Bioaccessibility of Folate in Faba Bean, Oat, Rye and Wheat Matrices

Abstract

Cereals and legumes are rich in folate. However, due to the instability of folate, processing and digestion can induce significant folate loss. In this paper, folate bioaccessibility of faba bean, oat, rye and wheat flours and pastes was studied using a static in vitro digestion model. Folate bioaccessibility depended on food matrices, varying from 42% to 67% in flours and from 40% to 123% in pastes. Digestion was associated with the interconversion of formyl folates, as well as the increase of oxidised vitamers and decrease of reduced vitamers. Especially in faba bean, 5-methyltetrahydrofolate showed surprisingly good stability both in digestion and heat treatment, resulting in high bioaccessibility. The physiological concentration of ascorbic acid did not stabilise folate in digestion; however, a higher level helped to maintain reduced vitamers. Heat treatment (10-min paste making) could improve folate bioaccessibility by liberating folate from the food matrices and by altering folate vitamer distribution.

Full text

Food Chemistry 350 (2021) 129259
Available online 10 February 2021
0308-8146/© 2021 Elsevier Ltd. All rights reserved.
Bioaccessibility of folate in faba bean, oat, rye and wheat matrices
Fengyuan Liu
*
, Susanna Kariluoto , Minnamari Edelmann , Vieno Piironen
Department of Food and Nutrition, University of Helsinki, Agnes Sj¨
obergin katu 2, FI-00014 Helsinki, Finland
ARTICLE INFO
Keywords:
Folate
In vitro bioaccessibility
Stability
Faba bean
Cereals
Flour
Heat treatment
Paste
ABSTRACT
Cereals and legumes are rich in folate. However, due to the instability of folate, processing and digestion can
induce signicant folate loss. In this paper, folate bioaccessibility of faba bean, oat, rye and wheat ours and
pastes was studied using a static in vitro digestion model. Folate bioaccessibility depended on food matrices,
varying from 42% to 67% in ours and from 40% to 123% in pastes. Digestion was associated with the inter-
conversion of formyl folates, as well as the increase of oxidised vitamers and decrease of reduced vitamers.
Especially in faba bean, 5-methyltetrahydrofolate showed surprisingly good stability both in digestion and heat
treatment, resulting in high bioaccessibility. The physiological concentration of ascorbic acid did not stabilise
folate in digestion; however, a higher level helped to maintain reduced vitamers. Heat treatment (10-min paste
making) could improve folate bioaccessibility by liberating folate from the food matrices and by altering folate
vitamer distribution.
1. Introduction
Folate describes a group of water-soluble vitamers that share a
similar structure with pteroyl-L-glutamic acid (folic acid). It acts as a
one-carbon donator, playing important roles in the methylation cycle
and amino acid and nucleotide metabolism. In addition to preventing
megaloblastic anaemia and neural tube defects, folate has also been
associated with the development of cardiovascular diseases (Wiebe
et al., 2018) and neurodevelopmental disorders (Lintas, 2019). Espe-
cially in countries where mandatory fortication is not practiced, such
as EU countries, it is important to study natural folate sources.
Cereals and legumes are rich in folate (Saini et al., 2016). In Finland,
cereal products account for 28% of the total dietary intake of folate for
men and 23% for women (Valsta et al., 2018). At the same time, legumes
have become popular in recent years. It is generally reported that 5-
methyltetrahydrofolate (5-CH
3
-H
4
folate) and 5-formyltetrahydrofolate
(5-HCO-H
4
folate) are the main folate vitamers in cereals and legumes,
and legumes usually contain more folate than cereals (Edelmann et al.,
2013; Jha et al., 2015).
From a nutritional point of view, and due to the instability of folate,
it is essential to study its fate during digestion, as food with high folate
contents may have low bioavailable folate and vice versa (Seyoum &
Selhub, 1998). Folate bioavailability varies considerably among
different foods, as well as among different human studies, ranging from
10% to 98% (Saini et al., 2016). In vivo bioavailability has been being the
gold standard to assess the availability of nutrients and bioactive com-
pounds in food. However, since in vivo studies are expensive and time-
consuming, bioaccessibility studies have been used to predict bioavail-
ability and for sample screening for bioavailability study.
Folate bioaccessibility, usually demonstrated by in vitro digestion
models, is dened as the proportion of folate present in the digesta
before absorption in the small intestine (Etcheverry et al., 2012). Unlike
bioavailability, bioaccessibility does not take the absorption of folate
into account. At present, studies on the effect of food components on
folate bioaccessibility often focus on added folic acid, whereas knowl-
edge of bioaccessibility of endogenous folate in staple foods is also
needed. In addition, only a few studies have investigated folate vitamers
that undergo interconversions during digestion. ¨
Ohrvik et al. (2010)
reported that around 80% of folate in breads was bioaccessible using a
dynamic in vitro digestion model (TIM, TNO Gastro-Intestinal Model).
Similarly, using the TIM system, Mo et al. (2013) reported 82% folate
bioaccessibility for tofu and around 100% for tempe. A recent compre-
hensive folate bioaccessibility study was carried out by Ringling and
Rychlik (2017) using a static in vitro model. They studied the folate
bioaccessibility of three food matrices, where wheat germ had the
lowest bioaccessibility, with around 30%. In addition, they assumed that
tetrahydrofolate (H
4
folate) had low bioavailability, and the bioavail-
ability of 5-CH
3
-H
4
folate varied according to the food matrix. Their
* Corresponding author.
E-mail addresses: fengyuan.liu@helsinki. (F. Liu), susanna.kariluoto@helsinki. (S. Kariluoto), minnamari.edelmann@helsinki. (M. Edelmann), vieno.
piironen@helsinki. (V. Piironen).
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
https://doi.org/10.1016/j.foodchem.2021.129259
Received 30 June 2020; Received in revised form 15 December 2020; Accepted 31 January 2021

Food Chemistry 350 (2021) 129259
2
results supported the idea that stability affects folate bioaccessibility.
Recently, a group of researchers introduced a standardised static in
vitro digestion method (Minekus et al., 2014). Use of this standardised
approach would increase the reliability of the bioaccessibility results
and decrease the variations among the results of different research
groups. Using this method, Bationo et al. (2020) reported folate bio-
accessibilities varying from 23% to 81% in seven cereal-based fermented
foods from West Africa. However, the folate quantication method they
used, as well as the complexity of the samples, made it difcult to
explain the discrepancy of folate bioaccessibility among the samples. In
addition, with this method, Hiolle et al. (2020) revealed that the release
of folic acid in the gastric phase was faster from biscuit and sponge cake
than it was from pudding and custard, indicating that food structures
generated by different processing methods can affect the bioaccessibility
of folate. Food density, texture and microstructure have been considered
to inuence the digestion of several nutrients, such as protein (Zahir
et al., 2020) and starch (Blazek & Gilbert, 2010). However, studies on
folate have been less common.
In this research, folate bioaccessibility of faba bean, oat, wheat and
rye ours, as well as pastes made of them was studied, using the
standardised static in vitro digestion method. Wheat was selected as a
point of comparison, whereas oat and rye are commonly consumed in
many parts of Europe. Interest towards legumes is increasing worldwide.
Faba bean is an important source of plant-based protein and can be
cultivated in boreal climate. The aims of this study were as follows: 1) to
determine folate bioaccessibility in different food matrices and study the
inuence of in vitro digestion on individual folate vitamers; and 2) to
investigate the effects of paste-making processing on folate stability and
bioaccessibility of the studied food matrices. It was hypothesised that
folate bioaccessibility differs among faba bean and cereals, and that heat
treatment could improve folate bioaccessibility.
2. Materials and methods
2.1. Enzymes and calibrants
The enzymes and bile extract used were obtained from Sigma-
Aldrich (St Louis, MO, USA), including:
α
-amylase from Aspergillus ory-
zae (A9857), protease (P8811), pepsin (P7125), bile from bovine and
ovine (B8381), chymotrypsin (C4129) and trypsin (T0303). For the
calibrants used for quantication, (6S)-Tetrahydrofolate (H
4
folate, so-
dium salt), (6S)-5-methyltetrahydrofolate (5-CH
3
-H
4
folate, calcium
salt), (6R,S)-5,10-methenyltetrahydrofolate hydrochloride (5,10-CH
+
-
H
4
folate) and (6S)-5-formyltetrahydrofolate (5-HCO-H
4
folate, sodium
salt) were purchased from Eprova AG (Schaffhausen, Switzerland). 10-
formylfolic acid (10-HCO-PGA) and folic acid (PGA) were purchased
from Schirck’s Laboratories (Jona, Switzerland). 10-formyldihydrofo-
late (10-HCO-H
2
folate) was synthesized from 5,10-CH
+
-H
4
folate ac-
cording to our previous publication (Kariluoto et al., 2004). In addition,
the standards were dissolved, and their concentrations were conrmed
by a spectroscopic method according to Kariluoto et al. (2004).
2.2. Sample preparation
Wholegrain ours (in 1-kg packages) were obtained from local
markets, including rye our (Helsinki Mills Ltd, J¨
arvenp¨
a¨
a, Finland), oat
our (Helsinki Mills Ltd), whole wheat our (Myllyn Paras, Hyvink¨
a¨
a,
Finland) and faba bean our (Vihre¨
a H¨
ark¨
a, Kalanti, Finland). The ours
were stored at −20 ◦C until further use. Two different paste samples
(labelled A and B) were prepared from ours, and the ratio of our to
water was 1:7 (w/v). Paste A samples were prepared as follows: 2.5 g of
our was mixed with 17.5 mL of Milli-Q water in a 50-mL centrifuge
tube. The tubes were ushed with nitrogen gas, closed, and placed in a
boiling water bath exactly for 10 min. During the incubation, tubes were
occasionally shaken using a vortex to avoid forming a clot. After the 10-
minute incubation, the temperature of paste A samples was from 91 to
96 ◦C, measured by an electronic thermometer with a probe (Testo Ltd,
Lenzkirch, Germany). To mimic a domestic cooking method, paste B
samples were prepared in a beaker. In brief, 5 g of our was mixed with
35 mL of Milli-Q water in a 100-mL beaker. Following this, the mixture
was placed on a hot plate. On average after 2 min, the our/water so-
lution began to boil and was kept boiling for 10 min with constant
stirring. The temperature was measured immediately after the boiling
and paste B samples reached the nal temperatures from 98 to 100 ◦C.
Pastes A and B were both analysed for folate bioaccessibility right after
they had cooled down to room temperature. In addition, the moisture
contents of pastes and ours were analysed using the AACC 44-15A
method (AACC, 2000) to report the data on a dry weight basis.
2.3. In vitro digestion
Simulated digestion of the ours and pastes was carried out in trip-
licate under subdued light using the static in vitro model described by
Minekus et al. (2014) with modications. Human salivary
α
-amylase
and porcine pancreatic
α
-amylase were replaced by
α
-amylase from
Aspergillus oryzae.
The activities of individual enzymes and the concentration of bile
salt were determined according to Minekus et al. (2014). Briey, the
approach included three different phases, which were as follows: the
oral phase (simulated salivary uid, SSF), gastric phase (simulated
gastric uid, SGF) and intestinal phase (simulated intestinal uid, SIF).
First, 5 g of sample was mixed with 4 mL of SSF (pH 7) containing
α
-amylase in a 50-mL centrifuge tube. Following this, CaCl
2
was added,
and the tube was lled with Milli-Q water to the volume of 10 mL (W/V
=5/5). The mixture was incubated at 37 ◦C for 2 min. Second, 8 mL of
SGF (pH 3) with pepsin and CaCl
2
was added. The pH of the solution was
adjusted to 3 before the volume was brought to 20 mL by Milli-Q water.
The gastric digesta was incubated at 37 ◦C with constant shaking for 2 h.
Finally, 10 mL of SIF (pH 7) with bile extract and 6 mL of SIF with
α
-amylase, as well as CaCl
2
, chymotrypsin and trypsin were added. After
the pH was adjusted to 7, the volume was brought to exact 40 mL by
Milli-Q water, and the nal mixture was incubated at 37 ◦C under
constant shaking for 2 h. The digesta was obtained after centrifugation
(10 000 rpm, 10 min) and stored at −20 ◦C until folate analysis. A blank
control (where the sample was replaced by 5 mL of Milli-Q water) was
carried out in each batch of in vitro digestion.
As ascorbic acid is secreted in the human stomach and can affect
folate stability (Ringling & Rychlik, 2017), the effect of ascorbic acid in
the gastric phase on the stability of folate and its bioaccessibility were
studied using faba bean, oat and rye ours. In vitro digestion with
ascorbic acid was carried out in the same way as previously described,
but ascorbic acid was included in the gastric phase. Two different con-
centration levels were applied, which were as follows: 0.1 µmol/mL
(pharmacological concentration) and 100 µmol/mL (excessive
concentration).
2.4. Extraction and purication of folate
The extraction was carried out via tri-enzyme treatment with
α
-amylase, hog kidney conjugase and protease under yellow light, as
described previously (Edelmann et al., 2012). In brief, samples (1 g of
our, 2 g of paste) were extracted in triplicate with 15 mL of CHES/
HEPES buffer (pH 7.85) containing 2% sodium ascorbate and 10 mM 2-
mercaptoethanol in a boiling water bath for 10 min. Then, the pH was
adjusted to 4.9, and
α
-amylase (20 mg) and hog kidney conjugase were
added. The extract was subsequently incubated for 3 h at 37 ◦C, after
which time, the pH was adjusted to 7 and protease (4 mg) was added.
The extract was incubated for 1 h at 37 ◦C, and the enzymes were
inactivated in a boiling water bath for 5 min. The supernatant was
collected after centrifugation (12 000 rpm, 10 min) and ltrated
through a 0.45 µm syringe lter. The extraction of digesta was carried
out in duplicate in a similar way except for the exclusion of
α
-amylase
F. Liu et al.

Food Chemistry 350 (2021) 129259
3
and protease. In short, 10 mL of digesta was mixed with 10 mL of
extraction buffer and boiled for 10 min. The pH was then adjusted to 4.9,
and the extract was incubated only with hog kidney conjugase for 3 h at
37 ◦C. The steps of enzyme inactivation and supernatant collection that
followed were as previously described. Duplicate blank controls were
carried out in each batch of extraction. The purication of folate extracts
was carried out by afnity chromatography as described previously
(Edelmann et al., 2012). Afnity agarose gel (Af-Gel 10, Bio-Rad
Laboratories, Richmond, CA, USA) coupled with folate-binding protein
(Scripps Laboratories, San Diego, CA, USA) was used. Folates were
eluted by 0.02 M triuoracetic acid/0.01 M dithiothreitol into a 5 mL
volumetric ask with 10 mg of ascorbic acid and 5 µL of 2-mercaptoetha-
nol. The eluent was ltered through a 0.2-µm syringe lter, ushed with
nitrogen, and stored at −20 ◦C for no more than 7 days.
2.5. Quantication of folate
The determination of folate vitamers was conducted using a
reversed-phase ultra-high performance liquid chromatography (UHPLC)
method developed and validated by our laboratory (Edelmann et al.,
2012). Vitamers were separated on the HSS T3 column (1.8 µm, 2.1 ×
150 mm; Waters, Milford, MA, USA) at 30 ◦C. During the run, samples
were stored in a dark autosampler at 4 ◦C, and the injection volume was
30 µL. The mobile phases were 30 mM potassium phosphate buffer
(Eluent A, pH 2.2) and acetonitrile (Eluent B). Gradient elution (ow
rate: 0.4 mL/min) conditions were as follows: 5% B at 0–2.16 min,
5–6.9% B at 2.16–4.71 min, 6.9–15.4% B at 4.71–7.47 min, 15.4% B at
7.47–7.87 min, and nally, to initial conditions from 15.4% to 5% B at
7.87–8.3 min, with reconditioning of the column to 5% B at 8.3–11 min.
Seven monoglutamate folate vitamers were determined and quanti-
ed using uorescence (FL) and a photodiode array (PDA) detectors as
follows: H
4
folate (tetrahydrofolate) and 5-CH
3
-H
4
folate (5-methylte-
trahydrofolate) were analysed using FL with an excitation wavelength of
290 nm and emission of 356 nm; 10-HCO-PGA (10-formylfolic acid) was
analysed using FL, with excitation of 360 nm and emission of 460 nm;
10-HCO-H
2
folate (10-formyldihydrofolate), PGA (folic acid) and 5-
HCO-H
4
folate (5-formyltetrahydrofolate) were analysed using PDA with
290 nm; and 5,10-CH
+
-H
4
folate (5,10-methenyltetrahydrofolate) was
analysed using PDA with 360 nm. The details of the preparation and
spectrophotometric purity determination of standard calibrants and
their mixture run on UHPLC have been illustrated by Edelmann et al.
(2012). The identication of folate vitamers was achieved by comparing
the retention times and the ultraviolet (UV) spectra of the sample peaks
to those of standard peaks. Quantication was based on calibration
curves with peak areas plotted against concentrations. Especially in rye,
5-HCO-H
4
folate and PGA peaks were often masked by unknown impu-
rities, hindering the accurate quantication of these vitamers.
2.6. Calculation and statistical analysis
The R Studio platform was used to generate bar plots and analyse the
differences among groups. Folate content is expressed as mean ±stan-
dard deviation (µg/100 g, n =3) on dry matter (DM) basis. Total folate
was expressed as the sum of folate vitamers (without conversion to folic
acid equivalent). Two sample t-tests were applied to study the differ-
ences between folate contents before and after the in vitro digestion. In
addition, one-way analysis of variance (ANOVA) and Tukey’s HSD
(honestly signicant difference) post hoc test were selected for multi-
group comparisons relating to folate contents. The formula used to
calculate folate bioaccessibility was:
FB(%) = 100 ×TFD(
μ
g/100gDM)
TF(
μ
g/100gDM)
where FB means folate bioaccessibility; TFD means total folate in
our or paste digesta; TF means total folate in our or paste; DM means
dry matter.
Theoretical folate contents in paste samples were calculated using
the following equation:
TFP(
μ
g/100gDM) = FWF(g) × FF(
μ
g/100gDM) × DF(%)
FP(g) × DP(%)
where TFP means theoretical folate content in paste; FWF means
fresh weight of our; FF means folate contents in our; DF means dry
matter of our; FP means fresh weight of paste; DP means dry matter of
paste; DM means dry matter.
3. Results
3.1. Folate content and bioaccessibility in ours
Faba bean our had the highest total folate content, with 142.0 ±
5.3 µg/100 g DM (Table 1), followed by rye (49.5 ±3.0 µg/100 g DM),
wheat (46.2 ±1.0 µg/100 g DM) and oat ours (41.6 ±2.3 µg/100 g
DM). In our digesta, signicantly (p <0.05) lower total folate contents
were measured, which brought the folate bioaccessibility values of faba
bean, oat, rye and wheat to 63%, 67%, 47% and 42%, respectively
(Table 1).
Fig. 1 demonstrates the differences in folate vitamer contents and
distributions between our and digesta samples. 5-HCO-H
4
folate was
one of the main vitamers for all the our samples, and in vitro digestion
decreased its content in all matrices, although the result was not sta-
tistically signicant in the case of oat. However, it remained one of the
main vitamers in faba bean (27%), oat (27%) and rye (55%) our
digesta. The changes of 5,10-CH
+
-H
4
folate levels were like those of 5-
HCO-H
4
folate, but the contribution of 5,10-CH
+
-H
4
folate to the total
folate contents was smaller in the digesta than it was in our. 5-CH
3
-
H
4
folate had a marked contribution to the total folate contents in oat
(35%), rye (21%) and wheat (19%) ours, but it was not found in the
our digesta. Nevertheless, in faba bean, the contents of 5-CH
3
-H
4
folate
accounted for around 16% of the total folate contents in both our (24.6
Table 1
Total Folate Contents and Bioaccessibilities of Flours and Pastes.
Material Total folate content (µg/100 g DM) Folate
bioaccessibility
(%)
Theoretical Before
digestion
After
digestion
Faba
bean
Flour −142.0 ±
5.3
96.1 ±
12.1 *
63 ±10
Paste
A
142.0 134.0 ±
21.7
165.3 ±
18.0
123 ±13
Paste
B
125.6 193.5 ±
35.6
231.7 ±
10.2
120 ±5
Oat Flour −41.6 ±2.3 28.7 ±
4.0 **
67 ±6
Paste
A
41.6 30.1 ±3.6 23.5 ±
2.0
78 ±7
Paste
B
32.9 37.1 ±9.0 37.5 ±
21.1
101 ±57
Rye Flour −49.5 ±3.0 22.6 ±
4.6 **
47 ±8
Paste
A
49.5 34.6 ±3.2 31.0 ±
5.1
90 ±15
Paste
B
41.3 58.2 ±4.6 31.2 ±
2.6 **
54 ±5
Wheat Flour −46.2 ±1.0 20.7 ±
2.3 **
42 ±4
Paste
A
46.2 37.1 ±3.9 34.5 ±
6.0
93 ±16
Paste
B
40.9 27.9 ±8.7 11.1 ±
3.2 *
40 ±12
Note: Values are expressed as mean ±standard deviation. For total folate con-
tents before digestion, the standard deviations represent variation among three
analytical replicates; for total folate contents after digestion and folate bio-
accessibility, the standard deviations represent variation among triplicate di-
gestions In addition, statistically signicant differences between the folate
contents before and after digestion are marked with * (p <0.05) or ** (p <0.01).
F. Liu et al.

Food Chemistry 350 (2021) 129259
4
±1.9) and our digesta (14.3 ±2.8 µg/100 g DM), although its content
decreased by 42% (p <0.05) during digestion. By contrast, an increase
of 10-HCO-PGA content was observed in all the our digesta samples,
especially in wheat (by 225%), increasing its contribution in all these
samples. At the same time, a signicant (p <0.05) change of 10-HCO-
H
2
folate content was only seen in wheat, with an 85% decrease from
our to digesta. In addition, H
4
folate had a small contribution to the
total folate content but only in our matrices.
3.2. Effect of ascorbic acid on folate in our digesta
Compared with the regular in vitro digestion, the addition of ascorbic
acid at the physiological level (0.1 µmol/mL) had only a limited effect on
total folate contents in all our digesta samples (Table 2). As for the
individual vitamers, signicant (p <0.05) changes were only observed
in oat. 5-CH
3
-H
4
folate was not detected in the regular oat digesta, but it
was present in small amounts (1.0 ±0.1 µg/100 g DM) in the oat digesta
with added ascorbic acid. The content of 10-HCO-PGA was higher in
regular oat digesta than it was in the digesta with added ascorbic acid.
Digesta samples with the addition of 100 µmol/mL ascorbic acid had
higher total folate contents than those with either 0.1 µmol/mL or
without ascorbic acid. Especially for oat and rye, the total folate contents
in digesta were close to those in ours when excessive concentration of
ascorbic acid was applied. Most of the individual vitamers in the original
our samples (before digestion) were detected in the respective digesta
with the addition of 100 µmol/mL of ascorbic acid; the exception was
10-HCO-H
2
folate in oat, which was found in oat our but not in the
digesta. H
4
folate, 5-CH
3
-H
4
folate and 5,10-CH
+
-H
4
folate contents were
notably higher in digesta with 100 µmol/mL ascorbic acid addition than
in digesta without ascorbic acid or with the physiological level of
ascorbic acid addition.
3.3. Folate content and bioaccessibility in pastes
3.3.1. Heat treatment in closed tubes (Paste A)
The highest total folate content in the Paste A samples was found in
faba bean paste, at 134.0 ±21.7 µg/100 g DM, while in wheat, rye and
oat pastes, the total folate contents were around 35 µg/100 g DM
(Table 1). No signicant difference in total folate level was observed
between the paste and paste digesta for any matrices. Faba bean paste
digesta contained 165.3 ±18.0 µg/100 g DM of total folate, resulting in
folate bioaccessibility of 123%. Meanwhile, folate bioaccessibility levels
from the wheat, rye and oat pastes were 93%, 90% and 78%, respec-
tively (Table 1).
5-HCO-H
4
folate was one of the most abundant vitamers both in all
Paste A and paste digesta samples (Fig. 2). In addition, it was stable in
cereal (oat, rye and wheat) pastes, in terms of both its content and
contribution to the total folate. However, in faba bean paste, the content
of 5-HCO-H
4
folate decreased markedly (52%) during digestion, and its
proportion to the total folate fell from 29% in the paste to 14% in the
paste digesta. In contrast, 5-CH
3
-H
4
folate was stable in faba bean sam-
ples, whereas it could not be detected in cereal paste digesta, even
though it had been the main vitamer in oat paste (37% contribution).
Considerably higher contents of 5,10-CH
+
-H
4
folate were measured
in all the paste digesta samples than in pastes, accounting for higher
contribution to total folate levels in paste digesta. 10-HCO-PGA was
stable in cereal samples, while a signicant increase of its content (58%)
was seen in faba bean paste digesta, making it the second dominant
vitamer (23%). Moreover, considerably higher amounts of PGA were
measured in almost all the paste digesta compared with pastes. In rye
samples, the PGA peak was unfortunately masked.
3.3.2. Heat treatment in open beakers (Paste B)
As displayed in Table 1, Paste B from faba bean contained the highest
amount of folate, with 193.5 ±35.6 µg/100 g DM, followed by rye (58.2
±4.6 µg/100 g DM), oat (37.1 ±9.0 µg/100 g DM) and wheat (27.9 ±
8.7 µg/100 g DM) pastes. In faba bean paste digesta, the total folate level
was higher than in the paste, bringing the folate bioaccessibility value to
120%. Folate bioaccessibility in oat paste was 101%, whereas signi-
cantly lower (p <0.05) total folate contents were measured in rye and
wheat paste digesta than in the respective pastes, resulting in folate
bioaccessibility of 54% in rye and 40% in wheat.
Fig. 3 shows the individual vitamer contents and vitamer distribu-
tions in paste B and its digesta for each material. 5-HCO-H
4
folate was the
major (30%–65%) folate vitamer in all the paste samples; however,
signicantly lower contents (p <0.05) were observed in faba bean (by
Fig. 1. a) Folate vitamer contents and b) folate vitamer distributions in the our samples before and after digestion. The error bars of our and our digesta represent
the standard deviation among triplicate analysis and among triplicate digestions, respectively. In addition, * or ** indicates a signicant difference in vitamer
contents before and after digestion at a level of p <0.05 or p <0.01, respectively.
F. Liu et al.

Food Chemistry 350 (2021) 129259
5
31%) and rye (by 58%) paste digesta. Meanwhile, digestion signicantly
decreased the levels of 5-CH
3
-H
4
folate in cereal pastes, while in faba
bean, this vitamer was stable, contributing to 16% and 12% of the total
folate in paste and paste digesta. In addition, only in faba bean digesta,
signicantly higher amounts of 10-HCO-PGA (2-fold increase) and 10-
HCO-H
2
folate (increased by 48%) were detected compared with paste.
In contrast, 10-HCO-H
2
folate was one of the main vitamers (32%) in
wheat paste, but it was not detected in wheat paste digesta.
3.4. Folate stability in heat treatments
According to Table 1, the effects of the heat treatments on total folate
contents differed in the studied materials. Folate contents in all Paste A
samples were somewhat lower than the folate contents in the respective
original ours (before digestion), and a decrease by 30% was shown for
rye paste. In Paste B samples, analysed total folate contents were
generally in line with folate contents in ours for faba bean, oat and rye,
whereas in Paste B from wheat, the folate content was considerably
lower than that in wheat our. Nevertheless, for all Paste B samples,
except wheat paste, the analysed folate contents were higher than the
theoretical total folate contents, where folate and moisture contents in
our and the loss of water during heat treatment were considered.
Finally, paste B usually had higher levels of analysed total folate than
paste A, with an exception for wheat, where lower folate level in paste B
than in paste A was observed.
Fig. 4 provides a direct view of the effect of heat treatments on in-
dividual vitamers. Lower 5-CH
3
-H
4
folate contents were observed in oat,
rye and wheat after heat treatments, while in faba bean samples, the
contents remained similar. In addition, lower levels of 5,10-CH
+
-
H
4
folate were found in all the matrices after the heat treatments. In
contrast, higher levels of 5-HCO-H
4
folate were detected in Paste B from
faba bean and rye than in our or Paste A. In addition, the content of 10-
HCO-H
2
folate in Paste B from faba bean increased substantially, by
almost 200% compared with faba bean our. As for 10-HCO-PGA, its
content decreased in Pastes A and B from oat compared with that of oat
our, whereas Paste A from wheat had a higher content of 10-HCO-PGA
compared with either Paste B or wheat our. Finally, the contents of
PGA in Paste B from faba bean and oat were higher than in our or Paste
A, while H
4
folate was only present in our samples.
4. Discussion
4.1. Folate bioaccessibility in ours and pastes
Folate bioaccessibility varied from 40% to 120% in all the studied
food matrices and from 42% to 67% in ours. Little is known about the
folate bioaccessibility of raw materials; however, Ringling and Rychlik
(2017) reported around 30% folate bioaccessibility in wheat germ.
Recent data on folate bioaccessibility or bioavailability in legumes is
also scarce. Gregory (1989) summarised a generally high folate
bioavailability in beans; however, the precision of the assays was low (0
– 181%). Mo et al. (2013) reported folate bioaccessibility of 81% and
around 100% in soybean-based tofu and tempe, respectively.
Paste samples generally had better folate bioaccessibility than our
samples did. One plausible explanation for this is that folate was more
easily liberated from the matrices after boiling, especially from faba
bean. In contrast, the secondary structure formed during boiling could
also have stabilised the folate. The main reaction during paste making is
starch gelatinisation. A gelatinised structure could perhaps protect
folate during acidic gastric digestion and release it in the intestinal
phase, leading to higher folate bioaccessibility for paste samples.
Bationo et al. (2020) studied the folate bioaccessibility of four different
African foods made from pearl millet with various structures. They re-
ported bioaccessibility values ranging from 24% to 81%, with the
highest in batter fritters, suggesting that a dense structure could protect
folate from degradation during digestion. In a recent study, different
foods were produced with various structures but from the same in-
gredients (Hiolle et al., 2020). For some reason, the release of added
folic acid in the gastric phase was faster from biscuit and sponge cake
than it was from pudding and custard, although there were no differ-
ences in folic acid release among the studied foods at the end of diges-
tion. Folic acid is the most stable folate vitamer, whereas for labile
endogenous folate, a rapid release from food structure in the gastric
phase could be detrimental.
Among cereals, oat exhibited better folate bioaccessibility than rye
and wheat in our and Paste B, but it was somewhat lower in Paste A.
The composition of oat our is quite different from that of rye or wheat
our, especially due to β-glucan. β-Glucan forms viscous solutions in
Table 2
Folate Contents in Flour and Flour Digesta with and without Ascorbic Acid
Addition.
Material Component Folate content (µg/100 g DM)
Digesta
without
ascorbic
acid
(regular)
Digesta
with 0.1
µmol/mL
ascorbic
acid
Digesta
with 100
µmol/mL
ascorbic
acid
Before
digestion
Faba
bean
PGA 5.5 ±1.2c 5.1 ±0.4c 2.7 ±1.2b 1.9 ±1.2
a
10-HCO-H
2
14.6 ±2.6c 12.9 ±2.4c 18.1 ±4.0b 14.5 ±
4.4 a
10-HCO-
PGA
29.3 ±
3.0b
28.2 ±5.2b 22.2 ±5.9b 17.5 ±
1.3 a
H
4
− − 1.3 ±0.5b 4.9 ±1.0
a
5-CH
3
-H
4
13.6 ±2.7c 16.3 ±2.3
bc
28.4 ±7.3
a
23.0 ±
2.6 ab
5-HCO-H
4
22.0 ±2.6
bc
16.5 ±1.9c 25.0 ±4.5b 36.4 ±
2.1 a
5,10-CH
+
-
H
4
5.9 ±3.3c 5.6 ±0.8c 14.4 ±3.0b 28.2 ±
2.3 a
Total 86.1 ±
13.5 bc
84.4 ±7.0c 112.1 ±
24.4b
126.3 ±
8.3 a
Oat PGA 2.6 ±0.9 a 4.1 ±0.5 a 4.0 ±0.9 a 2.6 ±0.7
a
10-HCO-H
2
− − − 4.1 ±1.3
a
10-HCO-
PGA
10.5 ±2.3
a
7.2 ±2.7
bc
9.7 ±0.8
abc
7.8 ±0.3
abc
H
4
− − 1.0 ±0.1 a 0.8 ±0.4
a
5-CH
3
-H
4
−1.0 ±0.1c 15.0 ±1.1b 12.6 ±
0.5 a
5-HCO-H
4
9.8 ±2.3 a 8.7 ±1.5 a 9.8 ±1.0 a 9.9 ±0.2
a
5,10-CH
+
-
H
4
1.4 ±0.5b 2.4 ±0.5b 3.7 ±0.8 a 2.9 ±1.1
ab
Total 23.6 ±
3.3b
24.1 ±1.9b 42.8 ±2.8
a
40.8 ±
2.4 a
Rye PGA − − − −
10-HCO-H
2
− − 2.2 ±0.8 a 1.9 ±1.2
a
10-HCO-
PGA
9.1 ±1.7b 8.3 ±1.9
ab
6.5 ±0.8 a 6.2 ±0.6
ab
H
4
− − 1.0 ±0.1b 1.9 ±0.6
a
5-CH
3
-H
4
− − 9.8 ±0.7 a 8.7 ±1.5
a
5-HCO-H
4
14.2 ±3.3
a
16.5 ±3.4
a
18.7 ±2.3
a
18.2 ±
1.3 a
5,10-CH
+
-
H
4
0.2 ±0.4c 1.3 ±0.5c 6.5 ±1.2b 8.2 ±1.0
a
Total 23.4 ±
5.0b
26.4 ±4.0b 44.1 ±2.1
a
44.9 ±
4.1 a
Note: Results are expressed as mean ±standard deviation. For folate contents
before digestion, the standard deviations represent variation among three
analytical replicates; for folate contents of digesta, the standard deviations
represent variation among triplicate digestions. −not detected. Values within
the same row with different letters differ signicantly (p <0.05).
F. Liu et al.

Food Chemistry 350 (2021) 129259
6
digestion and has been reported to be stable under gastric conditions of
low pH (Kumar et al., 2013). β-Glucan may be able to interact with folate
in the viscous structure, protecting it. However, the structure should be
loose enough for folate to be liberated in the intestinal phase. In addition
to changes in carbohydrates, heat treatment affects protein conforma-
tion, potentially releasing folate.
Faba bean exhibited high folate bioaccessibility, with values
exceeding 100% in pastes. Theoretically, bioaccessibility should always
be below or equal to 100%. In vitro digestion can be considered a folate
extraction process with higher extraction volume, more enzymes and
longer extraction time compared with the normal tri-enzyme extraction
used in folate determination. Hence, the high folate bioaccessibility of
faba bean pastes means that folate was efciently liberated during the in
vitro digestion. However, this indicates that the folate content in faba
bean our was underestimated, and consequently, the true bio-
accessibility may be lower. Hefni et al. (2015) reported the total folate
levels in faba bean our ranging from 92 to 140 µg/100 g DM, which is
in line with the results from this study (126–142 µg/100 g DM). The
Fig. 2. a) Folate vitamer contents and b) folate vitamer distributions in Paste A samples before and after digestion. The error bars of Paste A and Paste A digesta
represent the standard deviation among triplicate analysis and among triplicate digestions, respectively. In addition, * or ** indicates a signicant difference in
vitamer contents before and after digestion at a level of p <0.05 or p <0.01, respectively.
Fig. 3. a) Folate vitamer contents and b) folate vitamer distributions in Paste B samples before and after digestion. The error bars of Paste B and Paste B digesta
represent the standard deviation among triplicate analysis and among triplicate digestions, respectively. In addition, * or ** indicates a signicant difference in
vitamer contents before and after digestion at a level of p <0.05 or p <0.01, respectively.
F. Liu et al.

Food Chemistry 350 (2021) 129259
7
insufcient extraction of folate in faba bean our seems to be a common
problem, and a careful assessment should be made regarding the dietary
value of folate in faba bean. In addition, an alternative, or more effective
method for folate extraction in legumes should be developed. Never-
theless, differences between folate bioaccessibility in faba bean and
cereals could be attributed to different seed structures. Folate in legume
seeds is mostly located in cotyledons, which represents most of the total
seed mass (Cofgniez et al., 2019), whereas in cereals, folate is
concentrated in the bran and germ (Edelmann et al., 2013). Folate in
legumes may be more easily released from the starch and protein-rich
cotyledons than from the bran or germ structure during the in vitro
digestion. Faba bean is protein-rich, and high protein concentration
could result in a high buffer capacity (Mennah-Govela et al., 2019),
creating a relatively stable pH environment that could stabilise folate.
The high buffer capacity of faba bean samples was also noticed during
the experiments, as more NaOH or HCl was needed during the pH
adjustment for faba bean than for cereals.
Information on the individual vitamers suggested that folate in faba
bean was more stable than it was in cereals. For example, 5-CH
3
-
H
4
folate was stable—or alternatively, better liberated—in faba bean
samples during the in vitro digestion, especially in pastes. In all the cereal
digesta samples, the content of this vitamer decreased signicantly. The
relatively good stability of folate in faba bean could be explained by the
high antioxidant capacity, possibly mainly from phenolic compounds
(Lafarga et al., 2019), which could protect folate from oxidation during
in vitro digestion. However, similar or even higher capacities have also
been reported for cereals (Luo et al., 2015).
4.2. Folate vitamer interconversion and degradation induced by in vitro
digestion
Two major trends in the changes of folate vitamers due to in vitro
digestion could be identied for all matrices; these were in-
terconversions of formyl folates and decrease of reduced vitamers
(mainly H
4
folate and 5-CH
3
-H
4
folate). Generally, the interconversion
among formyl folates was characterised by the decrease of reduced and
intermediate vitamers (5–HCO–H
4
folate and 5,10-CH
+
-H
4
folate) and
the increase of oxidised vitamers (10–HCO–H
2
folate and 10-HCO-PGA).
One exception was 5,10-CH
+
-H
4
folate in Paste A, where an increased
level of this vitamer was observed in paste digesta. Formation of oxi-
dised vitamers during heat treatment could partly explain the generally
higher folate bioaccessibility in pastes compared with ours, as these
vitamers are more stable, and thus, likely to survive digestion. It is
noteworthy that the standard deviations of these vitamers were high.
One explanation for this is that variation tends to increase with smaller
concentrations. Nevertheless, the uctuation could also indicate that
these vitamers were unstable during the digestion and analysis, and
minor changes may have caused interconversion among them. The
stability of formyl folates has been thoroughly reviewed by other
scholars (J¨
agerstad & Jastrebova, 2013), who summarised that the re-
actions are mainly pH rather than temperature driven. 5-HCO-H
4
folate
is stable in a neutral environment and will convert to 5,10-CH
+
-H
4
folate
under an acidic pH. Therefore, these changes could have been caused by
pH changes during the digestion. In addition, one study indicated that
iron compounds can catalyse the oxidation of 10-HCO-H
4
folate to 10-
HCO-H
2
folate (Baggott et al., 1998), and 10–HCO–H
4
folate could be
obtained from conversion of 5,10-CH
+
-H
4
folate under neutral pH
(J¨
agerstad & Jastrebova, 2013). In our analytical system, 10-HCO-
H
4
folate could not be determined as such, but it was converted mainly to
10-HCO-PGA. Since the ours we studied are whole grain ours with
relatively high mineral contents, iron could have promoted in-
terconversions among formyl folates during in vitro digestion. In addi-
tion, it is worth mentioning that digestion also resulted in an increase in
another oxidised folate, PGA. However, in Paste B, where the intensive
heat treatment was applied, this vitamer may have suffered from further
degradation, and its content was smaller after digestion.
Among the reduced folate vitamers, H
4
folate was almost completely
lost during digestion due to its inherent instability. This nding agrees
well with the literature showing that H
4
folate is unstable even at 37 ◦C,
especially under acidic conditions (De Brouwer et al., 2007). Similarly,
5-CH
3
-H
4
folate degraded during digestion in all cereal samples. Even
under mild conditions, 5-CH
3
-H
4
folate is readily oxidised to 5-CH
3
-5,6-
H
2
folate, which can be reduced back to 5-CH
3
-H
4
folate. However, in
acidic media, 5-CH
3
-5,6-H
2
folate may degrade further (Lucock et al.,
1995). Since ascorbic acid/ascorbate and other antioxidants are
commonly used in folate analysis, it is difcult to estimate the propor-
tion of 5–CH
3
–5,6–H
2
folate naturally present in foods. Surprisingly, 5-
CH
3
-H
4
folate showed great stability (or better liberation) in the diges-
tion of faba bean pastes, and in our digesta, a considerable concen-
tration of this vitamer was still left. The reason for this unexpectedly
good stability remained unclear and warrants further study; however, it
may somehow be related to the antioxidant capacity of faba bean.
Especially in rye sample chromatograms, extra peaks often interfered
with 5-HCO-H
4
folate and PGA. Under more severe conditions, the
oxidative degradation of 5-CH
3
-H
4
folate produces p-amino-
benzoylglutamate (pABG), as well as 4
α
-hydroxy-5-
Fig. 4. Folate vitamer contents in a) faba bean, b) oat, c) rye and d) wheat our, Paste A and Paste B samples. The error bars represent variation among three
analytical replicates.
F. Liu et al.

Food Chemistry 350 (2021) 129259
8
methyltetrahydrofolate (MeFox). MeFox has been reported to be abun-
dant in cereal grains (Shahid et al., 2020) and to disturb the quanti-
cation of 5-HCO-H
4
folate, as well as PGA, because of similar retention
(Fazili & Pfeiffer, 2013). However, the coupling of stable isotope dilu-
tion assay and mass spectrum technology has been used to address these
issues by other groups (Ringling & Rychlik, 2017; Shahid et al., 2020).
Since ascorbic acid is secreted into the stomach (Sobala et al., 1991),
we explored the effect of inclusion of ascorbic acid on the folate bio-
accessibility of faba bean, oat and rye ours. When the physiological
amount of ascorbic acid was added, no signicant changes in folate
content were oberserved inthe studied samples, and thus, bio-
accessibility. However, when an excessive amount (1000 times the
physiological amount) of ascorbic acid was added, more folate vitamers,
especially 5-CH
3
-H
4
folate, were retained. This could indicate that the 5-
CH
3
-5,6-H
2
folate naturally present in the studied ours was efciently
reduced to 5-CH
3
-H
4
folate, as well as that 5-CH
3
-H
4
folate was stabilised
by ascorbic acid. Ringling and Rychlik (2017) included ascorbic acid in
their in vitro model and found that the inuence of ascorbic acid
depended on the food matrix. They observed a 94% loss of
5–CH
3
–H
4
folate in wheat germ even with added ascorbic acid and only a
small difference in folate bioaccessibility. This observation is consistent
with the results of cereal samples from our study. Another interesting
phenomenon was the absence of 10-HCO-H
2
folate in the oat our
digesta with 100 µmol/mL ascorbic acid as this vitamer was present in
other matrices. 10-HCO-H
2
folate could have been converted to 5,10-CH
+-H
4
folate under acidic environment (J¨
agerstad & Jastrebova, 2013),
resulting in an increase of 5,10-CH +-H
4
folate. In addition, 10-HCO-
H
2
folate can be oxidised to 10-HCO-PGA. The relatively good stability of
10-HCO-H
2
folate in faba bean and rye samples could be due to their rich
formyl folate pool (Table 2) or the endogenous antioxidants. As the
addition of ascorbic acid in a physiological concentration had little ef-
fect on the results, we decided to exclude ascorbic acid in our model.
Nevertheless, the results stress the importance of folate stability in the
context of folate bioaccessibility, and foods with high antioxidant ca-
pacity could result in high folate bioaccessibility.
4.3. Changes in folate content and vitamer distribution induced by heat
treatments
The evaluation of the effect of heat treatment on folate content and
the comparison between the treatments was complicated by the greater
than 100% calculated retention in most of the Paste B samples. As ex-
pected, the analysed total folate contents in Paste A samples were lower
compared with the theoretical values, conrming that the processing led
to folate losses (or perhaps that folate could not be liberated from the
studied our matrices). However, in Paste B samples, where a more
severe heat treatment was applied, the analysed values were higher than
the corresponding theoretical values except for wheat. Especially in faba
bean and rye, folate was more easily liberated from Paste B than it was
from the corresponding our, presumably due to the structural changes
discussed above. In addition, the heat treatment might improve the
liberation of antioxidants as well, which could protect folate in these
matrices (Ng & Tan, 2017). In contrast, processing of Paste B from wheat
may have been excessive, to the point where the liberation of folate
could no longer compensate for the loss. This was further conrmed by
the bioaccessibility data for Paste B from wheat, where an even lower
folate level was observed for the digesta.
The effect of heat treatments on folate in different foods has been
studied by several research groups. A 50-min cooking time did not
signicantly affect the folate content in pearl millet paste (Bationo et al.,
2019). Delchier et al. (2012) reported that folate lost in green beans
during boiling could be found in processing media and suggested that
leaching was also responsible for the folate loss during these treatments.
In addition, an article reported no loss of folate in faba bean after in-
dustrial blanching (Hefni & Witth¨
oft, 2014). In rye and wheat baking,
approximately 25% folate losses have been reported (Kariluoto et al.,
2004).
The information about individual folate vitamers could help us un-
derstand the effect of heat treatments in more detail. In most samples,
the contents of oxidised vitamers (PGA, 10–HCO–H
2
folate and 10-HCO-
PGA) increased, while the contents of reduced vitamers (H
4
folate, 5-
CH
3
-H
4
folate and 5,10-CH
+
-H
4
folate) decreased. It is interesting to note
that 5–CH
3
-H
4
folate in faba bean seemed to be highly stable compared
with that in the cereal samples during the thermal processing. 5,10-CH
+
-
H
4
folate, which is the intermediate form of 5–HCO–H
4
folate, was re-
ported to convert back to 5-HCO-H
4
folate under heating at neutral pH
(De Brouwer et al., 2007). This could explain the increase of 5-HCO-
H
4
folate and decrease of 5,10-CH
+
-H
4
folate observed in paste samples.
The increase of 5-HCO-H
4
folate contents in Paste B from faba bean and
rye could also be due to the liberation from the food matrices during the
paste making.
Motta et al. (2017) reported that boiled or steamed quinoa contained
more 5-CH
3
-H
4
folate than raw quinoa did, while no losses were found in
buckwheat; in amaranth, the folate content decreased. Similarly,
seemingly contradictory results where heat treatments have led to no
folate loss or even an increase in folate content have been reported for
broccoli (Stea et al., 2007), green beans (Delchier et al., 2012) and
lentils (Zhang et al., 2019). This again indicates that food matrices in-
uence folate liberation and stability.
5. Conclusion
This study provided new information about folate bioaccessibility in
legume and cereal matrices. Faba bean is a promising source of dietary
folate due to its high folate content and better folate bioaccessibility
than the levels observed in cereals. In addition,
5–methyltetrahydrofolate in faba bean showed exceptionally good sta-
bility in both digestion and heat treatment, which warrants further
study. However, other characteristics, such as sensory quality, might
hinder the popularity of faba bean among consumers. The physiological
concentration of ascorbic acid was not able to stabilise folate during
digestion; however, the better retention of reduced vitamers at a higher
level of ascorbic acid addition emphasises the great importance of folate
stability on bioaccessibility. Our results indicated that the structure
formed by the paste-making process can protect folate from oxidation
during in vitro digestion, enhance folate extractability, and thus, result in
better folate bioaccessibility from paste than from the respective our.
The prediction of folate bioaccessibility is complex. In addition to
inherent vitamer distribution in the raw material, processing may
improve bioaccessibility by changing folate vitamer distribution to-
wards more stable vitamers, destroying folate-binding structures or
forming secondary structures protecting folate. Therefore, further
studies about the effect of food macrocomponent structure on folate
bioaccessibility are needed. The role of antioxidants should also be
elucidated. Finally, from a nutritional point of view, in addition to the
determination of folate contents, bioaccessibility and bioavailability of
common foods should gain more attention in the future.
CRediT authorship contribution statement
Fengyuan Liu: Conceptualization, Investigation, Data curation,
Visualization, Writing - original draft, Writing - review & editing. Sus-
anna Kariluoto: Conceptualization, Writing - review & editing, Su-
pervision. Minnamari Edelmann: Writing - review & editing,
Supervision. Vieno Piironen: Writing - review & editing, Supervision.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
F. Liu et al.

Food Chemistry 350 (2021) 129259
9
Acknowledgements
The authors would like to thank Miikka Olin for his kind help in
UHPLC analysis, China Scholarship Council (CSC) for the nancial
support for the PhD project of Fengyuan Liu, and the Finnish Food
Research Foundation for research grant.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.foodchem.2021.129259.
References
AACC. (2000). AACC Method 44–15A Moisture –air-oven methods. Approved Methods of
the American Association of Cereal Chemists ((10th ed.).). Minnesota, USA: American
Association of Cereal Chemists Inc, St. Paul.
Baggott, J. E., Robinson, C. B., Eto, I., Johanning, G. L., & Cornwell, P. E. (1998). Iron
compounds catalyze the oxidation of 10-formyl-5, 6, 7, 8 tetrahydrofolic acid to 10-
formyl-7, 8 dihydrofolic acid. Journal of Inorganic Biochemistry, 71(3-4), 181-187.
https://doi.org/10.1016/S0162-0134(98)10052-1.
Bationo, F., Humblot, C., Songr´
e-Ouattara, L. T., Hama-Ba, F., Le Merrer, M.,
Chapron, M., Kariluoto, S., & Hemery, Y. M. (2020). Total folate in West African
cereal-based fermented foods: Bioaccessibility and inuence of processing. Journal of
Food Composition and Analysis, 85, 103309. https://doi.org/10.1016/j.
jfca.2019.103309.
Bationo, F., Songr´
e-Ouattara, L. T., Hemery, Y. M., Hama-Ba, F., Parkouda, C.,
Chapron, M., Le Merrer, M., Leconte, N., Sawadogo-Lingani, H., Diawara, B., &
Humblot, C. (2019). Improved processing for the production of cereal-based
fermented porridge enriched in folate using selected lactic acid bacteria and a back
slopping process. LWT, 106, 172–178. https://doi.org/10.1016/j.lwt.2019.02.048.
Blazek, J., & Gilbert, E. P. (2010). Effect of Enzymatic Hydrolysis on Native Starch
Granule Structure. Biomacromolecules, 11(12), 3275–3289. https://doi.org/10.1021/
bm101124t.
Cofgniez, F., Rychlik, M., Sanier, C., Mestres, C., Striegel, L., Bohuon, P., & Briffaz, A.
(2019). Localization and modeling of reaction and diffusion to explain folate
behavior during soaking of cowpea. Journal of Food Engineering, 253, 49–58. https://
doi.org/10.1016/j.jfoodeng.2019.02.012.
De Brouwer, V., Zhang, G.-F., Storozhenko, S., Van Der Straeten, D., & Lambert, W. E.
(2007). pH stability of individual folates during critical sample preparation steps in
prevision of the analysis of plant folates. Phytochem. Anal., 18(6), 496–508. https://
doi.org/10.1002/pca.1006.
Delchier, N., Reich, M., & Renard, C. M. G. C. (2012). Impact of cooking methods on
folates, ascorbic acid and lutein in green beans (Phaseolus vulgaris) and spinach
(Spinacea oleracea). LWT - Food Science and Technology, 49(2), 197–201. https://doi.
org/10.1016/j.lwt.2012.06.017.
Edelmann, M., Kariluoto, S., Nystr¨
om, L., & Piironen, V. (2012). Folate in oats and its
milling fractions. Food Chemistry, 135(3), 1938-1947. https://doi.org/ 10.1016/j.
foodchem.2012.06.064.
Edelmann, M., Kariluoto, S., Nystr¨
om, L., & Piironen, V. (2013). Folate in barley grain
and fractions. Journal of Cereal Science, 58(1), 37–44. https://doi.org/10.1016/j.
jcs.2013.04.005.
Etcheverry, P., Grusak, M. A., & Fleige, L. E. (2012). Application of in vitro
bioaccessibility and bioavailability methods for calcium, carotenoids, folate, iron,
magnesium polyphenols, zinc, and vitamins B-6, B-12, D, and E. Frontiers in
Physiology, 3, 1–22. https://doi.org/10.3389/fphys.2012.00317.
Fazili, Z., & Pfeiffer, C. M. (2013). Accounting for an isobaric interference allows correct
determination of folate vitamers in serum by isotope dilution-liquid
chromatography-tandem mass spectrometry. Journal of Nutrition, 143(1), 108–113.
https://doi.org/10.3945/jn.112.166769.
Gregory, J. F., III (1989). Chemical and nutritional aspects of folate research: Analytical
procedures, methods of folate synthesis, stability, and bioavailability of dietary
folates. In J. E. Kinsella (Ed.), Advances in Food and Nutrition Research (Vol. 33, pp.
1–101). Elsevier.
Hefni, M. E., Shalaby, M. T., & Witth¨
oft, C. M. (2015). Folate content in faba beans (Vicia
faba L.)—effects of cultivar, maturity stage, industrial processing, and bioprocessing.
Food Science and Nutrition, 3(1), 65–73. https://doi.org/10.1002/fsn3.192.
Hefni, M., & Witth¨
oft, C. M. (2014). Folate content in processed legume foods commonly
consumed in Egypt. LWT - Food Science and Technology, 57(1), 337–343. https://doi.
org/10.1016/j.lwt.2013.12.026.
Hiolle, M., Lechevalier, V., Floury, J., Boulier-Month´
ean, N., Prioul, C., Dupont, D., &
Nau, F. (2020). In vitro digestion of complex foods: How microstructure inuences
food disintegration and micronutrient bioaccessibility. Food Research International,
128, 108817. https://doi.org/10.1016/j.foodres.2019.108817.
J¨
agerstad, M., & Jastrebova, J. (2013). Occurrence, Stability, and Determination of
Formyl Folates in Foods. J. Agric. Food Chem., 61(41), 9758–9768. https://doi.org/
10.1021/jf4028427.
Jha, A. B., Ashokkumar, K., Diapari, M., Ambrose, S. J., Zhang, H., Tar’an, B., Bett, K. E.,
Vandenberg, A., Warkentin, T. D., & Purves, R. W. (2015). Genetic diversity of folate
proles in seeds of common bean, lentil, chickpea and pea. Journal of Food
Composition and Analysis, 42, 134–140. https://doi.org/10.1016/j.jfca.2015.03.006.
Kariluoto, S., Vahteristo, L., Salovaara, H., Katina, K., Liukkonen, K.-H., & Piironen, V.
(2004). Effect of Baking Method and Fermentation on Folate Content of Rye and
Wheat Breads. Cereal Chemistry Journal, 81(1), 134–139. https://doi.org/10.1094/
CCHEM.2004.81.1.134.
Kumar, H., Wen, J. G. Y., Shaw, J., Cornish, J., & Bunt, C. (2013). Physiochemical
characterization of beta-glucan and in vitro release of lactoferrin from beta-glucan
microparticles. Current Drug Delivery, 10(6), 713–721. https://doi.org/10.2174/
15672018113109990043.
Lafarga, T., Villar´
o, S., Bobo, G., Sim´
o, J., & Aguil´
o-Aguayo, I. (2019). Bioaccessibility
and antioxidant activity of phenolic compounds in cooked pulses. Int J Food Sci
Technol, 54(5), 1816–1823. https://doi.org/10.1111/ijfs.14082.
Lintas, C. (2019). Linking genetics to epigenetics: The role of folate and folate-related
pathways in neurodevelopmental disorders. Clin Genet, 95(2), 241–252. https://doi.
org/10.1111/cge.13421.
Lucock, M. D., Priestnall, M., Daskalakis, I., Schorah, C. J., Wild, J., & Levene, M. I.
(1995). Nonenzymatic Degradation and Salvage of Dietary Folate: Physicochemical
Factors Likely to Inuence Bioavailability. Biochemical and Molecular Medicine, 55
(1), 43–53. https://doi.org/10.1006/bmme.1995.1030.
Luo, Y. W., Wang, Q., Li, J., Jin, X. X., & Hao, Z. P. (2015). The relationship between
antioxidant activity and total phenolic content in cereals and legumes. Advance
Journal of Food Science and Technology, 8(3), 173–179.
Mennah-Govela, Y. A., Singh, R. P., & Bornhorst, G. M. (2019). Buffering capacity of
protein-based model food systems in the context of gastric digestion. Food Funct., 10
(9), 6074–6087. https://doi.org/10.1039/C9FO01160A.
Minekus, M., Alminger, M., Alvito, P., Ballance, S., Bohn, T., Bourlieu, C., Carri`
ere, F.,
Boutrou, R., Corredig, M., Dupont, D., Dufour, C., Egger, L., Golding, M.,
Karakaya, S., Kirkhus, B., Le Feunteun, S., Lesmes, U., Macierzanka, A., Mackie, A.,
Marze, S., McClements, D. J., M´
enard, O., Recio, I., Santos, C. N., Singh, R. P.,
Vegarud, G. E., Wickham, M. S. J., Weitschies, W., & Brodkorb, A. (2014).
A standardised static in vitro digestion method suitable for food – an international
consensus. Food Funct., 5(6), 1113–1124. https://doi.org/10.1039/C3FO60702J.
Mo, H., Kariluoto, S., Piironen, V., Zhu, Y., Sanders, M. G., Vincken, J.-P., Wolkers-
Rooijackers, J., & Nout, M. J. R. (2013). Effect of soybean processing on content and
bioaccessibility of folate, vitamin B12 and isoavones in tofu and tempe. Food
Chemistry, 141(3), 2418–2425. https://doi.org/10.1016/j.foodchem.2013.05.017.
Motta, C., Delgado, I., Matos, A. S., Gonzales, G. B., Torres, D., Santos, M., Chandra-
Hioe, M. V., Arcot, J., & Castanheira, I. (2017). Folates in quinoa (Chenopodium
quinoa), amaranth (Amaranthus sp.) and buckwheat (Fagopyrum esculentum):
Inuence of cooking and malting. Journal of Food Composition and Analysis, 64,
181–187. https://doi.org/10.1016/j.jfca.2017.09.003.
Ng, Z. X., & Tan, W. C. (2017). Impact of optimised cooking on the antioxidant activity in
edible mushrooms. J Food Sci Technol, 54(12), 4100–4111. https://doi.org/10.1007/
s13197-017-2885-0.
¨
Ohrvik, V., ¨
Ohrvik, H., Tallkvist, J., & Witth¨
oft, C. (2010). Folates in bread: Retention
during bread-making and in vitro bioaccessibility. Eur J Nutr, 49(6), 365–372.
https://doi.org/10.1007/s00394-010-0094-y.
Ringling, C., & Rychlik, M. (2017). Simulation of food folate digestion and bioavailability
of an oxidation product of 5-methyltetrahydrofolate. Nutrients, 9(9), 19. https://doi.
org/10.3390/nu9090969.
Saini, R. K., Nile, S. H., & Keum, Y.-S. (2016). Folates: Chemistry, analysis, occurrence,
biofortication and bioavailability. Food Research International, 89, 1–13. https://
doi.org/10.1016/j.foodres.2016.07.013.
Seyoum, E., & Selhub, J. (1998). Properties of food folates determined by stability and
susceptibility to intestinal pteroylpolyglutamate hydrolase action. The Journal of
Nutrition, 128(11), 1956 –1960. https://doi.org/doi:10.1093/jn/128.11.1956.
Shahid, M., Lian, T., Wan, X., Jiang, L., Han, L., Zhang, C., & Liang, Q. (2020). Folate
monoglutamate in cereal grains: Evaluation of extraction techniques and
determination by LC-MS/MS. Journal of Food Composition and Analysis, 91, 103510.
https://doi.org/10.1016/j.jfca.2020.103510.
Sobala, G. M., Pignatelli, B., Schorah, C. J., Bartsch, H., Sanderson, M., Dixon, M. F.,
Shires, S., King, R. F. G., & Axon, A. T. R. (1991). Levels of nitrite, nitrate, N -nitroso
compounds, ascorbic acid and total bile acids in gastric juice of patients with and
without precancerous conditions of the stomach. Carcinogenesis, 12(2), 193–198.
https://doi.org/10.1093/carcin/12.2.193.
Stea, T. H., Johansson, M., J¨
agerstad, M., & Frølich, W. (2007). Retention of folates in
cooked, stored and reheated peas, broccoli and potatoes for use in modern large-
scale service systems. Food Chemistry, 101(3), 1095–1107. https://doi.org/10.1016/
j.foodchem.2006.03.009.
Valsta, L., Kaartinen, N., Tapanainen, H., M¨
annist¨
o, S., & S¨
a¨
aksj¨
arvi, K. (Eds.). (2018).
Nutrition in Finland—The National FinDiet 2017 Survey. Report 12/2018, Institute
for Health and Welfare (THL). http://urn./URN:ISBN:978-952-343-238-3.
Wiebe, N., Field, C. J., & Tonelli, M. (2018). A systematic review of the vitamin B12,
folate and homocysteine triad across body mass index: Systematic review of B12
concentrations. Obesity Reviews, 19(11), 1608–1618. https://doi.org/10.1111/
obr.12724.
Zahir, M., Fogliano, V., & Capuano, E. (2020). Effect of soybean processing on cell wall
porosity and protein digestibility. Food Funct., 11(1), 285–296. https://doi.org/
10.1039/C9FO02167A.
Zhang, H., Jha, A. B., De Silva, D., Purves, R. W., Warkentin, T. D., & Vandenberg, A.
(2019). Improved folate monoglutamate extraction and application to folate
quantication from wild lentil seeds by ultra-performance liquid chromatography-
selective reaction monitoring mass spectrometry. Journal of Chromatography B, 1121,
39–47. https://doi.org/10.1016/j.jchromb.2019.05.007.
F. Liu et al.