RESEARCH ARTICLE www.mnf-journal.com Virgin Olive Oil Phenolic Compounds Modulate the HDL Lipidome in Hypercholesterolaemic Subjects: A Lipidomic Analysis of the VOHF Study Sara Fernández-Castillejo, Anna Pedret,* Úrsula Catalán, Rosa-Maria Valls, Marta Farràs, Laura Rubió, Olga Castañer, Alba Macià, Montse Fitó, Maria José Motilva, Maria-Isabel Covas, Martin Giera, Alan T. Remaley, and Rosa Solà 1. Introduction Scope: The lipidomic analysis of high-density lipoprotein (HDL) could be useful to identify new biomarkers of HDL function. Based on multiple intervention and gene association studies, high-density lipopro- Methods and results: A randomized, controlled, double-blind, crossover trial tein (HDL) cholesterol (HDL-C) is now (33 hypercholesterolaemic subjects) is performed with a control virgin olive oil recognized as an imperfect measure of (VOO), VOO enriched with its own phenolic compounds (FVOO), or VOO cardiovascular disease (CVD) risk, sug- enriched with additional phenolic compounds from thyme (FVOOT) for 3 gesting that emerging HDL-related pa- weeks. HDL lipidomic analyses are performed using the Lipidyzer platform. rameters may be useful for the clinical management of patients with CVD.[1] In VOO and FVOO intake increase monounsaturated-fatty acids (FAs) and particular, HDL functional quality seems decrease saturated and polyunsaturated FAs in triacylglyceride (TAG) species, to be a more important biomarker than among others species. In contrast, FVOOT intake does not induce these FAs HDL-C levels.[2] Therefore, the identifica- changes. The decrease in TAG52:3(FA16:0) after VOO intake and the decrease tion and quantification of additional HDL in TAG52:5(FA18:2) after FVOO intake are inversely associated with changes functionality biomarkers are vital to un- in HDL resistance to oxidation. After FVOO intake, the decrease in ravel the role of HDL in CVD. The constant metabolism and re- TAG54:6(FA18:2) in HDL is inversely associated with changes in HDL modeling of HDL particles results in cholesterol efflux capacity. circulating HDL particles that differ in Conclusion: VOO and FVOO consumption has an impact on the HDL functionality and composition irrespec- lipidome, in particular TAG species. Although TAGs are minor components of tive of cholesterol content.[2] One of the HDL mass, the observed changes in TAG modulated HDL functionality main functions of HDL appears to be its cholesterol efflux (ChE) capacity. ChE towards a cardioprotective mode. The assessment of the HDL lipidome is a from macrophages to HDL is inversely valuable approach to identify and characterize new biomarkers of HDL associated with prevalent coronary artery function. disease,[3] type 2 diabetes mellitus,[4] and incident atherosclerotic CVD.[5] S. Fernández-Castillejo, A. Pedret, Ú. Catalán S. Fernández-Castillejo, A. Pedret, Ú. Catalán, R.-M. Valls, R. Solà Eurecat, Centre Tecnològic de Catalunya Facultat de Medicina i Ciències de la Salut, Departament de Medicina i Unitat de Nutrició i Salut Cirurgia, Universitat Rovira i Virgili, Grup Nutrició Funcional, Oxidació i Reus 43204, Spain Malalties Cardiovasculars (NFOC-Salut) M. Farràs Reus 43201, Spain Institut de Recerca de l’Hospital Santa Creu i Sant Pau-Institut E-mail: anna.pedret@urv.cat d’Investigacions Biomèdiques (IIB) Sant Pau Barcelona 08041, Spain M. Farràs The ORCID identification number(s) for the author(s) of this article CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM) can be found under https://doi.org/10.1002/mnfr.202001192 Instituto de Salud Carlos III Madrid 28029, Spain © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH. This is an open access article under the terms of the L. Rubió, A. Macià Creative Commons Attribution License, which permits use, distribution Food Technology Department, XaRTA-TPV, Agrotecnio Center, Escola and reproduction in any medium, provided the original work is properly Tècnica Superior d’Enginyeria Agrària cited. University of Lleida. Avda/ Alcalde Rovira Roure 191 Catalonia, Lleida 25198, Spain DOI: 10.1002/mnfr.202001192 Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (1 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com Therefore, ChE may be a useful biomarker for cardiovascular risk when compared to those identified in the circulation.[8] There- assessment. In addition, HDL resistance to oxidation has been fore, lipidomics is a useful tool for understanding HDL lipid related to its ChE capacity[6] and HDL antioxidative activity[7] due metabolism and possibly for the discovery of novel lipid metabo- to the activity of the paraoxonase 1 enzyme. lites associated with CVD, serving as potential new biomarkers Lipidomics is an emerging “omic” science that allows the study for HDL functionality assessment.[9,12] of several lipid classes and species, in addition to the classi- The HDL lipidome is also significantly affected by nutritional cal cholesterol and triacylglyceride (TAG) levels determined rou- factors. Dietary fats from different food sources modify the fatty tinely in clinical laboratories.[8] The study of the HDL lipidome acid (FA) composition of the HDL surface due to the different FA may be a valuable approach for identifying new biomarkers of compositions of dietary fats, especially the high content of mo- HDL functionality beyond those traditionally used,[9] because the nounsaturated fatty acids (MUFAs) in olive oil (OO).[17] Virgin lipid content of HDL is estimated to be approximately 35–65% olive oil (VOO) contains not only MUFAs but also minor compo- by weight depending on the HDL subclass.[10] Its lipid content is nents, including phenolic compounds, which represent 1–2% of mainly located on the amphipathic surface of the HDL particle, the total content of VOO.[18] Even though phenolic compounds forming a monolayer of phospholipids (35–50% of the total HDL account for a small amount of VOO weight, they are responsi- lipid mass). This monolayer also contains minor sphingolipids ble for the cardioprotective benefits of VOO.[18–21] In particular, (5–7%), along with free cholesterol and other sterols (5–10%). our group has previously reported that the sustained intake of The hydrophobic core of HDL is mostly composed of cholesteryl functional VOOs enriched with phenolic compounds modifies esters (CE; 35–40%) and TAG (2–3%).[2,10,11] HDL function and composition, specifically the HDL proteome, The HDL lipidome is altered in several medical towards a cardioprotective mode.[22,23] However, to the best of our conditions,[12,13] such as metabolic syndrome,[14,15] coronary knowledge, no studies have aimed to discern the effects of VOO artery disease, and type 2 diabetes mellitus.[16] Different lipid phenolic compounds on the HDL lipidome. profiles have also been described in atherosclerotic lesions The main aim of this study is to assess the effects of the sus- tained intake of VOO and two different functional VOOs, en- riched with its own phenolic compounds (mainly secoiridoids) O. Castañer, M. Fitó, M.-I. Covas or with its own phenolic compounds plus complementary com- Cardiovascular Risk and Nutrition Research Group Hospital del Mar Medical Research Institute (IMIM) pounds from thyme (mainly flavonoids) on the HDL lipidome. Barcelona 08003, Spain Moreover, we also investigated whether such changes in the HDL O. Castañer, M. Fitó lipidome are associated with changes in HDL functionality, in PhD Program in Biomedicine particular HDL ChE capacity and its resistance to oxidation, as a Universitat Pompeu Fabra surrogate of HDL functionality. Barcelona 08005, Spain O. Castañer, M. Fitó Consorcio CIBER, M.P. Fisiopatología de la Obesidad y Nutrición 2. Results (CIBEROBN) Instituto de Salud Carlos III 2.1. Characteristics of the Study Participants and Dietary Madrid 28029, Spain Adherence M. J. Motilva Instituto de Ciencias de la Vid y del Vino-ICVV (Consejo Superior de Investigaciones Científicas-CSIC In the VOHF study, 62 participants were assessed for eligibil- Universidad de La Rioja, Gobierno de La Rioja), Finca La Grajera, Ctra. de ity and 33 were randomized and therefore allocated into one of Burgos Km. 6 (LO-20 - salida 13) the three sequences of intervention (Figure S2, Supporting Infor- Logroño (La Rioja) 26007, Spain mation). The baseline characteristics of the participants are de- M.-I. Covas scribed in Table S2, Supporting Information.[22,24] As previously NUPROAS Handelsbolag (NUPROAS HB) Nacka, Sweden published, participant adherence was good and no changes were M. Giera reported in the main nutrients and medication intake throughout Center for Proteomics and Metabolomics the study.[22,25] Leiden University Medical Center Albinusdreef 2, Leiden 2333ZA, The Netherlands 2.2. HDL Lipid and Protein Characterization A. T. Remaley Department of Laboratory Medicine Clinical Center National Institutes of Health The HDL isolated for this study was shown to be pure and to Bethesda, MD 20814, USA contain no remnants from other lipoproteins, as shown by the A. T. Remaley ApoB100 and albumin levels in Table S3, Supporting Informa- Lipoprotein Metabolism Section Cardio-Pulmonary Branch National tion. Moreover, some general changes in HDL composition were Heart Lung and Blood Institute National Institutes of Health observed after the three interventions (Table S4, Supporting In- Bethesda, MD 20814, USA formation) as previously published by our group.[23] R. Solà Institut d’Investigació Sanitaria Pere Virgili (IISPV) Reus 43204, Spain 2.3. Multivariate Analysis of the HDL Lipidome R. Solà Hospital Universitari Sant Joan de Reus Samples were randomly analyzed in three different batches run Reus 43204, Spain on 3 consecutive days. The multivariate principal component Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (2 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com analysis performed in every set of analyses discarded any batch ef- Table 1. HDL lipids significantly modified after VOO intervention versus fect (data not shown). Quality controls were included in the anal- its baseline. ysis every 10 samples. The analysis of such controls showed intra- assay coefficients of variation <12% and inter-assay coefficients Lipid Log fold p-value p-value change adjusted of variation <15%. by FDR A total of 13 lipid classes, 185 LC-FA combinations, and 792 lipid species were identified in HDL, as detailed in Table S5, Sup- LC-FA combination composition porting Information. TAG(FA18:1) 0.0940763 0.0001126 0.0124762 Multivariate analyses performed with Metaboanalyst SM(FA22:1) 0.1038876 0.0001543 0.0124762 software[26] allowed us to identify those lipids that were dif- CE(FA22:6) −0.2134798 0.0002163 0.0124762 ferentially present in the HDL lipidome before and after each PC(FA18:1) 0.1147135 0.0013702 0.0538892 intervention (within-intervention comparisons) and among the TAG(FA18:2) −0.1862806 0.0015575 0.0538892 interventions (between-intervention comparisons). In all the SM(FA26:0) −0.1462294 0.0026301 0.0703188 sets of multivariate analyses, the OPLS-DA showed a trend CE(FA18:1) 0.0616907 0.0029727 0.0703188 for separating two different groups of data, corresponding to TAG(FA20:1) 0.1643749 0.0032517 0.0703188 pre- and post-intervention values in the within-intervention Lipid species concentration comparisons, and to the different OO interventions in the between-intervention comparisons. Moreover, of all the models TAG56:8(FA18:2) −0.4777177 0.0000352 0.0270411 predicted by OPLS-DA, 12 models improved data classification PC(FA18:1/FA18:1) 0.3392268 0.0001713 0.0658497 as indicated by an R2 /Q2 > 0.7. These models correspond to TAG51:4(FA18:2) −0.3454150 0.0004444 0.0867678 changes in lipid class concentration and composition (after TAG51:5(FA18:3) −0.7895801 0.0004843 0.0867678 FVOO intake), lipid class composition (after FVOO intake, and TAG54:7(FA22:5) −0.7164132 0.0005642 0.0867678 comparing VOO vs FVOOT intake), LC-FA concentration (after Lipid species composition FVOO intake and after FVOOT intake), LC-FA composition TAG56:5(FA20:3) 0.2616832 0.0000099 0.0043817 (after VOO intake and after FVOOT intake), lipid species con- TAG51:4(FA18:2) −0.3755527 0.0000126 0.0043817 centration (after VOO intake and comparing VOO vs FVOOT TAG51:4(FA15:0) −0.3697035 0.0000171 0.0043817 intake), and lipid species composition (after VOO intake, after TAG54:2(FA20:1) 0.2707045 0.0000334 0.0064200 FVOOT intake, and comparing VOO vs FVOOT intake). TAG52:2(FA16:0) 0.1733400 0.0000996 0.0135271 Paired t-test analyses performed in these 12 data sets with R2 /Q2 >0.7 showed that significant changes were found in three TAG52:2(FA18:1) 0.1893192 0.0001055 0.0135271 of these data sets, as indicated by a p-value adjusted by FDR < PC(FA18:1/FA18:1) 0.2851526 0.0001326 0.0145688 0.1. The OPLS-DA plots of these three sets of analyses are shown SM(FA22:1) 0.1038876 0.0001543 0.0148317 in Figure S3, Supporting Information. These analyses revealed TAG54:7(FA22:5) −0.8269101 0.0002159 0.0166373 that no changes in LC-FA concentration, lipid class concentra- CE(FA22:6) −0.2134798 0.0002163 0.0166373 tion, or lipid class composition after any intervention were ob- TAG56:8(FA18:2) −0.5077355 0.0003964 0.0277134 served. However, the LC-FA composition of eight lipids, the con- TAG51:5(FA18:3) −0.9335297 0.0005247 0.0336250 centration of five lipid species, and the composition of 54 lipid TAG54:2(FA18:1) 0.1971140 0.0005983 0.0353917 species were significantly modified after VOO intake. Compre- TAG56:4(FA20:2) 0.2516185 0.0006653 0.0365420 hensive information on the fold change, the p-value, and the p- TAG54:4(FA20:3) 0.1716807 0.0007294 0.0373921 value adjusted by FDR of these lipids is detailed in Table 1. TAG56:4(FA20:3) 0.1471088 0.0008224 0.0395282 The data corresponding to the lipids with an R2 /Q2 > 0.7 in the models predicted by OPLS-DA and that were differently present TAG50:4(FA18:2) −0.3485138 0.0008820 0.0398996 (p-value adjusted by FDR < 0.1) were further analyzed by using TAG52:4(FA18:2) −0.3113580 0.0010850 0.0441748 both general linear models (within-intervention comparisons) TAG52:4(FA16:0) −0.3070039 0.0011523 0.0441748 and ANCOVA (between-intervention comparisons) adjusted by TAG53:3(FA18:2) −0.1628848 0.0011950 0.0441748 age, sex, baseline values, the sequence of intervention, and the TAG50:3(FA14:1) 0.2673740 0.0012636 0.0441748 number of HDL particles, as described below. TAG52:1(FA18:1) 0.1685961 0.0012638 0.0441748 TAG54:2(FA16:0) 0.1873376 0.0013449 0.0449658 2.4. Changes in LC-FA Combination Composition TAG54:3(FA20:2) 0.1731868 0.0014159 0.0453672 TAG54:3(FA18:1) 0.2509162 0.0019271 0.0574393 Table S6, Supporting Information, shows the changes (95% CI) TAG54:8(FA18:2) −0.8899613 0.0019420 0.0574393 in LC-FA combination composition (%), after 3-weeks of sus- TAG51:3(FA18:2) −0.1743374 0.0023050 0.0648026 tained intake of VOO, FVOO, and FVOOT. In general, all the LC-FA combinations that were modulated in the present study TAG51:3(FA15:0) −0.1886565 0.0023595 0.0648026 were significantly modified after VOO, and most of them were TAG54:2(FA18:0) 0.1473225 0.0024488 0.0649346 also modified after FVOO. However, these changes were not al- SM(FA26:0) −0.1462308 0.0026301 0.0674181 ways significant among them (between intervention differences). TAG52:5(FA16:0) −0.3024300 0.0028230 0.0692749 With regard to FA saturation, some MUFAs showed an in- TAG54:6(FA16:0) −0.2767250 0.0029219 0.0692749 crease after OO consumption. There was a significant increase (Continued) Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (3 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com Table 1. Continued. 2.5. Changes in Lipid Species Concentration Lipid Log fold p-value p-value Table S7, Supporting Information, shows the changes (95% CI) change adjusted in lipid species concentrations (nmol g–1 ) after 3-weeks of sus- by FDR tained intake of VOO, FVOO, and FVOOT. In general, an in- CE(FA18:1) 0.0616905 0.0029728 0.0692749 crease in some lipid species containing MUFAs and a decrease TAG52:6(FA16:1) −0.3444020 0.0033464 0.0721775 in lipid species containing PUFAs were observed after the inter- TAG52:5(FA18:3) −0.2470372 0.0033575 0.0721775 ventions. TAG52:5(FA18:2) −0.3449757 0.0033789 0.0721775 In particular, PC(FA18:1/FA18:1) increased after the three in- TAG54:7(FA18:2) −0.5761714 0.0035175 0.0731075 terventions (p < 0.05). Although the significant quadratic trend (p < 0.001) indicates that the increase after VOO or FVOO tends TAG54:7(FA20:4) −0.2252724 0.0036766 0.0734343 to be higher than that after FVOOT, no significant differences DAG(FA18:2/FA22:6) −3.7042677 0.0038030 0.0734343 were observed among interventions. TAG51:4(FA18:2) decreased TAG52:6(FA18:2) −0.4073001 0.0038197 0.0734343 significantly only after VOO and FVOO intake (p < 0.05), but TAG50:5(FA16:0) −0.7027341 0.0041405 0.0766631 no differences were observed among interventions. In the case TAG54:6(FA18:2) −0.4742435 0.0041871 0.0766631 of TAG56:8(FA18:2), a significant decrease was present only af- TAG52:6(FA16:0) −0.5787183 0.0044123 0.0788637 ter VOO intervention, and this decrease reached significance TAG56:7(FA18:2) −0.2425078 0.0045124 0.0788637 versus the changes observed after FVOOT intervention (p = TAG56:8(FA16:0) −0.4646922 0.0046963 0.0802540 0.007). No significant changes in either or between intervention TAG56:8(FA22:6) −0.5023336 0.0049195 0.0821218 TAG54:7(FA22:5) levels were observed, likely due to the minute TAG55:1(FA16:0) −0.4700962 0.0050191 0.0821218 abundance of this lipid species. TAG50:4(FA14:0) −0.3123558 0.0060180 0.0959950 TAG52:3(FA16:0) −0.0962283 0.0061255 0.0959950 2.6. Changes in Lipid Species Composition TAG56:7(FA22:5) −0.2748293 0.0065134 0.0959950 TAG54:5(FA18:0) −0.2489313 0.0065471 0.0959950 In general, changes were observed in SM, CE, DAG, and PC lipid TAG56:3(FA20:1) 0.2700999 0.0066015 0.0959950 species, but TAG was the lipid class that showed the most signif- TAG56:3(FA18:1) 0.2131561 0.0066160 0.0959950 icant changes after OO intervention. TAG54:6(FA20:4) −0.1866872 0.0068664 0.0977823 Data correspond to paired t test analyses of those data sets with R2 /Q2 > 0.7 in the models predicted by OPLS-DA and with p-value adjusted by FDR < 0.1 in the 2.6.1. SM, CE, DAG, and PC Lipid Species Composition multivariate analysis. Data is ordered according to the p-value adjusted by FDR. CE, cholesteryl esters; DAG, diacylglycerol; FA, fatty acid; FDR, false discovery rate; PC, Changes in SM and CE lipid species composition have been pre- phosphatidylcholine; TAG, triacylglycerol; SM, sphingomyelin. viously described in the “Changes in LC-FA combination com- position” section and in Table S6, Supporting Information. Table S8, Supporting Information, shows the changes (95% CI) in DAG in CE(FA18:1) after the VOO, FVOO, and FVOOT interventions. and PC lipid species composition (%) after 3 weeks of sustained Although the significant quadratic trend (p < 0.001) indicates intake of VOO, FVOO, and FVOOT. In general, an increase in that the increase after FVOO tends to be higher than that af- those lipid species containing MUFAs and a decrease in those ter FVOOT, no differences were observed among interventions. containing PUFAs or SFAs were observed. The same occurred for SM(FA22:1). The MUFA 18:1 presence Between-intervention comparisons showed an increase in in PC and TAG((PC(FA18:1) and TAG(FA18:1), respectively) in- DAG(FA18:2/FA22:6) after FVOOT versus VOO and FVOO (p < creased after VOO and FVOO intake, but not after FVOOT. The 0.05). As observed when the concentration data were examined, increase observed after VOO reached a borderline significance the PC(FA18:1/FA18:1) composition increased after all interven- (p < 0.09) versus the changes after FVOOT. TAG(FA20:1) showed tions (p < 0.05). Although the significant quadratic trend (p < only a borderline significant increase after VOO intervention 0.001) indicates that the increase after VOO or FVOO tends to (p = 0.091). be higher than that after FVOOT, no differences were observed In contrast, one saturated FA (SFA) and two PUFAs decreased among interventions. after all interventions. There was a significant decrease in the PUFA 22:6 presence in CE (CE(FA22:6)) after the three inter- ventions (p < 0.05), and the decrease after VOO was signifi- 2.6.2. TAG Lipid Species Composition cant versus that after FVOO intake (p = 0.015). The presence of PUFA 22:6 in SM and PUFA 18:2 in TAG (SM(FA22:6) and Increase After Consumption: Table S9, Supporting Informa- TAG(FA18:2), respectively) decreased reaching significance only tion, shows the increases (mean, 95% CI) in TAG lipid species after the VOO and FVOO interventions (p < 0.05). In the for- composition (%) after 3-weeks of sustained intake of VOO, mer, no differences among interventions were observed. In the FVOO, and FVOOT. latter, however, the decreases observed after VOO and FVOO in- In general, almost all TAGs that increased contained MUFAs take reached significance versus the changes after FVOOT (p < (n = 8) but some TAGs containing PUFAs (n = 5) and SFAs 0.05). (n = 3) also increased after OO intake. Moreover, all the TAG lipid Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (4 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com species that increased in the present study, increased after VOO The third pattern observed in Table S10, Supporting Infor- and some of them also increased after FVOO but these changes mation, is similar to the second pattern because some lipid were not always significant among them (between intervention species decreased after VOO and FVOO intake, but no differ- differences). No TAG specific significant increases were observed ences or borderline ones were observed between FVOOT in- after FVOOT intake. These changes can be comprehensively ex- take versus the other interventions. In this group, three lipid plained allocating them into three different patterns. species are included: TAG51:4(FA18:2); TAG52:3(FA16:0); and In the first pattern, some lipid species’ compositions increased TAG54:6(FA20:4). significantly after VOO and FVOO intake, but these changes In the fourth pattern, some lipid species decreased af- were not statistically significant compared to those observed ter the three interventions, including TAG51:3(FA18:2) and after FVOOT (between intervention differences). In this group TAG53:3(FA18:2). For the former, no differences between inter- the following lipid species are included: TAG50:3(FA14:1); ventions were observed, whereas for the latter decreases after TAG52:1(FA18:1); TAG54:2(FA18:1); TAG54:2(FA20:1); FVOO were significant versus those observed after FVOOT (p = TAG54:3(FA18:1); TAG56:3(FA20:1); TAG56:4(FA20:3), and 0.035). TAG56:5(FA20:3). In the fifth pattern, the changes in some lipid species reached In the second pattern, similar to the previous pattern, signifi- significance only in the between-intervention comparisons as cant increases occurred after VOO and FVOO intake, and these in the case of the following lipid species: TAG54:5(FA18:0); increases were significant versus the changes after FVOOT inter- TAG55:1(FA16:0); TAG56:7(FA22:5); and TAG56:8(FA16:0). The vention (p < 0.03). In this group, two lipid species are included: significant quadratic trends found in the within-intervention TAG52:2(FA16:0) and TAG52:2(FA18:1). comparisons, however, match the borderline significances In the third pattern, some lipid species’ compositions in- (p < 0.1) obtained in the between-intervention comparisons: in creased only after VOO intake, but these changes were not sig- general, a higher decrease after FVOOT intake was observed nificantly different from those observed after FVOO and FVOOT versus VOO consumption. In the case of TAG54:5(FA18:0), (between intervention differences). In this pattern the following between-intervention changes were also significant between lipid species are included: TAG54:2(FA16:0); TAG54:2(FA18:0); FVOOT and FVOO intake (p = 0.038). TAG52:6(FA16:0) could TAG54:3(FA20:2); TAG4:4 (FA20:3); TAG56:3(FA18:1), and be included in this pattern, although the between-intervention TAG56:4(FA20:2). differences after FVOOT versus VOO intake did not reach signif- icance, probably due to the small sample size. Decrease after Consumption: Table S10, Supporting Infor- mation, shows the decreases (mean, 95% CI) in TAG lipid species 2.7. Associations of Changes in the HDL Lipidome with Changes composition (%) after 3 weeks of sustained intake of VOO, in HDL Functionality: HDL ChE Capacity and HDL Resistance to FVOO, and FVOOT. Oxidation In general, almost all TAG that decreased contained PUFAs (n = 18) or SFAs (n = 11), and only one TAG containing MU- Lipids with an R2 /Q2 > 0.7 in the models predicted by OPLS-DA FAs decreased after OO intake. Moreover, most of the TAG lipid and that were differently present (p-value adjusted by FDR < 0.1) species that decreased in the present study, decreased after VOO, were further analyzed to assess their relationship with changes after FVOO, or after the intake of both OOs. These changes were in HDL functionality (ChE and HDL resistance to oxidation), as significant among them in some cases (between intervention dif- described below in depth. ferences). Different patterns of effects arise from these results as The Pearson’s correlation coefficients of the 3-week changes detailed below. in lipids differently modified after OO intake and the HDL func- In the first pattern, some lipid species decreased signif- tionality are detailed in Tables S11, Supporting Information, icantly only after FVOO intake. Between-intervention anal- to S15 (ChE) and Tables S16, Supporting Information, to S20 yses showed that the decreases, not only after FVOO in- (resistance to oxidation). The superscripts in these tables indi- take but also after VOO intake, were significant, or reached cate cases where Pearson’s correlations were nonsignificant, but a borderline significance, versus the changes after FVOOT Spearman’s correlations were significant. Variables with signifi- intake. In this pattern, the following lipid species are in- cant (p < 0.05) or borderline (p < 0.1) values in the correlation cluded: TAG50:4(FA14:0); TAG52:6(FA18:2); TAG54:6(FA16:0); coefficients were entered into a stepwise linear regression model TAG54:6(FA18:2), and TAG54:7(FA18:2). as detailed in the Statistics section and described below. In the second pattern, a decrease after VOO and FVOO occurred, and differences after FVOOT consumption reached significance versus VOO, FVOO, or both (p < 0.05). In this 2.8. HDL ChE Capacity group the following lipid species are included: TAG50:4(FA18:2); TAG51:3(FA15:0); TAG51:4(FA15:0); TAG52:4(FA16:0); TAG52: Table 2 shows the lipid associations of the 3-week changes in the 4(FA18:2); TAG52:5(FA16:0); TAG52:5(FA18:2); TAG52: HDL lipidome and ChE capacity in hypercholesterolaemic sub- 5(FA18:3); TAG52:6(FA16:1); TAG54:7(FA20:4); TAG56:7 jects after OO ingestion. As this table shows, different variables (FA18:2), and TAG56:8(FA18:2). In this group TAG54:8(FA18:2) were associated with the 3-week changes in ChE after each type could also be included, although intra-intervention differences of OO. after FVOO did not reach significance, probably due to the small Concerning VOO changes, the TAG51:3(FA15:0) 3 week sample size. changes showed an inverse association with those observed Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (5 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com Table 2. Associations of 3 week changes in ChE and HDL lipidome in hy- Table 3. Associations of 3-week changes in HDL resistance to oxidation percholesterolemic subjects after OOs ingestion. and HDL lipidome in hypercholesterolemic subjects after OOs ingestion. Predictor variable B coefficient SE Standardized B T p-value Predictor variable B coefficient SE Standardized B T p-value (3 week changes) (3-week changes) VOO (n = 30) VOO (n = 21) TAG51:3(FA15:0), % −42.7 11.4 −0.546 −3.72 0.001 SM(FA22:1), % 11.8 5.50 0.386 2.14 0.048 TAG52:2(FA18:1), % −0.19 0.08 −0.362 −2.31 0.029 TAG52:3(FA16:0), % −20.5 7.40 −0.504 −2.77 0.014 FVOO (n = 31) FVOO (n = 15) TAG54:6(FA18:2), % −0.89 0.47 −0.309 −1.89 0.070 TAG(FA18:2), % −7.64 1.96 −0.698 −3.89 0.003 FVOOT (n = 30) TAG52:5(FA18:2), % −150 63 −0.537 −2.40 0.035 TAG56:3(FA20:1), % 14.6 8.41 0.334 1.73 0.095 FVOOT (n = 22) TAG52:2(FA18:1), % −3.50 1.44 −0.471 −2.43 0.025 Stepwise linear regression model adjusted by age and sex. For all oils, a stepwise general linear mixed model was fitted adjusted by age and sex and individual level Stepwise linear regression model adjusted by age and sex. For all oils, a stepwise of test subjects as a random effect. Significant values in bold and borderline ones in general linear mixed model was fitted adjusted by age and sex and individual level of italic bold. SE, standard error; TAG, triacylglycerol. test subjects as a random effect. Significant values in bold. SE, standard error; TAG, triacylglycerol. for ChE (p = 0.001). Therefore, the decrease observed in sumption (Tables S6 and S10, Supporting Information) is asso- TAG51:3(FA15:0) after VOO intake (Table S10, Supporting ciated with an increase in the HDL resistance to oxidation. Information) is associated with an increase in ChE. The After FVOOT consumption only the increase in TAG52:2(FA18:1) 3 week changes also showed an inverse rela- TAG52:2(FA18:1) was inversely associated with changes in tionship with those observed for ChE (p = 0.029). This implies, HDL resistance to oxidation (p = 0.025). However, the increase however, that the increase observed in TAG52:2(FA18:1) after in this lipid species observed after FVOOT intake did not reach VOO may be associated with a decrease in ChE. significance (Table S10, Supporting Information). With regard to FVOO, only the lipid species TAG54:6(FA18:2) was inversely associated in a borderline way (p = 0.070) with changes in ChE after FVOO ingestion. Thus, the decrease ob- 3. Discussion served in TAG54:6(FA18:2) after FVOO ingestion (Table S10, This study shows that the sustained intake of VOO and two dif- Supporting Information) could be associated with an increase in ferent functional VOOs, enriched with its own phenolic com- ChE. pounds (mainly secoiridoids; FVOO) or with its phenolic com- Concerning FVOOT, TAG56:3(FA20:1) was directly associated pounds plus complementary compounds from thyme (mainly with changes in ChE with a borderline significance (p = 0.095). flavonoids; FVOOT) differentially modulates the HDL lipidome. Accordingly, we observed an increase, but without significance, In general, VOO and FVOO similarly increase the presence of after FVOOT consumption (Table S10, Supporting Information) MUFAs in several lipid classes and species, at the expense of which could be associated with an increase in ChE. decreasing SFAs and PUFAs, especially in CE, SM, PC, DAG, and TAG. The intake of FVOOT, did not induce the same lipid changes observed after VOO and FVOO intake, even though they 2.9. HDL Resistance to Oxidation had the same lipid matrix (Table S1, Supporting Information). In- terestingly, most of the significant changes associated with HDL Table 3 shows the lipid associations of the 3-week changes in were observed in TAG species, despite it being a minor lipid class HDL lipidome and HDL resistance to oxidation in hypercholes- present in the HDL particle as it accounts for only 2–3% of the terolaemic subjects after OO ingestion. As this table shows, dif- total HDL lipid mass.[27] ferent variables were associated with the 3-week changes in HDL According to the results observed after the sustained intake of resistance to oxidation after each type of OO. VOOs, dietary fats from different sources modify the HDL lipid Concerning VOO changes, the 3 week changes in SM(FA22:1) composition.[17] A diet rich in MUFAs modifies the chain length were directly associated with changes in HDL resistance to oxida- of the phospholipids embedded in the HDL which, in turn, tion (p = 0.048), whereas the changes in TAG52:3(FA16:0) were promotes changes in HDL monolayer fluidity and eventually en- inversely associated with changes in HDL resistance to oxidation hances HDL ChE capacity when compared to diets rich in SFAs (p = 0.014). Thus, the increase in SM(FA22:1) (Table S6, Sup- or PUFAs.[28] Moreover, OO intake also modifies the FA profile porting Information) and the decrease in TAG52:3(FA16:0) (Ta- of the PC present on the HDL surface, increasing MUFAs, and ble S10) observed after VOO consumption are associated with an decreasing SFAs.[17] These changes were assumed to be due increase in the HDL resistance to oxidation. to the different FA compositions of the dietary fats consumed, After FVOO consumption, the 3 week changes in the especially the high content of MUFAs in OO.[17,29] In the present LC-FA combination TAG(FA18:2) and in the lipid species study, the intake of different VOOs with the same FA profile but TAG52:5(FA18:2) both appear to be inversely associated with with different phenolic compound profiles modified the HDL changes in HDL resistance to oxidation (p = 0.003 and p = 0.035). lipidome. However, significant changes in particular LC-FA com- Thus, the decrease observed in both of them after FVOO con- binations and lipid species present in the HDL particles observed Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (6 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com after the intake of VOO and FVOO were not different between FVOOT at administered dosages could influence fat metabolism, these OOs. This fact suggests that a quantitative phenolic com- causing a decrease in MUFA incorporation into HDL lipid pound enrichment of VOO with its own phenolic compounds in species when compared with OOs enriched with their own phe- FVOO might not promote a differential remodeling of the HDL nolic compounds. lipidome. Likewise, our group reported an important matrix In the present study, we also aimed to elucidate whether the effect on the cardioprotective remodeling of the HDL proteome, changes in the lipid classes and LC-FA combination present in as the protein expression modifications observed in the HDL the HDL that occurred after VOO, FVOO, and FVOOT intake particles were common for the three VOO interventions.[23] were associated with changes in HDL functionality. The vast In reference to the FA present in the common matrix of the majority of studies aimed at studying the HDL lipidome are VOOs used in our study, 8.2% are PUFAs (mainly linoleic acid, limited to lipid class analysis,[14,35–38] while only a few have gone 7.4%), 13.75% SFAs (mainly palmitic acid, 11.2%), and 77.7% further relating lipid species and LC-FA combinations to HDL MUFAs (mainly oleic acid, 76.8%). Thus, oleic acid (18:1 𝜔9) is function.[16,39] Our results indicate that changes in specific LC-FA by far the most abundant FA in VOO. The FA that increased the combinations and lipid species are determinants of HDL ChE most in the different lipid species, especially TAG, in the HDL capacity and its resistance to oxidation, as a surrogate of HDL particles was FA18:1. Although the lipidomic approach used in functionality. On the one hand, the changes observed after the this study does not allow us to know the position of the double intake of VOO and FVOO could promote a beneficial increase bond, we assume that FA18:1 detected in our study is oleic acid in the HDL resistance to oxidation in hypercholesterolemic due to its high abundance in the common matrix. Other MUFAs patients, in particular the changes in SM(FA22:1), TAG(FA18:2), that also increased during the intervention with VOO and FVOO TAG52:3(FA16:0) and TAG52:5(FA18:2). On the other hand, were FA22:1, FA20:1, and FA14:1, to the detriment of some SFAs some of the changes observed after the intake of VOO and and PUFAs. Recently, a randomized crossover study assessed FVOO could enhance HDL ChE capacity in the same popula- the short-term effects (4 days) of a fast-food diet and a Mediter- tion, in particular TAG51:3(FA15:0) and TAG54:6(FA18:2). Once ranean diet intervention on HDL lipidomic composition.[30] Sim- more, these effects observed after VOO and FVOO intake differ ilar to our results, they observed that FAs in different classes, from those obtained after FVOOT intake, because some lipids specifically PC, TAG, and CE became shorter and had fewer were found to be determinants of ChE and HDL resistance to double bonds after fast-food diet intervention, which was rich oxidation after FVOOT intake, but no changes in these lipids in SFAs. In contrast, FAs became longer and had more double were observed after this OO intake. bonds after the Mediterranean diet intervention, which was rich Ståhlman et al. revealed that dyslipidemia is the major fac- in MUFAs.[30] tor affecting functional components in the HDL lipidome in Additionally, a study assessing the effects of diets rich in 𝜔-3 type 2 diabetes mellitus. In particular, an increase in TAG52:3, FAs and/or phenolic compounds on the HDL lipidome in over- TAG52:5, and TAG54:6, among others was reported in dys- weight human subjects with high CVD risk observed an increase lipidemic type 2 diabetes mellitus subjects when compared in long-chain PUFA-containing TAG after the high 𝜔-3 FA diet. with healthy and with normolipidemic type 2 diabetes melli- Moreover, the authors also observed a phospholipid profile char- tus subjects.[40] In our study carried out in hypercholesterolemic acterized by a significant increase of medium-chain FAs after subjects, the TAG52:3(FA16:0) and TAG52:5(FA18:2) decreased the diet exclusively rich in phenolic compounds (approximately after VOO and FVOO intake, and this decrease inversely cor- 2900 mg of phenolic compounds). However, the authors did not related with HDL resistance to oxidation. This relationship be- specify the family of phenolic compounds administered to the tween changes in these TAG species and HDL function was con- volunteers. Moreover, an inverse association between phenolic firmed by the fact that TAG52:3(FA16:0) and TAG52:5(FA18:2) compounds and TAGs containing long-chain low unsaturated were determinants of HDL resistance to oxidation. Moreover, the FAs was observed.[31] lipid species TAG54:6(FA18:2) decreased after FVOO intake, and The intervention with FVOOT resulted in differential effects this decrease inversely correlated with HDL ChE capacity and on the HDL lipidome when compared to VOO and FVOO. HDL resistance to oxidation. The relation between the changes This could likely be due to the enrichment of FVOOT with in this TAG species and HDL ChE capacity was confirmed by the phenolic compounds from thyme (mainly flavonoids). Thus, fact that TAG54:6(FA18:2) was a determinant of HDL function. the presence of flavonoids could have interfered with the VOO The decreases in these three lipid species (TAG52:3(FA16:0) after common matrix by reducing the incorporation of MUFAs into VOO, and TAG52:5(FA18:2) and TAG54:6(FA18:2) after FVOO) the different lipid classes of HDLs. For instance, phenolic com- observed in our study were associated with an increase in HDL pounds affect postprandial lipemia decreasing TAG absorption functionality. These species are increased in dyslipidemic type via inhibition of pancreatic lipase activity or decreasing bile acid 2 diabetes mellitus subjects.[40] Thus, our findings point to an reabsorption.[32–34] Moreover, phenolic compounds can decrease atherogenic role for these abovementioned lipid species. Conse- the activity of lipolytic enzymes by affecting the emulsification quently, the HDL targeted lipidomic assessment of these 3 FA process and thus decreasing fats absorption and inhibiting the lipid species, which account for only a 6.90% of the TAG fraction synthesis of lipids in the liver, including TAG and FA.[33] of the HDL lipidome, could be a good approach to assess HDL In summary, the specific type of phenolic compounds (mainly function, in particular ChE and resistance to oxidation as a surro- secoiridoids and hydroxytyrosol derivates) and the given dose gate of HDL functionality. However, more studies are warranted (25 mL day-1 for 3 weeks) could exert a key role in lipid on this issue. metabolism,[33,34] including HDL, leading to changes in the HDL We have previously reported that changes in lipids present in lipidome. Thus, we hypothesize that thyme flavonoids present in HDL (other than phospholipids), namely, free cholesterol and Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (7 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com TAG, are determinants of HDL fluidity and oxidative status, in laemic subjects (total cholesterol >200 mg dL−1 ). Participants ingested turn, determine HDL ChE capacity.[6] In concordance with such 25 mL day–1 of raw VOO (control VOO, FVOO, or FVOOT) during meals results, in the present study, we have reported that changes in for 3 weeks, according to the assigned sequence of intervention. All in- terventions were preceded by 2 week washout periods with common OO the SM and TAG composition are prone to modulate both HDL (Figure S1, Supporting Information). ChE capacity and HDL oxidative status. Concordantly, we have The institutional ethics committee (CEIC-IMAS 2009/3347/I) approved previously reported the enrichment of lipo- and hydrophilic the protocol. The present clinical trial was conducted in accordance with antioxidants linked to HDL, particularly phenolic compounds the Helsinki Declaration and the Good Clinical Practice for Trials on Med- (thymol sulfate, caffeic acid sulfate, hydroxyphenylpropionic acid ical Products in the European Community and International Conference sulfate, and hydroxytyrosol acetate sulfate) and fat-soluble antiox- of Harmonization. The subjects gave their written informed consent be- fore participation. The study was registered at the International Standard idants (alpha-tocopherol, lutein, ubiquinol and 𝛽-cryptoxanthin), Randomized Controlled Trial Register (Identifier: ISRCTN77500181). after the intake of FVOO and FVOOT.[41] The coexistence of these Detailed information on the study subjects, the design of the study, and lipophilic and hydrophilic antioxidants linked to HDL may confer dietary adherence is described in the Supporting Methods section. benefits by protecting HDL itself from oxidative damage via dif- HDL Isolation and Characterization: Blood was collected in Vacutainer ferent antioxidant pathways.[22] The proper oxidative balance in tubes with K2EDTA anticoagulant. Blood samples were centrifuged at the HDL lipidome may result in an increase in its ChE capacity. 1500 × g for 15 min and 2.8 mL of plasma were finally recovered. A pro- A limitation of this study could be the inability to evaluate tease inhibitor cocktail (Sigma-Aldrich, Tres Cantos, Spain) was added to plasma at a concentration 1/100 (1 µL of protease inhibitor cocktail for whether possible interactions of other dietary components or 100 µL of plasma). All samples were stored at −80 °C until HDL isola- medication with the intervention OOs modify the HDL lipidome, tion. The HDL fraction (1.036–1.21 g mL−1 ) was isolated from plasma by however controlling diet and medication during the study with- the sequential density gradient ultracentrifugation method as previously out any observed changes throughout the intervention could described.[23] To ensure the purity of the HDL, apolipoprotein(Apo) B100 limit these interactions. and albumin levels were determined in these samples by immunoturbidi- In conclusion, lipidomic analyses show that the intake of VOO, metric methods using a Cobas-Mira Plus automated analyzer. Moreover, HDL lipid and protein characterizations were also FVOO, and FVOOT, differentially modulates the presence of par- performed by enzymatic and immunoturbidimetric methods (ABX- ticular FA and lipid species, mostly TAG, in the HDL particle. HoribaDiagnostics, France; Roche Diagnostic System, Spain; Spinreact, The changes observed after VOO intake and after FVOO in- Spain) using the automatic analyzer Cobas-Mira Plus, as previously take were similar, whereas FVOOT did not induce the same FAs described.[23] changes. The type of phenolic compound present in the OOs, Detailed information on the HDL isolation and characterization is de- rather than the common matrix of the three OOs, could account scribed in the Supporting Methods section. Lipidomic Data Analyses: Quantitative lipidomic analysis was carried for this fact. Moreover, in the present study, we demonstrate that out using the Lipidyzer platform according to the manufacturers’ instruc- changes in the HDL lipidome in hypercholesterolemic patients tions. Lipid extraction was carried out by applying the methyl tert-butyl are associated with a modulation of HDL functionality promot- ether extraction method as described elsewhere.[43] Lipid analysis and ing an increase in ChE capacity and in HDL resistance to oxida- quantification were carried out as described in detail by Cao et al. and tion through a decrease in TAG52:3(FA16:0), TAG52:5(FA18:2), Contrepois et al.[44,45] The identified and quantified lipid classes, lipid and TAG54:6(FA18:2) lipid species. species, and FAs were reported as .xls sheets. In all the quantifications per- In summary, from our data, the type of phenolic compounds formed, both the concentration (expressed in nmol g–1 ) and composition (expressed as the percentage of the fraction) were analyzed and reported. present in the same matrix, in our case VOO, modulates the HDL Lipid classes refer to: a) four types of phospholipids: lysophosphatidyl- lipidome. Our data also indicate that the HDL lipidome and func- choline, lysophosphatidylethanolamine, phosphatidylcholine (PC), and tion are interconnected. Therefore, the assessment of the HDL phosphatidylethanolamine; b) five types of sphingolipids: ceramides, dihy- lipidome could be a good approach to identify and character- droceramides, hexosylceramides, lactosylceramides, and sphingomyelin ize new HDL functionality biomarkers.[9] Therefore, therapeu- (SM); and c) four neutral lipids: cholesteryl esters, free FAs, diacylglyc- tic strategies aimed at reducing CVD risk may take into account erides (DAGs), and TAGs. Within every particular lipid class, the lipid species refers to differ- not only changes in HDL-C levels and HDL functionality but also ent lipid subclasses that differ in any of the FAs that it contains. As an changes in the HDL lipidome. example, this is the case for the two lipid species DAG(12:0/18:1) and DAG(12:0/18:2). The Lipidyzer platform also allows us to quantify each FA in each partic- ular lipid class. That is, it quantifies the sum of all lipid species containing 4. Experimental Section a determined FA within a particular class of lipids, regardless of the other Phenol-Enriched OO Preparation and Composition: A natural VOO con- FAs it is combined with, if any. From now on, the term Lipid Class-FA (LC- taining 80 mg kg−1 of phenolic compounds was used as a control con- FA) combination will be used throughout the manuscript to refer to this dition. As described previously, this VOO was also used as a matrix to quantification. As an example, this is the case for the LC-FA combination prepare a functional VOO (FVOO) enriched with its own phenolic com- DAG(FA12:0) which encompasses both lipid species DAG(FA12:0/FA18:1) pounds, mainly secoiridoid derivatives (500 mg kg−1 ). The same parental and DAG(FA12:0/FA18:2). VOO was used to prepare a second functional OO (FVOOT; 500 mg HDL Functionality Assessment and Particle Number: HDL function- kg−1 ) enriched with both its own phenolic compounds (50%; mainly sec- ality was measured as HDL resistance to oxidation (assessed as the oiridoid derivatives) and complementary compounds from thyme (50% lag time of conjugated diene formation) and ChE capacity as previously mainly flavonoids, phenolic acids, and monoterpenes).[42] As previously described.[25,41,46] HDL particle number assessment was performed by reported, the OOs did not differ in fat and micronutrient composition, ex- nuclear magnetic resonance in a Vantera clinical spectrometer produced cept for the phenolic content (Table S1, Supporting Information).[25] by LipoScience (Raleigh, NC, USA).[25] Study Subjects, Design of the Study, and Dietary Adherence: The virgin Sample Size and Power Analysis: A sample size of 30 individuals allows olive oil and HDL functionality (VOHF) study is a randomized, controlled, a power of at least 80% to detect a statistically significant difference be- double-blind, crossover trial that was conducted in 33 hypercholestero- tween groups of 3 mg dL–1 of HDL-C (according to the main aim of the Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (8 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com VOHF study) and a standard deviation of 1.9. A dropout rate of 15% and Keywords a Type I error of 0.05 (2-sided) were assumed. Statistical Analyses: Lipidomic data were matched and aggregated by cholesterol efflux, HDL functionality, lipidomics, polyphenols, resistance metabolite sorting. Within- and between-intervention comparisons were to oxidation performed. For each comparison, only volunteers with values in the com- pared interventions were considered. Data were normalized using the Received: December 17, 2020 mean centering on log2 transformed metabolites. For each comparison, Published online: March 18, 2021 the following analyses were performed: 1) principal component analysis to visualize the global variance of the data sets, to reveal intrinsic simi- larities in the spectral profiles, and to identify outliers; 2) orthogonal par- tial least squares-siscriminant analysis (OPLS-DA) with a permutation test [1] P. Barter, M. Caulfield, M. Eriksson, S. Grundy, J. Kastelein, M. Ko- (N = 1000). Here, metabolomic data are the descriptor matrix (X) and the majda, J. Lopez-Sendon, L. Mosca, J. Tardif, D. Waters, C. Shear, J. interventions (each compared vs its baseline, and the differences between RevkiN, K. Buhr, M. Fisher, A. Tall, B. Brewer, N. Engl. J. Med. 2007, them) were used pairwise as the response variable (Y). Orthogonal signal 357, 2109. correction filters were used to remove the variation in the descriptor matrix [2] S. K. Karathanasis, L. A. Freeman, S. M. Gordon, A. T. Remaley, Clin. that it is unrelated to the response variable and thus assist in the interpre- tation of the model and the identification of metabolites associated with Chem. 2017, 63, 196. the response variable; 3) Paired t test for comparisons with two groups [3] A. Rohatgi, A. Khera, J. D. Berry, E. G. Givens, C. R. Ayers, K. E. Wedin, and analysis of variance (ANOVA) for comparisons of the three groups. p- I. J. Neeland, I. S. Yuhanna, D. R. Rader, J. A. de Lemos, P. W. Shaul, values were adjusted by the false discovery rate (FDR). Variables in which D. Ph, I. J. Neeland, I. S. Yuhanna, D. Ph, D. R. Rader, J. A. De Lemos, the FDR-adjusted p-value was <0.1 in the paired t test were selected to test P. W. Shaul, A. Rohatgi, A. Khera, J. D. Berry, E. G. Givens, C. R. Ayers, variables in a more advanced model, such as an ANCOVA model adjusted K. E. Wedin, I. J. Neeland, I. S. Yuhanna, D. R. Rader, J. a. de Lemos, by age sex, sequence, HDL particle number, and baseline values, for as- P. W. Shaul, N. Engl. J. Med. 2014, 371, 2383. sessing whether inter- and intra-treatment differences could be obtained. [4] D. Saleheen, R. Scott, S. Javad, W. Zhao, A. Rodrigues, A. Picataggi, Variables with an R2 /Q2 > 0.7 in the models predicted by OPLS-DA and D. Lukmanova, M. L. Mucksavage, R. Luben, J. Billheimer, J. J. P. with an FDR-adjusted p-value <0.1 in the paired t test were further ana- Kastelein, S. M. Boekholdt, K. T. Khaw, N. Wareham, D. J. Rader, lyzed. In these variables, we applied: a) a general linear model, for within- Lancet Diabetes Endocrinol. 2015, 3, 507. intervention comparisons; b) an ANCOVA model (adjusted by age, sex, [5] P. Mody, P. H. Joshi, A. Khera, C. R. Ayers, A. Rohatgi, J. Am. Coll. sequence of OO administration, number of HDL particles, and baseline Cardiol. 2016, 67, 2480. values) for between-intervention comparisons; and c) univariate associ- [6] S. Fernández-Castillejo, L. Rubió, Á. Hernáez, Ú. Catalán, A. Pedret, ations examined by Pearson’s and Spearman’s correlation coefficients to R.-M. Valls, J. I. Mosele, M.-I. Covas, A. T. Remaley, O. Castañer, M.-J. assess their relationship with changes in HDL functionality (ChE capacity) Motilva, R. Solá, Mol. Nutr. Food Res. 2017, 1700445. and HDL resistance to oxidation as a surrogate of HDL functionality. Vari- [7] B. J. Ansell, K. E. Watson, A. M. Fogelman, M. Navab, G. C. Fonarow, ables with significant (p < 0.05) or borderline (p < 0.1) values obtained J. Am. Coll. Cardiol. 2005, 46, 1792. in one of the two coefficients were entered in a stepwise linear regression [8] I. E. Hoefer, S. Steffens, M. Ala-Korpela, M. Bäck, L. Badimon, M.- model adjusted by age and sex with restricted collinearity (variation infla- tion factor [VIF] <2.5). Graphics and paired t tests were generated in the L. Bochaton-Piallat, C. M. Boulanger, G. Caligiuri, S. Dimmeler, J. R statistical environment (Version 3.5.2) with different packages from the Egido, P. C. Evans, T. Guzik, B. R. Kwak, U. Landmesser, M. Mayr, C. Comprehensive R Archive Network (CRAN 2017) and Metaboanalyst R.[47] Monaco, G. Pasterkamp, J. Tuñón, C. Weber, Eur. Heart J. 2015, 36, 2635. [9] F. M. Sacks, M. K. Jensen, Arterioscler. Thromb. Vasc. Biol. 2018, 38, Supporting Information 487. [10] A. Kontush, M. Lindahl, M. Lhomme, L. Calabresi, M. J. Chapman, W. Supporting Information is available from the Wiley Online Library or from S. Davidson, Contents, High Density Lipoproteins: From Biological Un- the author. derstanding to Clinical Exploitation, Springer International Publishing, London 2015. [11] H. Y. Choi, A. Hafiane, A. Schwertani, J. Genest, Can. J. Cardiol. 2016, Acknowledgements 33, 1. M.F., M.J.M., and R.S. are senior authors. [12] T. A. Lydic, Y.-H. Goo, Clin. Transl. Med. 2018, 7, 4. The VOHF project (AGL2009-13517-457 C03-01) and the AppleCOR [13] J. Salazar, L. C. Olivar, E. Ramos, M. Chávez-Castillo, J. Rojas, V. Project (AGL2016-76943-C2) were made possible with the support of the Bermúdez, Cholesterol 2015, 2015, 1. http://doi.org/10.1155/2015/ Ministerio de Economía, Indústria y Competitividad, the Agencia Estatal de In- 296417. vestigación, and the European Regional Development Fund. U.C. has a Pla [14] A. A. Khan, P. A. Mundra, N. E. Straznicky, P. J. Nestel, G. Wong, R. Estratègic de Recerca i Innovació en Salut (PERIS) postdoctoral grant (no. Tan, K. Huynh, T. W. Ng, N. A. Mellett, J. M. Weir, C. K. Barlow, Z. SLT002/16/00239; Catalunya, Spain) from the Generalitat de Catalunya. H. Alshehry, G. W. Lambert, B. A. Kingwell, P. J. Meikle, Arterioscler. A.P. has a Torres Quevedo postdoctoral grant with the Subprograma Es- Thromb. Vasc. Biol. 2018, 38, 438. tatal de Incorporación, Plan Estatal de Investigación Científica Técnica y de [15] M. Jové, A. Naudí, M. Portero-Otin, R. Cabré, S. Rovira-Llopis, C. Innovación. M.F. has a Sara Borrell postdoctoral contract (CD17/00233 to Bañuls, M. Rocha, A. Hernández-Mijares, V. M. Victor, R. Pamplona, M.F.-St.Pau). O.C. has a JR17/00022 contract from Instituto de Salud Carlos FASEB J. 2014, 28, 5163. III. CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM) [16] W. Pruzanski, E. Stefanski, F. C. de Beer, M. C. de Beer, A. Ravandi, is a project of Instituto de Salud Carlos III (Madrid, Spain). The NFOC-Salut A. Kuksis, J. Lipid Res. 2000, 41, 1035. group is a consolidated research group of the Generalitat de Catalunya, [17] R. Solà, M. F. Baudet, C. Motta, M. Maillé, C. Boisnier, B. Ja- Spain (reference no. 2017 SGR 522). cotot, Biochim. Biophys. Acta, Lipids Lipid Metab. 1990, 1043, 43. Conflict of Interest [18] M. I. Covas, R. De la Torre, A. Kafatos, R. M. Lamuela-Raventos, J. Osada, Nutr. Rev. 2006, 64, 20. The authors declare no conflict of interest [19] M. I. Covas, Pharmacol. Res. 2007, 55, 175. Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (9 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH 16134133, 2021, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/mnfr.202001192 by Readcube (Labtiva Inc.), Wiley Online Library on [23/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.mnf-journal.com [20] J. M. Lou-Bonafonte, C. Gabás-Rivera, M. A. Navarro, J. Osada, Nu- [35] J. Salazar, L. C. Olivar, E. Ramos, M. Chávez-Castillo, J. Rojas, V. trients 2015, 7, 4068. Bermúdez, Cholesterol 2015, 2015, 296417. [21] M. Fitó, R. de la Torre, M. Farré-Albaladejo, O. Khymenetz, J. Marru- [36] M. Holzer, R. Birner-Gruenberger, T. Stojakovic, D. El-Gamal, V. gat, M. Covas, Eur. J. Clin. Nutr. 2004, 58, 955. Binder, C. Wadsack, A. Heinemann, G. Marsche, J. Am. Soc. Nephrol. [22] A. Pedret, S. Fernández-Castillejo, R.-M. Valls, Ú. Catalán, L. Rubió, 2011, 22, 1631. M. Romeu, A. Macià, M. C. López de Las Hazas, M. Farràs, M. Giralt, [37] F. Rached, R. D. Santos, L. Camont, M. H. Miname, M. Lhomme, C. J. I. Mosele, S. Martín-Peláez, A. T. Remaley, M.-I. Covas, M. Fitó, Dauteuille, S. Lecocq, C. V. Serrano, M. J. Chapman, A. Kontush, J. M.-J. Motilva, R. Solà, Mol. Nutr. Food Res. 2018, 1800456. Lipid Res. 2015, 55, 2509. [23] A. Pedret, Ú. Catalán, S. Fernández-Castillejo, M. Farràs, R.-M. Valls, [38] F. Brites, M. Martin, I. Guillas, A. Kontush, BBA Clin. 2017, 8, L. Rubió, N. Canela, G. Aragonés, M. Romeu, O. Castañer, R. de la 66. Torre, M.-I. Covas, M. Fitó, M.-J. Motilva, R. Solà, PLoS One 2015, 10, [39] E. Zakiev, F. Rached, M. Lhomme, M. Darabi-Amin, M. Ponnaiah, P. e0129160. H. Becker, P. Therond, C. V. Serrano, R. D. Santos, M. J. Chapman, [24] M. Farràs, O. Castañer, S. Martín-Peláez, Á. Hernáez, H. Schröder, A. Orekhov, A. Kontush, J. Clin. Lipidol. 2019, 13, 468. http://doi.org/ I. Subirana, D. Muñoz-Aguayo, S. Gaixas, R. de la Torre, M. Farré, L. 10.1016/j.jacl.2019.02.004. Rubió, Ó. Díaz, S. Fernández-Castillejo, R. Solà, M. J. Motilva, M. Fitó [40] M. Ståhlman, B. Fagerberg, M. Adiels, K. Ekroos, J. M. Chapman, A. Colomer, Mol. Nutr. Food Res. 2015, 59, 1758. Kontush, J. Borén, Biochim. Biophys. Acta, Mol. Cell Biol. Lipids 2013, [25] S. Fernández-Castillejo, R.-M. Valls, O. Castañer, L. Rubió, Ú. Catalán, 1831, 1609. A. Pedret, A. Macià, M. L. Sampson, M.-I. Covas, M. Fitó, M.-J. [41] M. Farràs, S. Fernández-Castillejo, L. Rubió, S. Arranz, Ú. Catalán, Motilva, A. T. Remaley, R. Solà, Ú. Catalan, A. Pedret, A. Macià, M. L. I. Subirana, M.-P. Romero, O. Castañer, A. Pedret, G. Blanchart, D. Sampson, M.-I. Covas, M. Fitó, M.-J. Motilva, A. T. Remaley, R. Solà, Muñoz-Aguayo, H. Schröder, M.-I. Cova, R. de la Torre, M.-J. Motilva, Mol. Nutr. Food Res. 2016, 60, 1544. R. Solà, M. Fitó, M.-I. Covas, R. de la Torre, M.-J. Motilva, R. Solà, M. [26] Z. Pang, J. Chong, S. Li, J. Xia, Metabolites 2020, 10. https://doi.org/ Fitó, J. Nutr. Biochem. 2018, 51, 99. http://doi.org/10.1016/j.jnutbio. 10.3390/metabo10050186. 2017.09.010. [27] A. Kontush, M. Lhomme, J. Glycomics Lipidomics 2015, 05. https:// [42] L. Rubió, M. J. Motilva, A. MacIà, T. Ramo, M. P. Romero, J. Agric. doi.org/10.4172/2153-0637.1000133. Food Chem. 2012, 60, 3105. [28] R. Sola, C. Motta, M. Maille, M. T. Bargallo, C. Boisnier, J. L. Richard, [43] A. Körner, E. Zhou, C. Müller, Y. Mohammed, S. Herceg, F. Bracher, B. Jacotot, Arterioscler. Thromb. 1993, 13, 958. P. C. N. Rensen, Y. Wang, V. Mirakaj, M. Giera, Proc. Natl. Acad. Sci. [29] R. Solà, C. Motta, M. Maille, M. Bargalló, J. Richard, B. Jacotot, USA 2019, 116, 20623. Atheroscler. Thromobosisbosis 1993, 13, 958. [44] Z. Cao, T. C. Schmitt, V. Varma, D. Sloper, R. D. Beger, J. Sun, J. Pro- [30] C. Zhu, L. Sawrey-Kubicek, E. Beals, R. L. Hughes, C. H. Rhodes, R. teome Res. 2019. https://doi.org/10.1021/acs.jproteome.9b00289. Sacchi, A. M. Zivkovic, Metabolomics 2019, 15, 114. http://doi.org/ [45] K. Contrepois, S. Mahmoudi, B. K. Ubhi, K. Papsdorf, D. Hornburg, 10.1007/s11306-019-1579-1. A. Brunet, M. Snyder, Sci. Rep. 2018, 8, 1. [31] I. Bondia-Pons, P. Pöhö, L. Bozzetto, C. Vetrani, L. Patti, A.-M. Aura, [46] S. Fernández-Castillejo, L. Rubió, Á. Hernáez, Ú. Catalán, A. Pedret, G. Annuzzi, T. Hyötyläinen, A. A. Rivellese, M. Orešič, Mol. Nutr. Food R.-M. R.-M. Valls, J. I. J. I. J. I. Mosele, M.-I. M.-I. M.-I. Covas, A. T. A. Res. 2014, 58, 1873. http://doi.org/10.1002/mnfr.201400155. T. A. T. Remaley, O. Castañer, M.-J. M.-J. Motilva, R. Solá, A. Hernáez, [32] C. Desmarchelier, P. Borel, D. Lairon, M. Maraninchi, R. Valéro, Nu- U. Catalán, A. Pedret, R.-M. R.-M. Valls, J. I. J. I. J. I. Mosele, M.-I. M.-I. trients 2019, 11. https://doi.org/10.3390/nu11061299. M.-I. Covas, A. T. A. T. A. T. Remaley, O. Castañer, M.-J. M.-J. Motilva, [33] N. Kardum, M. Glibetic, in Adv. Food Nutr. Res., Vol. 84, Academic R. Solá, Á. Hernáez, Ú. Catalán, A. Pedret, R.-M. R.-M. Valls, J. I. J. Press Inc. 2018, 103. I. J. I. Mosele, M.-I. M.-I. M.-I. Covas, A. T. A. T. A. T. Remaley, O. [34] B. Pokimica, M. T. García-Conesa, M. Zec, J. Debeljak-Martačić, Castañer, M.-J. M.-J. Motilva, R. Solá, Mol. Nutr. Food Res. 2017, 61. S. Ranković, N. Vidović, G. Petrović-Oggiano, A. Konić-Ristić, https://doi.org/10.1002/mnfr.201700445. M. Glibetić, Nutrients 2019, 11, 850. http://dx.doi.org/10.3390/ [47] J. Chong, M. Yamamoto, J. Xia, Metabolites 2019, 9. https://doi.org/ nu11040850. 10.3390/metabo9030057. Mol. Nutr. Food Res. 2021, 65, 2001192 2001192 (10 of 10) © 2021 The Authors. Molecular Nutrition & Food Research published by Wiley-VCH GmbH