Comparative Study of Phytosterol Derivatives in Monovarietal OliveOilsRaquel B. Gomez-Coca,* Gabriel D. Fernandes, Chellah del Aguila-Sanchez,María del Carmen Perez-Camino, and Wenceslao Moreda
Instituto de la Grasa-CSIC, Avenida Padre García Tejero 4, E-41012 Sevilla, Spain
ABSTRACT: Plant sterols and their derivatives are minor compounds that have been extensively studied in vegetable oils,mainly in olive oil, where they are closely related with its identity. The objective of this work is to determine the content of freeand esterified steryl glucosides and their profiles in olive oil in relation to different geographical situation of olive orchards,cultivar, farming modality, and sampling time. The orchards under study were located in the outer ring of the submetropolitanarea of Madrid (Spain), where olives from Cornicabra, Manzanilla Cacerena, Manzanilla Castellana, and Picual varieties weregrown under traditional and organic modes, and harvested in four different samplings. Conclusions state that cultivar, farmingmode, and light exposure do not have outstanding effects, whereas pedoclimate might affect the steryl glucoside presence in asubstantial way. Further studies are being carried out presently in order to confirm such statement. Also glucoside derivativeprofiles are discussed, and reasons for differences with results in previous studies pointed out.
Phytosterols, as well as their derivatives, are compounds of theunsaponifiable matter of vegetable oils, whose amount andprofile are important parameters of oil identity. In the case ofsterol derivatives, SG consist of a glucose moiety bound at the C3position of the sterol residue via an acetyl bond. If additionally afatty acid is esterified at the C6 position of the sugar, ESG areformed.The presence of free and esterified steryl glucosides in edible
matrices has been widely studied, although their synthesispathway and influencing factors are not clear yet.1 The healthpromoting effects of steryl glucosides have been unknown for along time. Their use for the treatment of pathologies related toelevated cholesterol in blood was never compared to those of FSsolutions2 or functional foods,3 and it was not until 2009 thatstudies in this field demonstrated the potential health benefits ofglucosylated phytosterols in both free and esterified (acylated)forms.4,5 Actually, SG could get reductions of around 37% incholesterol absorption in the gut of human subjects, a valuesimilar to the 30% obtained with SE.4 As a consequence, researchon possible steryl glucoside sources became more relevant andgroups started publishing detailed data on the content andcomposition of glucosylated sterols in edible plant matrices tofacilitate comparisons of the potential contribution of differentfoods as sources of steryl glucosides when included in the diet.6
Those studies did not usually include olive oil, which justifies ourstudies on the presence of both SG and ESG in this matrix.One would expect the SG and ESG composition to be closely
related to the free sterol pool, and therefore the contribution ofboth SG and ESG to be bound to the oil identity. The procedurefor isolating steryl glucosides from olive oil had been developedearlier. Initially the method was applied to a small number of oilsamples of different categories and origins,7 and then to a widernumber also from different cultivars and treated with differentrefining processes,8 reaching the conclusion that the only SG that
might be present was BSSG, whose concentration in olive oil wasnever higher than 3 mg·kg−1. Regarding ESG, mass spectrometryresults confirmed that in the majority of the samples the onlyESG present -at concentration sometimes above 8 mg·kg−1- wasEBSSG, possibly eluting just before small quantities of EAvSG.In this case we have determined the composition and content
of free and esterified steryl glucosides in a more systematic way,paying special attention to the contribution of each sterylglucoside to the total glucoside concentration. We have focusedon monovarietal olive oils from different geographical situations,farming modalities, and sampling times. Our intention has beennot only to start determining the factors affecting the sterylglucoside presence but also to contribute in a direct way tonutrient databases.
■ MATERIALS AND METHODSReagents. The internal standard ChSG, commercial mixtures of
ESG and SG, and the analytical-grade chemical reagents used in thecourse of this research were the same as those utilized previously.8
Samples and Standard Solutions. Virgin olive oils were obtainedfrom Cornicabra, Manzanilla Cacerena, Manzanilla Castellana, andPicual olive varieties. They were directly extracted in the laboratory ofDoctor Perez Jimenez group (IMIDRA) using an Abencor systemidentical to that described earlier.8 Our blank sample (chemically refinedolive-pomace oil) was obtained directly from the producers.
Stock solutions (100 μg·mL−1) and samples were treated as describedformerly,7,8 in short: solutions of ESG standard were prepared bydissolving the standard mixture in chloroform. Stocks solutions of SGand of ChSG were made with a chloroform:methanol (2:1, by volume)blend. Samples consisted of 2 g of oil dissolved in 4 mL of chloroform,spiked with 100 μL of ChSG stock solution. The hydroxyl groups were
Received: March 18, 2014Revised: May 26, 2014Accepted: May 27, 2014
derivatized (1 h at 60 °C) with 200 μL of a mixture of Py:HMDS:TMCS(9:3:1, by volume) before carrying out the GC analysis.Olive Orchards and Olive Oil Extraction. The orchards were
situated in Madrid, the center part of Spain, near the Manzanares River(Mediterranean climate), where they could be kept under optimalcultivation conditions. Specifically they were located in eight municipal-ities in the southern outer ring of the submetropolitan area (Table 1).
In these regions winters are cold due to the altitude (average of 667 mabove sea level), January being the coldest month with temperaturesbetween 3 and 10 °C, and 148 sunshine hours. These are dry climateareas; actually the common annual precipitations are between 400 and500 mm per year in AC, AG, CR, CH, and VG, and below 400 mm peryear in AZ, TI, and VS, which concentrate from October to May,October and December being the wettest months (56 mm).
Fruits from four different cultivarsCornicabra, ManzanillaCacerena, Manzanilla Castellana, and Picual cv.were obtained fromirrigated traditional orchards (drip irrigation). Additionally, olives fromManzanilla Castellana and Cornicabra cultivars were also obtained fromorganic orchards (Table 1).
The olive plantations were in a classical frame with around 300 treesper hectare. Fruits were handpicked according to the sampling timeshown in Table 1.
To ensure maximum oil quality, olives were kept in darkness at 4 °C,and oil was extracted within 24 h after harvesting using an Abencorsystem and a small-quantity mill simulating commercial oil-extractionprocedures (MC2 Ingenieria de Sistemas, Sevilla, Spain): the olives werecrushed with an Abencor hammer mill equipped with a 4 mm sieve; 700g of paste was processed using the system’s malaxer and centrifugeduring 1 min at 3500 rpm. Conditions: malaxation temperature, 27 °C;malaxation time, 20 min, plus 10 min after water addition; amount ofdistilled water added to the paste, 300 mL; amount of talc added, 20 g.The mixtures were left to decant during 24 h before proceeding to oilseparation and filtration. To facilitate the work comprehension, the oilsamples were coded according to Table 1.
Instrumentation. GC analyses of the sterol glucosides were carriedout with an Agilent 6890N gas chromatograph (Agilent Technologies,Santa Clara, California) equipped with an Agilent 7683B automaticliquid sampler and FID. Acquisition of data was done with the AgilentChemStation for GC System program. The conditions for the GC assayswere as follows: TRB-5HT column (5% diphenyl−95% dimethylpoly-
Table 1. Identification and Coding System of Studied Samples
municipality AC AnchueloAG ArgandaAZ AranjuezCR Campo RealCH ChinchonTI TielmesVG ValdelagunaVS Villarejo de Salvanes
sampling time 1 Nov 142 Nov 283 Dec 124 Dec 26
Table 2. Total Steryl Glucoside Contents (mg·kg−1) of Olive Oil from Different Cultivars, and Relative Composition RegardingCampesteryl Glucoside (CSG), Stigmasteryl Glucoside (SSG), β-Sitosteryl Glucoside (BSSG), and Δ5-Avenasteryl Glucoside(AvSG) (Part 1)a
aA minimum of two independent measures was made in each case. The standard deviation is also given; ND (nondetected). bThe use of the sameletters (a−e) indicates the absence of statistical difference between samples (ANOVA/TUKEY test).
Journal of Agricultural and Food Chemistry Article
dx.doi.org/10.1021/jf501340q | J. Agric. Food Chem. XXXX, XXX, XXX−XXXB
siloxane; 30 m × 0.32 mm i.d. × 0.10 μm film; Teknokroma, Sant Cugatdel Valles, Barcelona, Spain), 1.0 μL injection volume, hydrogen carriergas at 3 mL·min−1, and ECP cool on-column injection. The oventemperature program was as follows: 80 °C (1min), 50 °C·min−1 to 320°C (12 min), and 40 °C·min−1 to 360 °C (19 min). The detectortemperature was 360 °C.Steryl Glucoside Analysis. The SPE procedure developed
previously8 was utilized. The method consists of the conditioning ofthe cartridges (ExtraBond Si 1 g; Scharlab S.L., Barcelona, Spain) with 5mL tert-butyl methyl ether, 5 mL hexane, 5 mL chloroform, and 10 mLof a newly prepared blend consisting of chloroform:methanol 4:1, byvolume. Samples are then quantitatively transferred using 1 mLchloroform to wash the walls of the containers. The loaded cartridgesare washed with 10 mL hexane, 10 mL of tert-butyl methyl ether, and 10mL chloroform, applied one after the other. The steryl glucosides areeluted with 5 mL of the chloroform:methanol 4:1, by volume, solution.The procedure was carried out under negative pressure (1 mmHg) in avacuum manifold. The eluted fractions are evaporated to dryness undernitrogen flux, derivatized and chromatographed using cool on-columninjection.The quantitative evaluation of the SG was carried out using ChSG as
internal standard, whereas we quantified the ESG using three-pointcalibration curves (one specific curve for each ESG) built by spiking thein-house blank sample with the commercial standard mixture. Thisevaluation together with the purity, qualitative, and quantitativecomposition of the commercial standards is comprehensively describedin our former work.8
All statistical analyses were run using the GraphPad Prism software,version 5.01 for Windows. The samples were compared by an ANOVA,with statistical significance of p < 0.05. After that they were differed by aTUKEY average comparison test.
■ RESULTS AND DISCUSSION
In the present study we have analyzed 44 oil samples, each ofthem at least in triplicate, in a relatively short time (sixdeterminations/hour) and with fairly little solvent volumes, incomparison with former techniques, e.g., column chromatog-raphy.9 This has proved the suitability of the system to be appliedon a routine basis.The quantitative results of the present study are shown in
Tables 2−5. For the SG, the total amount is within the range 0.59± 0.01 to 2.62 ± 0.04 mg·kg−1, the lowest and the highestamounts being those of Manzanilla Castellana samples fromValdelaguna, MCLL_T_VG_3 and MCLL_O_VG_4, respec-tively. If we compare them with those obtained previously,8 it isclear that the SG concentration in virgin olive oils, devoid ofdefects, keeps constant within certain ranges. Consequently, wecould assume that the SG concentration in virgin olive oils will bewithin the 0.6−2.7 mg·kg−1 range.In the case of the SG profile, BSSG is always the main SG
present, contributing at least with 70.52 ± 1.18% to the total.Actually, MCLL_T_VG_3 and MCLL_T_VS_3 only showBSSG. On the other hand, CSG and AvSG have great importancein almost all samples. MCAC_T_CR_4 has the highest CSGconcentration (13.38 ± 0.03%) whereas MCLL_T_AC_4exhibits the highest amounts of AvSG (14.96 ± 0.28%). SSGwas present only in two Manzanilla Cacerena samples:MCAC_T_VS_3 (11.91 ± 1.45%) and MCAC_T_CR_4(3.29 ± 0.82%). If we compare these results with the FS profilefrom olive oil, it is clear that there is no relationship betweenthem beyond the fact that β-sitosterol derivatives are alwayspresent.
Table 3. Total Steryl Glucoside Contents (mg·kg−1) of Olive Oil from Different Cultivars, and Relative Composition RegardingCampesteryl Glucoside (CSG), Stigmasteryl Glucoside (SSG), β-Sitosteryl Glucoside (BSSG), and Δ5-Avenasteryl Glucoside(AvSG) (Part 2)a
aA minimum of two independent measures was made in each case. The standard deviation is also given; ND (nondetected). bThe use of the sameletters (a−e) indicates the absence of statistical difference between samples according to ANOVA and TUKEY test.
Journal of Agricultural and Food Chemistry Article
dx.doi.org/10.1021/jf501340q | J. Agric. Food Chem. XXXX, XXX, XXX−XXXC
ESG contents are within a range from 0.18 ± 0.01 to 3.07 ±0.07 mg·kg−1. Previous studies8 show varieties with very highESG concentrations (e.g., Uovo di Pichone, 11.29 ± 2.75mg·kg−1, and Blanqueta, 9.63 ± 0.71 mg·kg−1), which are notdiscussed here. If we compare the samples that were common toboth studies (Cornicabra and Picual), ESG concentrations arenearly the same. Compiling global results from both studies, it ispossible to establish a concentration interval from 0.2 to 14.0mg·kg−1 for ESG presence in olive oil. As far as the profiles areconcern, ECSG was never detected, while ESSG was seen only intwo samples, which where those where SSG could bedetermined. EAvSG was detected in 70% of the samples(3.23−74.82% range). Except in the case of MCLL_T_VS_3,where EBSSG was 25.18 ± 2.34% of total ESG, in all the otherssamples, EBSSG was the main ESG, showing concentrationsabove 69%. Although it has already been demonstrated that thereis no correspondence between the SG profile and that of theiresterified counterparts,6,8 this lack of any equivalence is stillstriking. Once again, it becomes clear that there is no connectionbetween both SG and ESG concentrations, and that the SG:ESGrate does not have to be lower than 1 (Tables 4 and 5). Actually itis above 1 in more than 77% of the instances. Furthermore, thereis not a correlation between global steryl glucoside quantities andtheir corresponding profiles. The fact of having high SG and/orelevated ESG concentration(s) does not imply richer profile(s),and vice versa. Actually, all four SG have been found in sampleMCAC_T_CR_4 (Figure 1), together with three other ESG,and the respective concentrations of both kinds of sterolglucoside derivatives are not significantly higher than with“poorer” sets.
The cultivars tested in the present study cannot be considereddifferent as far as the (E)SG content is concerned. However, wemay observe that the Cornicabra variety shows the highestSG:ESG relationship, being in some cases surprisingly elevated(samples COR_O_VG_4, COR_T_CH_4, and COR_O_-AZ_3, with values of 8.6, 6.1, and 4.7, respectively). On the otherhand the values are the lowest (around 1) in ManzanillaCacerena cv. Finally, Picual cv. presents a more even distributionof the SG:ESG correlation.Although correlation and cluster analysis, and other statistical
analysis (PCA and HCA) did not let us get a clear distributionpattern to correlate sample characteristics with their (E)SGcomposition and content, calculation of three times the standarddeviation allowed us to group results provisionally. Table 6shows the SG and ESG results grouped according to the origin ofthe olive fruits. The fact of having somehow acceptable relativeSD values (between 1 and 30%) when observing the SGconcentration as a function of the location of the olive tree, andnot on dependence of the cultivar, drives us to think that it mightactually be the pedoclimate (soil water, nutrients, and aeration)that has any real influence on such values and not the fruit variety,as one might think a priori. Wider studies on this aspect arepresently being carried out. On the other hand, the ESG behaviorremains quite random, supporting our previous statement thatboth kinds of phytosterols (SG and ESG) behave independentlyof each other. The same trend is observed when classifying theresults according to the sampling time (Table 7), whereas resultson the two different farming modes (traditional and organic)cannot be conclusive, mainly due to the lack of enough samplematerial.
Table 4. Total Esterified Steryl Glucoside Contents (mg·kg−1) of Olive Oil from Different Cultivars, and Relative CompositionRegarding Esterified Stigmasteryl Glucoside (ESSG), Esterified β-Sitosteryl Glucoside (EBSSG), and Esterified Δ5-AvenasterylGlucoside (EAvSG) (Part 1)a
aA minimum of two independent measures was made in each case. The standard deviation is also given; ND (nondetected). bThe use of the sameletters (a−e) indicates the absence of statistical difference between samples according to ANOVA and TUKEY test.
Journal of Agricultural and Food Chemistry Article
dx.doi.org/10.1021/jf501340q | J. Agric. Food Chem. XXXX, XXX, XXX−XXXD
To summarize we will conclude as follows: The performanceof more than one hundred analyses in a relatively short time andwith no waste of solvent has demonstrated the suitability of theprocedure to be applied on a routine basis.The presence of steryl glucosides in olive oil is not just limited
to BSSG and EBSSG as thought previously, but also AvSG,
EAvSG, and CSG have been found in an extensive number ofsamples. This might also be symptomatic either of differentsample treatment or of differences during the oil extractionprocess. Actually it seems that industrial samples,8 which havebeen obtained using a two or three phase decanter (higher wateraddition), have poorer (E)SG profiles than those samples
Table 5. Total Esterified Steryl Glucoside Contents (mg·kg−1) of Olive Oil from Different Cultivars, and Relative CompositionRegarding Esterified Stigmasteryl Glucoside (ESSG), Esterified β-Sitosteryl Glucoside (EBSSG), and Esterified Δ5-AvenasterylGlucoside (EAvSG) (Part 2)a
aA minimum of two independent measures was made in each case. The standard deviation is also given; ND (nondetected). bThe use of the sameletters (a−e) indicates the absence of statistical difference between samples according to ANOVA and TUKEY test.
Figure 1. GC-FID chromatogram for steryl glucosides in virgin olive oil extracted from olive fruits of the Manzanilla Cacerena cultivar (sampleMCAC_T_CR_4). This chromatogram has been obtained after analyzing the sample according to the proposed method and it shows the following: 1,the internal standard cholesterol glucoside; 2, campesteryl glucoside; 3, stigmasteryl glucoside; 4, β-sitosteryl glucoside; 5, (presumptive) Δ5-avenasteryl glucoside; 6, esterified stigmasteryl glucoside; 7, esterified β-sitosteryl glucoside; 8, (presumptive) esterified Δ5-avenasteryl glucoside.
Journal of Agricultural and Food Chemistry Article
dx.doi.org/10.1021/jf501340q | J. Agric. Food Chem. XXXX, XXX, XXX−XXXE
obtained with the Abencor system, where water addition hasbeen scarce. This reasoning seems to be supported by the fact offinding certain molecules (e.g., CSG) in Abencor samples, andnot finding them in industrial oils. In any case a larger number ofdeterminations under controlled circumstances must be carriedout before drawing definite conclusions.Olive cultivar, sampling time, and farming mode have no
influence on the parameters under study. Pedoclimate (soilaeration, micronutrient presence, and water content) seems tohave some influence on the glucoside presence. In this wayfurther studies are being carried out to confirm such arelationship.
■ ACKNOWLEDGMENTSThe authors would like to thank Ms. Diana Gomez Castillo forher assistance in the laboratory and Doctor Marıa de los AngelesPerez Jimemez and Ms. Carmen Mena Uriarte from IMIDRA(InstitutoMadrileno de Investigacion y Desarrollo Rural, Agrarioy Alimentario) for providing the varietal olive oil samples.
■ REFERENCES(1) Grille, S.; Zaslawski, A.; Thiele, S.; Plat, J.; Warnecke, D. Thefunctions of steryl glucosides come to those who wait: recent advancesin plants, fungi, bacteria and animals. Prog. Lipid Res. 2010, 49, 262−268.(2) Pollak, O. J. Reduction of blood cholesterol in man. Circulation1953, 7, 702−706.(3) MacKay, D. S.; Jones, P. J. H. Phytosterols in human nutrition:type, formulation, delivery, and physiological function. Eur. J. Lipid Sci.Technol. 2011, 113, 1427−1432.(4) Lin, X.; Ma, L.; Racette, S. B.; Anderson Spearie, C. L.; Ostlund, R.E. J. Phytosterol glycosides reduce cholesterol absorption in humans.Am. J. Physiol. 2009, 296, G931−G935.(5) Lin, X.; Ma, L.; Moreau, R. A.; Ostlund, R. E., Jr. Glycosidic bondcleavage is not required for phytosteryl glycoside-induced reduction ofcholesterol absorption in mice. Lipids 2011, 46, 701−708.(6) Nystrom, L.; Schar, A.; Lampi, A.-M. Steryl glycosides in plantfoods reflect unique sterol patterns. Eur. J. Lipid Sci. Technol. 2012, 114,656−669.(7) Gomez-Coca, R. B.; Perez-Camino, M. C.; Moreda, W. Specificprocedure for analysing steryl glucosides in olive oil. Eur. J. Lipid Sci.Technol. 2012, 114, 1417−1426.(8) Gomez-Coca, R. B.; Perez-Camino, M. C.; Moreda, W. On theglucoside analysis: simultaneous determination of free and esterifiedsteryl glucosides in olive oil. Detailed analysis of standards ascompulsory first step. Food Chem. 2013, 15, 1273−1280.(9) Lacoste, F.; Dejean, F.; Griffon, H.; Rouquette, C. Quantification offree and esterified steryl glucosides in vegetable oils and biodiesel. Eur. J.Lipid Sci. Technol. 2009, 111, 822−828.
Table 6. SG and ESGContents of the Olive Oil Samples underStudy Classified According to the Municipality of Origina
