B12 in Foods
B12 in Foods
LGC/R/2011/171
© LGC Limited 2011
GC11/0003
A Government Chemist Programme Report
Report number LGC/R/2011/171
August 2011
Page 1 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
Contents
1. Executive Summary........................................................................................................ 3
2. Background .................................................................................................................... 3
3. Objective ........................................................................................................................ 4
4. Sample extraction........................................................................................................... 4
5. Use of the Immunoaffinity cartidges (IAC) ...................................................................... 4
5.1 IAC recovery............................................................................................................... 5
6. Analysis of Foods ........................................................................................................... 5
6.1 HPLC Conditions ........................................................................................................ 6
6.2 Linearity, Range and injection repeatability ................................................................ 7
6.3 Limit of detection ........................................................................................................ 7
6.4 Sample Extraction ...................................................................................................... 7
6.5 Method Performance .................................................................................................. 7
6.6 Method specificity ....................................................................................................... 8
6.7 Method applicability .................................................................................................... 8
7. Analysis of vitamin premixes .......................................................................................... 9
7.1 UPLC conditions......................................................................................................... 9
7.2 Linearity, Range and injection repeatability .............................................................. 10
7.3 Limit of detection ...................................................................................................... 10
7.4 Sample Extraction .................................................................................................... 10
7.5 Method Performance ................................................................................................ 10
8. Conclusion.................................................................................................................... 11
Page 2 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
1. Executive Summary
Procedures for the determination of Vitamin B12 in foods, supplements and vitamin premixes using
HPLC and UPLC were evaluated in this study. The study was undertaken to underpin the capability
of the Government Chemist to perform referee analysis of vitamins by developing appropriate
analytical methods.
The report includes an evaluation of a commercial, immunoaffinity cartridge used for the selective
isolation of vitamin B12. The evaluation does not imply any endorsement of the product by LGC or
the Government Chemist and alternative products may be available.
2. Background
Vitamin B12 is the collective name for cobalt-containing corrinoids that have biological activity in
humans. The main cobalamins in humans and animals are hydroxocobalamin, adenosylcobalamin
and methylcobalamin, the last two being the active coenzyme forms. Cyanocobalamin (CN-Cbl) is a
form of vitamin B12 that is widely used clinically and for food supplementation due to its availability
and stability.
Vitamin B12 in the diet is largely derived from microbiological synthesis. Small amounts can
sometimes be found in plant products due to the presence of bacteria or due to microbial
contamination but in the main it is found naturally, only in animal products, as a product of gut
microflora synthesis. The richest food sources are meats (particularly offal), seafoods, eggs and
dairy products. In addition, vitamin B12 can be added to foods such as breakfast cereals, infant
formulae and dietary supplements.
Vitamin B12 naturally present in foods is bound to proteins and glycoproteins and must be released
by acid or enzymatic digestion before analysis. It can be measured by chemical,
spectrophotometric, microbiological, chromatographic, radioassay or immunoassay methods.
Microbiogical assays have traditionally been used for the determination of Vitamin B12 in foods and
much of the available data on this vitamin has been obtained using this technique. These assays
are sensitive but are non-specific, require careful handling and can be prone to contamination.
Incubation times of 24 hours or more are required. Assay of vitamin B12 in biological matrices by
chromatographic methods is difficult due to the low concentrations usually present and the relatively
high detection limits obtainable with typical chromatographic detectors. HPLC can be used for the
analysis of fortified foods and supplements although the determination of vitamin B12 at natural
levels remains a challenge.
In many samples, there is a need to enhance the vitamin B12 concentration prior to HPLC analysis
and to reduce the amount of other sample components which may interfere with the
chromatographic analysis. This has been made possible in recent years by the manufacture of
commercial immunoaffinity cartridges which use immobilised antibodies to selectively extract
cobalamins from food extracts. Such columns are available from a number of manufacturer’s as
below:
The Easi-Extract cartridges were used for this evaluation however this does not imply any
endorsement of this product by LGC or the Government Chemist.
Page 3 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
3. Objective
Procedures for the determination of vitamin B12 in foods and vitamin premixes using aqueous
extraction, immunoaffinity clean-up / concentration and HPLC were evaluated and assessed in this
study. The combination of HPLC or UPLC and sample preparation using immunoaffinity media
offers a rapid means of determining vitamin B12 in foods and should improve the analytical
uncertainty of the measurement compared to the microbiological assay procedures currently used.
4. Sample extraction
Cyanocobalamin (CN-Cbl) is water-soluble; therefore the free form can be extracted into aqueous
solution. It is most stable at pH 4.0 – 4.5 therefore most extraction procedures are carried out at this
pH. Vitamin B12 is stable to heat between pH 4 and 7 but should be protected from extremes of pH,
strong light and oxidising agents.
In foods, vitamin B12 is bound to proteins and these must be denatured using acid or by digestion
with proteases during extraction. For starchy samples, amylases are often used to break down
starch and assist filtration. When determining natural vitamin B12 forms, excess cyanide is usually
added to convert the cobalamins to the cyano form.
When carrying out analysis of vitamin B12 in foods by microbiological assay at LGC, samples are
autoclaved at 121 °C for 10 minutes in pH 4.5 sodium acetate buffer. Potassium cyanide is added to
ensure conversion of the active forms to CN-Cbl. (Note that potassium cyanide is a Schedule 1
poison and specific safety requirements must be implemented for its use). For starchy samples, a
further incubation with amylase is carried out to aid filtration. Samples are then filtered or
centrifuged before dilution to the required assay concentration.
In the Easi-Extract instructions, there are a range of extraction conditions given. For analysis of
added cyanocobalamin in tablets, drinks and juices, extraction with water is all that is required. For
infant formulae, and other foods that contain protein, an extraction with pepsin is used if total
vitamin B12 is required in order to release the protein bound forms. An amylase digestion is
suggested for starchy foods and potassium cyanide is added to ensure conversion to CN-Cbl. The
extraction conditions vary but all except the tablets and drinks require a digestion with enzymes
(pepsin and/or amylase), a heating step and control of the pH. Cyanide is used for determination of
natural cobalamins forms. Vitamin B12 is most stable at ~ pH 4 – 5 therefore extraction is usually
B
carried out in this range. Extraction conditions are given for tablets, drinks and juices, protein drinks
and premixes, infant formula and a range of other foods. In general, similar extraction principles can
be applied to all foods but the precise conditions required should be confirmed by the user in the
matrix to be analysed.
It is important to ensure that the extract pH is between 4.5 and 7.0 as the analyte may not bind
sufficiently to the IAC cartridge outside of this range. Sample extracts should be filtered or
centrifuged before the IAC procedure to avoid blocking of the cartridge and to ensure reproducible
elution characteristics.
The columns were used with gravity elution although positive or negative pressure may be applied if
required. Use of the IAC columns in these modes was not evaluated. A 10 mL plastic reservoir
was attached to each cartridge before use. Sample loading should be between 2-3 mL/min to
ensure adequate binding but in practice, the drip rate obtained by simple gravity elution was
normally sufficient. Using this technique, there was some variation in elution times between
cartidges but column performance was not adversely affected if the columns were allowed to “go
dry”.
Page 4 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
Following elution in methanol, the extracts were evaporated to dryness under nitrogen using a
Turbovap evaporator and were redissolved in mobile phase.
An important consideration is that the cartidges have limited capacity to bind vitamin B12. The
maximum capacity is quoted as 1 µg of cyanocobalamin however low recovery of vitamin B12 was
sometimes noted when the amount of vitamin B12 loaded, approached this value. Although this has
not been fully investigated, normal practice at LGC is to limit the maximum amount of vitamin B12 to
~ 0.5 µg.
To standardise the elution characteristics, the “loading volume” was standardised to 5 mL. If the
Vitamin B12 content in 5 mL of extract exceeded 0.5 µg, the extract was either diluted or the volume
of extract used was reduced and an appropriate amount of buffer was added to the reservoir to
bring the volume to 5 mL (e.g. 1 mL extract + 4 mL buffer). For samples containing lower
concentrations of vitamin B12, larger volumes of sample extract were used up to a maximum of 30
mL. (Note that higher volumes can be used i.e. up to 100 mL is suggested for carbonated drinks,
but this was not tested.)
The binding capacity of cartridges is a limitation as a maximum of 0.5 (or 1) µg of vitamin B12 is
obtained in the cartridge eluent. Since the minimum volume for redissolution is ~ 300 uL and
preferably 0.5 – 1 mL for normal HPLC vials, the vitamin B12 concentration in the final extract is still
quite low. A sensitive HPLC system and reasonably large injection volumes are required for
subsequent analysis. In addition, for samples where the expected vitamin B12 concentration is
unknown, care is required when interpreting the analytical results, to ensure that the IAC capacity
has not been exceeded as the results for such samples will be low.
The IAC instructions recommend that the IAC recovery should be tested using 10 mL of a 0.025
µg/mL solution of vitamin B12 in extraction buffer. When this was done, the recovery of vitamin B12
was around 97 %. However, lower and more variable recoveries were obtained when the amount of
vitamin B12 on the column was between 0.5 µg and 1 µg. In addition, variable recoveries (between
65 and 130 %) have been noted when recovery was assessed using sample extracts to which
vitamin B12 spikes had been added, even though the volumes were adjusted to keep the total
expected vitamin B12 concentration within the theoretical limit for the cartridges. The latter
recoveries would of course, include losses from other parts of the analytical procedure and not just
the IAC cartridge. Although the general IAC column performance can be checked using a standard
as described, acceptable recovery ranges should ideally be established using spiked sample
extracts and an analytical procedure (loading volumes etc) as close as possible to that normally
used.
6. Analysis of Foods
High performance liquid chromatography (HPLC) and Ultra performance liquid chromatography
(UPLC) systems were evaluated. The choice between these techniques was based only on
availability of equipment but analysis using UPLC is much faster, where available. HPLC was used
for the determination of the foods and supplements whilst UPLC was used for determination of
vitamin premixes.
Page 5 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
Page 6 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
30000
y = 2778.3x + 42.65
25000
R2 = 1
20000
Peak Area
15000
10000
5000
0
0 2 4 6 8 10 12
µg/m l
The calibration was linear over the range tested. The injection repeatability (RSD %) at 0.1 µg/mL
and 0.8 µg/mL was 3.6 % and 1.4 % respectively.
Page 7 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
The certified value for CRM 487 is 1.12 ± 0.09 mg/kg, therefore the analytical mean result was very
close to the expected value. The expected, within–laboratory repeatability (%RSD) calculated from
the Horwitz equation at 1 µg/g is 10.72, therefore the experimental %RSD of 4.94 was acceptable.
A cereal bar with fruit and chocolate however, gave problems due to poor filtration indicating that a
modified extraction was necessary.
The IAC procedure provides a rapid and selective means of isolating vitamin B12 from foods and
supplements however; the analyst must establish the optimum extraction procedure for the sample
to be examined and the sample weights, extraction volumes and IAC volumes to be used. This can
be problematic for unknown samples and may require some investigation before an analysis can be
carried out although the manufacturer’s instructions give guidance for some sample types.
Page 8 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
Fig F: UPLC chromatogram of Vitamin B12 standard @ 0.5µg/ml Fig G: Chromatogram of vitamin premix sample
Page 9 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
30
y = 26.266x + 0.0886
25
R2 = 1
20
Pk Area
15
10
0
0.000 0.200 0.400 0.600 0.800 1.000
ug/m l
The calibration was linear over the range tested. The injection repeatability (RSD %) at ~0.5 µg/mL
was 1.3 %.
Page 10 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
Sample ID 1 2 3 4 5 6 7 8 9
Declaration 948 - 1158 315 - 385 69 - 79 12.3 - 15.1 10.0 - 11.9 2.9 - 3.7 1.4 - 1.7
1130 1080 1026 367 76 14.1 10.8 3.2 1.6
1106 1065 1038 348 - - 11.1 3.2 1.5
Vitamin B12
1061 1092 1055 360 - - - - -
µg/g
1153 1096 1018 367 - - - - -
1122 1056 1017 355 - - - - -
Mean 1114 1078 1031 360 76 14.1 10.9 3.2 1.5
Std Dev. 34.3 17.2 15.9 8.3 - - - - -
%RSD 3.1 1.6 1.5 2.3 - - - - -
The results obtained were within the expected ranges for all concentrations and samples tested.
The repeatability was acceptable for samples 1 – 4 (see table) and although this has not been fully
assessed for the samples containing lower concentrations of Vitamin B12, the duplicates for samples
7 -9 were in good agreement.
8. Conclusion
The analysis of vitamin B12 in foods using liquid chromatography has, until recently been hampered
by the low concentrations found in natural foodstuffs. Although radioassays or protein binding
assays are available for analysis of vitamin B12 in blood and sera, these are less suitable for use
with foods. Much of the available data on foods has therefore been obtained using microbiological
assay but these are difficult to perform and require long analysis times. HPLC has been less used
as the concentration of vitamin B12 in foods at natural levels is too low to enable reliable detection
using UV detection at 361 nm without analyte concentration.
The availability of the immunoaffinity extraction cartridges enables the selective extraction of vitamin
B12 from foods. In addition to removing other food components which could interfere with
subsequent analysis, the technology allows the vitamin B12 content of the sample extract to be
concentrated before analysis. Analysis by HPLC using UV detection then becomes a realistic
option.
The cartridges tested worked well with vitamin premixes, vitamin tablets and infant formulae and
gave good results using a certified reference material prepared from pig liver. Its use with other
foods has not been fully tested although the selective nature of the product should enable its use
with many different food matrices. The instructions provided give a range of extraction options and
users should ensure that the extraction used is suitable for the product being determined.
The cartridges have a limited binding capacity which is not normally a problem when analysing
foods at natural levels but care is needed when analysing unknown products with higher
concentrations of vitamin B12 to ensure that the binding capacity is not exceeded. Recovery of
vitamin B12 from the matrix of interest should also be checked.
Chromatographic analysis using HPLC or UPLC can be used although the latter technique is faster.
Acceptable performance was obtained using both techniques with the conditions described.
The cartridges are easy to use and provide a rapid method for sample clean-up and analyte
concentration, thereby facilitating HPLC analysis. Combined with a suitable extraction protocol, the
technology enables rapid and reliable analysis of vitamin B12 in foods to be undertaken.
Page 11 of 12
A Government Chemist Programme Report
Report number LGC/R/2011/171
References
1. Eitenmiller R, Landen W, Jr. 1999, Vitamin Analysis for the Health and Food Sciences, 1999,
CRC Press. ISBN 0-8493-2668-0.
Page 12 of 12