Journal of Chromatographic Science, Vol.

46, April 2008

Simultaneous Determination of Five Fat-Soluble

Vitamins in Feed by High-Performance Liquid
Chromatography Following Solid-Phase Extraction

Downloaded from https://academic.oup.com/chromsci/article-abstract/46/4/345/312729 by guest on 30 November 2018

Xiuping Xue, Jinming You, and Pingli He*
College of Animal Science and Technology, China Agricultural University, No. 2 Yuanmingyuan West Road, Beijing, 100094, P.R. China

Abstract (9,10). Recently, a single vitamin analysis (11–14) has been

developed. Rusing et al. (15) reported a simultaneous analytical
A high-performance liquid chromatography method is developed method for vitamins A and E in rodent feed by HPLC. In 1993, a
for the simultaneous determination of menadione, retinyl acetate, simultaneous HPLC method to determine vitamins K1, K2, and
cholecalciferol, α-tocopherol, and α-tocopherol acetate in feed. K3 in animal feed was developed (16). In addition, HPLC has been
The present study uses an enzyme to destroy the coating film, used for the simultaneous determination of fat-soluble vitamins
ethanol to extract free vitamins, and Oasis HLB cartridges to purify. A, D, and E in animal feed and foodstuff (17). However, the simul-
Vitamins are separated using an Atlantis dC18 column. The mobile taneous determination of five vitamins (A, D3, E, E-ac, and K3) in
phase is methanol–water (98:2 v/v). Detection is performed with a feed is rarely reported.
UV–vis detector at 230 and 265 nm. The linearity, accuracy, and
The previously mentioned methods mainly consist of the
repeatability of this method are all satisfactory. Application of the
method is suitable for the determination of the fat-soluble vitamins
process of sample saponification, liquid–liquid extraction, and
in general feed. HPLC analysis. In them, the regular alkaline saponification pro-
cedure readily causes the oxidation of vitamins because vitamins
are unstable under the experimental conditions. Moreover, it has
the risk of large variation, low recovery and reproducibility, and
Introduction is also a time-consuming and complicated operation. Therefore,
Qian et al. (18) reported a new method of one-step extraction
Biologically active vitamins are a class of low molecular weight instead of the traditional saponification for the simultaneous
organic compounds which maintain normal physiological func- determination of vitamins A, D, E, and pro-vitamin D2 in animal
tions for animal organisms, containing both the characteristics feeds. However, concerning the sample preparation, the method
of water- and fat-solubility. Fat-soluble vitamins play an impor- recommended that the whole determination process should be
tant role in the stimulation of synthesis and degradation of nutri- accomplished in a short period and under gentle operating con-
ents, enhancing immune function and growth performance. In ditions. Recently, enzymolysis was developed as an alternative to
vitamin deficiency, animals tend to develop disease (1). However, saponification; for example, a lipase-catalyzed reaction was used
an overdose of vitamins results in poisoning, evident in animals’ to determine the fat-soluble vitamins in milk power and infant
legs (2). Although some ruminants and grazing monogastric formula (19). This method with hydrolytic enzymes is rarely
animals can obtain autonomically vitamins B and K from bac- reported for vitamin detection in animal feed.
teria, most animals passively acquire vitamins from feed intake. Another challenge for the detection of vitamins in feed is
Currently, feed supplemented with vitamin additive is growing in multiple interference from the components in feed, which leads
popularity, which requires a parallel determination method for to inaccurate determination. The traditional purification
vitamin detection in animal feed. method is to extract the fat-soluble vitamins with an organic
In the past two decades, various analytical methods have been solution, such as hexane, heptane, ether, or chloroform, which
developed for this purpose, for example, capillary electrophoresis is solvent-consuming and causes incomplete extraction
(3,4), spectrophotometry (5), fluorimetry (6), colorimetry (7), (16,20,21). Solid-phase extraction is efficient enough to
and chromatography (8). Among them, determination by high- conquer this in terms of its rapidity, high efficiency, and use of
performance liquid chromatography methods (HPLC), due to its less solvent, which reasonably makes it a qualified pre-
rapid separation, high sensitivity, and accurate quantitation, has treatment of feed samples for HPLC determination, to exclude
become popular in the detection of vitamins in various matrices the interferents (22).
In this study, we developed an HPLC method with a new pre-
* Author to whom correspondence should be addressed: email hepl@mafic.ac.cn. treatment for the simultaneous determination of menadione

Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission. 345
 Journal of Chromatographic Science, Vol. 46, April 2008

(K3), retinyl acetate (A-ac), cholecalciferol (D3), α-tocopherol Chromatographic conditions

(E), and α-tocopherol acetate (E-ac) in animal feeds. Their The chromatographic conditions were chosen in terms of peak
chemical structures are shown in Figure 1. The fat-soluble vita- shape, column efficiency, retention time, resolution, and sensi-
mins were extracted with basic proteinase and ethanol, and tivity. For the separation of the fat-soluble vitamins, a Waters
purified with Oasis HLB cartridges. The five vitamins were sepa- Atlantis dC18 column (particle diameter 5 µm, 150 × 4.6 mm i.d.)
rated using reversed-phase HPLC combined with a UV–vis with a matching guard cartridge was used. The mobile phase was
detector. Application of the method is suitable for determination methanol–water (98:2, v/v). The mobile phase flow rate was 1.0
of all fat-soluble vitamins in broad animal feed. As we know, few mL/min, and the injection volume was 10 µL. The column oven
studies on the determination of vitamins in feed with proteinase temperature was 35°C. Detection with a UV–vis detector was car-
extraction and SPE purification have been reported. ried out at 230 nm for E and E-ac, and 265 nm for K3, A, and D3.

Extraction of feed sample

Downloaded from https://academic.oup.com/chromsci/article-abstract/46/4/345/312729 by guest on 30 November 2018

Firstly, the feed sample (0.1 g vitamin additive, 1–2 g multi-
Experimental vitamin, 5 g premix, and 10 g complete feed), spiked with five fat-
soluble vitamins, was weighed in a 100-mL volumetric amber
Reagents and materials flask, and the Savinase proteinases (approximately 100 mg, 200
Vitamin A-ac, E, K3, and D3 standards were purchased from mg, 500 mg, and 1 g corresponding to the different samples)
Sigma-Aldrich (St. Louis, MO), and vitamin E-ac from Supelco were added. In order to release vitamins easily, 10 mL of freshly
(Bellefonte, MA). These standards were used without further prepared 0.2% ammonia solution was added to the flask. The
purification. Basic proteinase named “Savinase” was obtained mixture was shaken in an ultrasonic bath at 40–50°C for 20–30
from Roche (Shanghai, China). HPLC-grade methanol was sup- min. Then approximately 65 mL of ethanol was added to extract
plied by Fisher Scientific International (Hampton, NH). All other the free vitamins. The solution was allowed to cool to room tem-
chemicals were analytical grade and water used in the study was perature in the dark, and the solution volume was made up to
double-deionized water (Milli-Q, Millipore Corp., Milford, MA) of 100 mL by the addition of ethanol. The mixture was shaken vig-
18.2 MΩ/cm resistivity. orously for 1 min and was centrifuged at 5000 rpm for 10 min.
Finally, the supernatant was cleaned with SPE as follows.
Apparatus and procedures
HPLC analysis was performed on an LC-10A HPLC system Solid-phase extraction
(Shimadzu, Kyoto, Japan) with an Atlantis dC18 column (4.6 mm A 1 mL aliquot of the supernatant was decanted into a 5 mL
× 150 mm; 5 µm) (Waters Corp., Dublin, Ireland). OASIS HLB tube and evaporated to near dryness in a water bath at 55°C,
cartridges (1 cc), used for purification of samples, were pur- under a stream of nitrogen. The extract was reconstituted in 1
chased from Waters Corp. The solid-phase extraction system was mL of 65% ethanol–water solution (ethanol–water, v/v), then the
a vacuum manifold processing station obtained from Agilent solution was purified by the Oasis HLB cartridge according to
Technologies (Palo Alto, CA). Ultrasonic cleaner (Kunshang, the optimal process of SPE. Firstly, the OASIS HLB column was
China) was used for promoting the sample dissolution. preconditioned by passing through 1 mL of methanol, followed
by 1 mL of double-deonized water. Secondly, 1 mL of the extract
Standard solutions solution was slowly passed through the OASIS HLB column at a
Stock standard solution (1 mg/mL) was prepared separately by flow rate of 1 mL/min. After washing with 1 mL of 5% methanol,
dissolving 10 mg of individual vitamin in 10 mL of methanol and the analyte was eluted with 1 mL of ethanol.
stored in darkness at 4°C. Working standard solutions were pre-
pared daily by methanol dilution of the stock standard solutions Determination of inter- and intra-assay variation
in appropriate propotions into the concentrations of 5 µg/mL for The mixed standards of five fat-soluble vitamins were spiked in
K3, 20 µg/mL for A-ac, 10 µg/mL for D3, 40 µg/mL for E, and 40 feed samples. Intra-assay variation was measured for five repli-
µg/mL for E-ac. cates of mixed vitamins, while for inter-assay variation, the solu-
tion was determined for seven consecutive days. Linear analysis
revealed calibration equations with the expected correlation
coefficient.

Results and Discussion

Chromatographic parameters
To obtain the best peak shape, resolution, and retention time,
the chromatographic conditions of standard solutions, such as
analytical column, mobile phase composition, flow rate, column
temperature, and detector wavelength, were optimized. For
example, several analytical columns, such as Separon C18 (150 × 3
Figure 1. Chemical structures of the five fat-soluble vitamins.

346
Journal of Chromatographic Science, Vol. 46, April 2008

mm, 7 µm) (23), Nova-Pack C18 (150 × 3.9 mm, 4 µm) (24), and gelatin beadlets. Three enzymolysis times (10, 30, and 60 min)
MetaChem Polaris C18-A (150 × 4.6 mm, 3 µm) (25) were used to were assayed. Considering the higher recoveries for most vita-
separate the water- and fat-soluble vitamins with HPLC. Here mins and faster extaction time, 30 min enzymolysis was chosen.
Atlantis dC18 was chosen as the analytical column, which seemed In accordance with the optimal bio-action of Savinase pro-
a better fit to separate the vitamins. Methanol and water were teinase, 40°C bath temperature was set. During the ultrasonic
chosen as the mobile phase. Nine ratios of methanol–water process, a variation of less than 5–10°C was allowed.
(50:50, 60:40, 80:20, 85:15, 90:10, 92:8, 95:5, 98:2, and 99:1, v/v) Three ammonia concentrations (0.02%, 0.1%, and 0.2%) and
were tested. Considering the optimal resolution, the ratio of 98:2 three ammonia volumes (10 mL, 15 mL, and 20 mL) were
was selected. Because the peak of vitamins K3 was earlier, three assayed. The results suggested that the ammonia concentration
flow rates (1.0, 0.8, and 0.6 mL/min) were tested in order to adjust did not play an important role in the method, and the volume
the retention time. The results showed that the lower flow rates was chosen according to the sample weight. The sample must be
did not delay the retention time of vitamins K3, but increased the kept in a liquid status during the whole ultrasonic process. If

Downloaded from https://academic.oup.com/chromsci/article-abstract/46/4/345/312729 by guest on 30 November 2018

retention time of vitamins E and E-ac. Thus 1 mL/min was con- ammonia volume was insufficient to wet the whole sample, the
sidered sufficient. For the detector wavelength, photodiode array extraction would be incomplete. Ethanol was then compared
detector was used to scan the wavelength from 200 to 400 nm, with methanol to extract the free vitamins, and the results
and a larger absorbance peak was observed at 230 nm for vitamins showed that higher recoveries of vitamins were obtained with
E and E-ac, and at 265 nm for vitamin K3, D3, and A-ac; then both ethanol. Moreover, ethanol has a relatively low toxicity exposure
wavelengths 230 and 265 nm were monitored. for operators. Therefore, ethanol was selected as the extractant.
A typical chromatogram of the mixture of five fat-soluble In addition, according to previous studies, most investigators
vitamins under the optimal HPLC conditions is presented in used butylated hydroxytoluene (BHT) (27) or ascorbic acid (13)
Figure 2. as the antioxidant to avoid vitamin oxidation. However, in our
study, similar results were acquired with or without BHT.
Extraction step
Vitamins used for animal feed are either contained in the ini- Purification of samples by SPE
tial raw plant material or introduced as special additives with An Oasis HLB cartridge containing poly divinylbenzene-co-N-
solid preparations. Because vitamins are unstable and readily vinylpyrrolidone sorbents was chosen as the SPE column,
oxidized, the synthetic vitamin supplements are often coated because the copolymer which exhibits both hydrophilic and
with collagen or gelatin beadlets to enhance their stability. The lipophilic retention characteristics plays a valid role in the
beadlet must be penetrated to dissolve the vitamins. Previous extraction of medium-polar and non-polar organic compounds
studies suggested that organic solvents alone did not extract vita- from mixtures of water and organic solvent (28). Thus, this
mins completely from vitamin-fortified feeds and supplements. cartridge is suitable for the clean-up of feed, including fat- or
Hung (26) and Qian et al. (18) added 1% and 5%, respectively, of water-soluble vitamins.
20% sodium phosphate tribasic solution to the organic solvent If the extraction solution containing 90% ethanol as loading
to dissolve the protective covering film. In the present work, the solution was directly transferred into the cartridge, the vitamins
basic proteinase named Savinase was chosen to disrupt the would almost run down. To optimize the SPE loading condi-
tions, various proportions of ethanol and water as the loading
solution were assessed. Ten concentrations of ethanol ranging
from 15% to 95% were assayed. The retention abilities of

Figure 2. High-performance liquid chromatograms of standard mixture of five

fat-soluble vitamins. Peaks: K3, tR = 2.59 min; A-ac, tR = 5.14 min; D3, tR =
8.57 min; E, tR = 10.50 min; E-ac, tR = 14.04 min. Chromatographic condi-
tions: mobile phase, methanol–water (98:2, v/v); flow rate, 1 mL/min; detec- Figure 3. Dependence of the retention abilities of five vitamins on the Oasis
tion wavelength, 230 nm. HLB cartridge on different concentrations of ethanol as the loading solvent.

347
 Journal of Chromatographic Science, Vol. 46, April 2008

vitamins on the HLB column under different concentrations of methanol–water were assayed as the detergent. The results
ethanol solution are presented in Figure 3. The results indicate showed no difference. The 5% methanol solution was chosen as
that all vitamins can be held on the cartridge under 45%, 55%, a common detergent. Several solvents with different polarities,
and 65% ethanol solvent. Finally, in view of better solubility for such as tetrahydrofuran, acetonitrile, ethyl acetate, acetone,
the five vitamins, 65% was selected. In addition, water and 5% cyclohexane, ethanol, and methanol, were chosen as the eluent.
Three volumes (1, 2, and 3 mL) were assayed. Taking into con-
sideration the volatility, highest recovery, and smallest volume, 1
Table I. The Step and Purpose of Solid-Phase Extraction mL of ethanol was set as eluent. Summarily, the optimal clean-
in Feed Clean-up up steps are presented in Table I. The chromatograms were com-
pared before and after an SPE-treated concentrate feed spiked
Step Operating Purpose
vitamins (Figure 4), and clearly showed the importance of the
utilization of the Oasis HLB cartridge, especially for vitamin K3.

Downloaded from https://academic.oup.com/chromsci/article-abstract/46/4/345/312729 by guest on 30 November 2018

Conditioning 1 mL MeOH + 1mL water Preparation of sorbent

Loading 1 mL of 65% ethanol–water Retention of analytes Method validation

extraction solution on cartridge The method was validated with respect to linearity and sensi-
tivity, and also precision and accuracy. The linearity was studied
Washing 1 mL of 5% MeOH Exclusion of salts and other
organic interference

Elution 1 mL ethanol Elution of fat-soluble

vitamins

Figure 5. Calibration curves of standard solutions containing five fat-soluble

vitamins.

Table II. Recovery of Fat-Soluble Vitamins Spiked in Feed

Samples

Added Measured Recovery C.V.

Vitamin (mg/g) (mg/g) (%) (%)

K3 0.01 0.013 ± 0.001 129.6 8.8

0.1 0.095 ± 0.005 94.6 4.8
1 0.876 ± 0.076 87.64 8.6
A-ac 0.1 0.092 ± 0.007 91.66 7.8
0.4 0.427 ± 0.004 106.66 1.0
2.0 1.788 ± 0.149 89.41 8.4
D3 0.02 0.021 ± 0.002 104.64 7.5
0.24 0.234 ± 0.005 97.66 2.0
1.8 1.756 ± 0.065 97.53 3.7
E 0.08 0.079 ± 0.004 99.33 4.9
0.4 0.403 ± 0.013 100.74 3.3
1.6 1.662 ± 0.048 103.88 2.9
Figure 4. High-performance liquid chromatograms of vitamins K3, A-ac, D3, E-ac 0.08 0.079 ± 0.006 99.31 7.9
E, and E-ac spiked in concentrate feed without (A) and with (B) prepurifica- 0.8 0.809 ± 0.035 101.1 4.3
tion with Oasis HLB cartridge. 2.4 2.329 ± 0.068 97.05 2.9

348
Journal of Chromatographic Science, Vol. 46, April 2008

triplicate analyses are summarized in Table III.

Table III. Declared and Determined Concentrations of Vitamins in Feed
Samples (n = 3)
Because the amount of vitamin D3 was trace in
those samples, the samples were pre-concen-
Premix feed Ferment feed trated 10 times. The results suggest that
vitamin K3 had a great deviation in the present
Declared Measured C.V. Declared Measured C.V. method and the other vitamins were well
Vitamin (mg/kg) (mg/kg) (%) (mg/kg) (mg/kg) (%)
determined.
K3 35 31.9 ± 1.6 4.9 8 9.84 ± 0.8 8.1
A-ac 258 251.0 ± 10.2 4.1 5.16 4.62 ± 0.25 5.4
D3 5 4.7 ± 0.08 1.8 0.05 ND† –
E nd* ND† – nd* ND† –
Conclusion
E-ac 5900 5906.8 ± 12.8 0.2 20 19.28 ± 0.51 2.6

Downloaded from https://academic.oup.com/chromsci/article-abstract/46/4/345/312729 by guest on 30 November 2018

This work successfully developed a new
* Not declared.
† Not found. method including the procedure of emzymol-
ysis, extraction, and purification for the simul-
taneous determination of five fat-soluble
by a series of mixed standard solutions of each vitamin at six vitamins (K3, A-ac, D3, E, and E-ac) using HPLC combined with
levels. The method was linear in the ranges as follows: from 0.26 UV detector under the double wavelength conditions. Instead of
to 26.0 µg/mL for menadione, from 2.0 to 100.0 µg/mL for saponification, emzymolysis made the operation easy and saved
retinyl acetate, from 0.8 to 86.8 µg/mL for cholecalciferol, from time. Solid-phase extraction proved to be a useful way for puri-
3.5 to 390.0 µg/mL for α-tocopherol, and from 4.5 to 213.0 fying the five fat-soluble vitamins from interferences in complex
µg/mL for α-tocopherol acetate. A satisfactory linearity was feed samples, especially for vitamin K3, which was well-separated
observed (Figure 5), and indicated the correlation coefficients from the feed matrix. Moreover, HPLC separation was excellent
(r2) ranging from 0.99 to 1. The limit of detection (LOD), defined with a retention time less than 16 min. The accuracy of this
as the compound concentrations which produced a signal-to- method was tested and obtained average recoveries ranging from
noise ratio generally greater than three, ranged from 0.075 87% to 130%, and coefficients of variation were less than 10%.
µg/mL to 1.26 µg/mL. The limit of quantitation (LOQ) of the Thus, the developed method is proficient in the determination of
assay was evaluated as the concentration greater than 10 times fat-soluble vitamins in feed samples, such as multi-vitamin sup-
the value of the signal-to-noise ratio. LOQ values ranged from plements, premixed feed, and formula feed.
0.25 µg/mL to 4.2 µg/mL.
To evaluate the precision and accuracy of the method, inter-
and intra-assay variation repeatability was estimated. The
recovery of vitamins spiked in concentrated feed was determined Acknowledgments
by five replications at the medium concentration: 5.56 µg/mL for
menadione, 20.34 µg/mL for retinyl acetate, 15.36 µg/mL for The financial support from the National Natural Science
cholecalciferol, 41.2 µg/mL for α-tocopherol, and 45.6 µg/mL for Foundation of P. R. China (30121004) and the National Basic
α-tocopherol acetate. The results indicated that recoveries of all Research Program (2004 CB 117503) are gratefully acknowl-
vitamins ranged from 91% to 103% and coefficients of variation edged.
were less than 6% for both intra-assay and inter-assay determi-
nations.

Analysis of spiked feed samples References

Recoveries from spiked feed samples were determined at three
different concentrations by comparing the peak area ratios for 1. H.A. Lehr, P. Vajkoczy, M.D. Menger, and K.E. Arfors. Do vitamin E
extracted vitamins from the feed samples and the respective supplements in diets for laboratory animals jeopardize findings in
animal models of disease? Free Radical Bio. Med. 26: 472–481
values derived from the vitamins’ calibration curves (Table II). (1999).
The statistical results show satisfied recoveries that ranged from 2. L. Waldenstedt. Nutritional factors of importance for optimal leg
87% to 130% and coefficients of variation were less than 10%. health in broilers: A review. Anim. Feed Sci. Tech. 126: 291–307
Therefore, combined with the extraction and clean-up proce- (2006).
dures, this HPLC method was ideal to accurately determine vita- 3. H.L. Shi, Y.F. Ma, J.H. Humphrey, and N.E. Craft. Determination of
vitamin A in dried human blood spots by high-performance capil-
mins in feed samples. lary electrophoresis with laser-excited fluorescence detection.
J. Chromatogr. B 665: 89–96 (1995).
Identification and determination of vitamins in real samples 4. M.M. Delgado-Zamarreño, I. González-Maza, A. Sánchez-Pérez,
This method was suitable for the simultaneous determination and R. Carabias-Martinez. Separation and simultaneous determina-
of vitamins K3, A-ac, D3, E, and E-ac in vitamin additives, multi- tion of water-soluble and fat-soluble vitamins by electrokinetic cap-
illary chromatography. J. Chromatogr. A 953: 257–262 (2002).
vitamin, premix, and some complete feed. Two feed samples were 5. E. Tütem, R. Apak, E. Günayd, and K. Sözgen. Spectrophotometric
prepared and analyzed in order to check the applicability of the determination of vitamin E (α-tocopherol) using copper (II)-
method in real feed samples. Concentrations measured after neocuproine reagent. Talanta 44: 249–255 (1997).

349
 Journal of Chromatographic Science, Vol. 46, April 2008

6. T. Pérez-Ruiz, C. Martínez-Lozano, V. Tomás, and J. Martín. Flow- 18. H. Qian and M. Sheng. Simultaneous determination of fat-soluble
injection fluorimetric determination of vitamin K1 based on a pho- vitamins A, D and E and pro-vitamin D2 in animal feeds by one-step
tochemical reaction. Talanta 50: 49–56 (1999). extraction and high-performance liquid chromatography analysis.
7. A.S. Amin. Colorimetric determination of tocopheryl acetate J. Chromatogr. A 825: 127–133 (1998).
(vitamin E) in pure form. Eur. J. Pharm. Biopharm. 51: 267–272 19. C. Turner, M. Persson, L. Mathiasson, P. Adlercreutz, and J.W. King.
(2001). Lipase-catalyzed reactions in organic and supercritical solvents:
8. O. Heudi, M.J. Trisconi, and C.J. Blake. Simultaneous quantification application to fat-soluble vitamin determination in milk powder
of Vitamins A, D3 and E in fortified infant formulae by liquid chro- and infant formula. Enzyme Microb. Tech. 29: 111–121 (2001).
matography-mass spectrometry. J. Chromatogr. A 1022: 115–123 20. L. Yakushina and A. Taranova. Rapid HPLC simultaneous determi-
(2004). nation of fat-soluble vitamins, including carotenoids, in human
9. P. Salo-Väänänen, V. Ollilainen, P. Mattila, K. Lehikoinen, serum. J. Pharm. Biomed. Anal. 13: 715–718 (1995).
E. Salmela-Mölsä, and V. Piironen. Simultaneous HPLC analysis of 21. S. Wieli ski and A. Olszanowski. Development and validation of
fat-soluble vitamins in selected animal products after small-scale HPLC method for simultaneous determination of fat-vitamins in
extraction. Food Chem. 71: 535–543 (2000). capsules. J. Liq. Chromatogr. Related Technol. 24: 201–213 (2001).

Downloaded from https://academic.oup.com/chromsci/article-abstract/46/4/345/312729 by guest on 30 November 2018

10. J.M. Mata-Granados, M.D. Luque de Castro, and J.M. Quesada. 22. P.F. Chatzimichalakis, V.F. Samanidou, and I.N. Papadoyannis.
Fully automated method for the determination of 24, 25(OH)2 and Development of a validated liquid chromatography method for the
25(OH) D3 hydroxyvitamins, and Vitamins A and E in human serum simultaneous determination of eight fat-soluble vitamins in biolog-
by HPLC. J. Pharm. Biomed. 35: 575–582 (2004). ical fluids after solid-phase extraction. J. Chromatogr. B 805:
11. J.N. Thompson. Problems of official methods and new techniques 289–296 (2004).
for analysis of foods and feeds for vitamin A. J.—Assoc. Off. Anal. 23. S. Scalia, G. Ruberto, and F. Bonina. Determination of vitamin A,
Chem. 69: 727–738 (1986). vitamin E, and their esters in tablet preparations using supercritical
12. J.Z. Huo, H.J. Nelis, P. Lavens, P. Sorgeloos, and A.P. De Leenheer. fluid extraction and HPLC. J. Pharm. Sci. 84: 433–436 (1995).
Determination of vitamin E in aquatic organisms by high-perfor- 24. P. Moreno and V. Salvadó. Determination of eight water- and fat-
mance liquid chromatography with fluorescence detection. soluble vitamins in multi-vitamin pharmaceutical formulations by
Anal. Biochem. 242: 123–128 (1996). high-performance liquid chromatogramphy. J. Chromatog. A 870:
13. H. Iwase. Determination of vitamin D in emulsified nutritional 207–215 (2000).
supplements by solid-phase extraction and column-switching high- 25. B. Klejdus, J. Petrlová, D. Potešil, V. Adam, R. Mikelová, J. Vacek,
performance liquid chromatography with UV detection. R. Kizek, and V. Kubán. Simultaneous determination of water- and
J. Chromatogr. A 881: 189–196 (2000). fat-soluble vitamins in pharmaceutical preparations by high-perfor-
14. L.Y. Wang, C.J. Bates, L. Yan, D.J. Harrington, M.J. Shearer, and mance liquid chromatography coupled with diode array detection.
A. Prentice. Determination of phylloquinone (vitamin K1) in plasma Anal. Chim. Acta. 520: 57–67 (2004).
and serum by HPLC with fluorescence detection. Clin. Chim. Acta 26. G.W.C. Hung. Determination of vitamin D2 and D3 in feedingstuffs
347: 199–207 (2004). by high-performance liquid chromatography. J. Liq. Chromatogr.
15. L.G. Rushing, W.M. Cooper, and H.C. Thompson. Simultaneous 11: 953–969 (1988).
analysis of vitamins A and E in rodent feed by high-pressure liquid 27. R.G. Pozo, E.S. Saitua, I. Uncilla, and J.A. Montoya. Simultaneous
chromatography. J. Agric. Food Chem. 39: 296–299 (1991). determination by HPLC of fat-soluble vitamins in albacore
16. S. White. Determination of vitamins K1, K2 (MK-4), and K3 in animal (Thunnus alalunga). J. Food Sci. 55: 77–78 (1990).
feeds by high-performance liquid chromatography. Anal. Proc. 30: 28. D.E. Ponde. New polymeric sorbents: enhancing solid-phase
266–267 (1993). extraction. Aust. J. Chem. 58: 375 (2005).
17. A.P. Arbatskii, G.N. Afon’shin, and V.M. Vostokov. Determination of
vitamins in feed and foodstuffs by high-performance liquid chro- Manuscript received November 14, 2006;
matography. J. Anal. Chem. 59: 1304–1307 (2004). revision received February 8, 2007.

350