Introduction

Vitamins are a vital group of food components that must be provided to the human body in sufficient amounts. They support metabolism processes and improve the efficiency of proteins and enzymes. Vitamins are provided in final form or as provitamins that are subsequently transformed into vitamins. The main sources of vitamins are fruits, vegetables, meats, and fish. However, when acquisition from food is insufficient, the vitamins can be acquired from pharmaceutical preparations. Currently, the increased interest in a balanced and wholesome diet has caused increased demand for multivitamin diet supplements.

Vitamins are divided into two groups based on their solubility. The fat-soluble vitamins A, D3, E, and K are non-polar, hydrophobic compounds that are isoprene derivatives. They are absorbed by the human organism also with fat and can be stored in liver, kidneys, and fat tissue. Fat-soluble vitamins perform essential roles in the regulation of height, calcium and phosphorus absorption, and the development and functionality of bones. They also act as anticoagulants. The water-soluble vitamin group contains vitamin C (ascorbic acid), B1 (thiamine), B2 (riboflavin), B3 (nicotinamide), B5 (pantothenic acid), B6 (pyridoxine), B9 (folic acid), and B12 (cyanocobalamin). They have different chemical structures because they represent acid, pyrimidine, and imidazole derivatives as well as acid amides. They are responsible for the proper functioning of the nervous and respiratory systems, synthesis of nucleic and fatty acids, and creation of red blood cells (National Academy of Sciences 1989; Sikorski 2007). Lack or deficiency of vitamins in consumed food can lead to deficiency states and diseases.

A widely used analytical method for the determination of vitamins in food and diet supplements is chromatography, especially high-performance liquid chromatography (HPLC). HPLC is described by high resolution and selectivity, short time of analysis, and ability to couple with other techniques. Literature publications describe numerous methods for vitamin determination; however, particular procedures allow only determining one, two, or several vitamins at the same time, but only from either the fat- or water-soluble group. Table 1 shows parameters of chromatographic procedures for the determination of fat- and water-soluble vitamins. Literature data allow the selection of a chromatographic procedure that enables the determination of vitamins in food samples with adequate concentration levels with the use of both gradient and isocratic systems. The disadvantage of the described procedures is the small number of analytes that can be determined at the same time. Most of the methods allow the determination of several vitamins from one group at once. Most often, it is vitamin A together with E or three to four vitamins of the B group. Only in single particular cases does a chromatographic procedure allow for the simultaneous separation of several compounds of both fat- and water-soluble groups without requirement of stationary or mobile phase changes.

Table 1 Procedures for the selected chromatographic determination methods of vitamins in foodstuff
Full size table

Another issue is proper food sample preparation, which would allow the simultaneous extraction of all analytes from a matrix. Solid–liquid and solid phase extraction are the most often used extraction techniques. Most of the procedures described in the literature concern single vitamins of water-soluble vitamins (Hussein et al. 2000; Kurilich et al. 1999; Wang et al. 1996; Franke et al. 2004; Frenich et al. 2005; Gil et al. 2006; Mahattanatawee et al. 2006; Wall 2006; Singh et al. 2007; Lopez-Berenguer et al. 2009) and fat-soluble vitamins (Kurilich et al. 1999; Mahattanatawee et al. 2006; Singh et al. 2007; Ching and Mohamed 2001). Only two papers (Ndaw et al. 2000; Lebiedzińska and Szefer 2006) describe procedures that allow the simultaneous extraction of three to four vitamins from one group. Other studies (Moreno and Salvadó 2000; Munzuroglu et al. 2003) show that it is possible to extract vitamins from both groups; however, this requires a few additional changes in the procedures, like adding a secondary solvent.

Our study presents results of the development of an analytical procedure for the simultaneous extraction as well as simultaneous chromatographic determination of fat- and water-soluble vitamins. After an examination of standard samples, application to real samples was performed. The developed method allows reduction in time for the analysis of fruits, vegetables, and juices, as well as multivitamin diet supplements.

Materials and Methods

Reagents

Standard solutions (10 mg/mL) of retinol (A), cholecalciferol (D3), α-tocopherol (E), thiamine (B1), riboflavin (B2), nicotinamide (B3), pantothenic acid (B5), pyridoxine (B6), folic acid (B9), cyanocobalamin (B12), and ascorbic acid (C) (all from Sigma, St. Louis, MO, USA) were prepared in methanol. Methanol, trifluoroacetic acid (TFA), and water of HPLC grade used for the mobile phase were purchased from Merck (Darmstadt, Germany). Acetic acid 99%, zinc acetate, and potassium ferricyanide (all from POCH, Gliwice, Poland) of analytical grade were used. Standard solutions were prepared directly before use.

High-Performance Liquid Chromatography Conditions

HPLC analysis was performed using a Merck-Hitachi chromatograph equipped with an L6200A pump, L-7360 thermostat, and L4500A diode array detector. Chromatographic separation was carried out on a TSKGel ODS-100V, 150 × 4.6 mm (5 μm) column (Tosoh Bioscience). Examination was performed at a temperature of 30 °C. A gradient elution with 0.01% TFA in water (A) and methanol (B) was applied. The gradient elution parameters are shown in Table 2.

Table 2 Gradient elution parameters
Full size table

Detection of vitamins was performed using a diode array detector (DAD) with analytical wavelengths typical for each analyte: retinol, 320 nm; cholecalciferol, 275 nm; thiamine, 253 nm; riboflavin, α-tocopherol, and cyanocobalamin, 290 nm; nicotinamide, 258 nm; pantothenic acid, 218 nm; pyridoxine, 289 nm; folic acid, 360 nm; and ascorbic acid, 262 nm.

Calibration Curves

The calibration curves for the examined compounds were prepared in a matrix of methanolic food extract. Matrix of food samples was purified by solid phase extraction with the same parameters as the extraction of analytes (see “Sample Preparation”). The effluent formed during sample passing through solid phase was collected. Then, appropriate amounts of standard solutions were added.

Curves were determined in a concentration range appropriate for the expected content of analytes in real food samples. The number of experimental points taken for regression was n = 6. Every analyte was injected three times. The volume of each injection was 20 μL. Values of standard deviation of slope and intercept for calibration curves were calculated.

Carrez I and Carrez II Solution Preparation

Carrez I solution was prepared by dissolving 10.6 g of potassium ferricyanide (K4Fe(CN)6⋅3H2O) in 100 mL of distilled water. Carrez II solution was prepared by dissolving 21.95 g of zinc acetate (Zn(CH3COO)2⋅2H2O) and 30 mL of 99% acetic acid in 100 mL of distilled water.

Sample Preparation

Samples were taken from fruits (apricot, avocado), fruit–vegetable juices, and multivitamin diet supplements (powders and tablets). Apricots and avocados were peeled and cut into small pieces. Two grams for each product was weighed and then intermixed with 4 mL of methanol. Subsequently, samples were centrifuged for 5 min (2,000 rpm). The examined material was filtered using a nylon filter with a pore size of 0.45 μm.

One hundred milliliters of fruit–vegetable juice was taken and then 10 mL of Carrez I and 10 mL of Carrez II solutions were added for sedimentation and subsequently centrifuged for 10 min (7,000 rpm). The examined material was filtered using a nylon filter with a pore size of 0.45 μm, and then solid phase extraction (SPE) was performed. A RP18 Bakerbond column (500 mg, 3 mL) was used. The column was conditioned with methanol (1 × 3 mL) and water (1 × 3 mL). Analytes were eluted with portions of the solvents: 3 mL of 60% methanol in water, 3 mL of methanol, and 3 mL of chloroform. The SPE procedure was based on vitamins from multivitamin preparations extraction method (Kozhanova et al. 2002).

Two grams of diet supplements (tablet or powder) were ground, diluted in methanol, and filtered using a nylon filter with a pore size of 0.2 μm. Volumes of methanol used for dilution were matched with declared contents of analytes in the supplement. The volume of the filtered solution for every injection was 20 μL. This volume was taken three consecutive times from each analyzed sample. Matrices for standards which were used for calibration curves were prepared in the same manner as the samples.

Results and Discussion

The developed chromatographic system allows the simultaneous determination of 11 vitamins (both water- and fat-soluble). Dissociation of analytes was determined by a pK a value for each compound and mobile phase pH value. Thus, a solution of TFA with the proper pH value was used. The pH value of the mobile phase was chosen with pK a values of particular analytes taken into consideration to allow analyte separation. The pH value for trifluoroacetic acid with concentration of 0.01% is equal to 3. At the start of the mobile phase, gradient proportion of this acid to methanol is 95:5 (v/v).

With regard to the non-polar stationary phase, B1 and C vitamins eluted first from the chromatographic column. The pK a values of those compounds are 3.8 and 4.2, respectively. Applying a 0.01% solution of TFA allowed increasing the affinity of the compounds to the stationary phase in the initial part of the analysis. It allowed one to determine and elute outside of the dwell time of vitamin B1 and C. Both flow rate of the mobile phase and temperature of the stationary phase column were also subject to examinations. Optimal analytical conditions were obtained at a temperature of 30 °C. The flow rate gradient shown in Table 2 was applied. Detection was performed with the use of a diode array detector. Thanks to this, it was possible to optimize the parameters of analytical wavelengths, which allowed decreasing interferences from the matrix. Calibration curve parameters are shown in Table 3.

Table 3 Retention times and calibration curve parameters for vitamins (n = 6)
Full size table

The retention parameters obtained for particular analytes, capacity factor (k′), selectivity (α), and limits of detection (LOD) and quantification (LOQ) values, are shown in Table 4. Limits of detection and quantification values were determined from the signal-to-noise ratio on the assumption that the signal should be three times more intensive than the noise and the LOQ is three times higher than the LOD.

Table 4 Retention parameters for water- and fat-soluble vitamins
Full size table

Comparing the obtained LOD and LOQ values with the literature, it can be noticed that in the works which present the application of the same detector, the LOD and LOQ values are slightly lower or comparable (Ghorbani et al. 2004; Vidović et al. 2008; Leporatia et al. 2005). However, these chromatographic systems do not allow the simultaneous determination of such number of analytes. It should also be noted that the obtained LOD and LOQ values are sufficient for the determination of the examined compounds in fruits and juices as well as in diet supplements.

The obtained selectivity factor (α) values above 1 (Table 4) mean total separation of signals for particular vitamins. The obtained LOD and LOQ values were in the range from 0.7 to 2.9 μg in 1 mL of sample for water-soluble vitamins and from 1.85 to 15.84 μg in 1 mL of sample for fat-soluble vitamins. When comparing the obtained LOD values with those from the literature for DAD, one can see that the developed chromatographic system has suitable sensitivity and can be used for the determination of all examined compounds in the chosen food and multivitamin diet supplement samples. Figure 1a shows a chromatogram example of methanol solution of a diet supplement.

Fig. 1
figure 1

a Chromatogram of multivitamin diet supplement solution. b Chromatogram of apricot extract. c Chromatogram of fruit–vegetable juice extract

Full size image

Procedures for the simultaneous extraction of both fat-soluble and water-soluble vitamins from fruit and juice samples were developed. Fresh fruits were purchased in stores in Poland, multivitamin juice came from the domestic market, and diet supplements were purchased from pharmacies. For analytes, the extraction from solid matrices, solid–liquid extraction was used. Parameters of this extraction are given above (“Sample Preparation”). Examinations included apricot and avocado because they are rich in fat-soluble and water-soluble vitamins. Avocado samples contained nine vitamins: vitamin C, five vitamins of group B (B1, B3, B6, B9, B2), and A, D3, and E vitamins. Apricot samples additionally contained vitamin B12.

Samples of fruit–vegetable juices were prepared according to the procedure specified above (“Sample Preparation”). Solid phase extraction procedure allowed extracting all eight vitamins (water-soluble C, B1, B3, B6, B9, and B2 as well as fat-soluble A and E), which were present in the juices. Figure 1b, c shows a chromatogram example of apricot extract and a chromatogram of juice extract, respectively. The presence of analytes, in particular with real samples, was confirmed by comparing the absorption spectra in the range of 200–600 nm and adding standard solutions. Signals without labels are derived from the matrix.

Method precision was examined by analysis of recoveries from apricot, avocado, and juice matrices. The procedure for determining the value of recoveries from the solid matrix consisted of the addition of all the vitamin standards to samples of fresh fruit and then extraction with methanol. For juice samples, standard solutions were added to portions of juices, thoroughly mixed, and then representative samples were collected. Afterwards, Carrez solutions were added; the procedure was the same as for real samples. Simultaneously, whole analytical procedure for “sample without added standards” was conducted. A “sample without added standards” was a real sample of the product for which no standards solutions were added as well as no purification from the analytes was made.

The applied solid–liquid extraction (SLE) procedure achieved recoveries of water-soluble vitamins from 78% to 93% for fruit matrices. The SPE procedure achieved recoveries of water-soluble vitamins from 83% to 94% for juices matrices. For fat-soluble vitamins, recoveries were from 80% to 90% for fruits (SLE) and from 86% to 95% for multivitamin juices (SPE). Recoveries are shown in Tables 5 and 6.

Table 5 Precision for vitamins from fruit matrix using SLE
Full size table
Table 6 Precision for vitamins from juice matrix using SPE
Full size table

Analyte contents determined in real samples of fruit, juice, and multivitamin preparation are presented in Table 7. Additionally, for the multivitamin preparation, this table contains the manufacturer’s declared values. The obtained results show that the developed chromatographic procedure can be used for the simultaneous separation and determination of water- and fat-soluble vitamins in fruits, juices, and diet supplements. Problems with the determination of B12 and D3 vitamins on the levels of concentration that are present in the juices and diet supplements can be eliminated by concentrating the sample before chromatographic analysis.

Table 7 Contents of water- and fat-soluble vitamins in real samples of fruits, juice, and multivitamin preparation
Full size table

Conclusion

The developed chromatographic system allows the simultaneous analysis of 11 vitamins from both groups (water- and fat-soluble) in various food samples. The procedures of sample preparation are rapid and easy to perform. They allow the simultaneous extraction of all analytes and show high recovery values. Altogether, the developed chromatographic system along with the sample preparation procedures can be routinely used in food analysis. It should be emphasized that a chromatographic system together with a sample preparation procedure allows the simultaneous determination of all examined compounds in a very short time, which has never been described before.