Chapter 17 extraction techniques for the water soluble vitamins

17
Extraction Techniques for the
Water-Soluble Vitamins

In vitro analytical techniques require prior extraction of the vitamins from
the food matrix in order to facilitate their measurement. The appropriate
method of extraction depends upon the following criteria: the analytical
information required, the nature of the food matrix, the form in which
the vitamin occurs naturally or is added (different bound forms of vitamins are often found in meat, plant, and dairy products), the nature
and relative amounts of potentially interfering substances, the stability
of the vitamin towards heat and extremes of pH, and the selectivity and
specificity of the analytical method to be used. Extraction procedures
for water-soluble vitamins include hydrolysis of the sample with a
mineral acid [hydrochloric acid (HCl) or sulfuric acid (H2SO4)], alkaline
hydrolysis with calcium hydroxide, deproteinization with trichloroacetic acid or similarly acting agent, and digestion with an appropriate
enzyme.

17.1 Vitamin B1
The extraction procedure generally used for the determination of total
vitamin B1 by fluorometry, GC, HPLC, and microbiological assay involves
hot mineral acid digestion to release the thiamin and thiamin phosphate
esters from their association with proteins, followed by enzymatic
hydrolysis of the phosphate esters to complete the liberation of thiamin.
Food samples of animal origin can be autoclaved at 1218C for 30 min
with 0.1 N HCl, as the phosphorylated forms of thiamin present in such
samples are not degraded under these conditions. For the majority of
cereals and cereal products, which contain mostly nonphosphorylated
thiamin, it is necessary to lower the autoclaving temperature to 1088C
in order to avoid vitamin loss.
A commercial diastatic enzyme preparation of fungal origin (e.g.,
Takadiastase, Claradiastase, or Mylase) is suitable for the hydrolysis

step, as such preparations contain phosphatase activity in addition to
© 2006 by Taylor & Francis Group, LLC

321


322

Extraction Techniques for the Water-Soluble Vitamins

a-amylase and other enzymes [1]. The enzyme treatment can be omitted
for the analysis of those grain products that do not contain phosphorylated thiamin. For proteinaceous samples such as meat, the proteolytic
enzyme papain is sometimes added to the diastase in order to dissolve
the proteins that have been denatured during the previous acid digestion.
Instead of using an enzyme hydrolysis procedure for thiamin extraction
prior to HPLC, rice flour samples can be refluxed with a mixture of hydrochloric acid and methanol (0.1 N HCl –40% aqueous methanol) for 30 min
at 608C [2]. For the analysis of milk, the extraction procedure simply
entails precipitation of the protein by acidification at room temperature,
and filtration. This nonhydrolytic extraction procedure has the advantage
of leaving the biologically inactive thiamin monophosphate intact, so this
compound can be excluded from the measurement.

17.2 Vitamin B2
When carrying out physicochemical or microbiological assays for vitamin
B2, it is necessary to release the flavins from their intimate association with
proteins and to completely convert the FAD to FMN. Both of these
requirements are readily accomplished (for noncovalently bound
flavins) by autoclaving food samples at 1218C for 30 min with dilute
mineral acid (usually 0.1 N HCl) at a pH of ,3. During acid digestion
some of the FMN is hydrolyzed to riboflavin, and a small fraction of

the FMN is converted to the isomeric 20 -, 30 -, and 40 -phosphates [3].
The complete conversion of FMN to riboflavin can only be achieved by
subsequent enzymatic hydrolysis, for which a standardized diastatic
enzyme preparation such as Takadiastase or Claradiastase is used.
Watada and Tran [4] reported that Mylase was as effective as Takadiastase, the latter being unobtainable at that time. These are relatively
inexpensive and crude preparations that contain varying degrees of phosphatase activity. In practice, the complete enzymatic conversion of FMN
to riboflavin may not always be achieved, the degree of hydrolysis
depending on the source and batch-to-batch phosphatase activity of the
enzyme and on the incubation conditions.
For the analysis of milk, eggs, and dairy products, it is common practice
to determine the riboflavin specifically, on the assumption that free or
loosely bound riboflavin is the predominant naturally occurring flavin
present. In this case, the extraction procedure simply entails precipitation
of the protein by acidification and filtration, omitting the acid and enzyme
digestion steps. Rashid and Potts [5] removed the protein from milk and
milk products by filtration after treatment with acidified lead acetate
solution.
© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

323

Acid and enzymatic hydrolysis carried out successively are incapable
of liberating the covalently bound FAD of certain enzymes, and hence
this source of FAD will not be measured. This is perhaps fortuitous
when the nutritional value of the food sample is under assessment, as
there is evidence that covalently bound FAD is largely unavailable to
the host.

17.3 Niacin
In order to assess the nutritional value of a foodstuff with respect to its
niacin content, it is necessary to determine the niacin that is biologically
available. As discussed in Section 9.5.1, the majority of the niacin in
mature cereal grains exists in chemically bound forms of nicotinic acid
that are not biologically available. Therefore, measurement of total
niacin (i.e., free plus bound) provides a gross overestimate of the biologically available niacin of several staple cereal-based foods.
The terms “total” and “free” (bioavailable) niacin are defined by the
extraction methods employed in the analysis. Total niacin generally
refers to the niacin that is extractable by autoclaving the sample with
alkali or 1 N mineral acid; free niacin is frequently defined as the niacin
extractable by heating or autoclaving with 0.1 N mineral acid.
In the AOAC colorimetric method for determining total niacin [6],
noncereal foods are extracted by autoclaving for 30 min at 1218C in the
presence of 1 N (0.5 M) H2SO4. This same procedure is used in the
AOAC microbiological method for determining niacin in milk-based
infant formulas [7]. The acid treatment liberates nicotinamide from its
coenzyme forms and simultaneously hydrolyzes it to nicotinic acid; it
does not, however, completely liberate the bound nicotinic acid from
cereal products. A procedure that has been used for extracting total
niacin from cereal products is autoclaving at 1218C for 1 h in the presence
of 0.22 M calcium hydroxide [8,9]. This alkali treatment readily liberates
the nicotinic acid from its chemically bound forms; it also converts nicotinamide to nicotinic acid, but with a yield lower than 80%. Sodium
hydroxide, although more effective at hydrolyzing nicotinamide, is not
used because it induces gelation of the cereal sample. If the microbiological assay with Lactobacillus plantarum or the AOAC colorimetric assay are
to be used, complete conversion of nicotinamide to nicotinic acid is not
necessary, as these procedures account for both vitamers.
Autoclaving meat samples with 1 N HCl in the presence of urea
resulted in a significant increase in the niacin content when compared

with extraction using 1 N acid alone [10]. This suggests the release of
© 2006 by Taylor & Francis Group, LLC


324

Extraction Techniques for the Water-Soluble Vitamins

niacin from nonester conjugates by the acid – urea combination, possibly
from amide-linked forms.
Windahl et al. [11] found that autoclaving food samples at 1218C for
30 min in the presence of 1 N H2SO4 did not completely hydrolyze nicotinamide to nicotinic acid. These authors ensured complete hydrolysis
by autoclaving samples in the presence of 1.6 N (0.8 M) H2SO4 for 2 h at
1218C. They also performed alkaline extraction by autoclaving samples
in the presence of saturated calcium hydroxide for 2 h at 1218C. Acid
and alkali extractions gave similar levels of niacin in foods as determined
by capillary electrophoresis and HPLC. In meat samples, acid extraction
resulted in slightly higher niacin values compared with alkali extraction.
Conversely, in cereal samples, alkali extraction yielded slightly higher
values compared with acid extraction.
Among the many published HPLC methods for determining niacin in
foods, several have used extraction procedures designed to yield a
value for bioavailable niacin. Lahe´ly et al. [12] added 0.1 N HCl to
ground food samples and heated the suspensions in a water-bath at
1008C for 1 h. A portion of the diluted and filtered digest was then autoclaved at 1208C in a medium of 0.8 N NaOH for 1 h to ensure complete
conversion of nicotinamide to nicotinic acid. Thus, ultimately, only nicotinic acid needed to be measured chromatographically. The application
of this method to beef liver and yeast gave comparable niacin values to
those obtained when simulating gastric digestion conditions (0.1 N HCl
hydrolysis at 378C for 3 h, followed by an alkaline treatment). However,
when the method was applied to cereal products, the alkaline treatment

induced the formation of impurities, which interfered with the chromatography. Rose-Sallin et al. [13] found that a one-step acid hydrolysis (0.1 N
HCl, 1 h, 1008C water-bath) yielded similar concentrations of niacin to
those following two-step acid-alkaline or acid-enzymatic hydrolysis in a
range of fortified foods, including cereal products. The one-step
procedure also yielded slightly better recoveries for niacin compared to
the two-step methods. Rose-Sallin et al. [13] adopted the one-step extraction and calculated bioavailable niacin from the nicotinic acid and nicotinamide peaks in the chromatogram. Vidal-Valverde and Reche [14] found
that treatment of acid hydrolysates with Takadiastase was absolutely
necessary in the case of legume samples, because the high starch
content made the hydrolysate extremely viscous.
Ndaw et al. [15] replaced the usual 0.1 N acid extraction by enzymatic
hydrolysis, using a NADase that hydrolyses only the bound forms of
niacin clearly bioavailable (i.e., NAD and NADP). This enzymatic
hydrolysis (incubation at 378C for 18 h) did not induce any subsequent
conversion of nicotinamide into nicotinic acid. The one-step enzymatic
treatment was always sufficient, even when the foodstuff contained
large quantities of starch (rice, wheat flour) or proteins (wheat germ,
© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

325

TABLE 17.1
Influence of the Extraction Protocol on the Niacin Concentration in Various
Foodstuffs as Determined by HPLC with Fluorometric Detection

Food
Peas
Spinach

French beans
Sweet corn
Rice
Wheat flour
Wheat germ
Peanuts
Yeast
Beef fillet
Pork escalope

Extraction
Protocola

Nicotinic Acid
(mg/g)

Nicotinamide
(mg/g)

Niacin (mg/g of
Nicotinic Acid
Equivalents)

NADase
0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl
NADase

0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl
NADase
0.1 N HCl

0.29 (0.01)
1.22 (0.08)
0
0
0.19 (0.01)
0.37 (0.06)
3.6 (0.2)
4.3 (0.4)
10.3 (0.2)
10.0 (0.5)
3.4 (0.1)
5.7 (0.4)
10.8 (0.2)
13.8 (0.5)

26.5 (0.9)
93.4 (0.7)
17 (1)
22.0 (0.3)
3.8 (0.2)
3.6 (0.7)
0
0.2 (0.1)

11.0 (0.1)
10.2 (0.4)
0.72 (0.06)
0.69 (0.04)
2.8 (0.2)
2.6 (0.2)
13.8 (1.0)
12.7 (0.9)
0
0
1.7 (0.1)
1.9 (0.1)
0
0
3.7 (0.3)
1.9 (0.3)
182 (5)
174 (5)
53 (1)
50 (2)
64 (2)

57 (1)

11.3 (0.1)
11.4 (0.4)
0.72 (0.06)
0.69 (0.04)
3.0 (0.2)
3.0 (0.2)
17.4 (1.0)
17.0 (1.0)
10.3 (0.2)
10.0 (0.5)
5.2 (0.1)
7.6 (0.4)
10.8 (0.2)
13.8 (0.5)
30.2 (1.0)
95.8 (0.8)
199 (5)
196 (5)
57 (1)
54 (2)
64 (2)
58 (1)

Note: Concentrations are averages of three determinations (standard deviations in
parentheses).
a
(1) NADase (pH 4.5, 18 h, 378C); (2) 0.1 N HCl (water-bath at 1008C during 1 h).
Source: From Ndaw, S. et al. Food Chem, 78, 129–134, 2002. With permission from Elsevier.

peanuts, beef fillet). Table 17.1 compares the niacin contents of various
foods extracted either by NADase treatment or acid hydrolysis (0.1 N
HCl, 1 h, 1008C water-bath). Acid hydrolysis led to significantly higher
niacin contents in the analysis of wheat flour, wheat germ, and peanuts,
attributable to the release of nicotinic acid from bound forms that are
probably nonbioavailable. On analysis of peas, French beans, and yeast
(foods in which nicotinamide is by far the major vitamer), nicotinic acid
contents were slightly higher after acid hydrolysis than they were after
enzymatic hydrolysis. This increase most probably resulted from a
partial conversion of the nicotinamide to nicotinic acid. When the acid
hydrolysis was applied to standard solutions of NAD (1.35 mM) and
NADP (1.17 mM), about 10% of the nicotinamide liberated was converted
to nicotinic acid.
© 2006 by Taylor & Francis Group, LLC


326

Extraction Techniques for the Water-Soluble Vitamins

For the determination of added nicotinic acid as a color fixative in fresh
meat (illegal in Japan), meat samples have been extracted by boiling with
96% ethanol [16] and blending with water [17 –19], acetonitrile [20],
methanol [21], or methanol after addition of a small amount of phosphoric
acid [22].

17.4 Vitamin B6
Because animal and plant tissues differ greatly with respect to the forms
of vitamin B6 contained in them, there is no single set of conditions that

can quantitatively extract vitamin B6 from both plant and animal products. In the AOAC microbiological method [23] for determining total
vitamin B6 in food extracts, animal-derived foods are autoclaved with
0.055 N HCl for 5 h at 1218C. This treatment hydrolyzes phosphorylated
forms of vitamin B6, whilst also liberating PL from its Schiff base and substituted aldamine bound forms. Plant-derived foods are autoclaved with
0.44 N HCl for 2 h at 1218C, the stronger acid environment being necessary to liberate PN from its glycosylated form. Autoclaving whole-wheat
samples with 0.055 N HCl, instead of 0.44 N HCl, yielded a similar PL
value, but lower values of PN and PM [24]. Conversely, autoclaving
meat products with 0.44 N HCl, instead of 0.055 N HCl, gave approximately the same PN and PL values, but only about half of the PM [25].
The superiority of the lower concentration of acid used for animal products does not result from destruction of vitamin B6 by the stronger
acid; rather, it is due to the incomplete liberation of the vitamin by the
more concentrated acid [26]. The optimum release occurs between pH
1.5 and 2.0, with a maximum at pH 1.7 –1.8 [27]. To satisfy these strict
pH criteria, one must always ensure that the acid is added in amounts
that exceed the buffering capacity of the sample. Another factor to
consider is that PMP is more resistant to acid hydrolysis than is PLP.
Autoclaving for 3 h at 1258C in 0.055 N HCl was required for complete
hydrolysis of PMP, while PLP was completely hydrolyzed in 30 min
under the same conditions [28].
The possibility of interaction of PL or PLP with amino acids during the
AOAC extraction procedure for animal foods has been investigated [29].
No loss of activity for Saccharomyces cerevisiae was observed when PL or
PLP was autoclaved in the presence of a relatively high concentration
of glutamic acid, which indicated that transamination does not occur
under these conditions.
PN-glucoside exhibits around 60% bioavailability relative to PN in
humans [30]. Since the AOAC extraction procedure for plant foods hydrolyzes glycosylated forms of PN, analyses based on this procedure
© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

327

would overestimate the biologically available vitamin B6 in foods that
contain significant quantities of b-glucoside conjugates.
The AOAC acid hydrolysis procedures have no effect upon the peptidebound 1-pyridoxyllysine and its 50 -phosphate derivative, which are
formed during the heat-sterilization of evaporated milk and other
animal-derived canned foods (Section 10.3.2). These conjugates,
which possess anti-vitamin B6 activity under certain conditions, exhibit
75 –80% stability when subjected to 6 N HCl at 1058C for 48 h [31].
Bogna˚r and Ollilainen [32] investigated the use of hydrochloric acid
and trichloroacetic acid alone, and in combination with several commercial enzyme preparations, as extractants for the determination of total
vitamin B6 in food by HPLC. Three reference materials were tested:
CRM 121 (wholemeal flour), CRM 485 (lyophilized mixed vegetables)
and CRM 487 (lyophilized pig liver). Also included in the investigation
were broccoli, Brussels sprouts, kidney beans, spelt (a kind of wheat),
potatoes, sunflower seeds, pork meat, cod, and milk. The highest values
of total vitamin B6 were achieved by autoclaving samples at 1208C for
30 min in 0.1 N HCl, followed by incubation with acid phosphatase and
b-glucosidase at 378C for 18 h after adjustment to pH 4.8. Enzymatic
hydrolysis of food by Takadiastase, degraded PL distinctly and also produced a compound that interfered with the PN peak during gradient
elution. The content of glycosylated PN could be determined by analyzing the acid hydrolysate before and after the double enzyme treatment.
The difference in PN content before and after enzyme treatment gives
an estimate of glycosylated PN.
The simultaneous separation of all six B6 vitamers, plus pyridoxic acid,
can be achieved using HPLC. Treatment of samples with deproteinizing
agents such as metaphosphoric, perchloric, trichloroacetic, or sulfosalicylic acid at ambient temperature readily hydrolyzes Schiff bases,
whilst preserving the phosphorylated vitamers. These acids also preserve
PN-glucoside, and hence their use provides better estimates of available
vitamin B6 than the use of mineral acids. The high efficiency of extraction

using these acidic reagents is partly due to the conversion of the pyridine
bases to quaternary ammonium salts, thereby increasing their solubility
in water. Their use as extracting agents also prevents enzymatic interconversion of B6 vitamers during homogenization of samples. In such
procedures it is usually necessary to remove excess reagent, which
might otherwise interfere with the analytical chromatography. Trichloroacetic acid can be removed by extraction with diethyl ether; perchloric acid
by reaction with 6 M potassium hydroxide and precipitation as insoluble
potassium perchlorate; and sulfosalicylic acid by chromatography on an
anion exchange column [33]. An extraction procedure using 5% sulfosalicylic acid has been successfully applied to such complex foods as
pork, dry milk, and cereals [34]. Recoveries of B6 vitamers added to
© 2006 by Taylor & Francis Group, LLC


328

Extraction Techniques for the Water-Soluble Vitamins

samples were 95– 105% for all vitamers except for PNP, where the
recovery was 85%. Other workers [35,36] have found perchloric acid to
be a better extracting agent of the B6 vitamers for animal tissues than
sulfosalicylic acid.

17.5 Pantothenic Acid
Before pantothenic acid can be determined by methods other than an
animal bioassay, it is necessary to liberate the vitamin from its bound
forms, chiefly coenzyme A. Neither acid nor alkaline hydrolysis can be
used, as the pantothenic acid is degraded by such treatments. The only
practicable alternative is enzymatic hydrolysis, and this was successfully
accomplished through the simultaneous action of intestinal phosphatase
and an avian liver enzyme [37]. This double enzyme combination liberates practically all of the pantothenic acid from coenzyme A, but it does
not release the vitamin from acyl carrier protein [38]. The phosphatase

splits the coenzyme A molecule between the phosphate-containing
moiety and pantethiene, while the liver enzyme breaks the link in
pantethiene between the pantothenic acid and b-mercaptoethylamine
moieties. The double enzyme combination is used in the AOAC microbiological method for determining pantothenic acid in milk-based
infant formula [39].

17.6 Biotin
Bound forms of biotin, including biocytin, cannot be utilized by
L. plantarum, the organism usually employed in microbiological biotin
assays, and strong mineral acid hydrolysis at elevated temperature is
required to liberate biotin completely from natural materials [40].
Animal tissues require more stringent hydrolysis conditions than do
plant tissues, because the latter contain a higher proportion of free
water-extractable biotin [41]. Experimental studies with meat and meat
products [42] and feedstuffs of animal origin [43] showed that
maximum liberation of biotin in animal-derived products is obtained by
autoclaving with 6 N H2SO4 for 2 h at 1218C. This procedure promotes
losses of biotin in plant materials, which are extracted more efficiently
by autoclaving with 4 N H2SO4 for 1 h at 1218C [41] or with 2 N H2SO4
for 2 h at 1218C [43]. Because of the differences in extractability between
animal and plant tissues, a single acid extraction procedure to cover all
food commodities must be a compromise, and no such procedure has
© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

329

been universally adopted. Representative methods for extracting foods of

any type entail autoclaving with 2 N H2SO4 at 1218C for 2 h [41,44] or 3 N
H2SO4 at 1218C for 1 h [45] or 30 min [46].
Hydrolysis with 6 N H2SO4 destroys the synthetic sodium salt of biotin
added to feed premixes. A suggested procedure for extracting feed
premixes with biotin potencies up to 1 g/lb entailed the addition of
50 ml of 0.1 N NaOH and 250 ml of water to 5 g of sample, shaking vigorously, and then standing for 30 min at room temperature with occasional
swirling [47].
Sulfuric acid, rather than hydrochloric acid, is invariably used for
sample hydrolysis, as the biotin content of dilute (30 ng/ml) solutions is
almost completely destroyed by autoclaving with 2 N HCl [48]. Evidence
from differential microbiological assay points to the oxidation of biotin to
a mixture of its sulfoxide and sulfone derivatives, possibly caused by
trace impurities (e.g., chlorine) in the acid. This loss of vitamin activity
does not necessarily occur when autoclaving actual food samples, as
many natural products are capable of preventing this oxidation [49].
Finglas et al. [50] reported no loss of biotin from liver using 3 N HCl.
It is evident from the foregoing that sulfuric acid hydrolysis is an unreliable way of extracting biotin from food. The results depend on both the
concentration of acid and the duration of autoclaving. This makes the
microbiological assay of biotin problematic, since acid hydrolysis is
used to convert biocytin to biotin. A proposed HPLC method [51]
solves the problems associated with acids by eliminating acid extraction.
Instead, food samples are digested with papain for 18 h, a treatment that
releases biotin from its association with proteins, but leaves biocytin
intact. There is no degradation of biotin during the digestion at 378C.
Biotin and biocytin are measured separately after postcolumn conversion
to fluorescent derivatives. The addition of Takadiastase is necessary for
starchy foods such as cereals and yeast.

17.7 Folate
The AOAC microbiological method for determining folic acid in infant

formula [52] employs a single-enzyme digestion with folate conjugase
(pteroylpoly-g-glutamyl hydrolase; EC 3.4.22.12). The chicken pancreas
conjugase specified in the method converts folylpolyglutamates to diglutamates, which can be utilized by the assay organism, L. rhamnosus.
HPLC methods for determining folate require deconjugation of folylpolyglutamates to monoglutamates, and therefore chicken pancreas conjugase
is unsuitable. Conjugases from hog kidney and human or rat plasma
do yield folylmonoglutamates and can be used in HPLC and other
© 2006 by Taylor & Francis Group, LLC


330

Extraction Techniques for the Water-Soluble Vitamins

nonmicrobiological methods. Chicken pancreas conjugase is most
active at neutral pH, in contrast to hog kidney conjugase and plasma
(human or rat) conjugase whose pH optimum is 4.5 [53]. The various
conjugases are not commercially available in purified form and enzyme
solutions have to be prepared in the laboratory from their crude
sources, such as lyophilized human or rat plasma and hog kidney
acetone powder.
In 1990, DeSouza and Eitenmiller [54] reported that increased folate
levels could be obtained in microbiological and radioassays by including
protease (EC 3.4.24.31) and a-amylase (3.2.1.1) with the conjugase treatment. Martin et al. [55] then published a tri-enzyme digestion procedure
using chicken pancreas conjugase, a-amylase, and protease in the microbiological determination of total folate in foods. This was followed by
reports from other laboratories advocating tri-enzyme treatment as a
means of extracting the maximum possible amount of folate from foods
as diverse as cereal-grain products [56], American fast foods [57], dairy
products [58], foods commonly consumed in Korea [59], and complete
food composites [60]. Folate values in 8 of 16 fortified bakery products,
and 4 of 13 fortified products in the rice, macaroni, and noodle category

were significantly higher following the additional protease and
a-amylase treatments [61].
In order to achieve maximum extraction of bound folate from the food
matrix, food samples suspended in buffered aqueous medium are first
autoclaved to break up particles, gelatinize starch, and denature folatebinding proteins and enzymes that may catalyze folate degradation or
interconversion. The inclusion of an antioxidant is essential in preventing
the destruction of labile folates during heat treatment. The most effective
reducing conditions are provided by the presence of both ascorbic acid
and mercaptoethanol, with the air displaced by nitrogen. The autoclaved
samples are digested with protease to liberate the folate bound to proteins, then heated to inactivate the protease. Digestion with a-amylase
then follows to liberate the folate bound to starch. Prolonged digestion
with conjugase completes the tri-enzyme treatment. A variety of foods
cause detectable inhibition of conjugase activity [62], but this problem
can be partly overcome by extracting at near neutral pH using a large
excess of conjugase [63].
In 2000, the American Association of Cereal Chemists (AACC) [64]
published a microbiological assay using tri-enzyme extraction for the
determination of total folate in cereal products. The extraction procedure
is as follows: Weigh an amount of ground sample equal to 0.25 – 1.0 g dry
solids and containing about 1 mg folic acid into a 125-ml conical flask. Add
20 ml 0.1 M phosphate buffer (pH 7.8) containing 1% ascorbic acid, mix
thoroughly, then add enough water to bring the total volume to 50 ml.
Add 0.1– 1.0 ml octanol (antifoaming agent), cover flasks with 50-ml
© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

331

beakers, and autoclave for 15 min at 121 – 1238C. Cool and add a further
10 ml of the pH 7.8 buffer. Add 1 ml protease solution, cover the flask,
and incubate for 3 h at 378C. Autoclave for 3 min at 1008C, then cool.
Add 1 ml a-amylase solution and incubate for 2 h at 378C. Add 4 ml
chicken pancreas conjugase solution and incubate for 16 h (or overnight)
at 378C. Inactivate the enzymes by autoclaving for 3 min, then cool.
Adjust to pH 4.5, dilute to 100 ml with water, and filter approximately
20 ml through 2V filter paper. Dilute an aliquot of the clear filtrate with
0.1 M phosphate buffer (pH 6.7 + 0.1) to a final volume such that the
folate concentration is about 0.2 ng/ml, and assay microbiologically
using L. rhamnosus.
Rader et al. [61] tested the efficiency of the tri-enzyme extraction using
chicken pancreas conjugase at four different pHs (pH 4.3, 6.0, 6.8, and 7.8)
for the microbiological assay of four cereal-grain products and found no
significant differences among folate values. A pH 7.8 buffer is used in
the AACC method, but this pH is not optimal for the tri-enzyme extraction of all food types. Tamura et al. [60], for example, found that
complex food composites were extracted more efficiently at pH 4.1 than
at pH 6.3 or 7.85. The pH optima and incubation times for protease and
a-amylase can vary, depending on the substrates present in the foods
[65], and this creates a dilemma in deciding which conditions should be
used for the tri-enzyme treatment.

17.8 Vitamin B12
Procedures for extracting vitamin B12 generally have the dual purpose of
liberating protein-bound cobalamins and converting the labile naturally
occurring forms to a single, stable form — cyanocobalamin or sulfitocobalamin. Conversion to the sulfitocobalamin by reaction with metabisulfite
avoids the use of lethally toxic cyanide solutions required to form
cyanocobalamin.
The extraction procedure employed in the AOAC microbiological
method for determining vitamin B12 activity in vitamin preparations

[66] is also applicable to foods, having been found satisfactory by interlaboratory collaborative analysis of a crude liver paste, condensed fish
solubles, and a crude vitamin B12 fermentation product [67]. The procedure entails homogenizing the sample with 0.1 M phosphate – citrate
buffer at pH 4.5 containing freshly prepared sodium metabisulfite
(Na2S2O5), and then autoclaving the mixture for 10 min at 1218C.
The heat treatment denatures the proteins, inactivates the enzymes, and
accelerates the conversion of liberated cobalamins to sulfitocobalamin.
© 2006 by Taylor & Francis Group, LLC


332

Extraction Techniques for the Water-Soluble Vitamins

In the AOAC method for determining vitamin B12 in milk-based infant
formula [68], protein is removed by filtration after adjustment of the autoclaved extract to the point of maximum precipitation (ca. pH 4.5).
Methods, in which the sample is heated on a boiling water-bath, rather
than autoclaved, may not completely extract all of the bound vitamin [69].

17.9 Vitamin C
An effective means of extracting vitamin C from foods is homogenization
with a solution of 3% (w/v) metaphosphoric acid dissolved in 8% glacial
acetic acid [70]. This extracting solution denatures and precipitates
proteins (thereby inactivating all enzymes) and provides a medium
below pH 4, which favors the stability of ascorbic acid and dehydroacorbic acid. Furthermore, metaphosphoric acid prevents catalysis of the
oxidation of ascorbic acid by copper(II) or iron(III) ions [71]. Addition
of ethanol or acetone to the metaphosphoric extract precipitates solubilized starch [72]. Dilute (5 or 6%) metaphosphoric acid forms precipitates
when mixed with certain ion-pairing reagents [73], which warrants
caution in its use for ion-pair chromatography. It is recommended to
deoxygenate extracting solutions by bubbling an inert gas (e.g., oxygenfree nitrogen) through the solution before use.
Krall and Andersen [74] extracted fruits and vegetables with an

aqueous solution of 1% (w/v) metaphosphoric acid and 0.5% (w/v)
oxalic acid adjusted to pH 2. Homogenization of high-starch food
samples with aqueous 2% metaphosphoric acid and 1% oxalic acid
mixed 1 : 1 with ethanol resulted in precipitation of starch. Use of these
extractants was compatible with the reversed-phase ion-pair HPLC
system.
Bogna´r and Daood [75] compared two solvent systems, A and B, for
their effect on the stability of vitamin C derivatives in standard solutions
and spiked extracts of fruits and vegetables. Solvent A was the classic
solution of 3% metaphosphoric acid in 8% acetic acid; solvent B was
solvent A mixed with acetonitrile (1 : 2). In solvent A, dehydroascorbic
acid was unstable in standard solution: only 64 and 40% of the initial concentration was retained after 8 and 24 h of ambient storage, respectively.
In contrast, the corresponding retentions in solvent B were 100 and 98%.
In fruit and vegetable extracts prepared in solvent A, 36 –54% of the
spiked quantity of dehydroascorbic acid was lost after 12 h of storage
time. These decreases in dehydroascorbic acid were accompanied by
remarkable increases in ascorbic acid concentration (29 – 53%), most probably due to the presence of reducing agents in the extracts of fruits and
vegetables. There was little or no loss of dehydroascorbic acid added to
© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

333

food extracts prepared in solvent B. After a 24-h storage, ascorbic acid was
highly stable in all of the standard solutions and food extracts tested.
Taking these findings into account, Bogna´r and Daood [75] added
acetonitrile to the standard solutions or to food extracts immediately
after preparation or extraction with 3% metaphosphoric acid in 8%

acetic acid. This modification of the extraction procedure resulted in complete recoveries of dehydroascorbic acid and total vitamin C.
In an interlaboratory study [76], fruit juices and processed foods were
extracted by diluting or blending with water, then immediately adding
dithiothreitol to reduce dehydroascorbic acid to ascorbic acid, thereby
stabilizing the vitamin C. Proteinaceous samples were treated with 5%
trichloroacetic acid to precipitate the proteins.

References
1. Defibaugh, P.W., Evaluation of selected enzymes for thiamine determination,
J. Assoc. Off. Anal. Chem., 70, 514, 1987.
2. Ohta, H., Maeda, M., Nogata, Y., Yoza, K.-I., Takeda, Y., and Osajima, Y., A
simple determination of thiamine in rice (Oryza sativa) by high-performance
liquid chromatography with post-column derivatization, J. Liq. Chromatogr.,
16, 2617, 1993.
3. Nielsen, P., Rauschenbach, P., and Bacher, A., Preparation, properties, and
separation by high-performance liquid chromatography of riboflavin
phosphates, Meth. Enzymol., 122G, 209, 1986.
4. Watada, A.E. and Tran, T.T., A sensitive high-performance liquid chromatography method for analyzing riboflavin in fresh fruits and vegetables,
J. Liq. Chromatogr., 8, 1651, 1985.
5. Rashid, I. and Potts, D., Riboflavin determination in milk, J. Food Sci., 45, 744,
1980.
6. AOAC official method 961.14. Niacin and niacinamide in drugs, foods, and
feeds. Colorimetric method. Final action 1962, in Official Methods of Analysis
of AOAC International, Indyk, H. and Konings, E., Eds., 17th ed., AOAC
International, Gaithersburg, MD, 2000, p. 45-12.
7. AOAC official method 985.34. Niacin and niacinamide (nicotinic acid
and nicotinamide) in ready-to-feed milk-based infant formula. Microbiological-turbidimetric method. Final action 1988, in Official Methods of Analysis
of AOAC International, Phifer, E., Ed., 17th ed., AOAC International,
Gaithersburg, MD, 2000, p. 50-21.
8. Tyler, T.A. and Shrago, R.R., Determination of niacin in cereal samples by

HPLC, J. Liq. Chromatogr., 3, 269, 1980.
9. van Niekerk, P.J., Smit, S.C.C., Strydom, E.S.P., and Armbruster, G., Comparison of a high-performance liquid chromatographic and microbiological
method for the determination of niacin in foods, J. Agric. Food Chem., 32,
304, 1984.
© 2006 by Taylor & Francis Group, LLC


334

Extraction Techniques for the Water-Soluble Vitamins

10. Roy, R.B. and Merten, J.J., Evaluation of urea-acid system as medium of extraction for the B-group vitamins. Part II. Simplified semi-automated chemical
analysis for niacin and niacinamide in cereal products, J. Assoc. Off. Anal.
Chem., 66, 291, 1983.
11. Windahl, K.L., Trenerry, V.C., and Ward, C.M., The determination of niacin in
selected foods by capillary electrophoresis and high performance liquid
chromatography: acid extraction, Food Chem., 65, 263, 1998.
12. Lahe´ly, S., Bergaentzle´, M., and Hasselmann, C., Fluorimetric determination of
niacin in foods by high-performance liquid chromatography with postcolumn derivatization, Food Chem., 65, 129, 1999.
13. Rose-Sallin, C., Blake, C.L., Genoud, D., and Tagliaferri, E.G., Comparison of
microbiological and HPLC — fluorescence detection methods for determination of niacin in fortified food products, Food Chem., 73, 473, 2001.
14. Vidal-Valverde, C. and Reche, A., Determination of available niacin in
legumes and meat by high-performance liquid chromatography, J. Agric.
Food Chem., 39, 116, 1991.
15. Ndaw, S., Bergaentzle´, M., Aoude´-Werner, D., and Hasselmann, C., Enzymatic
extraction procedure for the liquid chromatographic determination of niacin
in foodstuffs, Food Chem., 78, 129, 2002.
16. Gorin, N. and Schu¨tz, G.P., Comparison of a microbiological and a spectrophotometric method for the determination of nicotinic acid in fresh meat, J. Sci.
Food Agric., 21, 423, 1970.
17. van Gend, H.W., An automated colorimetric method for the determination

of free nicotinic acid in minced meat, Z. Lebensm. Unters. Forsch., 153, 73, 1973.
18. Takatsuki, K., Suzuki, S., Sato, M., Sakai, M., and Ushizawa, I., Liquid chromatographic determination of free and added niacin and niacinamide in beef and
pork, J. Assoc. Off. Anal. Chem., 70, 698, 1987.
19. Hamano, T., Mitsuhashi, Y., Aoki, N., and Yamamoto, S., Simultaneous determination of niacin and niacinamide in meats by high-performance liquid
chromatography, J. Chromatogr., 457, 403, 1988.
20. Tanaka, A., Iijima, M., Kikuchi, Y., Hoshino, Y., and Nose, N., Gas chromatographic determination of nicotinamide in meats and meat products as
3-cyanopyridine, J. Chromatogr., 466, 307, 1989.
21. Oishi, M., Amakawa, E., Ogiwara, T., Taguchi, N., Onishi, K., and Nishijima, M.,
Determination of nicotinic acid and nicotinamide in meats by high performance liquid chromatography and conversion of nicotinamide in meats
during storage, J. Food Hyg. Soc., Jpn., 29, 32, 1988 (in Japanese).
22. Tsunoda, K., Inoue, N., Iwasaki, H., Akiya, M., and Hasebe, A., Rapid
simultaneous analysis of nicotinic acid and nicotinamide in foods, and
their behaviour during storage, J. Food Hyg. Soc., Jpn., 29, 262, 1988 (in
Japanese).
23. AOAC official method 961.15. Vitamin B6 (pyridoxine, pyridoxal, pyridoxamine) in food extracts. Microbiological method. Final action 1975, in Official
Methods of Analysis of AOAC International, Indyk, H. and Konings, E., Eds.,
17th ed., AOAC International, Gaithersburg, MD, 2000, pp. 45 – 55.
24. Polansky, M.M., Murphy. E.W., and Toepfer, E.R., Components of vitamin B6
in grains and cereal products, J. Assoc. Off. Anal. Chem., 47, 750, 1964.

© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

335

25. Polansky, M.M. and Toepfer, E.W., Vitamin B-6 components in some meats,
fish, dairy products, and commercial infant formulas, J. Agric. Food Chem.,
17, 1394, 1969.

26. Rubin, S.H. and Scheiner, J., The availability of vitamin B6 in yeast to Sacchararomyces carlsbergensis, J. Biol. Chem., 162, 389, 1946.
27. Rubin, H., Scheiner, J., and Hirschberg, E., The availability of vitamin B6 in
yeast and liver for growth of Saccharomyces carlsbergensis, J. Biol. Chem., 167,
599, 1947.
28. Gregory, M.E. and Mabbitt, L.A., The occurrence of pyridoxamine phosphate
in milk, J. Dairy Res., 28, 293, 1961.
29. Rabinowitz, J.C., Mondy, N.I., and Snell, E.E., The vitamin B6 group. XIII. An
improved procedure for determination of pyridoxal with Lactobacillus casei,
J. Biol. Chem., 175, 147, 1948.
30. Gregory, J.F., III, Trumbo, P.R., Bailey, L.B., Toth, J.P., Baumgartner, T.G., and
Cerda, J.J., Bioavailability of pyridoxine-50 -b-glucoside determined in
humans by stable isotopic methods, J. Nutr., 121, 177, 1991.
31. Gregory, J.F., III, Ink, S.L., and Sartain, D.B., Degradation and binding to food
proteins of vitamin B-6 compounds during thermal processing, J. Food Sci., 51,
1345, 1986.
32. Bogna˚r, A. and Ollilainen, V., Influence of extraction on the determination of
vitamin B6 in food by HPLC, Z. Lebens. Unters. Forsch., A, 204, 327, 1997.
33. Ollilainen, V., HPLC analysis of vitamin B6 in foods, Agric. Food Sci. Finland,
8 (6), 515, 1999.
34. Vanderslice, J.T., Maire, C.E., Doherty, R.F., and Beecher, G.R., Sulfosalicylic
acid as an extraction agent for vitamin B6 in food, J. Agric. Food Chem., 28,
1145, 1980.
35. Bitch, R. and Mo¨ller, J., Analysis of B6 vitamers in foods using a modified
high-performance liquid chromatographic method, J. Chromatogr., 463, 207,
1989.
36. Toukairin-Oda, T., Sakamoto, E., Hirose, N., Mori, M., Itoh, T., and Tsuge, H.,
Determination of vitamin B6 derivatives in foods and biological materials by
reversed-phase HPLC, J. Nutr. Sci. Vitaminol., 35, 171, 1989.
37. Novelli, G.D., Kaplan, N.O., and Lipmann, F., The liberation of pantothenic
acid from coenzyme A, J. Biol. Chem., 177, 97, 1949.

38. Wyse, B.W., Song, W.O., Walsh, J.H., and Hansen, R.G., Pantothenic acid,
in Methods of Vitamin Assay, Augustin, J., Klein, B.P., Becker, D., and
Venugopal, P.B., Eds., 4th ed., John Wiley & Sons, New York, 1985, pp. 399.
39. AOAC official method 992.07. Pantothenic acid in milk-based infant formula.
Microbiological turbidimetric method. Final action 1995, in Official Methods of
Analysis of AOAC International, 17th ed., Phifer, E., Ed., AOAC International,
Gaithersburg, MD, 2000, p. 50-26.
40. Thompson, R.C., Eakin, R.E., and Williams, R.J., The extraction of biotin from
tissues, Science, 94, 589, 1941.
41. Lampen, J.O., Bahler, G.P., and Peterson, W.H., The occurrence of free and
bound biotin, J. Nutr., 23, 11, 1942.
42. Schweigert, B.S., Nielsen, E., McIntire, J.M., and Elvehjem, C.A., Biotin content
of meat and meat products, J. Nutr., 26, 65, 1943.

© 2006 by Taylor & Francis Group, LLC


336

Extraction Techniques for the Water-Soluble Vitamins

43. Scheiner, J. and De Ritter, E., Biotin content of feedstuffs, J. Agric. Food Chem.,
23, 1157, 1975.
44. Strohecker, R. and Henning, H.M., Vitamin Assay. Tested Methods, Verlag
Chemie, Weinheim, 1966, pp. 173.
45. Shull, G.M., Hutchings, B.L., and Peterson, W.H., A microbiological assay for
biotin, J. Biol. Chem., 142, 913, 1942.
46. Bell, J.G., Microbiological assay of vitamins of the B group in foodstuffs, Lab.
Pract., 23, 235, 1974.
47. Scheiner, J., Extraction of added biotin from animal feed premixes, J. Assoc. Off.

Anal. Chem., 49, 882, 1966.
48. Axelrod, A.E. and Hofmann, K., The inactivation of biotin by hydrochloric
acid, J. Biol. Chem., 187, 23, 1950.
49. Langer, B.W., Jr. and Gyo¨rgy, P., Biotin. VIII. Active compounds and antagonists, in The Vitamins. Chemistry, Physiology, Pathology, Methods, Sebrell, W.H., Jr.
and Harris, R.S., Eds., 2nd ed., Vol. II, Academic Press, New York, 1968,
pp. 294.
50. Finglas, P.M., Faulks, R.M., and Morgan, M.R.A., The analysis of biotin in liver
using a protein-binding assay, J. Micronutr. Anal., 2, 247, 1986.
51. Lahe´ly, S., Ndaw, S., Arella, F., and Hassellmann, C., Determination of biotin in
foods by high-performance liquid chromatography with post-column derivatization and fluorimetric detection, Food Chem., 65, 253, 1999.
52. AOAC official method 992.05. Folic acid (pteroylglutamic acid) in infant
formula. Microbiological methods. Final action 1995, in Official Methods of
Analysis of AOAC International, Phifer, E., Ed., 17th ed., AOAC International,
Gaithersburg, MD, 2000, p. 50-21.
53. Gregory, J.F., III, Chemical and nutritional aspects of folate research: analytical
procedures, methods of folate synthesis, stability, and bioavailability of
dietary folates, Adv. Food Nutr. Res., 33, 1, 1989.
54. De Souza, S. and Eitenmiller, R., Effects of different enzyme treatments on
extraction of total folate from various foods prior to microbiological assay
and radioassay, J. Micronutr. Anal., 7, 37, 1990.
55. Martin, J.I., Landen, W.O., Jr., Soliman, A.-G.M., and Eitenmiller, R.R.,
Application of a tri-enzyme extraction for total folate determination in
foods, J. Assoc. Off. Anal. Chem., 73, 805, 1990.
56. Pfeiffer, C.M., Rogers, L.M., and Gregory, J.F., III, Determination of folate in
cereal-grain food products using trienzyme extraction and combined affinity
and reversed-phase liquid chromatography, J. Agric. Food Chem., 45, 407, 1997.
57. Johnston, K.E., Lofgren, P.A., and Tamura, T., Folate concentrations of fast
foods measured by trienzyme extraction method, Food Res. Int., 35, 565, 2002.
58. Johnston, K.E., DiRienzo, D.B., and Tamura, T., Folate content of dairy
products measured by microbiological assay with trienzyme treatment,

J. Food Sci., 67, 817, 2001.
59. Yon, M. and Hyun, T.H., Folate content of foods commonly consumed in
Korea measured after trienzyme extraction, Nutr. Res., 23, 735, 2003.
60. Tamura, T., Mizuno, Y., Johnston, K.E., and Jacob, R.A., Food folate assay with
protease, a-amylase, and folate conjugase treatments, J. Agric. Food Chem., 45,
135, 1997.

© 2006 by Taylor & Francis Group, LLC


Vitamins in Foods: Analysis, Bioavailability, and Stability

337

61. Rader, J.I., Weaver, C.M., and Angyal, G., Use of a microbiological assay with
tri-enzyme extraction for measurement of pre-fortification levels of folates in
enriched cereal-grain products, Food Chem., 62, 451, 1998.
62. Engelhardt, R. and Gregory, J.F., III, Adequacy of enzymatic deconjugation in
quantification of folate in foods, J. Agric. Food Chem., 38, 154, 1990.
63. Goli, D.M. and Vanderslice, J.T., Investigation of the conjugase treatment
procedure in the microbiological assay of folate, Food Chem., 43, 57, 1992.
64. AACC method 86-47. Total folate in cereal products — microbiological assay
using trienzyme extraction, in Approved Methods of the American Association of
Cereal Chemists, 10th ed., Vol. II, Association of Cereal Chemists, Inc., St.
Paul, MN, 2000.
65. Aiso, K. and Tamura, T., Trienzyme treatment for food folate analysis: optimal
pH and incubation time for a-amylase and protease treatments, J. Nutr. Sci.
Vitaminol., 44, 361, 1998.
66. AOAC official method 952.20. Cobalamin (vitamin B12 activity) in vitamin
preparations. Microbiological methods. Final action 1960, in Official Methods

of Analysis of AOAC International, Indyk, H. and Konings, E., Eds., 17th ed.,
AOAC International, Gaithersburg, MD, 2000, pp. 45 –47.
67. Krieger, C.H., Report on vitamin B12. Microbiological method, J. Assoc. Off.
Agric. Chem., 37, 781, 1954.
68. AOAC official method 986.23. Cobalamin (vitamin B12 activity) in milk-based
infant formula. Turbidimetric method. Final action 1988, in Official Methods of
Analysis of AOAC International, Phifer, E., Ed., 17th ed., AOAC International,
Gaithersburg, MD, 2000, p. 50-21.
69. Casey, P.J., Speckman, K.R., Ebert, F.J., and Hobbs, W.E., Radioisotope dilution
technique for determination of vitamin B12 in foods, J. Assoc. Off. Anal. Chem.,
65, 85, 1982.
70. International Organization for Standardization. Fruits, vegetables and derived
products. Determination of ascorbic acid. Part 1: reference method.
International Standard ISO 6557/1, 1986.
71. Ponting, J.D., Extraction of ascorbic acid from plant materials, Ind. Eng. Chem.
Anal. Edu., 15, 389, 1943.
72. Eitenmiller, R.R. and Landen, W.O., Jr., Vitamin Analysis for the Health and Food
Sciences, CRC Press, Boca Raton, FL, 1999, pp. 223.
73. Liau, L.S., Lee, B.L., New, A.L., and Ong, C.N., Determination of plasma
ascorbic acid by high-performance liquid chromatography with ultraviolet
and electrochemical detection, J. Chromatogr., Biomed. Appl., 612, 63, 1993.
74. Kall, M.A. and Andersen, C., Improved method for simultaneous determination of ascorbic acid and dehydroascorbic acid, isoascorbic acid, and
dehydroisoascorbic acid in food and biological samples, J. Chromatogr. B.,
730, 101, 1999.
75. Bogna´r, A. and Daood, H.G., Simple in-line postcolumn oxidation and derivatization for the simultaneous analysis of ascorbic and dehydroascorbic acids in
foods, J. Chromatogr. Sci., 38, 162, 2000.
76. Brause, A.R., Woollard, D.C., and Indyk, H.E., Determination of total vitamin
C in fruit juices and related products by liquid chromatography: interlaboratory study, J. AOAC Int., 86, 367, 2003.

© 2006 by Taylor & Francis Group, LLC