Data on the endogenous conversion of tyrosol into hydroxytyrosol in humans

Here we present new and original data on the endogenous conversion of tyrosol (Tyr) into hydroxytyrosol (OHTyr) in humans and its effects on the cardiovascular system. A randomized, crossover, controlled clinical trial was performed with individuals at cardiovascular risk (n = 33). They received white wine (WW) (females 1, males 2 standard drinks/day), WW plus Tyr capsules (WW + Tyr) (25mg Tyr capsule, one per WW drink), and water (control) ad libitum. Intervention periods were of 4 weeks preceded by three-week wash-out periods. We assessed the conversion of Tyr to OHTyr, its interaction with a polygenic activity score (PAS) from CYP2A6 and CYP2D6 genotypes, and the effects on cardiovascular risk markers. For further details and experimental findings please refer to the article “Cardiovascular benefits of tyrosol and its endogenous conversion into hydroxytyrosol in humans. A randomized, controlled trial” [1].


Specifications
Raw data collection and analysis Parameters for data collection Before and after each one of the three interventions with 1) white wine (WW) (females 1, males 2 standard drinks/day), 2) WW plus Tyr capsules (WW þ Tyr) (25mg Tyr capsule, one per WW drink), and 3) water (control) ad libitum. Intervention periods were of 4 weeks preceded by 3-weeks washout periods.

Description of data collection
Biological samples were collected and processed in the context of a randomized controlled intervention trial by field investigators Data source location Barcelona

Value of the Data
The presented data provide further details on how the polygenic activity score to evaluate the efficiency of Tyr to OHTyr conversion was generated. This report describes the effects on cardiovascular biomarkers of an intervention with white wine and white wine enriched with Tyr. They are useful to predict the effects of nutritional interventions rich in Tyr (olive oil, wine, beer …) considering the interaction with individual's genetic background.
The present data are of interest for the effects of Tyr alone and in the future within Tyr rich foods These data can be used for the design of new nutraceutical based on Tyr ingestion in humans

Data
Thirty-three participants (21 men and 12 women) were randomly allocated to participate in a clinical trial, and 32 participants completed the trial. Initially, 192 subjects were assessed for eligibility, 157 were excluded for 1) not meeting the inclusion criteria, 2) refusing to participate, 3) taking medication non compatible with the interventions, 4) suffering from a coronary heart disease, 5) having undergone bariatric surgery, 6) intestinal alterations, 7) mobility problems, 8) chronic inflammatory diseases, 9) dysregulated hypertension, 10) illicit drug consumption, 11) heavy alcohol consumption, and 12) hepatic alterations. The 33 volunteers were randomly allocated to receive the following treatments for 4 weeks: WW, WW þ Tyr and control intervention [1] (Fig. 1).
Baseline characteristics of the participants are shown in Table 1. No dietary differences were observed among interventions ( Table 2). Volunteers were genotyped for multiple allelic variants of CYP2A6 and CYP2D6 (Table 3). For each enzyme, an activity score was given to each volunteer according to the alleles identified (Table 4). A pooled polygenic activity score (PAS) was calculated by adding the two activity scores together.
On the basis of PAS, 11 individuals were categorized as low (LA, PAS range: 1e2.5), 19 as normal (NA, PAS range: 3e4), and 2 as rapid (RA, PAS range ¼ 5) activity metabolizers. Due to their low number, RAs were excluded for analyses. Age and gender were equally distributed among the three groups. The conversion of Tyr into OHTyr was assessed by measuring OHTyr and Tyr urinary recovery following each treatment. No changes were observed in lipid biomarkers ( Table 5) with exception of HDLcholesterol (HDLc) which increased after WW and WW þ Tyr. HDL-c increased in a dose-dependent manner with the content of alcohol plus Tyr administered in all participants (p ¼ 0.027 for linear trend), among men (p ¼ 0.001 for linear trend), and with a borderline significance in the NA group (p ¼ 0.082) (Fig. 2). Endothelin-1 levels at the end of WW þ Tyr were lower than at the end of WW intervention (Table 6). Table 7 outlines the transcriptomic changes observed in genes related with endothelial function. Changes are grouped by sex and PAS. Fig. 3 compares the effects observed in WW and WW þ Tyr intervention.

Study design
A randomized, controlled, clinical trial with 33 individuals at cardiovascular risk (21 men and 12 women) was performed (Fig. 1). Inclusion criteria were to be at high risk for coronary heart disease (CHD) with 3 or more risk factors including: current smoking (>1 cig/day during the last month), hypertension (!140/90 mmHg or antihypertensive medication), high LDL cholesterol (>130 mg/dl or lipid-lowering therapy), low HDL-cholesterol ( 40 mg/dl in men and 50 mg/dl in women), overweight/obesity (body mass index !25 kg/m 2 ), a family history of premature CHD, and/or type II diabetes treated with oral hypoglycemic agents; and to have a social or recreational use of ethanol/wine consumption at least once during lifetime. Exclusion criteria were participants with a history of cardiovascular disease or severe chronic illness, chronic inflammatory diseases, BMI > 40 kg/m 2 , suffered from any severe illness or undergone major surgery in the last three months prior to the clinical trial, an alcohol consumption exceeding 8 units or 80 g per day, a history of alcohol hypersensitivity/intolerance, illicit drug consumption, intake of antioxidant supplement(s), the taking of sedative drugs that could potentially interact with alcohol, multiple allergies or intestinal diseases, being vegetarian of following special diets, a history of food allergies or intolerances, illiteracy, and any condition that limited mobility making trial visits impossible or worsening adherence to treatments.
Participants were asked to follow a controlled diet with a moderate content of antioxidants and to abstain of any alcoholic drinks (except in the framework of treatment allocations) thorough the trial. The consumption of certain food was limited to a maximum of 1) vegetables (including pulses): one serving/day, 2) fruits (or juices): 2 pieces/day, 3) ordinary olive oil: maximum 25 mL/day, 4) drinks containing xanthines (coffee, tea, cola, energy drinks …): maximum 3 cups/day, 5) chocolate: maximum one piece (small 15 gr)/day, 6) nuts: maximum 30 g (a small handful)/week, and 7) fish: maximum 3 times per week (150 g/serving). During the control intervention participants were allowed to only drink water (no alcohol. wine or supplemented Tyr or OHTyr). The reason for different WW doses for men and women was in order to follow the American Heart Association (AHA) guidelines, which limit alcohol consumption to one drink in women and two in men, preferably taken at meals [2]. The doses of wine administered are within those recommended by the AHA.
A 24 h food recall was performed to assess dietary intake before and after each intervention period. Physical activity was recorded at the beginning and end of the clinical trial and assessed by the Minnesota Leisure Time Physical Activity Questionnaire, validated for the Spanish population [3]. A general physical examination, and routine urine, blood chemical and hematological analyses, were performed at the beginning and end of the trial. Blood and 24h-urine samples were collected at fasting state before and after each intervention period. Blood was collected into 10 mL tubes containing EDTA and centrifuged (1700g, 15 min, 4 C), and plasma and buffy coat samples were then isolated. Peripheral blood mononuclear cells (PBMC) were isolated using a Vacutainer Cell Preparation Tube 2989G > A Intron Variant C_34816116_20 Table 4 Activity score assigned to each tested variant in the PAS model. No function 0 1847 G > A a *5 No expression 0 Gene deletion *9 Decreased þ0.5 2616 del AGG *10 Decreased Multiple copies *2xN *35xN a *4 sub-alleles can commonly present other SNPs such as 100C > T, 4181 G > C and/or 2851C > T.  (CPT™) and kept for RNA extraction. Plasma, urine, and PBMC samples were frozen at À80 C until analysis. Genomic DNA isolation from buffy coat was performed with QIAamp DNA Blood Midi Kit (Qiagen, Dusseldorf, Germany).

Tyr and OHTyr metabolites analysis
The urinary concentrations of Tyr and OHTyr metabolites were determined from samples collected before and after each intervention following a validated methodology [4]. Briefly, 0.5 mL of urine was diluted with 0.5 mL of purified water, spiked with 10 mL of internal standard mixture (containing 10 mg/mL of 3-(4-hydroxyphenyl)-1-propanol, 3-(4-hydroxyphenyl)-1-propanol glucuronide and 10 mg/ mL HT-1 0 -O-sulfate) and stabilized with 1 mL of phosphoric acid 4%. Thereafter, samples went under a solid-phase extraction using Oasis HLB columns 3 mL, 60-mg cartridges from Waters Corporation (Milford, MA USA). First, samples were loaded into cartridges then washed with 2 mL of purified water. Thereafter, the compounds of interest were eluted from the cartridge with 2 mL of pure methanol. The methanol extracts were then evaporated until dryness under a stream of nitrogen (29 C, 10e15 psi). Finally, the dried extracts were reconstituted with a mixture of mobile phases (95% A and 5% B v/v), transferred into HPLC microvials, and analyzed using LC-MS/MS. To prepare blank samples and calibration curves, urine from volunteers after consuming a diet poor in Tyr and OHTyr-Tyr and OHTyr metabolite concentrations were below quantification limits in these blank samples. Blank urine was spiked with increasing concentrations of the metabolites of interest, and then processed in the same manner as samples (described above). Identification and quantification of the metabolites was performed using an Agilent 1200 series HPLC system coupled to a triple quadrupole (6410 Triple Quad LC/MS) mass spectrometer with an electrospray interface from Agilent Technologies (Santa Clara, CA, USA). For the chromatographic separation, an Acquity UPLC®BEH C18 column (100 mm Â 3.0 mm i.d., 1.7 mm particle size) from Waters Corporation (Milford, MA, USA) was used at 40 C. The composition of mobile phase A was 0.01% (v/v) formic acid in water, and mobile phase B was acetonitrile with 0.01% (v/v) of formic acid. Injection volume was 10 mL and the flow rate was set at 0.25 mL/min. The ion source operated in negative ionization for 27 minutes. Finally, urinary concentrations of each metabolite were standardized with the total urinary excretion volume to obtain the total recovery of each metabolite. Quantified Tyr metabolites included Tyr -4-sulfate and Tyr -4-glucuronide. OHTyr metabolites quantified were OHTyr -3-sulfate, OHTyr-4-sulfate, OHTyr-acetate-3-sulfate, OHTyr-3glucuronide, OHTyr-4-glucuronide, and homovanillyl alcohol (HVAL)-4-glucuronide. Total Tyr and total OHTyr correspond to the molar sum of their respective quantified metabolites.

Copy number variation (CNV) detection analysis
TaqMan CNV assays were used to analyze CYP2A6 allelic variants *4, *12 (Hs07545274_cn; Hs07545275_cn), and CYP2D6 allelic variants *5 (deletion),*1xN, *2xN, and *35xN (Hs00010001_cn). Real time qPCR was performed using the specific TaqMan assays. Quantitative PCR was performed in QuantStudio™ 12K Flex Real-Time PCR System (Applied Biosystems, Foster City, USA). Reaction was carried in 384-well plates with a mixture of TaqMan Master Mix (Applied Biosystems, Foster City, CA, USA), CNV assays, 10 ng DNA/well and RNase P as reference (Applied Biosystems, Foster City, CA, USA). Reactions were performed in duplicates. Copy number calls were made with the Expression Suite Software v1.0.3 (Applied Biosystems, Foster City, CA USA). CYP2D6 gene duplications as previously described [5]. First, a specific 6.6 kb long piece of CYP2D6 was amplified. Second, a 3.5 kb fragment was amplified from alleles carrying gene duplications. Every duplication-positive sample was further analyzed using two long-range PCR reactions that allow to discriminate among CYP2D6*1xN, *2xN, and *4xN duplications to determine allele-defining SNPs.

Endothelial function measurement
Endothelial function was assessed before and after interventions by monitoring endotheliummediated changes (reactive hyperemia index, RHI) in the digital pulse waveform, known as the Peripheral Arterial Tone (PAT) signal (EndoPAT 2000; Itamar Medical Inc., Caesarea, Israel). Specially designed finger probes were placed on the middle finger of each subject's dominant hand. The probes comprised a system of inflatable latex air cuffs connected by pneumatic tubes to an inflating device controlled through a computer algorithm. A constant counter pressure (pre-determined by baseline DBP) was applied through the air cushions. Pulsatile volume changes of the distal digit induced pressure alterations in the finger cuff, which were sensed by pressure transducers and transmitted to and recorded by the EndoPAT 2000 device. EndoPAT 2000 also provides the augmentation index (AI), a measurement of arterial stiffness via pulse-wave analysis, which was normalized to 75 bpm heart rate. Measurements were performed by a trained professional with the participants in resting supine conditions, in a quiet room at a constant temperature after 10 minutes of stabilization. Hyperemic reactivity index measured by EndoPAT 2000 has been shown to predict cardiovascular disease [7].
Gene expression was measured, before and after interventions, by a real-time polymerase chain reaction with a QuantStudio™ 12K Flex Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and SYBR Green dye-based analysis. Samples were analyzed in duplicate. Results were obtained with the Expression Suite Software v1.0.3 (Applied Biosystems, Foster City, CA, USA). Changes in gene expression were assessed first by calculating the relative quantification, applying the 2 eDDCT of each sample. Thereafter, the fold change of each intervention was extracted by calculating the ratio between values at the end and at baseline of each intervention period.

Sample size and power analyses
A total sample of 32 participants would allow at least 80% power to detect a statistically significant difference among groups of 0.205 units in the RHI measurement, assuming a dropout rate of 5% and type I error of 0.005 (2-sided). The standard deviation of the measurement was assumed 0.4 [7].

Statistical analyses
Normality of continuous variables was assessed by normal probability plots and data were log transformed when required. Intra-treatment comparisons were assessed by Student's t-test for paired samples. Comparisons among treatments were made by an ANOVA for repeated measures and adjusted by age, gender, smoking, AAS medication, and baseline concentrations. In the case of lipids an additional adjustment for LDL cholesterol values at the beginning of the clinical trial was performed. A general lineal model was used to assess linear and quadratic trends. For the post-hoc pairwise comparison, the Tuckey test was used. Statistical analyses were performed with R (R Foundation for Statistical Computing, Vienna, Austria). version 3.0.2., and R package multcomp. Significance was defined as p < 0.05.