Food Chemistry 134 (2012) 559–563

Contents lists available at SciVerse ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

The simultaneous determination of vitamins A, E and b-carotene

in bovine milk by high performance liquid chromatography–ion trap mass
spectrometry (HPLC–MSn)
Tim Plozza a,⇑, V. Craige Trenerry a, Domenico Caridi b
a
Department of Primary Industries, 621 Sneydes Road, Werribee, Victoria 3030, Australia
b
School of Engineering and Science, Victoria University, P.O. Box 14428, Melbourne, Victoria 8001, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Bovine milk is considered to be a good dietary source of vitamins A, E and b-carotene. This paper
Received 12 August 2011 describes a robust method for the simultaneous determination of these compounds using high perfor-
Received in revised form 23 December 2011 mance liquid chromatography–ion trap mass spectrometry (HPLC–MSn) as the determinative step. The
Accepted 19 February 2012
compounds were separated with a 5 lm Polaris 2.1  150 mm C18-A column and an aqueous-methanol
Available online 27 February 2012
mobile phase and the levels determined using the MS2 fragments (vitamin A, m/z 213 and m/z 199, vita-
min E m/z 165 and b-carotene m/z 413). The HPLC–MSn was operated in the positive ion APCI mode. The
Keywords:
method was validated using repeatability studies, duplicate analyses, recovery experiments, proficiency
Quantification
Vitamin A
study data and comparison with previously validated high performance liquid chromatographic ultravi-
Vitamin E olet/visible (HPLC–UV/Vis) and fluorescence (HPLC–Fl) procedures used in our laboratory. Expanded
b-Carotene measurement uncertainties were similar for HPLC–MSn, HPLC–UV/Vis and HPLC–Fl, e.g. vitamin A
Bovine milk 45 lg/100 ml, HPLC–MSn ±8%, HPLC–UV ±9%; vitamin E 150 lg/100 ml, HPLC–MSn ±11%, HPLC–Fl ±9%;
Liquid chromatography–ion trap mass b-carotene 12 lg/100 ml, HPLC–MSn ±18%, HPLC–Vis ±21%.
spectrometry (HPLC–MSn) Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction 2002). Vitamin E is defined as a-tocopherol, but the other tocophe-

Vitamins are organic compounds which are essential for human
life, and are differentiated into ‘‘fat’’ and ‘‘water’’ soluble com-
pounds. The fat soluble vitamins (A, E, D and K) have a range of
important biological functions in the body including regulation of
cell and tissue growth and differentiation (vitamin A), lipid antiox-
idant (vitamin E), absorption of calcium and phosphorus and bone
mineralisation (vitamin D) and blood clotting factor (vitamin K).
Vitamins A and E cannot be synthesised in our bodies (Coultate,
2002) and must be sourced through the diet. Vitamin D is produced
through exposure to UV-B radiation (sunlight), and vitamin K is
produced by bacteria in the large intestine. In some circumstances
this may be insufficient to meet the body’s needs, in which case the
vitamins must also be sourced from the diet.
Vitamin A is defined as all trans-retinol, however, other reti-
noids have vitamin A activity to a lesser extent (Ball, 1998), as do
carotenoids, e.g. b-carotene, which are metabolised to retinoids
in the liver (Ball, 1998). Carotenoids also act as important antioxi-
dants with a preventative effect for various diseases (Coultate,
⇑ Corresponding author. Address: Department of Primary Industries, 621
Burwood Highway, Knoxfield, Victoria 3180, Australia. Tel.: +61 3 9210 9222; fax:
+61 3 9210 9202.
E-mail address: tim.plozza@dpi.vic.gov.au (T. Plozza).
rols also possess some vitamin E activity. Vitamin D exists in two
main forms, with equal activity: cholecalciferol (D3), derived from
animals, and ergocalciferol (D2), derived from plants. Vitamin K oc-
curs naturally as phylloquinone (K1) and the many derivatives of
menaquinone (K2 series).
Bovine milk is a good dietary source of all trans-retinol, b-
carotene and a-tocopherol (Bergamo, Fedele, Iannibelli, & Marzillo,
2003; Jensen & Robert, 1995) but contains relatively little vitamin
D3 or vitamin K (Jensen & Robert, 1995). All trans-retinol, b-carotene
and a-tocopherol come from the cow’s diet, and variations in
the levels of these compounds have been attributed to variations
in diet (Bergamo et al., 2003; Hulshof, van Roekel-Jansen, van de
Bovenkamp, & West, 2006). b-Carotene is the predominant
(90%) carotenoid found in milk, followed by lutein (Hulshof
et al., 2006; Nozière et al., 2006). b and c-tocopherol are also present
in bovine milk, although at much lower levels than a-tocopherol
(Lanina, Toledo, Sampels, Kamal-Eldin, & Jastrebova, 2007).
To date, high performance liquid chromatographic ultraviolet/
visible (HPLC–UV/Vis) detection for vitamins A, D and carotenes,
and fluorescence (HPLC–Fl) detection for vitamins A, E and K have
been the methods of choice for the analysis of fat soluble vitamins
in milk (Ball, 1998; Bergamo et al., 2003; Byrdwell et al., 2008;
Hulshof et al., 2006; Kamao et al., 2007a; Kurmann & Indyk,
1994). The compounds are usually assayed individually due to

0308-8146/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2012.02.121
560 T. Plozza et al. / Food Chemistry 134 (2012) 559–563
their chemical diversity and varying levels within samples,
however, recently an UPLC–UV/Vis method was reported for the
simultaneous determination of all trans-retinol, a-tocopherol and
b-carotene in milk (Chauveau-Duriot, Doreau, Noziere, & Graulet,
2010). During the past decade, high performance liquid chroma-
tography coupled with mass spectrometry (HPLC–MS) has
emerged as a major breakthrough in analytical science (Wil-
loughby, Sheehan, & Mitrovich, 2002). High performance liquid
chromatography–tandem mass spectrometry (HPLC–MS/MS) is
widely accepted as the benchmark for quantification of a wide
range of non-volatile compounds at the parts-per-million (mg/
kg) and parts-per-billion (lg/kg) levels (Zhang, Rose, & Trenerry,
2009) due to its superior sensitivity and selectivity compared to
single quadrupole (MS), 3D (Paul) ion trap (MSn) and time of flight
(Tof) instruments. Ion trap mass spectrometers (MSn) cover a sim-
ilar m/z range as tandem mass spectrometers and can also operate
in full scan mode at high and low resolution. The major advantage
of MSn instruments is the capability of performing successive frag-
mentation steps which provides more structural information than
other mass analysers, which can then be used to assist in the iden-
tification of unknown compounds (Willoughby, Sheehan, & Mitrovic,
2002). However, the sensitivity of traditional 3D ion trap mass
spectrometers is an order of magnitude less than tandem mass
spectrometers. The linear ion trap mass spectrometer exhibits sen-
sitivity approaching that of tandem mass spectrometers that are
routinely used in testing laboratories and the wide range of capa-
bilities make HPLC–MSn a popular choice for more detailed exam-
ination of sample extracts as well as metabolomic studies.
HPLC–MS/MS has been used for the analyses of vitamins A, D, E,
and K in breast milk (Kamao et al., 2007b), fortified and other foods
(Byrdwell, 2009; Dimartino, 2009; Huang, 2011; Huang, LaLuzerne,
Winters, & Sullivan, 2009), infant formula (Heudi, Trisconi, & Blake,
2004; Lim, Kim, Ahn, & Kim, 2011), and human blood serum (Blum
et al., 2008; Capote, Jimenez, Granados, & De Castro, 2007).
The aim of this work was to develop a method for the simulta-
neous determination of endogenous levels of all trans-retinol, a-
tocopherol and b-carotene in bovine milk using HPLC–MSn as the
determinative step that could be used in conjunction with our pre-
viously reported method for determining vitamin D3 in bovine
milk by LC–MSn (Trenerry, Plozza, Caridi, & Murphy, 2011). This
would greatly simplify the process of monitoring the nutritional
quality of bovine milk as well as providing quantitative/qualitative
data on other fat soluble compounds present in the extract, e.g.
cholesterol and other phytosterols.
2. Materials and methods
2.1. Chemicals and reagents
DL-a-tocopherol was supplied by Fluka, 30% fluid suspension of
b-carotene was supplied by Roche and all trans-retinol acetate was
sourced from USP, Rockville, MD, USA. All other chemicals and re-
agents were AR, HPLC or HPLC–MS grade.
2.2. Preparation of standards
A stock standard solution of all trans-retinol in methanol was pre-
pared from all trans-retinol acetate by saponification as described
in Section 2.4 and the concentration determined by UV–Visible spec-
trophotometry. Stock standard solutions of a-tocopherol and b-
carotene were prepared in hexane and the concentrations checked
weekly by UV–Visible spectrophotometry (Eitenmiller & Landen,
1999). Intermediate and working standard solutions were prepared
daily in methanol. A series of standards containing approximately
0.05–1.2 lg/100 ml all trans-retinol, 0.3–6.5 lg/100 ml a-tocoph-
erol and 0.03–0.7 lg/100 ml b-carotene were prepared for quantifi-
cation. Linear regression coefficients (R2) were greater than 0.999 for
all analytes assayed on all instruments.
2.3. Sampling, preparation and storage
Fresh bovine milk was sourced from the Department of Primary
Industries (Victoria) research dairy based at Ellinbank, Victoria,
Australia. All samples were stored at 4 °C for one day, sub-sampled,
and the sub-samples stored at 70 °C prior to analysis. The thawed
milk samples were mixed thoroughly before assaying. A powdered
infant formula sample received from Proficiency Testing Australia
was used as a control sample for all trans-retinol and a-tocopherol.
The infant formula was dissolved in water (1 g with 9 ml water)
immediately before analysis. Commercial full cream milk was pur-
chased from a local retail outlet, stored at 4 °C and assayed before
‘‘the best before’’ date. All experiments were conducted under
yellow fluorescent lighting and amber glassware used where
appropriate.
2.4. Sample extraction
A 10 ml aliquot of bovine milk was mixed with 0.5 g of ascorbic
acid, 40 ml of ethanol and 10 ml of 1:1 potassium hydroxide in
water and heated at reflux with stirring for 30 min. The mixture
was cooled in an ice bath and quantitatively transferred to a sepa-
rating funnel with 50 ml water, 10 ml ethanol and 50 ml hexane
containing 1.5 mg/100 ml butylated hydroxytoluene (BHT). The
separating funnel was shaken vigorously for 2 min and the phases
allowed to separate. The aqueous phase was removed and ex-
tracted twice more with 20 ml portions of hexane containing
1.5 mg/100 ml BHT. The hexane extracts were combined, washed
three times with 100 ml of water and then made to 100 ml with
hexane. 10 ml of the hexane solution was then transferred to a
glass tube and the solvent removed under a flow of nitrogen at
room temperature. The residue was reconstituted with 1.0 ml of
methanol and filtered through a 13 mm 0.45 lm Teflon filter disc
into three separate vials for concurrent analysis by HPLC–MSn,
HPLC–UV/Vis and HPLC–Fl.
2.5. Apparatus
2.5.1. HPLC–MSn
The analyses were performed with an Agilent 1100 series qua-
ternary solvent delivery system with a cooled autosampler (4 °C)
(Agilent, Waldron Germany) connected to a Thermo Fisher LTQ
ion trap mass spectrometer operating in the positive ion Atmo-
spheric Pressure Chemical Ionisation mode (Thermo Fisher, San
Jose, USA). The compounds were separated with a 5 lm Polaris
2.1  150 mm C18-A column (Varian, Mulgrave, Australia) fitted
with a C18 guard column and maintained at 30 °C. The mobile
phase consisted of (A) water and (B) methanol with a flow rate
of 0.2 ml/min. The elution profile was: 0 min, 95% B; 5 min, 95%
B; 10 min, 100% B (linear gradient); 25 min, 100% B, 26 min, 95%
B and the column equilibrated for 5 min at a flow rate of 0.4 ml/
min before the next injection.
The mass spectrometer was tuned by infusing a 10 lg/ml solu-
tion of each compound at a rate of 10 ll/min, mixed with 0.2 ml/
min HPLC mobile phase via a T-piece before entering the mass
spectrometer.
The APCI vaporiser temperature was set at 250 °C, the ion cur-
rent 4.5 lA, the heated capillary was maintained at 200 °C and
the sheath and auxiliary gases were at 20 and 5 units, respectively.
Source voltage was set to 4.4 kV and the capillary voltage set at
16 V. A full scan ms2 protocol was used with the following settings;
all trans-retinol, parent mass m/z 269, normalised collision energy
 T. Plozza et al. / Food Chemistry 134 (2012) 559–563 561

CH 3

HO
OH CH3

H3C O C16 H33

CH 3

all trans-retinol α-tocopherol β-carotene

Fig. 1. Structures of all trans-retinol, a-tocopherol and b-carotene.

of 30, scan range of m/z 70–250, quantification ions m/z 213 and improve the chromatographic performance of HPLC–MS by virtue
m/z 199, a-tocopherol, parent mass m/z 431, normalised collision of their greater peak capacity and resolution (Byrdwell, 2009).
energy of 20, scan range of m/z 130–180, quantification ion m/z All trans-retinol, a-tocopherol and b-carotene are structurally
165 and b-carotene, parent mass m/z 537, normalised collision en- quite different (Fig. 1) and exhibit varying degrees of polarity, mak-
ergy of 22, scan range of m/z 150–490, quantification ion m/z 413. ing chromatographic separation with a single set of conditions dif-
Excalibur V2.1 software was used to process the data. ficult. For this study, a suitable separation was achieved using a
5 lm Polaris C18-A HPLC column and a methanol/water gradient
as described in the Experimental section 2.5.1-see Fig. 2. Using this
2.5.2. HPLC–UV/Vis and HPLC–Fl
gradient, the polar all trans-retinol eluted at 4.4 min and was well
2.5.2.1. All trans-retinol and a-tocopherol. The analyses were per-
separated from other early eluting compounds that may have
formed with a Waters series 2695 quaternary solvent delivery sys-
caused ionisation suppression/enhancement of the molecular ion.
tem with a cooled autosampler at 4 °C and heated column
a-Tocopherol and the non polar b-carotene eluted at 11.7 and
compartment at 30 °C (Waters, Milford, MA, USA), coupled in ser-
ies to a Shimadzu RF-10Axl fluorescence detector (a-tocopherol Ex
295 nm Em 330 nm) and a Waters 996 photodiode array detector
(all trans-retinol 320 nm). The compounds were separated on a RT: 0.00 - 30.04 SM: 7G
100 NL: 4.83E3
10 lm Bondclone 3.9  300 mm C18 column (Phenomenex, Syd- m/z= 198.60-199.60 F:
ney, Australia) fitted with a C18 guard column. The mobile phase 90 ITMS + c APCI corona Full
ms2 269.10@30.00 [
consisted of water:methanol (5:95), at a flow rate of 1 ml/min. 80 70.00-250.00] MS Avalon
all trans-retinol FSol03031012
70
60
2.5.2.2. b-Carotene. The analyses were performed with a Waters
series 2695 quaternary solvent delivery system with a cooled auto- 50
sampler at 4 °C and heated column compartment at 30 °C (Waters, 40
Milford, MA, USA), coupled to a Waters 996 photodiode array 30
detector (450 nm). The compounds were separated on a 5 lm
20
Sphereclone 4.6  250 mm ODS2 C18 column (Phenomenex, Syd-
10
ney, Australia) fitted with a C18 guard column. The mobile phase
consisted of acetonitrile:50 mM ammonium acetate in metha- 100 NL: 1.74E5
nol:dichloromethane (75:20:5 v/v) and containing 0.05% triethyl- 90 m/z= 164.60-165.60 F:
ITMS + c APCI corona Full
amine and 0.1% butylated hydroxytoluene, at a flow rate of 2 ml/ ms2 431.40@20.00 [
80
min. -tocopherol
130.00-180.00] MS Avalon
Relative Abundance

70 FSol03031012

60
3. Results and discussion 50
40
The standard method for the analysis of all trans-retinol a-
30
tocopherol, vitamin D3 and b-carotene in bovine milk involves
digesting a portion of the sample in hot ethanolic KOH solution fol- 20

lowed by solvent extraction with hexane or similar solvent, re- 10

moval of the solvent and the re-dissolving the residue in a small
100 NL: 4.74E2
amount of methanol or mobile phase prior to analysis. During the m/z= 412.90-413.90 F:
90
digestion process, the esterified analytes, as well as any triglycer- ITMS + c APCI corona Full
80 ms2 537.30@22.00 [
ides and proteins in the sample matrix, are hydrolysed to the free 150.00-490.00] MS
all-trans- -carotene
compounds (Ball, 1998; Byrdwell, 2009; Byrdwell et al., 2008; 70 FSol03031012

Heudi et al., 2004; Hulshof et al., 2006; Kurmann & Indyk, 1994; 60
Lanina et al., 2007). The final extract is then assayed by reverse 50
phase HPLC–UV/Vis for b-carotene and vitamin D3, and reverse 13-cis- -carotene
40
phase HPLC–UV and HPLC–FL for all trans-retinol and a-tocopherol.
30
Separate HPLC conditions are usually used for the analytes.
Chromatographic separation of the positional isomers b- and c- 20
tocopherol, which have identical mass and can not be easily differ- 10
entiated by MS, has traditionally required the use of normal phase
chromatography, however, the introduction of new stationary 0 5 10 15 20 25 30
Time (min)
phases has allowed this separation under reversed phase condi-
tions, which are more amenable to HPLC–MS (Lanina et al., 2007). Fig. 2. HPLC–MS2 chromatogram of a bovine milk sample containing all trans-
Other recent advances in chromatographic column technology such retinol (50 lg/100 ml), a-tocopherol (200 lg/100 ml) and b-carotene (12 lg/
as small particle size stationary phases have the potential to further 100 ml) using the conditions described in the Section 2.
562 T. Plozza et al. / Food Chemistry 134 (2012) 559–563

Table 1
Instrument Repeatability (%CV) determined by sequential assay (n = 8) of a standard solution containing all trans-retinol (0.6 lg/ml), a-tocopherol (3.3 lg/ml) and b-carotene
(0.34 lg/ml), and a sample extract containing all trans-retinol (0.51 lg/ml), a-tocopherol (2.13 lg/ml) and b-carotene (0.13 lg/ml).

Analyte HPLC–Fl (a-tocopherol) %CV HPLC–UV/Vis (all trans-retinol and b-carotene) %CV HPLC–MS2 %CV
Std Sample Std Sample Std (n = 7) Sample (n = 8)
All trans-retinol 1.0 0.4 m/z 213 1.1 m/z 199 1.0 m/z 213 0.7 m/z 199 0.8
a-Tocopherol 0.5 0.4 m/z 165 1.6 m/z 165 1.0
b-Carotene 1.9 2.6 m/z 413 2.3 m/z 413 3.1

Table 2
Method repeatability (%CV) determined by assaying separate aliquots (n = 8) of a full cream milk sample containing all trans-retinol (41 lg/100 ml), a-tocopherol (100 lg/100 ml)
and b-carotene (15 lg/100 ml).

Analyte HPLC–Fl (a-tocopherol) %CV HPLC–UV/Vis (all trans-retinol and b-carotene) %CV HPLC–MS2 %CV
All trans-retinol 1.4 m/z 213 1.2 m/z 199 2.2
a-Tocopherol 1.1 m/z 165 1.2
b-Carotene 2.8 m/z 413 2.4

Table 3
Levels of all trans-retinol, a-tocopherol and b-carotene in a selection of bovine milk samples determined by HPLC–MS2 and HPLC–UV (all trans-retinol), HPLC–Fl (a-tocopherol)
and HPLC–Vis (b-carotene).

Sample All trans-retinol (lg/100 ml) a-Tocopherol (lg/100 ml) b-Carotene (lg/100 ml)
m/z 199 m/z 213 HPLC–UV m/z 165 HPLC–Fl m/z 413 HPLC–Vis
1 59 58 55 145 145 25 26
2 52 54 50 147 150 23 26
3 41 40 38 154 156 21 23
4 55 56 52 155 137 24 27
5 35 36 33 131 135 18 19

24.4 min, respectively, giving an overall run time of 30 min, which
was suitable for routine use. The analytes could be ionised by either
positive or negative mode APCI. Although negative ionisation gave
larger peak areas for all trans-retinol and a-tocopherol, positive ion-
isation gave a similar signal to noise ratio and a much lower coeffi-
cient of variation for the analysis. For these reasons, positive
ionisation was chosen for this study. MS2 fragmentation of the
molecular ions produced fragment ions which were also suitable
for quantification: e.g. all trans-retinol; molecular ion [M+H H2O]+
m/z 269, fragment ions m/z 213 and m/z 199; a-tocopherol, molec-
ular ion m/z 431, fragment ion m/z 165; b-carotene, molecular ion
m/z 537, fragment ion m/z 413. The structure of the compound elut-
ing at 26.5 min was assigned as 13-cis-b-carotene based on the ms
data ([M+H]+ m/z 537) and comparison of the UV spectra with liter-
ature spectra (Nyambaka & Ryley, 1996). This compound is
produced via isomerisation of b-carotene during the hot saponifica-
tion procedure and is also present in the standard.
3.1. Comparison with validated methods
Instrument repeatability was determined by repetitive injection
of standard and sample solutions. The data are presented in Table
1. Method repeatability was determined by assaying eight aliquots
of a sample of commercial full cream milk and the data presented
in Table 2. The HPLC–MSn instrument repeatability for each com-
pound was similar to those for the standard HPLC–UV/Vis (all
trans-retinol and b-carotene) and HPLC–Fl (a-tocopherol) instru-
ments. Seventy-two bovine milk samples sourced from a feeding
trial conducted at DPI-Ellinbank were assayed as part of the valida-
tion process. Excellent correlation was achieved between the levels
determined by HPLC–MSn and the levels determined by validated
HPLC–UV/Vis and HPLC–Fl methods, e.g. vitamin A: HPLC–MSn /
HPLC–UV/Vis correlation 1.09, 2.6% CV; vitamin E: HPLC–MSn /
HPLC–Fl correlation 1.06, 3.5% CV, b-carotene HPLC–MSn /HPLC–
UV/Vis correlation 1.10, 8% CV (n = 72). A representative set of data
are displayed in Table 3. The levels determined by HPLC–MSn for all
trans-retinol were measured at both m/z 213 and m/z 199 and were
similar, indicating that either ion could be used for quantification.
Recoveries of spiked milk samples (n = 8) assayed by LC–MSn were
excellent; all trans-retinol; range 91–114%, mean 101%, CV 7.6%, a-
tocopherol; range 91–115%, mean 100%, CV 8.3%, b-carotene; range
83–112%, mean 93%, CV 12.4%.
3.2. Proficiency sample data
A fortified infant formula supplied by Proficiency Testing Aus-
tralia as part of a proficiency study was used to further validate
the method. The infant formula was used as an ‘‘in house’’ refer-
ence material and assayed for vitamin A and vitamin E content in
every batch of samples (n = 9). The levels determined by LC–MSn
compared favourably with the proficiency study data: e.g. vitamin
A; this study, range 0.61–0.74 mg/100 g, CV 7.1%, n = 9, proficiency
study range 0.78–1.2 mg/100 g, CV 19.7%, n = 7, vitamin E; this
study, range 9.5–11.8 mg/100 g, CV 6.2%, n = 9, proficiency study
range 10.5–14.3 mg/100 g, CV 8.9%, n = 6 (Bunt, 2010). The levels
determined by LC–MSn were slightly lower than those from the
proficiency study, however, the data were suitable for monitoring
method performance.
3.3. Limit of quantification
The limits of quantification (LOQ) for LC–MSn were determined at
a S/N ratio of 10:1 from standard solutions of each compound and
were equivalent to 0.1 lg/100 ml for all trans-retinol and a-tocoph-
erol and 1 lg/100 ml for b-carotene in the actual milk samples.
 T. Plozza et al. / Food Chemistry 134 (2012) 559–563
3.4. Measurement uncertainty
In house validation data were used to estimate the measure-
ment uncertainty (MU) for each analyte at three different levels
(Plozza, Trenerry, Zeglinski, Nguyen, & Johnstone, 2011). The major
sources of uncertainty were homogeneity, method recovery and
the analytical standard calibration curves. Homogeneity uncer-
tainty was estimated from the relative standard deviation (RSD)
of duplicate analyses. The purity of the analytical standards was
shown to be insignificant and not included. The expanded MU for
the analytes at three different levels are similar for HPLC–MSn
and the traditional HPLC procedures e.g. all trans-retinol: 27 lg/
100 ml HPLC–MSn ±10%, HPLC–UV ±12%, 45 lg/100 ml HPLC–MSn
±8%, HPLC–UV ±9%, 68 lg/100 ml HPLC–MSn ±8%, HPLC–UV ±7%;
a-tocopherol: 110 lg/100 ml HPLC–MSn ±13%, HPLC–Fl ±12%, 150
lg/100 ml HPLC–MSn ±11%, HPLC–UV ±9%, 205 lg/100 ml HPLC–
MSn ±9%, HPLC–UV ±8%; b-carotene: 7 lg/100 ml HPLC–MSn
±27%, HPLC–Vis ±29%, 12 lg/100 ml HPLC–MSn ±18%, HPLC–Vis
±21%, 23 lg/100 ml HPLC–MSn ±14%, HPLC–Vis ±19%.
This method was subsequently used to measure the levels of
these compounds in bovine milk collected from a feeding trial
conducted by the Department of Primary Industries, Victoria. The
results of this trial will be reported elsewhere.
4. Conclusion
A straight forward, robust method for determining the levels of
all trans-retinol, a-tocopherol and b-carotene in bovine milk using
HPLC–MSn has been developed and can be used in conjunction
with our previously reported method for determining vitamin D3
in bovine milk by LC–MSn. The method was validated using repeat-
ability studies, duplicate analyses, recovery experiments, profi-
ciency study data and comparison with previously validated high
performance liquid chromatographic ultraviolet/visible (HPLC–
UV/Vis) and fluorescence (HPLC–Fl) procedures used in our
laboratory.
Acknowledgement
The authors thank Dr. Martin Auldist (DPI-Ellinbank) for provid-
ing the fresh bovine milk samples.
References
Ball, G. F. M. (1998). Fat-soluble vitamin assays in food analysis. A comprehensive
review. Essex: Elsevier Science Publishers Ltd.
Bergamo, P., Fedele, E., Iannibelli, L., & Marzillo, G. (2003). Fat-soluble vitamin
contents and fatty acid composition in organic and conventional Italian dairy
products. Food Chemistry, 82(4), 625–631.
Blum, M., Dolnikowski, G., Seyoum, E., Harris, S. S., Booth, S. L., Peterson, J., et al.
(2008). Vitamin D3 in fat tissue. Endocrine, 33(1), 90–94.
Bunt, M. (2010). Food proficiency testing program round 31-vitamins. Silverwater,
NSW, Australia: Proficiency Testing Australia.
Byrdwell, W. C. (2009). Comparison of analysis of vitamin D3 in foods using
ultraviolet and mass spectrometric detection. Journal of Agricultural and Food
Chemistry, 57(6), 2135–2146.
563
Byrdwell, W. C., DeVries, J., Exler, J., Harnly, J. M., Holden, J. M., Holick, M. F., et al.
(2008). Analyzing vitamin D in foods and supplements: Methodologic
challenges. American Journal of Clinical Nutrition, 88(2).
Capote, F. P., Jimenez, J. R., Granados, J. M. M., & De Castro, M. D. L. (2007).
Identification and determination of fat-soluble vitamins and metabolites in
human serum by liquid chromatography/triple quadrupole mass spectrometry
with multiple reaction monitoring. Rapid Communications in Mass Spectrometry,
21(11), 1745–1754.
Chauveau-Duriot, B., Doreau, M., Noziere, P., & Graulet, B. (2010). Simultaneous
quantification of carotenoids, retinol, and tocopherols in forages, bovine
plasma, and milk: Validation of a novel UPLC method. Analytical and
Bioanalytical Chemistry, 397(2), 777–790.
Coultate, T. P. (2002). Food. The Chemistry of its Components. Cambridge: The Royal
Society of Chemistry.
Dimartino, G. (2009). Simultaneous determination of cholecalciferol (vitamin D3)
and ergocalciferol (vitamin D2) in foods by selected reaction monitoring. Journal
of AOAC International, 92(2), 511–517.
Eitenmiller, R. R., & Landen, W. O. (1999). Vitamin analysis for the health and food
sciences. Boca Raton: CRC Press LLC.
Heudi, O., Trisconi, M. J., & Blake, C. J. (2004). Simultaneous quantification of
Vitamins A, D3 and E in fortified infant formulae by liquid chromatography–
mass spectrometry. Journal of Chromatography A, 1022(1–2), 115–123.
Huang, M. (2011). Application of ultra-performance liquid chromatography/tandem
mass spectrometry for the measurement of vitamin D in foods and nutritional
supplements. Journal of AOAC International, 94(1), 211–223.
Huang, M., LaLuzerne, P., Winters, D., & Sullivan, D. (2009). Measurement of vitamin
d in foods and nutritional supplements by liquid chromatography/tandem mass
spectrometry. Journal of AOAC International, 92(5), 1327–1335.
Hulshof, P. J. M., van Roekel-Jansen, T., van de Bovenkamp, P., & West, C. E. (2006).
Variation in retinol and carotenoid content of milk and milk products in The
Netherlands. Journal of Food Composition and Analysis, 19(1), 67–75.
Jensen, R. G., & Robert, G. J. (1995). Fat-soluble vitamins in bovine milk. Handbook of
milk composition (pp. 718–725). San Diego: Academic Press.
Kamao, M., Suhara, Y., Tsugawa, N., Uwano, M., Yamaguchi, N., Uenishi, K., et al.
(2007a). Vitamin K content of foods and dietary vitamin K intake in Japanese
young women. Journal of Nutritional Science and Vitaminology, 53(6), 464–470.
Kamao, M., Tsugawa, N., Suhara, Y., Wada, A., Mori, T., Murata, K., et al. (2007b).
Quantification of fat-soluble vitamins in human breast milk by liquid
chromatography–tandem mass spectrometry. Journal of Chromatography B:
Analytical Technologies in the Biomedical and Life Sciences, 859(2), 192–200.
Kurmann, A., & Indyk, H. (1994). The endogenous vitamin D content of bovine milk:
Influence of season. Food Chemistry, 50(1), 75–81.
Lanina, S. A., Toledo, P., Sampels, S., Kamal-Eldin, A., & Jastrebova, J. A. (2007).
Comparison of reversed-phase liquid chromatography–mass spectrometry with
electrospray and atmospheric pressure chemical ionization for analysis of
dietary tocopherols. Journal of Chromatography A, 1157(1–2), 159–170.
Lim, Y. O., Kim, B., Ahn, S., & Kim, J. (2011). Improvement of accuracy for the
determination of vitamin A in infant formula by isotope dilution-liquid
chromatography/tandem mass spectrometry. Food Chemistry, 126(3),
1393–1398.
Nozière, P., Graulet, B., Lucas, A., Martin, B., Grolier, P., & Doreau, M. (2006).
Carotenoids for ruminants: From forages to dairy products. Animal Feed Science
and Technology, 131(3–4), 418–450.
Nyambaka, H., & Ryley, J. (1996). An isocratic reversed-phase HPLC separation of the
stereoisomers of the provitamin A carotenoids (a- and b-carotene) in dark green
vegetables. Food Chemistry, 55(1), 63–72.
Plozza, T., Trenerry, V. C., Zeglinski, P., Nguyen, H., & Johnstone, P. (2011). The
confirmation and quantification of selected aminoglycoside residues in animal
tissue and bovine milk by liquid chromatography mass spectrometry.
International Food Research Journal, 18(3), 59–66.
Trenerry, V. C., Plozza, T., Caridi, D., & Murphy, S. (2011). The determination of
vitamin D3 in bovine milk by liquid chromatography mass spectrometry. Food
Chemistry, 125(4), 1314–1319.
Willoughby, R., Sheehan, E., & Mitrovich, S. (2002). A global view of LC/MS.
Pittsburgh: Global View Publishing.
Zhang, P., Rose, G., & Trenerry, V. C. (2009). Advances in pesticide monitoring: The
impact of liquid chromatography–tandem mass spectrometry on the Victorian
Produce Monitoring Program. Food Australia, 61(4), 137–141.