Determination of Retinol , αα-Tocopherol , Lycopene , and ββ-Carotene in Human Plasma Using HPLC with UV-Vis Detection : Application to a Clinical Study

A method is described here for the simultaneous determination of retinol, αα-tocopherol, lycopene, and ββ-carotene in human plasma. e effectiveness of various protein precipitants and extraction solvents was tested. Aer adequate sample preparation, the samples were injected directly into the HPLC system. e separation was realized on an analytical reversed-phase column with a UV-Vis detection. e analytical performance of this method was satisfactory. e intraassay and interassay coefficients of variation were below 10%. e recoveries were as follows: 97.0% (CV 2.4%) for retinol, 94.6% (CV 1.7%) for αα-tocopherol, 91.9% (CV 3.6%) for lycopene, and 93.9% (CV 4.2%) for ββ-carotene. e levels of selected fat-soluble vitamins in plasma of patients with cardiovascular disease were measured and discussed.


Introduction
Fat-soluble vitamins such as -tocopherol, -carotene, and lycopene are very important antioxidants [1][2][3].Reduced levels of these antioxidants in plasma have been reported to be associated with numerous diseases [3][4][5][6].Association between high plasma levels of fat-soluble antioxidant vitamins and lower risk of atherosclerosis was described in several studies [6].Vitamin E is constituted by a group of eight isomers that include both tocopherols and tocotrienols.-Tocopherol is the most biologically active form of vitamin E [7].Lycopene is a powerful quencher of singlet oxygen that produced during exposure to ultraviolet light [8,9].-Carotene, the main dietary source of Vitamin A, is also a quencher of singlet oxygen [10,11].Numerous methods have been described for the analysis of fat-soluble antioxidant vitamins in various biological matrices.Up to date, several reversed-phase high-performance liquid chromatographic (RP-HPLC) methods using C 18 or C 30 stationary phase coupled with spectrophotometric, �uorometric, electrochemical, or mass spectrometric detection were developed [1,2,[12][13][14][15][16]. e instability of fat-soluble antioxidant vitamins during sample storage and preparation is the major problem at their measurement.erefore, sample preparation is essential for accurate analysis of such compounds.While -tocopherol together with retinol are relative stable, -carotene and lycopene are easily subject to degradation.Degradation of lycopene and -carotene is in�uenced by temperature, light, and dissolved oxygen [17][18][19].
e aim of this study was to develop a reliable RP-HPLC method for the measurement of fat-soluble antioxidant vitamins in human plasma and to prevent their degradation during sample preparation.

Instrumentation.
Chromatographic analysis was performed with a liquid chromatograph (Ecom, Prague, Czech Republic) equipped with a LCP 4100 solvent delivery system, an AS 54 autosampler, a LCO 101 column oven, and a LCD 2084 variable wavelength spectrophotometric detector.Data were collected digitally using Clarity chromatography soware (DataApex, Prague, Czech Republic).Spectrophotometric analyses were carried out on a Shimadzu (Kyoto, Japan) UV-1700 PharmaSpec spectrophotometer.

Subjects.
A total of 129 patients that diagnosed with nonacute coronary angiography for chest pain (40 women in the age 61 ± 7 years and 89 men in the age 59 ± 7 years) were included in this study.All patients gave written informed consent to participate in this research study, which was approved by the Hospital Committee on Human Research (Regional Hospital of Pardubice, Czech Republic, protocol 336/2010) according to the Helsinki Declaration.

Patients' Inclusion and Exclusion
Criteria.Criteria for the submission into the study were (1) clinically signi�cant coronary stenosis without subsequent percutaneous coronary intervention (PCI) treatment (coronary stenosis of >50% of the le main coronary artery or >70% of the epicardial coronary arteries without PCI treatment); (2) stenosis with PCI treatment (stenosis of the same grade with subsequent PCI treatment); (3) patients without any stenosis (symptoms imitating coronary artery disease with normal angiographic �nding and no stenosis).All participants underwent angiography and completed a questionnaire.Patients who had any serious health complications and patients with increased high-sensitivity C-reactive protein levels (>10 mg/L) were excluded.None of the studied subjects exhibited renal, hepatic, gastrointestinal, pulmonary, or oncological diseases.
2.6.Blood Samples Collection.Venous blood samples were obtained under standard conditions, from 7 to 8 a.m.aer fasting for at least 12 hours the day aer coronarographic examination.Blood was collected into tubes with EDTA (the Vacuette Detection Tube, No. 454246, Greiner Labortechnik Co., Kremsmünster, Austria) that covered with an aluminum foil to minimize exposure of blood samples to daylight.Plasma was separated from blood cells by centrifugation (1700 ×g, 15 min, 8 ∘ C) and immediately stored at −80 ∘ C in 1.5-mL amber polypropylene tubes.

Sample Preparation.
Retinyl acetate and -tocopheryl acetate were used as internal standards.ey are not present in human plasma and their chromatographic properties are quite similar.On the other hand, retinyl palmitate and -apo-8 ′ -carotenal were detected in most of the samples.
All procedures were performed in a darkened room.
For the analysis of blood plasma, 10 L of the internal standard of retinyl acetate solution in ethanol (≈40 mol/L; 13 g/mL) and 10 L of the internal standard -tocopheryl acetate solution in ethanol (≈2 mmol/L; 945 g/mL) were pipetted into a well-capped 1.5-mL amber polypropylene tube.200 L of plasma were added and the content mixed vigorously on a vortex mixer for 10 min.200 L of cold ethanol were added and the solution was vortexed for 60 s.500 L of cold n-hexane containing butylated hydroxytoluene (1 g/L) were then added, the solution was vortexed for 5 min, and centrifuged (22 000 ×g, 5 min, 4 ∘ C). e upper hexane layer was transferred into 5-mL amber glass tubes.e hexane extraction process was repeated two times.Combined hexane extracts were evaporated to dryness, under nitrogen, at 4 ∘ C (Linde Gas, Prague, Czech Republic).e dried residue was resuspended in 200 L of cold ethanol and vortexed for 60 s.e sample was then �ltered through a nylon �lter (pore size 0.20 m, 4 mm diameter, Supelco, Bellefonte, PA, USA), transferred into 0.2-mL cramped amber vial, and purged with pure nitrogen for 10 s.
e stock solutions of fat-soluble vitamins were diluted with ethanol (retinol, -tocopherol) or n-hexane (lycopene, -carotene) to give a series of mixed working standards.To 200 L of mixed solution of standards, 10 L of the internal standard of retinyl acetate solution in ethanol (≈40 mol/L; 13 g/mL), 10 L of the internal standard -tocopheryl acetate solution in ethanol (≈2 mmol/L; 945 g/mL), 200 L of deionized water, and 500 L of cold n-hexane containing butylated hydroxytoluene (1 g/L) were carefully added.Prepared standards were subjected to the same procedure as described above for plasma samples.
For recovery experiments, 10 L of mixed solution of individual vitamins at different concentrations was added to 190 L of plasma.e next steps were the same as for plasma sample preparation.
2.8.Chromatographic Analysis.e chromatographic analysis of selected fat-soluble vitamins was accomplished using an isocratic elution on a Discovery HS C 18 , 150 × 4 mm i.d., 5 m analytical column that �tted with a Discovery C 18 , 20 × 4 mm i.d., 5 m guard column (Supelco, Bellefonte, PA, USA) at 40 ∘ C. e mobile phase was a mixture of methanol and ethanol (75 : 25, v/v).e �ow rate was kept constant at 0.8 mL/min.Optimum response of -tocopherol and -tocopheryl acetate was observed when wavelength was set at 292 nm, while retinol together with retinyl acetate were monitored at 325 nm, -carotene at 450 nm, and lycopene at 468 nm.e amount of individual vitamins was quanti�ed from the corresponding peak area ratio of vitamin/internal standard (-tocopherol/-tocopheryl acetate; retinol, lycopene, and -carotene/retinyl acetate) using Clarity chromatography soware (DataApex).e concentration of each analyte in the samples was determined from its calibration curve.
2.9.Statistical Analysis.e data are presented as mean ± S.D. Differences between studied groups were analyzed using the Student's -test.e correlation analysis was carried out using Spearman Rank Order Correlation, and regression analysis was carried out using the least squares method (soware QCexpert, Trilobyte, Pardubice, Czech Republic).A    value was considered statistically signi�cant.

Sample Preparation, Stability.
Sample preparation is essential for accurate analysis.e stability of selected fat-soluble vitamins in human plasma during the sample treatment was investigated.Several protein precipitants were tested.Protein precipitant of the different temperature (methanol, acetonitrile, ethanol, 1-propanol, and 2propanol) was carefully added to plasma.en n-hexane containing or not an antioxidant (butylated hydroxytoluene) was carefully added.Fat-soluble vitamins and internal standards in plasma samples were found to be stable at 4 ∘ C for at least 8 h under these conditions: a protein precipitation with ethanol, cooled to −20 ∘ C; extraction with n-hexane containing butylated hydroxytoluene, cooled to −20 ∘ C; centrifugation at 4 ∘ C; evaporation to dryness, under pure nitrogen, at 4 ∘ C; resuspension of a residue in ethanol, cooled to −20 ∘ C. It is very important that extracts were to be dried as quickly as possible.Fat-soluble vitamins, especially lycopene, are not stable in n-hexane.Fat-soluble vitamins in human plasma samples that stored at −80 ∘ C are stable for at least one year.It is necessary that all procedures were to be performed in a darkened room.

High-Performance Liquid Chromatographic Assay of
Fat-Soluble Vitamins.Fat-soluble vitamins were separated on a reversed-phase column using an isocratic system of methanol and ethanol.e mobile phase was optimized in order to obtain the best separation of the analytes in the shortest time.Standard solutions of fat-soluble vitamins and internal standards as well as pooled plasma samples were used for studying the mobile phase composition.Several eluents (mixtures of organic solvents such as acetonitrile, methanol, ethanol, 1-propanol, 2-propanol, and n-hexane) and several gradients were assessed.e best results were obtained for the conditions that described in "Chromatographic Analysis." Column temperature was changed from 25 to 45 ∘ C. Optimal temperature interval was from 40 to 45 ∘ C. e criteria were resolution, stability of the absorbance, and analysis duration.According to our results, we can conclude that the presented method is highly robust.HPLC chromatograms of fat-soluble vitamins in standard solution and human plasma are shown in Figures 1 and 2. e precision of fat-soluble vitamin analysis in plasma samples is shown in Table 1.To determine the same-day precision, the plasma samples were analyzed ten times in the same day under the same conditions.Similarly, data on the betweenday precision were obtained using the same plasma samples, analyzed on ten different days.e coefficients of variation were below 10%.e spike recoveries ranged between 93.8 and 99.4% for retinol, 92.2-95.9%for -tocopherol, 87.7-95.8%for lycopene, and 88.6-98.8%for -carotene (Figure 3).Calibration curves (9-point for determining analytical parameters and 7-point for routine analysis) were linear over the whole tested range (Figure 4).e calibration curve parameters obtained as an average from ten standard curves are shown in Table 2. e lowest concentration that could be quanti�ed with acceptable accuracy and precision was 0.  (0.46 pmol/inject) for -carotene.Furthermore, the limits of detection for retinol, -tocopherol, lycopene, and -carotene, de�ned as a signal-to-noise (S/�) ratio of 3 : 1, were 0.03 mol/L (0.7 pmol/inject), 0.63 mol/L (13.9 pmol/inject), 0.006 mol/L (0.13 pmol/inject), and 0.006 mol/L (0.13 pmol/inject), respectively.

Determination of Selected Fat-Soluble Vitamins in Human
Plasma of Patients with Cardiovascular Diseases.e role of retinol, -tocopherol, and carotenoids in both physiological and pathological processes has been widely discussed.
Many of the biological actions of these fat-soluble vitamins have been attributed to their antioxidant properties.Several
). T 1: Precision of retinol, -tocopherol, lycopene, and -carotene in human plasma.1 Nine-point for the determination of analytical parameters and seven-point for routine analysis.2e-intercept (in mol/L) is the point at which the line crosses the -axis (where the  value equals 0).