). Therefore, it is advisable to develop a cholesterol standard curve with the chosen concentration of vegetable extract in order to ascertain linearity, even in the presence of the cholesterol independent fluorescence blank. It should be noted that in specific cases the interference of the vegetable extracts precluded the use of cholesterol standard curve even at a lower doses (from 1 mg/ml in tea).
Vegetable extracts induce an apparent decrease in cholesterol incorporation in bile micelles. Chromatographic approach for cholesterol analysis
In order to further characterize those interferences in duodenal-like conditions, we used a chromatographic approach. Figure 3E shows that the cholesterol concentration does not change with the addition of cholesterol esterase, probably due to the virtual absence of cholesteryl esters in porcine bile. The addition of cholesterol oxidase in the mixture resulted in a strong decrease in the cholesterol detection because most of the cholesterol was oxidized (Figure 3E). However, the addition of vegetable derived extracts caused an increase in cholesterol content in those conditions, suggesting an inhibitory activity of those extracts over cholesterol oxidase in the presence of bile.
There are several methods described to measure cholesterol content in foodstuff using chromatographic approaches [13, 14]. However, when it is necessary to analyze a high number of samples, an enzymatic method based on the cholesterol oxidase-peroxidase coupled reaction could be the best option [10–12] if we take into account some interferences reported here.
With the objective of modeling the capability of vegetable extracts decreasing the bioavailable micellar cholesterol, we found different interferences in the measurement of cholesterol using an enzymatic method. In phosphate buffered saline, we evidenced a highly intense cholesterol independent fluorescence that we could attribute partially (50%) to an exogenous H2O2 production, as in those conditions it was diminished by catalase preincubation. The rest of the interference could be explained by a different mechanism recently described in our group , showing an interaction of vegetable antioxidants with peroxidases. Horseradish peroxidase is often used as final step for enzymatic-coupled reactions. Briefly, in the absence of hydrogen peroxide but in aerobic conditions, the antioxidant compound of the vegetable extracts could reduce the ferric-horseradish peroxidase to ferrous-horseradish peroxidase and then interact with the O2 of the medium producing horseradish peroxidase-Compound III. This compound can undergo spontaneous decay to ferriperoxidase with the generation of O2- which may interact with the antioxidant producing an antioxidant radical. This antioxidant radical may then react with the Amplex Red® and horseradish peroxidase and produce resorufin. Thus, all peroxidase based enzymatic methods may show interferences by those vegetable derived compounds. To solve these interferences we propose two different and complementary methods: i) the use of cholesterol independent condition as a blank of the reaction to eliminate the fluorescence when the vegetable extracts are diluted in PBS and ii) the application of cholesterol standard curves including the working concentration of the extract.
For the studies of "in vitro" cholesterol absorption we used porcine bile in order to reproduce a physiologically relevant and a stable source of micellar cholesterol. We first analyze the interference of the extracts diluted in bile and we found a novel interference that decreases the apparent cholesterol concentration. In this case, the use of the cholesterol independent fluorescence is not as useful as in the case of buffered saline-based systems because in this latter case the interference is lower. However, the cholesterol and/or cholesterol ester standard curve addition with the extracts could be a good method to reduce the interferences detecting cholesterol at low concentrations of vegetable extracts. After incubating the extracts in the presence or in the absence of the enzymes cholesterol esterase and cholesterol oxidase we discover that vegetable extracts can interact directly and/or indirectly decrease the activity of cholesterol oxidase. We cannot exclude that other vegetable derived extracts could inhibits cholesterol esterase. The mechanisms behind this inhibitory effect are outside the scope of this work, but they may comprise displacement of cholesterol oxidase outside of lipid bilayers needed for efficient catalytic turnover  and they may be part of the defensive properties of polyphenols presents in vegetable extracts . In any case, cholesterol oxidase is among the more frequently used enzymes and its interference should be accounted when extending the use from clinical chemistry to food chemistry .
Finally, the development of a chromatographic method in order to measure the cholesterol concentration in biological models of food digestion is presented as a solution when the used concentration of foodstuff is high enough to interfere with the standard addition method. Both detection methods were used to test the effect of these extracts in lowering the concentration of potentially bioavailable cholesterol. Moreover, it is described that sterol, polyphenols and other nutrients can block the entry of most cholesterol into micelles [8, 18–20]. Although there were evidenced a blocking effect of cholesterol entry into micellar phases by both extracts using both methods the magnitude of the change differed. Using the enzymatic method we demostrated that the cholesterol independent fluorescence in the case of tea extract was very high, quite similar to the fluorescence of complete system. This high fluorescence resulted in a small detection of real cholesterol dependent fluorescence. In the case of cocoa extract the decrease in apparent bioavailability is about 20%. Analyzing the cholesterol concentration using a chromatographic method both extracts at tested concentrations decreased micellar cholesterol about a 50%.
The use of cholesterol-oxidase and/or peroxidase based systems for cholesterol analyses in foodstuffs should be accurately monitored, as important interferences in all the components of the enzymatic chain were evident. The use of adequate controls, standard addition and finally, chromatographic analyses should solve these issues.
Ethylenediaminetetraacetic acid (EDTA), diethylenetriaminepentaacetic acid (DTPAC), ergosterol, catalase, chloroform and cholesterol were from Sigma (Sigma-Aldrich, Saint Louis, MO, USA). Acetonitrile, 2-propanol, ammonium acetate and formic acid were from Baker (Mallinckrodt Baker, Phillipsburg, NJ, USA). Millex GP filters 0,22 μm and Ultrafree-MC filtres 30,000 from Millipore (Millipore, Billerica, MA, USA), methanol from Carlo Erba (Carlo Erba, Milano, Italy) and [25,26,26,26,26,27,27,27-2H7]cholesterol (cholesterol-D7) from Avanti Polar Lipids (Avanti Polar Lipids Inc, Alabaster, AL, USA).
Bile from porcine biliary vesicle was collected from a local abattoir and placed on ice. Immediately after collection, bile was centrifuged at 2000 g at 4°C for 10 min to remove debris. The cholesterol content of the bile was measured (see below) and then the bile was diluted with phosphate buffered saline to obtain a 0.5 mM of cholesterol concentration, as described by Kirana et al . It was aliquoted and stored at -80°C.
Green tea and cocoa vegetable extracts were provided by La Morella Nuts SA (La Morella Nuts, Reus, Spain).
Cholesterol measurement using an enzymatic method
The concentration of cholesterol in porcine bile, food extracts and mixes was analyzed by Amplex Red Cholesterol Assay Kit (A12216) based on an enzymatic reaction depicted in figure 1. Briefly, reactions took place in a 96 well plate by the addition of 50 μL of Amplex Red working solution with 50 μL of assay sample. Five mL of working solution, prepared prior the analysis, contained 75 μL of a 300 μM of Amplex Red reagent and 2 U/mL of HRP. The working solution volume was adjusted to 5 mL with reaction buffer, which contained 25 mM potassium phosphate, pH 7.4, 12.5 mM NaCl, 1.25 mM cholic acid and 0.025% Triton X-100. The reactions were incubated for 30 min at 37°C, protected from light. After incubation, fluorescence was measured in a fluorescence microplate reader (Tecan Infinite M200, Männedorf, Switzerland) using excitation wavelength at 560 nm and emission detection at 590 nm.
The samples were suspended with either phosphate buffered saline or porcine bile to obtain the initial desired concentration of nutrient (from 0 to 20 mg/ml). After a dilution in PBS (1:25) cholesterol amount was quantified by fluorescence according the kit instructions.
To model the cholesterol entry into micellar phases and its potential inhibition by vegetable extracts, the method of Kirana et al was used . Briefly, samples suspended in pig's bile at 10 mg/ml were incubated at 37°C for 1 h with continuous shaking at 160 r.p.m. (Unitron, Infors HT, Headquarter, Switzerland). The solution was then centrifuged at 1000 × g for 10 min, filtered through a 0.22 μm Millex GP and diluted 25 fold. The concentration of cholesterol was analyzed by Amplex Red Cholesterol Assay Kit (A12216).
Cholesterol analysis by liquid chromatography coupled to mass-spectrometry
To unequivocally quantify cholesterol in food extract-bile mixtures, cholesterol was extracted using chloroform:methanol (2:1) as described previously [21, 22]. Previous to extraction cholesterol-D7 was added, as an internal standard, to a final concentration of 125 μg/mL. The mixture was vortexed and centrifuged at 4400 × g at room temperature for 15 min. The organic phase was conserved and the procedure was repeated. Combined organic phases were dried in a SpeedVac (Thermo Fisher Scientific, Madrid, Spain) and dissolved with methanol. The samples were filtered in an UltraFree 5 kDa filter (Millipore, Billerica, MA, USA) before liquid chromatography analysis.
Liquid chromatography was done in an Agilent LC model G2226A coupled to an ESI-QTOF MS 6520 (Agilent Technologies, Barcelona, Spain). For this purpose, extracted sample was applied onto a reverse-phase column (C18 Luna 3 micron pfp(2) 100 A 150 × 2 mm, Phenomenex, California, USA), equipped with a guard-column kept at 50°C. The flow rate was 100 μl/min with solvent A composed of water containing 1% ammonium acetate 1 M, 0.1% formic acid, and solvent B composed of acetonitrile:2-propanol (5:2; v/v) containing 1% ammonium acetate 1 M, 0.1% formic acid. The gradient started from 65% A/35% B, reached 100% B in 35 min, held there for the next 10 min and re-equilibrated for 15 min . The temperature of the sample organizer was set at 10°C.
Data were collected in positive electrospray mode TOF operated in full-scan mode at 100 to 3000 m/z. The capillary voltage was 3500 V with a scan rate of 1 scan/s. N2 was used as a gas nebulizer (Flow: 5 L/min and T = 350°C). The Masshunter Software was used for integration and extraction of peak intensities of the cholesterol and cholesterol-D7. The m/z values used for quantification were: m/z 369.35 [M+H-H2O]+ for cholesterol and m/z 376.3955 [M+H-H2O]+ for cholesterol-D7. Interassay and intraassay variation < 5%, L.O.Q was 20 pmol, RSQ = 0.98.
We thank the support of Saray Arenas-Soria, Laura Arcal, David Argilès and Meritxell Martín for excellent technical assistance. Supported by I+D grants from the Spanish Ministry of Education and Science (BFU2009-11879/BFI; AGL2006-12433), the Generalitat of Catalunya (2009SGR-735) and the Spanish Ministry of Health (FIS 08-1843, RD06/0013/0012). Supported by the CENIT program from the Spanish Ministry of Industry and by a consortium of companies led by La Morella Nuts (Reus, Catalonia, Spain) with the following companies: KRAFT, BTSA (Biotecnologías Aplicadas), Selecció Batallé, Industrial Técnica Pecuaria, Neuron BioPharma, Shirota Functional Foods, Grupo Leche Pascual, Innaves. Supported also by the COST B-35 Action.
- Poli G, Sottero B, Gargiulo S, Leonarduzzi G: Cholesterol oxidation products in the vascular remodeling due to atherosclerosis. Mol Aspects Med. 2009, 30 (3): 180-189. 10.1016/j.mam.2009.02.003View ArticlePubMedGoogle Scholar
- Temel RE, Rudel LL: Diet effects on atherosclerosis in mice. Curr Drug Targets. 2007, 8 (11): 1150-1160. 10.2174/138945007782403847View ArticlePubMedGoogle Scholar
- Brown MS, Goldstein JL: Cholesterol feedback: from Schoenheimer's bottle to Scap's MELADL. J Lipid Res. 2009, 50 (Suppl): S15-27. 10.1194/jlr.R800054-JLR200PubMed CentralPubMedGoogle Scholar
- Lu K, Lee MH, Patel SB: Dietary cholesterol absorption; more than just bile. Trends Endocrinol Metab. 2001, 12 (7): 314-320. 10.1016/S1043-2760(01)00433-7View ArticlePubMedGoogle Scholar
- Nagarajan S, Burris RL, Stewart BW, Wilkerson JE, Badger TM: Dietary soy protein isolate ameliorates atherosclerotic lesions in apolipoprotein E-deficient mice potentially by inhibiting monocyte chemoattractant protein-1 expression. J Nutr. 2008, 138 (2): 332-337.PubMedGoogle Scholar
- Sutra T, Decorde K, Riss J, Dallas C, Cristol JP, Rouanet JM: A commercial extract of fruits and vegetables, Oxxynea, acts as a powerful antiatherosclerotic supplement in an animal model by reducing cholesterolemia, oxidative stress, and NADPH oxidase expression. J Agric Food Chem. 2007, 55 (10): 4258-4263. 10.1021/jf070029nView ArticlePubMedGoogle Scholar
- Ostlund RE: Phytosterols in human nutrition. Annu Rev Nutr. 2002, 22: 533-549. 10.1146/annurev.nutr.22.020702.075220View ArticlePubMedGoogle Scholar
- Kirana C, Rogers PF, Bennett LE, Abeywardena MY, Patten GS: Naturally derived micelles for rapid in vitro screening of potential cholesterol-lowering bioactives. J Agric Food Chem. 2005, 53 (11): 4623-4627. 10.1021/jf050447xView ArticlePubMedGoogle Scholar
- Serrano J, Jove M, Boada J, Bellmunt MJ, Pamplona R, Portero-Otin M: Dietary antioxidants interfere with Amplex Red-coupled-fluorescence assays. Biochem Biophys Res Commun. 2009, 388 (2): 443-449. 10.1016/j.bbrc.2009.08.041View ArticlePubMedGoogle Scholar
- Amundson DM, Zhou M: Fluorometric method for the enzymatic determination of cholesterol. J Biochem Biophys Methods. 1999, 38 (1): 43-52. 10.1016/S0165-022X(98)00036-0View ArticlePubMedGoogle Scholar
- Mizoguchi T, Edano T, Koshi T: A method of direct measurement for the enzymatic determination of cholesteryl esters. J Lipid Res. 2004, 45 (2): 396-401. 10.1194/jlr.D300024-JLR200View ArticlePubMedGoogle Scholar
- Situmorang M, Alexander PW, Hibbert DB: Flow injection potentiometry for enzymatic assay of cholesterol with a tungsten electrode sensor. Talanta. 1999, 49 (3): 639-649. 10.1016/S0039-9140(99)00057-0View ArticlePubMedGoogle Scholar
- Lin YT, Wu SS, Wu HL: Highly sensitive analysis of cholesterol and sitosterol in foods and human biosamples by liquid chromatography with fluorescence detection. J Chromatogr A. 2007, 1156 (1-2): 280-287. 10.1016/j.chroma.2007.01.091View ArticlePubMedGoogle Scholar
- Daneshfar A, Khezeli T, Lotfi HJ: Determination of cholesterol in food samples using dispersive liquid-liquid microextraction followed by HPLC-UV. J Chromatogr B Analyt Technol Biomed Life Sci. 2009, 877 (4): 456-460. 10.1016/j.jchromb.2008.12.050View ArticlePubMedGoogle Scholar
- Kreit J, Sampson NS: Cholesterol oxidase: physiological functions. FEBS J. 2009, 276 (23): 6844-6856. 10.1111/j.1742-4658.2009.07378.xPubMed CentralView ArticlePubMedGoogle Scholar
- Si W, Gong J, Tsao R, Kalab M, Yang R, Yin Y: Bioassay-guided purification and identification of antimicrobial components in Chinese green tea extract. J Chromatogr A. 2006, 1125 (2): 204-210. 10.1016/j.chroma.2006.05.061View ArticlePubMedGoogle Scholar
- Pollegioni L, Piubelli L, Molla G: Cholesterol oxidase: biotechnological applications. FEBS J. 2009, 276 (23): 6857-6870. 10.1111/j.1742-4658.2009.07379.xView ArticlePubMedGoogle Scholar
- Yasuda A, Natsume M, Sasaki K, Baba S, Nakamura Y, Kanegae M, Nagaoka S: Cacao procyanidins reduce plasma cholesterol and increase fecal steroid excretion in rats fed a high-cholesterol diet. Biofactors. 2008, 33 (3): 211-223. 10.1002/biof.5520330307View ArticlePubMedGoogle Scholar
- Shibata S, Oda K, Onodera-Masuoka N, Matsubara S, Kikuchi-Hayakawa H, Ishikawa F, Iwabuchi A, Sansawa H: Hypocholesterolemic effect of indigestible fraction of Chlorella regularis in cholesterol-fed rats. J Nutr Sci Vitaminol (Tokyo). 2001, 47 (6): 373-377.View ArticleGoogle Scholar
- Osada K, Suzuki T, Kawakami Y, Senda M, Kasai A, Sami M, Ohta Y, Kanda T, Ikeda M: Dose-dependent hypocholesterolemic actions of dietary apple polyphenol in rats fed cholesterol. Lipids. 2006, 41 (2): 133-139. 10.1007/s11745-006-5081-yView ArticlePubMedGoogle Scholar
- Ostrow JD, Murphy NH: Isolation and properties of conjugated bilirubin from bile. Biochem J. 1970, 120 (2): 311-327.PubMed CentralView ArticlePubMedGoogle Scholar
- Stark M, Jornvall H, Johansson J: Isolation and characterization of hydrophobic polypeptides in human bile. Eur J Biochem. 1999, 266 (1): 209-214. 10.1046/j.1432-1327.1999.00845.xView ArticlePubMedGoogle Scholar
- Laaksonen R, Katajamaa M, Paiva H, Sysi-Aho M, Saarinen L, Junni P, Lutjohann D, Smet J, Van Coster R, Seppanen-Laakso T, Lehtimaki T, Soini J, Oresic M: A systems biology strategy reveals biological pathways and plasma biomarker candidates for potentially toxic statin-induced changes in muscle. PLoS One. 2006, 1: e97- 10.1371/journal.pone.0000097PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.