The Antioxidant Effects of Hydroxytyrosol and Vitamin E on Pediatric Nonalcoholic Fatty Liver Disease,in a Clinical Trial: A New Treatment?

Study population and design

Eighty-one white European patients (aged 4–16 years) with NAFLD, referred to the Hepato-Metabolic Department of “Bambino Gesù” Children's Hospital (Rome, Italy) between March 2017 and April 2018, were enrolled in the present study (Supplementary Fig. S1). The diagnosis of NAFLD was performed according to the guidelines of the European Society of Pediatric Gastroenterology, Hepatology and Nutrition. The children were eligible for the study if they were between 4 and 16 years of age, had a liver steatosis, alanine aminotransferase (ALT) levels <10 upper limit of normal, and no laboratory and/or clinical signs of liver decompensation. Moreover, in all children, other causes of liver disease (i.e., viral liver disease, autoimmune hepatitis, Wilson's disease, and α-1-antitrypsin deficiency), celiac disease, alcohol consumption, use of drugs known to induce fatty liver were excluded.

The RCT was undertaken to examine the efficacy and safety of a mixture of HXT and vitamin E orally versus identical placebo for 16 weeks on hepatic steatosis and metabolic parameters in children and adolescents with biopsy-proven NAFLD.

The pharmaceutical formulation was composed of a hard gelatin enteric-coated (Eudraguard^® natural; Evonik Industries AG, Essen, Germany) capsule (Fenolia™; P&P Farma Srl, Turin, Italy) containing 3.75 mg of HXT from a standardized olive extract (elaVida™; DSM, Heerlen, The Netherlands) and 5 mg of DL-α-tocopheryl acetate, conveyed in extra virgin organic olive oil. Placebo capsule formulation was the same of what described above but without the elaVida extract.

Patients were randomized to receive two capsules combining 7.5 mg of HXT and 10 mg of vitamin E (treatment arm) or identical capsules as placebo (placebo arm) per day. The dosage of the HXT and vitamin E intervention was determined based on the available evidence in obese patients with NAFLD.

The computer-generated randomization sequence assigned the participants in a 1:1 ratio to treatment arm or placebo arm. A statistician, who was blinded to participants' clinical data and did not participate in patients' clinical care, generated the allocation sequence and assigned participants to their group. Only the statistician had access to the treatment codes, and participants and investigators were blinded to the treatment for the duration of the study.

Additionally, all patients were included in a lifestyle intervention program consisting of a hypocaloric diet [25–30 kcal/(kg·d)] and regular physical exercise (twice-weekly 2-h physical activity).

The compliance with treatment was monitored at each visit by counting the returned capsules. Moreover, adverse effects were recorded by the principal investigator. Anthropometric measurements and laboratory data were collected at baseline and 4 months. We defined changes in oxidative stress parameters and “marker” of the evolution of NAFLD as the primary outcome of the present proof-of-concept trial. The secondary outcomes were the improvement of hepatic steatosis.

This study was registered on July 26, 2016, in ClinicalTrials.gov (Registration No.: NCT02842567). (Supplementary Table S1).

Anthropometrical and biochemical measurements

Anthropometric measurements and laboratory tests including liver enzymes and glucoinsulinemic and lipids profiles with liver echography were performed at baseline and repeated at 4 months.

The body weight and height were measured with the patients wearing underwear. BMI (kg/m²) and waist circumference were calculated as described previously.

Serum glucose, lipid profile (triglycerides, cholesterol-total, high-density lipoprotein cholesterol, and LDL), liver function tests (aspartate aminotransferase and ALT, gamma-glutamyl transpeptidase [GGT], albumin, and international normalized ratio), fasting plasma glucose, and insulin were measured in all patients after an overnight 12-h fasting. IR was assessed by the HOMA: HOMA-IR = [insulin0 (μIU/mL) × glucose0 (mM)]/22.5. A cutoff value of >2.5 was considered as an index of IR.

Hepatic ultrasound

Liver steatosis was identified by abdominal ultrasound evaluation by the same experienced radiologist, who was blinded to the health conditions of the patients, using an Acuson Sequoia C512 scanner equipped with a 15L8 transducer (Universal Diagnostic Solutions, Oceanside, CA). Liver steatosis was defined according to the criteria (mild = 0, moderate = 1, and severe = 2), a bright hepatic echo pattern compared with echo response of the right kidney.

Chemicals

Physiological saline solution (NaCl 0.9%), trichoroacetic acid (TCA), 2,4-dinitro-phenylhydrazine (DNPH), guanidine, potassium phosphate monobasic (KH₂PO₄), Bradford reagent, ammonium sulfamate, mercuric chloride (HgCl₂), sulfanilamide, N-1-naphthylethylendiamine, S-nitrosoglutathione, phosphate-buffered saline (PBS), potassium iodide, chloramine-T, GSH, GSSG, HXT, gallic acid, α-tocopherol, and tocopheryl acetate were purchased from Sigma–Aldrich (Saint Louis, MI). Chloridric acid (HCl), sulfuric acid (H₂SO₄), meta-phosphoric acid, phosphoric acid (H₃PO₄), glacial acetic acid, ethyl acetate, and ethanol were purchased from Merck (Darmstadt, Germany). Trifluoroacetic acid (TFA), acetonitrile, n-heptane, tetrahydrofuran, and glacial acetic acid were high-performance liquid chromatography (HPLC) grade and were purchased from Merck. Other reagents and solvents were of analytical grade.

Analysis of protein carbonyl groups

Briefly, 100 μL of serum diluted 1:10 with physiological saline solution was added with 20% TCA and centrifuged at 13,000 g for 2 min at 4°C. The protein pellet was incubated with 100 μL of 10 mM DNPH (HCl 2 N) for 60 min at room temperature (RT) vortex mixing every 15 min. Then, the proteins were precipitated by adding 20% TCA solution, vortex mixed, and centrifuged as described previously. The protein pellet was washed three times with 200 μL of ethyl acetate/ethanol solution (1:1, v/v); then, it was resuspended in 120 μL of guanidine solution 6 M (KH₂PO₄ 20 mM, pH 2.3 by TFA) at 37°C for 15 min.

Sample solution was brought to RT, centrifuged at 13,000 g for 5 min at 4°C, and the supernatant read at 375 nm by a Multiskan™ GO reader plate (Thermo Fisher Scientific, Waltham, MA) against a blank (physiological saline solution instead of sample) as control. The serum concentration of carbonyl groups was quantified by the following equation: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} C \left( {nmol / well} \right) = ( A / 6.364 ) \; \times V , \end{align*} \end{document}

where A is the absorbance of sample at 375 nm, 6.364 is the mM extinction coefficient (ɛ^mM  = 22 mM ⁻¹ cm⁻¹) considering the optical path length (0.2893 cm), and V is the volume of sample within the well.

Results were normalized to the total protein amount determined by the Bradford reagent and expressed in nanomoles (nmol) of carbonyl groups per milligram of protein by the following equation: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} CP \left( {nmols / mg \;protein} \right) = \left( {C / P} \right) \times D , \end{align*} \end{document}

where C is the nanomoles per well, P is the total protein amount, and D is the sample dilution factor.

Analysis of SNOPs

Serum sample (60 μL) was pipetted directly into a multiwell plate and 50 μL of ammonium sulfamate solution 0.5% was added. After incubation time (3 min), 40 μL of HgCl₂/sulfanilamide solution (0.25% and 2.55% v/v, respectively) and 40 μL of N-1-naphthylethylendiamine 0.38% in HCl 0.4 N were added. The absorbance was read at 540 nm by a Multiskan GO reader plate (Thermo Fisher Scientific) after an incubation time of 5 min.

SNOPs were quantified by using a standard calibration curve of S-nitrosoglutathione solubilized in H₂SO₄ 0.2 N (12.5–100 μM). Results were normalized to the total protein amount determined by the Bradford reagent and expressed in nanomoles of carbonyl groups per milligram of protein.

Analysis of AGEs and AOPPs

Determination of AGE and AOPP was carried out by spectrofluorimetric detection.

AGE determination was carried out on serum diluted 1:5 with PBS (pH 7.4). The fluorescence intensity was recorded (λ_em 440 nm, λ_ex 350 nm), and results are expressed in arbitrary units per gram of protein (AU/g protein).

For AOPP determination, 10 μL of KI (1.16 M) and 20 μL of glacial acetic acid were added to 200 μL of serum diluted 1:5 with PBS 20 mM. The absorbance was read immediately at 340 nm. Chloramine-T (12.5–100 μM) was used as reference compound while PBS was used, instead of sample, as negative control. AOPP concentration was expressed in nanomoles of Chloramine-T per milligram of protein.

Analysis of GSH and GSSG

The GSH and GSSG simultaneous quantification in plasma samples was carried out by solid-phase extraction (SPE) followed by HPLC-DAD (HPLC with diode array detection). Briefly, 400 μL of meta-phosphoric acid was added to 200 μL of plasma sample and incubated at RT for 15 min. Sample was then centrifuged at 10,000 g for 5 min at 4°C and the supernatant processed by SPE using a BOND Elut C18 cartridge 50 mg/3 mL (Agilent Technologies, Santa Clara, CA) pre-conditioned with methanol (2 mL). The elution was carried out with TFA 0.1% (1 mL). Sample was filtered using a 0.22 μm nylon syringe filter and injected into an Agilent 1100 series HPLC system (Santa Clara, CA). The separation was achieved on Prodigy™ 5 μm ODS-3 100 Å, LC Column 250 × 4.6 mm (Phenomenex, Torrance, CA) by isocratic elution (30 min) using TFA 0.1%/acetonitrile (95:5, v/v) as mobile phase. The flow rate was 1 mL/min, and the detection wavelength was set at 215 nm. The column oven was kept at RT while the auto-sampler temperature was maintained at 4°C. The quali-quantitative analysis of glutathione derivatives was carried out by comparing retention times and UV-visible spectra (UV-VIS) (range 190–400 nm) and using external standard calibration curves (5–40 μg/mL) of the commercial available reference compounds (GSH and GSSG) solubilized in mobile phase.

Analysis of bioactive compounds HXT and α-tocopherol

The SPE and analytical evaluation of HXT in plasma samples was carried out according to Colica et al. (1) with some modifications. Briefly, 10 μL of 8.5% H₃PO₄ and 10 μL of gallic acid methanol solution 1 mg/mL (internal standard) were added to 100 μL of plasma sample, which was passed through a Strata™-X 33 μm Polymeric Reversed Phase, 30 mg/mL (Phenomenex), previously activated with 1 mL of methanol followed by 1 mL of deionized water. After a washing step with 500 μL of deionized water, the HXT was eluted with 1.5 mL of 5% methanol solution, filtered using a 0.22 μm nylon syringe filter, and injected into the HPLC system.

The chromatographic separation was carried out by Eclipse plus C18 column (150 × 4.6 mm, 5 μm) (Agilent Technologies) using (A) CH₃COOH 0.2% (pH 3.1) and (B) methanol, as mobile phase, according to the following gradient elution: 0–2 min 90% B, 2–10 min 60% B, 10–13 min 90% B, 13–15 min 90% B. The injection volume was 20 μL, the flow rate was set at 1.5 mL/min, and the auto-sampler temperature was maintained at 4°C.

HXT was recognized by comparing the retention time and relative UV-VIS (range 200–400 nm) with that of the commercial available standard and quantified at 280 nm by an external calibration curve (3.125–50 μg/mL). Results are expressed in milligrams of HXT per milliliter of plasma sample.

The α-tocopherol determination was carried out by liquid–liquid extraction followed by HPLC with fluorescence detection analysis. Briefly, 10 μL of α-tocopheryl acetate 0.4 mg/mL in ethanol (internal standard) was added to 200 μL of plasma sample, vortex mixing for 10 min. After that, 200 μL of cold ethanol was added, vortex mixing for 1 min, and the sample extracted with 500 μL of cold hexane solution containing butylated hydroxytoluene 1 mg/mL. The sample was then vortex mixed for 5 min and centrifuged at 12,000 g for 5 min at 4°C. The above extraction procedure was repeated two times and the combined hexane layers were brought to dryness by nitrogen. Sample was resuspended with 200 μL of cold ethanol, vortex mixed for 1 min, filtered using a 0.22 μm nylon syringe filter, and injected into HPLC system.

The chromatographic elution was performed on a LiChrosorb^® 5 μm Si 60 Å, LC Column 250 × 4 mm (Phenomenex) maintained at 25°C, using a mobile phase consisting of n-heptane/tetrahydrofuran mixture (96.15/3.85, v/v). The injection volume was 10 μL, the flow rate was set at 1.0 mL/min, and the auto-sampler temperature was maintained at 4°C. The acquisition was set at λ_em 330 nm and λ_ex 295 nm.

The quantification was carried out by external calibration curve of the commercially available standard (0.5–8.0 μg/mL), and results are expressed in micromoles of α-tocopherol per liter of plasma sample.

All sample preparation steps for both bioactive compounds were carried out in the dark, using burnished glass vials, and maintaining a temperature of ≤4°C to avoid any analytes degradation.

Statistical analyses

Statistical analyses were undertaken using STATISTICA software (version 2010, Chicago, IL). The Kolmogorov–Smirnov test was used to determine whether variables were normally distributed. Normally distributed data are expressed in means ± standard deviations, and nonnormally distributed data are expressed in medians and interquartile ranges. Data were analyzed using the intention-to-treat principle, and the values recorded at baseline were compared with values recorded at 4 months in all patients, regardless of treatment duration. Baseline and follow-up characteristics were tested for differences by Student's t-test (p < 0.05). The change of anthropometrical and laboratory values, between placebo and treatment groups, was evaluated using analysis of variance with repeated measures. Difference between proportions was tested using the chi-square test. Pearson's correlation test was used to evaluate a possible correlation between biochemical variables and histological patterns in NAFLD patients.