ArticlePDF Available


Untargeted metabolomics analysis was applied to evaluate the phenolic profile of whole wheat bread with yerba mate (YM) during the bread-making process (flour, dough and bread). The free, bound and total phenolic contents of the samples evaluated by the Folin-Ciocalteu method showed the highest values for the flour, dough and bread samples prepared with 4.5% YM in fine and medium particle sizes (flour 181.48 – 175.26 mg GAE/g; dough 149.62 – 141.40 mg GAE/g; and bread 148.32 – 147.00 mg GAE/g). Globally, 104 phenolic compounds were tentatively identified, belonging to the five subclasses: flavonoids (35), phenolic acids (32), other polyphenols (10), stilbenes (2) and lignan (1). Of these compounds, 24 had the same m/z but showed different fragmentation profiles. A higher number of polyphenols was identified in the bound extracts (77%) than in the free extracts (59%). The addition of 4.5% of YM promoted an improved and more abundant profile of phenolic compounds in the dough and bread. The major compounds found in the samples containing YM were 5-caffeoylquinic acid and caffeic acid. The baking process did not adversely affect the abundance of phenolic compounds. The bread-making process positively affected the phenolic profile due to the release of bound phenolic compounds. At the same time, the addition of YM as a natural ingredient promoted an increase in the polyphenols in the bread.
Food Research International 159 (2022) 111635
Available online 8 July 2022
0963-9969/© 2022 Elsevier Ltd. All rights reserved.
Untargeted metabolomics analysis reveals improved phenolic prole in
whole wheat bread with yerba mate and the effects of the
bread-making process
Gabriela Soster Santetti
, Luciana Ribeiro da Silva Lima
, Barbara Biduski
, Millena Cristina
Barros Santos
, Carolina Thomaz dos Santos DAlmeida
, Luiz Claudio Cameron
Luiz Carlos Gutkoski
, Mariana Sim˜
oes Larraz Ferreira
, Renata Dias de Mello Castanho
Department of Food Science and Technology, Federal University of Santa Catarina, Rodovia Admar Gonzaga 1346, 88034-000, Florian´
opolis, SC, Brazil
Laboratory of Bioactives, Food and Nutrition Graduate Program (PPGAN), Federal University of State of Rio de Janeiro, UNIRIO, Av. Pasteur, 296, Urca, 22290-240
Rio de Janeiro, Brazil
Food Science and Technology Graduate Program, University of Passo Fundo, Rio Grande do Sul, Passo Fundo, RS 99052-900, Brazil
Laboratory of Protein Biochemistry, Center for Innovation in Mass Spectrometry, UNIRIO, Av. Pasteur, 296, Urca, 22290-240 Rio de Janeiro, Brazil
Mass spectrometry
Bioactive compounds
Bakery products
Bread-making process
Multivariate analysis
Untargeted metabolomics analysis was applied to evaluate the phenolic prole of whole wheat bread with yerba
mate (YM) during the bread-making process (our, dough and bread). The free, bound and total phenolic
contents of the samples evaluated by the Folin-Ciocalteu method showed the highest values for the our, dough
and bread samples prepared with 4.5% YM in ne and medium particle sizes (our 181.48 175.26 mg GAE/g;
dough 149.62 141.40 mg GAE/g; and bread 148.32 147.00 mg GAE/g). Globally, 104 phenolic compounds
were tentatively identied, belonging to the ve subclasses: avonoids (35), phenolic acids (32), other poly-
phenols (10), stilbenes (2) and lignan (1). Of these compounds, 24 had the same m/z but showed different
fragmentation proles. A higher number of polyphenols was identied in the bound extracts (77%) than in the
free extracts (59%). The addition of 4.5% of YM promoted an improved and more abundant prole of phenolic
compounds in the dough and bread. The major compounds found in the samples containing YM were 5-caffeoyl-
quinic acid and caffeic acid. The baking process did not adversely affect the abundance of phenolic compounds.
The bread-making process positively affected the phenolic prole due to the release of bound phenolic com-
pounds. At the same time, the addition of YM as a natural ingredient promoted an increase in the polyphenols in
the bread.
1. Introduction
The addition of natural ingredients in food products has become
more frequent nowadays since consumers are increasingly aware of their
benets. In general, natural ingredients are chosen according to their
phytochemical composition.
Ilex paraguariensis A. St. Hil., a native plant of the subtropical region
of South America, is widely consumed in several countries, mainly as a
hot or cold tea known as yerba mate (YM). It is also added to different
food products as a natural antioxidant (Mateos, Baeza, Sarri´
a, & Bravo,
2018; Cheminet, Baroni, & Wunderlin, 2021) since the leaves have well-
known pharmacological properties, including anti-inammatory and
antitumoral bioactivity (Bracesco, Sanchez, Contreras, Menini, &
Gugliucci, 2011; Gan, Zhang, Wang, & Corke, 2018). These properties
are associated with a high content of phenolic compounds (Santetti
et al., 2021a). Some phenolic compounds in yerba mate, such as 5-caf-
feolquinic acid, rutin, isoquercetin and caffeic acid, can increase the
functionality of food products due to their biological activity (Mateos
et al., 2018).
Phenolic compounds from different plant sources have been
* Corresponding author.
E-mail addresses: (L. Claudio Cameron), (M. Sim˜
oes Larraz Ferreira), (R. Dias de Mello
Castanho Amboni).
Contents lists available at ScienceDirect
Food Research International
journal homepage:
Received 26 April 2022; Received in revised form 29 June 2022; Accepted 5 July 2022
Food Research International 159 (2022) 111635
incorporated into bakery products to intensify the benecial health ef-
fects when consumed. This is usually carried out during the production
of bread, cakes and cookies, since it improves the nutritional and
functional properties (Martins, Pinho, & Ferreira, 2017; Ou, Wang,
Zheng, & Ou, 2019). Thus, incorporating YM leaves is an interesting
alternative for the bakery industry.
Bread is the main bakery product for incorporating bioactive in-
gredients due to its high level of consumption, being the most consumed
type of food in the world, sensorial acceptance and diversity of products
(Geng, Harnly, & Chen, 2016; Valli, Taccari, Di Nunzio, Danesi, &
Bordoni, 2018). However, in the bread-making process, fermentation
and baking are crucial steps and they determine the functional and
quality characteristics of the end product. The baking process can lead to
a decrease or increase in the phenolic content when compared to the
our and dough, since phenomena such as thermal degradation, poly-
merization/depolymerization, the release of bound phenolics and the
formation of Maillard reaction products can occur during the process
(Drakula et al., 2021).
Therefore, the potential health effects of bioactive compounds are
strongly dependent on their stability or transformation during the
baking process (Germ et al., 2019). Some phenolic compounds are not
thermally stable, for instance, high temperatures during baking can
cause the cleavage of the glycosyl portion of aglycone, leading to a
change in the bioavailability of the original compound (Ou et al., 2019).
In contrast, a previous study by our group veried that the fermentation
process can improve the antioxidant properties by increasing the free
phenolic acids content (Santetti et al., 2021a). Furthermore, adding YM
can enhance the prole of phenolic compounds, leading to a bakery
product with high bioactivity. Thus, the phenolic prole during the
bread-making process should be investigated since physical and chem-
ical changes can occur, leading to structural changes in the original
compounds with reductions or even transformation into other com-
pounds. Also, such studies can demonstrate the importance of the
incorporation a regional product that is rich in polyphenols of different
classes, such as avonoids and phenolic acids (Mateos et al., 2018), into
bakery products. To date, no investigations focused on changes in the
phenolic prole during the production of bread prepared with YM have
been reported.
In this context, we hypothesized that the bread-making process can
increase the content of phenolic compounds, leading to high potential
benets on consuming whole wheat bread prepared with yerba mate.
Thus, the aim of this study was to investigate the phenolic prole,
throughout the bread-making process, of our, dough and bread sam-
ples with YM added. Metabolomics tools based on an untargeted
approach were used to identify and relatively quantify the phenolic
compounds in ground YM leaves with two different particle sizes.
2. Material and methods
2.1. Material and chemicals
Wheat grains (Triticum aestivum L.), of the Toruk cultivar, 2017/18
harvest, were kindly provided by Biotrigo Gen´
etica-Ltda (Passo Fundo,
Brazil). The YM leaves from the 2017 harvest were kindly provided by
Inovamente (Il´
opolis, Brazil). The ingredients used for bread-making
were purchased in the local market.
Solvents (acetonitrile, ethanol and methanol) of LC-MS grade and
standards of phenolic compounds (benzoic acid, caffeic acid, catechin,
chlorogenic acid, ellagic acid, epicatechin, epigallocatechin, epicatechin
gallate, avonone, gallic acid, kaempferol, L-(-)-3-phenylacetic acid,
myricetin, p-coumaric acid, pyrogallol, quercetin, quercetin 3 glycoside,
synapinic acid, trans-cinnamic acid, isoferulic acid, vanillic acid, 4-
hydroxybenzyl alcohol, 4-hydroxy benzaldehyde, 4-hydroxybenzoic
acid, 4-phenylacetic acid, 3,4-dihydroxy phenylacetic acid, 2,5-dihy-
droxy benzoic acid, 4-methoxycinnamic acid, 2-hydroxycinnamic acid,
3-hydroxy-4-methoxycinnamic acid and 3-methoxycinnamic acid) were
purchased from Sigma-Aldrich (St. Louis, MO, USA). Celite® 545 and
formic acid, eluent additive for LC-MS, obtained from Fluka
(Switzerland) and the other reagents (ethyl acetate, hydrochloric acid
and sodium hydroxide) were of analytical grade. A Barnstead
Smart2Pure(Thermo Fisher Scientic, USA) system was used to purify
the water employed in the experiments.
2.2. Whole wheat our, yerba mate leaves and bread preparation
The whole wheat our (WWF) and YM leaves (Ilex paraguariensis St.
Hil) were prepared as previously described (Santetti et al., 2021b).
Briey, wheat grains were milled in an experimental mill (CD1, Chopin,
France) to obtain the fractions (rened our, bran and germ). The bran
and germ were then milled until they passed through a 30 mesh sieve.
Finally, the fractions were reincorporated into the rened our to obtain
the WWF. For the YM, the leaves were blanched (95 C for 30 s) and
dried in an air-forced oven (45 C for 24 h). Lastly, the YM was milled in
a knife mill (Marconi, M048, Brazil) and sieved to obtain particle sizes of
245 and 415.5 µm (Fig. S1, Supplementary Material), which were named
ne YM (FYM) and medium YM (MYM), respectively.
An experimental design was dened, where YM leaves were added to
the WWF (g YM.100 g
WWF; w/w) in two different proportions for
each particle size (ne and medium). Five treatments were prepared:
Fine YM (2.5 and 4.5%, w/w), Medium YM (2.5 and 4.5%, w/w) and the
control sample WWF (without YM). The our (F), dough (D) and bread
(B) samples were named: F (control), F1 (2.5% FYM), F2 (4.5% FYM), F3
(2.5% MYM) and F4 (4.5% MYM) for the ours; D (control), D1 (2.5%
FYM), D2 (4.5% FYM), D3 (2.5% MYM), and D4 (4.5% MYM) for the
dough; and B (control), B1 (2.5% FYM), B2 (4.5% FYM), B3 (2.5% MYM)
and B4 (4.5% MYM) for the bread. The amount of YM incorporated into
the WWF was determined from preliminary tests and the chemical
composition of the FYM, MYM and WWF has been previously reported
by Santetti et al. (2021b).
The bread samples were prepared according to a method described
by Bressiani et al. (2017) using a small-scale baking instrumental.
Briey, the formulation was composed of our, hydrogenated vegetable
oil, sodium chloride, ascorbic acid, sugar, yeast and water. Portions
(35.0 g) of dough were divided and fermented in a chamber (Multip˜
Brazil) at 30 ±1 C for 60 min. The bread baking was carried out in a
laboratory oven at 150 C for 12 min (Santetti et al., 2021b).
2.3. Extraction of free and bound phenolic compounds
The dough and bread samples were lyophilized for 48 h, milled and
sieved through a 30 mesh (Ning, Hou, Sun, Wan, & Dubat, 2017). The
free phenolic compounds in the our, dough and bread samples with YM
added were extracted with 80% ethanol, while the bound phenolic
compounds were obtained from the pellets and after alkaline and acid
hydrolysis, as described by Santos et al. (2019). The dried extracts ob-
tained were resuspended in 1.5 mL of methanol (2%), acetonitrile (5%)
and MilliQ water (93%), ltered through analytical hydrophilic lters
(13 mm diameter; 0.22 µm) (Analítica, Brazil), transferred to vials
(Waters, USA) and stored at 20 C until the UPLC-MS
2.4. Total phenolic content
The free and bound extracts of the our, dough and bread samples
were evaluated to determine the phenolic content. The total phenolic
content was determined from the sum of the values for the free and
bound extracts obtained by the Folin-Ciocalteu method (Singleton,
Orthofer, & Lamuella-Ravent´
os, 1999). The quantication was based on
a gallic acid standard curve (y =0.0161x +0.0133; R
=0.9953) and
the results were expressed in mg gallic acid equivalents per g of sample
on a dry basis (mg GAE/g).
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
2.5. Determination of the phenolic prole by UPLC-MS
The phenolic prole was determined based on the method described
by Santos et al. (2019) with some modications. Briey, 5
L of each
sample and a mix containing 33 analytical standards of phenolic com-
pounds (10 mg/L) were injected into a UPLC Acquity system (Waters
Co., Milford, MA) coupled to a XEVO G2S Q-Tof MS system (Waters Co.,
Manchester, UK) equipped with an ESI ionization source operating in
negative mode. The chromatographic gradient method was performed
as follows: 0 min 97% A; 11.80 min 50% A; 12.38 min 15% A;
14.23 min 15% A; 14.70 min 97% A (phase A MilliQ water with
0.3% formic acid and 5 mM ammonium formate and phase B acetonitrile
with 0.3% formic acid). The desolvation gas (N
) was set at 1.000 L.h
and 500 C, while the cone gas was set at 50 L.h
and the source at
150 C. The mass scanning range in the MS/MS acquisition was from m/
z 50 to 1200 Da and the ramp of the collision energy was from 20 to 45
eV for fragmentation, using ultrapure argon (Ar) as the collision gas. In
addition, pooled quality control (QC) samples were injected after each
set of 9 injections of the samples analyzed to monitor the instrument
performance and the data quality.
2.6. Data processing
The raw data obtained by UPLC-MS
was processed and submitted to
a normalization step in the Progenesis QI software, v. 2.1 (NonLinear
Dynamics, Waters Co.), respecting the following conditions: centroid
data, full-width resolution at half maximum of 30.000 and ionization in
negative mode. The identication of compounds was then performed by
comparison of the data obtained with the analytical standards as well as
a customized database for polyphenols obtained from PubChem (http
s:// and an online database (Phenol-Ex-
plorer v. 3.6), based on the exact mass error (<10 ppm), neutral mass
isotope distribution (>80%) and retention time, according to recom-
mendations given by Sumner et al. (2007). The MS/MS fragments prole
was also considered in the identication through the use of specic
databanks, such as MassBank of North America (MoNA) (https://mona.
 and NORMAN MassBank (
2.7. Statistical analysis
Data generated by Progenesis QI were exported to EZinfo 3.0 (Wa-
ters) to perform the multivariate principal component analysis (PCA)
and partial least squares-discriminant analysis (PLS-DA). Hierarchical
cluster analysis (HCA) and heat map analysis were performed using the
Metaboanalyst 5.0 web server ( The
results were submitted to one-way analysis of variance (ANOVA) with a
95% condence interval and the means were compared by applying the
Tukey test (p <0.05) using XLSTAT software (Addinsoft, France). All
analyses were carried out in triplicate.
3. Results and discussion
3.1. Total phenolic content of our, dough and bread
Table 1 shows the free, bound and total phenolic compounds in the
our, dough and bread samples. The highest content of free phenolic
compounds (p <0.05) was observed in the our, dough and bread
samples with the highest YM content (4.5%) added and using the ne
particle size (F2, D2 and B2) (Table 1). As expected, a greater amount of
yerba mate promotes a higher phenolic content, while the ne particle
size enables greater exposure and release of the compounds (Butiuk
et al., 2021; Santetti et al., 2021a). In addition, the free phenolic com-
pounds in the our samples with YM added were slightly higher (p <
0.05) than the levels in the dough and bread. This can be explained by
the fact that YM is more available and exposed in the matrix and thus the
organic solvent used extracts a greater amount of free phenolics (Alves &
Perrone, 2015). On the other hand, the Folin-Ciocalteu method may
overestimate the value, since some compounds detected by the reagent
may not be classied as phenolic compounds. For the dough and bread
samples D4 and B4, respectively, comprising 4.5% MYM, showed a high
free phenolic contents. However, the control our, dough and bread
samples had the lowest (p <0.05) phenolic content in both forms (free
and bound).
Regarding the results obtained for samples D1 and B1, the concen-
tration of 2.5% may not have been sufcient to detect possible changes
during processing due to the dilution factor resulting from the low
concentration of YM in the WWF compared with the concentration of
4.5%. A lower free phenolic content was observed (p <0.05) in the
bread sample B3 compared with the dough sample D3, for 2.5% MYM,
indicating that the phenolic compounds were affected by the action of
the oven temperature. In contrast, the samples D2 and B4 (4.5% ne and
medium YM, respectively) had a higher content of bound phenolics
compared to the other samples (p <0.05). On the other hand, D4 and B4
showed no statistical difference (Table 1).
For the bread sample B4, this could be related to the high baking
temperatures, which favor the hydrolysis of polysaccharides, thus
making bound phenolic compounds more available (C˘
alinoiu & Vodnar,
2020). Shahidi and Yeo (2016) reported that baking affects the phenolic
content, especially the bound phenolic compounds, since under high
temperatures these compounds can be released into the matrix. In
addition, the fermentation process increases the content of free phenolic
compounds (Santetti et al., 2021a). Also, the presence of YM in a higher
proportion facilitates greater extraction and quantication of the free
In general, the free phenolic content was higher than the bound
phenolic content, representing 62% of the total phenolic content of the
dough and bread samples, except for the control sample. In the control
sample, the bound phenolic content represented 53% of the total
(Table 1), possibly because most of the phenolic compounds in wheat
grains are found in the bound form (Santos et al., 2019).
Table 1
Free, bound and total phenolic compounds in dough and bread samples prepared
with the addition of YM.
Sample Free (mg/GAE g) Bound (mg/GAE g) Total (mg/GAE g)
F 54.55 ±0.83
63.05 ±0.82
116.81 ±0.94
F1 80.79 ±0.82
57.18 ±0.50
137.97 ±1.02
F2 122.68 ±0.16
65.07 ±0.15
181.48 ±0.08
F3 98.26 ±0.19
58.80 ±1.00
163.32 ±1.18
F4 113.04 ±0.53
62.22 ±1.41
175.26 ±1.93
D 29.85 ±0.33
38.52 ±0.56
68.37 ±0.70
D1 69.18 ±2.53
44.38 ±2.24
113.56 ±3.46
D2 98.75 ±0.25
50.88 ±0.56
149.62 ±0.69
D3 68.57 ±0.19
44.06 ±0.73
112.63 ±0.56
D4 95.24 ±1.00
46.16 ±0.83
141.40 ±1.09
B 29.03 ±2.99
37.4 ±0.83
66.43 ±3.68
B1 68.79 ±4.60
34.81 ±3.67
103.60 ±8.22
B2 101.21 ±2.15
47.11 ±1.04
148.32 ±2.88
B3 64.25 ±2.63
43.97 ±2.68
108.22 ±3.40
B4 95.74 ±0.16
51.27 ±0.88
147.00 ±0.91
Each value is presented as mean ±standard deviation (n =3). Uppercase letters
in the same column differs statistically (p <0.05) between our, dough and
bread. The lowercase in the same column differs statistically (p <0.05) from the
our, dough and bread, separately. The averages submitted by Tukey test, for
each form (free, bound and total). F: whole wheat our. D: dough whole wheat
our. B: bread whole wheat our. F1 and F2: Flour elaborated with 2.5 and 4.5%
of ne YM (w/w). F3 and F4: Flour elaborated with 2.5 and 4.5% of medium YM
(w/w). D1 and D2: Dough elaborated with 2.5 and 4.5% of ne YM (w/w). D3
and D4: Dough elaborated with 2.5 and 4.5% of medium YM (w/w). B1 and B2:
Bread elaborated with 2.5 and 4.5% of ne YM (w/w). B3 and B4: Bread elab-
orated with 2.5 and 4.5% of medium YM (w/w).
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
Phenolic compounds typically occur in plants in soluble form (free or
conjugated to soluble carbohydrates via ester/ether bonds) and are thus
easier to quantify. Other compounds are present in insoluble forms (e.g.,
bound via ester/ether bonds to cell wall constituents), requiring chem-
ical, enzymatic and/or physical actions for their release (Arig`
oa, ˇ
a, Calabr`
o, &
a, 2018). Although in cereals the bound
fraction is predominant (Acosta-Estrada, Guti´
errez-Uribe, & Serna-
Saldivar 2014), in this study, after the incorporation of YM, a greater
phenolic compounds content was observed in the free fraction, showing
the important contribution of the YM to the phenolic content.
As previously mentioned, the high free phenolic content may be due
to the effect of the fermentation process and the baking temperature,
which can lead to some complex phenols being hydrolyzed. Further-
more, the Maillard reaction in the bread-making process may, to some
extent, inuence the phenolic content (Verardo et al., 2018; Ou et al.,
2019; Antognoni, Mandrioli, Potente, Saa, & Gianotti, 2019). After the
thermal release of some phenolic acids during baking, certain com-
pounds, such as ferulic acid, can be incorporated into the structure of
melanoidins, end products of the Maillard reaction (Çelik & G¨
2020). According to ˇ
c et al. (2013), these compounds may be present
in free or bound form. Generally, the amount in bound form is greater
since glycosidic bonds between phenolics acids and melanoidins can
occur. Thus, these compounds can be identied in the matrix and pre-
sent antioxidant activity in the baked bread (Nooshkam, Varidi, &
Bashash, 2019).
These results support the hypothesis that the baking temperature can
affect the phenolic content in bread in two ways: i) promoting oxidation
and ii) releasing new phenolic compounds into the matrix. Since
phenolic compounds are highly reactive, baking can alter their physical
and chemical nature through, for instance, degradation, polymerization,
oxidation and the release of bound forms of phenols (Palermo, Pellegrini
& Fogliano 2014). The addition of YM promotes an increase in the
phenolic content in both forms (free and bound), favoring the devel-
opment of bread with functional appeal. The health benets associated
with these phytochemicals are due to their biological properties, such as
antioxidant and anti-inammatory activity. Thus, robust tools, such as
liquid chromatography coupled with mass spectrometry, can help
elucidate the bioactive compounds present in dough and bread, since
they can be changed, degraded or even released during the bread-
making process and at the oven temperature.
3.2. Phenolic prole during bread-making process determined by UPLC-
In order to obtain a comprehensive characterization of the phenolic
compounds prole for samples with YM incorporated into the whole
wheat our (in different proportions and particle sizes), as well as the
effect of baking the bread on this prole, the samples were submitted to
untargeted metabolomics analysis, using the UPLC-MS
A total of 104 phenolic compounds were putatively identied,
including some isomers, encompassing different classes. These com-
pounds and the parameters used for identication are described in
Table S1 (Supplementary Material). Of these compounds, 24 had the
same m/z but showed different fragmentation proles. Although it was
not possible to conrm their identity based on the fragmented ions
generated by the UPLC-MS
analysis and the other data considered in
the identication, it was possible to classify these unknown compounds
into avonoid and phenolic acid derivates. Thirteen phenolic com-
pounds were fully conrmed from analytical standards, including
quercetin, ferulic acid, 5-caffeoylquinic acid and vanillin (compounds in
Fig. 1. Number of phenolic compounds identied in each (A) class and sample and (B) extract and (C) the evolution of the number compounds identied during the
bread making stages.
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
bold in Table S1 of the Supplementary Material).
Fig. 1A shows the identication number for the phenolic compounds
in each class for the different samples (our, dough and bread). Glob-
ally, 5 subclasses were identied for all samples: the avonoid class had
the highest number of compounds identied (35), followed by phenolic
acids (32), other polyphenols (10), stilbenes (2) and lignans (1). The
bound extracts had a higher number of phenolic compounds for all
samples, representing 77% of the total, compared to the free extracts
(Fig. 1B).
The subclass of avonoids and their derivates represented 49% of the
total identications, followed by phenolic acids and their derivates
(38%), other polyphenols (10%), stilbenes (2%) and lignans (1%). Fla-
vonoids and phenolic acids are commonly found in YM and whole wheat
our (Mateos et al., 2018; Tian, Chen, Tilley, & Li, 2021), corroborating
these results and conrming their predominance. Moreover, the avo-
noids are more stable during the baking process and their structures are
linked by glycosidic bonds formed through sugar ligands (such as
glucose, rhamnose and galactose) (Ou et al., 2019; Li et al., 2020). In our
study, quercetin showed a notable increase in the bread samples (com-
pound number 32 in Table S1 of the Supplementary Material). This in-
crease can be explained by the transformation of rutin into quercetin
during baking bread, where the temperature and also the addition of
water and yeast may alter the medium and consequently the compound
is released (Brites et al., 2022).
This explains the greater abundance of avonoids compared with
phenolic acids, which are more susceptible to interaction with gluten
proteins, forming covalent complexes with sulfhydryl (S-H) groups,
leading to a low quantication of free compounds in the matrix (Xu,
Table 2
The most abundant free and bound phenolic compounds identied in our, dough and bread samples with addition of YM.
Name Molecular
m/z RT
Fragment data (% relative intensity) ME
Caffeoylquinic acid isomer I C
353.0867 4.53 57.5 91.9 191.0561 (100.00%), 135.0451 (57.58%), 179.0349
(55.07%), 353.0878 (53.58%), 161.0244 (6.20%),
85.0295 (4.45%), 127.0400 (2.07%), 111.0451
(1.97%), 155.0350 (1.72%)
3.1 99.3 PA
Caffeic acid C
179.0338 6.28 52.1 69.4 135.0451 (100.00%), 117.0345 (2.36%) ¡6.5 98.5 PA
5-Caffeoylquinic acid C
353.0869 5.91 56.5 86.3 191.0561 (100.00%), 353.0878 (8.68%), 85.0295
¡2.5 99.3 PA
Dicaffeoylquinic acid isomer II C
515.1190 8.73 53.5 69.2 135.0451 (100.00%), 161.0244 (9.88%), 335.0772
1.0 99.5 PA
Isoferulic acid C
193.0495 8.03 44.2 28.6 193.0506 (52.08%), 149.0608 (61.72%),
151.0400 (33.02%), 163.0400 (29.60%),
117.0346 (18.59%), 109.0295 (17.75%),
123.0451 (15.34%), 129.0346 (13.34%),
175.0400 (13.04%), 161.0608 (12.29%)
¡5.8 99.2 PA
Caffeoylquinic acid isomer II C
353.0858 6.14 55.4 84.9 173.0455 (100.00%), 179.0349 (89.73%), 135.0451
(55.21%), 191.0561 (44.35%), 155.0350 (5.83%),
161.0244 (4.35%), 85.0295 (2.62%), 109.0295
5.8 98.8 PA
515.1194 8.99 48.8 45.1 311.0772 (87.28%), 375.0721 (26.44%) 0.2 99.0 PAD
Kaempferol 3,7-o-diglucoside C
609.1467 8.07 53.1 71.1 300.0275 (100.00%), 151.0037 (3.44%), 271.0278
(3.14%), 255.0299 (2.01%), 343.0459 (1.20%),
273.0404 (0.48%)
1.0 95.5 F
Dicaffeoylquinic acid isomer I C
515.1190 8.55 52.6 64.8 179.0349 (76.03%), 515.1195 (33.27%), 353.0878
(29.56%), 161.0244 (17.88%), 335.0772 (7.36%),
109.0295 (5.40%), 123.0451 (2.99%), 203.0350
(1.94%), 91.0189 (1.38%)
1.0 99.6 PA
Esculetin C
177.0179 6.14 52.4 72.4 153.0193 (100.00%), 148.0166 (6.56%), 104.0267
(5.92%), 135.0087 (4.85%), 67.0189 (4.35%),
120.0216 (2.05%)
8.2 98.6 OP
Feruloylquinic acid isomer I C
367.1020 6.08 56.4 87.8 193.0506 (100.00%), 111.04551 (35.65%),
367.1034 (33.93%), 149.0608 (11.32%), 117.0608
3.8 98.7 PA
p-Coumaric acid C
163.0386 7.51 55.4 88.6 119.0502 (100.00%), 117.0346 (2.45%) ¡9.1 98.7 PA
Feruloylquinic acid isomer II C
367.1017 7.28 46.9 41.1 173.0455 (100.00%), 193.0506 (16.23%), 93.0345
(15.62%), 111.0451 (4.31%), 155.0350 (3.80%),
149.0608 (2.42%), 117.0346 (1.27%)
4.7 99.0 PA
Ferulic acid C
193.0491 8.28 41.8 20.2 117.0346 (22.08%), 106.0424 (3.46%) ¡7.8 97.2 PA
Luteolin 7-o-rutinoside C
593.1503 8.55 53.9 72.0 173.0455 (100.00%), 515.1192 (33.27%), 353.0878
(29.56%), 159.0451 (10.87%), 109.0295 (5.40%),
123.0451 (2.99%), 93.0345 (2.72%), 111.0451
(2.20%), 91.0189 (1.38%), 257.0455 (0.51%),
327.0510 (0.34%)
1.5 99.3 F
Apigenin 7-o-apiosyl-glucoside
isomer II
563.1406 7.53 57.7 90.3 563.1406 (100.00%), 353.0667 (74.95%), 383.0772
(45.92%), 443.0983 (12.86%), 473.1089 (6.94%),
365.0667 (2.18%), 503.1195 (2.00%), 413.08787
(1.83%), 425.0878 (1.82%), 397.0928 (0.82%)
0.0 98.1 F
p-Coumaroylquinic acid C
337.0907 6.90 43.1 22.8 119.0502 (100.00%), 117.0502 (11.92%) 6.4 99.7 PA
4-Hydroxyphenylglycolic acid /
167.0338 3.01 56.3 90.6 166.0271 (100.00%), 123.0451 (76.93%) 6.9 98.7 PA
Apigenin 7-o-apiosyl-glucoside
isomer I
563.1379 7.32 50.8 61.7 353.0667 (60.44%), 443.0983 (12.31%), 473.1089
(7.96%), 325.0717 (2.07%), 365.0667 (1.43%)
4.9 97.6 F
Rosmarinic acid C
359.0753 6.30 40.8 13.9 135.0451 (100.00%), 117.0345 (2.36%) 5.3 96.0 PA
m/z: mass/charge. RT: retention time. FS: fragmentation score. IS: isotopic similarity. ME: mass error. Bold represent reference standards. Class: PA: phenolic acids. F:
avonoids; PAD: phenolic acid derivatives. OP: other polyphenol.
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
Wang, & Li, 2019).
In assessing the number of compounds identied, the samples with
YM added presented a greater number of phenolic compounds than the
control samples at all stages of the bread-making process (our, dough
and bread). In the case of the our samples, F4 (4.5% MYM, w/w)
showed the highest number of compounds identied, while for bread,
the sample B2 (4.5% FYM, w/w) had the highest number (Fig. 1A). The
dough D1-D4 samples showed no statistical difference (p >0.05)
regarding the number of phenolic compounds identied, except for the
control dough, which had the lowest number. The D2 and B2 samples,
with the highest proportion of YM (4.5%) of the smaller particle size
(FYM), presented the highest number of compounds identied. This can
be explained by the greater amount of YM with a large contact surface
(FYM) favoring chemical interactions during the bread-making process
(Santetti et al., 2021b; Butiuk et al., 2021). Furthermore, these results
corroborate the total phenolic compounds data (Table 1), determined
using the Folin-Ciocalteu reduction capacity method.
The increase in phenolic compounds with the addition of YM was
expected given its high levels of bioactive components, especially
polyphenols. Although it was not analyzed, the phenolic contribution of
isolated YM can be mainly attributed to the phenolic acids class, with a
value four times higher than that of WWF (Fig. 1B). This is consistent
with results reported by Mateos et al. (2018), who found that the
phenolic acids class comprises 90% of Ilex paraguariensis phenols.
Among these, 3-caffeoylquinic acid and its isomers are the main com-
pounds present in YM (Mateos et al., 2018), corroborating the results of
the present study, since these compounds were only identied after YM
addition (compounds number 55, 60 and 62 in Table S1 of the Supple-
mentary Material). Finally, these YM phenolic acids are mostly present
in free extracts (Fig. 1A), which can positively affect their bioavailability
during human digestion.
Considering the total relative ion abundance, 20 compounds were
highlighted as the most abundant in all our, dough and bread samples
(Table 2). The B1, B2 and B3 showed a progressive increase (bread >
dough >our), while sample B4 (4.5% MYM) showed an increase
compared with the our and dough samples; however, the total relative
abundance was constant compared to the dough (Fig. 1C). This indicates
that the abundance of the compounds remained constant, which is
probably also due to the greater proportion and larger contact surface of
the YM added. On the other hand, the decrease is pronounced in the
bread sample prepared with 4.5% MYM (Fig. 1C) compared with the
our and dough samples. This could be directly related to the small
changes that occur in the structure of the original compounds, which
cannot be detected by the Folin-Ciocalteu method (Table 1) but was
reected in the decrease detected by mass spectrometry.
According to Santetti et al. (2021a), during the dough fermentation,
new compounds are released in the matrix in the presence of YM and this
may have contributed to the maintenance of the compounds after
baking. Tian et al. (2021) also found that baking can favor the release of
phenolic compounds, explaining the changes in the phenolic composi-
tion during the bread-making process. In addition to the covalent di-
sulde bonds (S-S), polyphenols can form complexes with wheat
proteins via hydrogen bonds and/or hydrophobic interactions (Yuan
et al., 2021). The formation of these protein-based complexes can
explain the presence of phenolic compounds even after bread baking
(Angelino et al., 2017).
Fig. 2A compares the relative abundance of free and bound phenolics
for all samples. The bound fraction is more abundant in the control
samples (WWF), while with YM incorporated the phenolic compounds
are mostly in the free fraction (Fig. 2B). These data corroborate the re-
sults found in the analysis of reducing compounds using Folin-Ciocalteu
reagent and demonstrate that the greatest effect of YM incorporation is
on the phenolic compounds found in the free form. Besides the positive
effect on the free fraction, samples D2, D4, B2 and B4 (4.5% YM) have a
Fig. 2. Total relative ion abundance of free and bound classes and extracts (A) and sum of the relative ion abundance for free and bound extracts of our, dough and
bread samples (B). Legend: Different letters mean signicant difference (p <0.05) and bars represent standard deviation.
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
higher relative abundance of phenolic compounds (p <0.05), regardless
of the YM particle size (Fig. 2A). These results evidence the importance
of adding an external source of phenolic compounds to the matrix to
improve the free phenolic content.
Considering the abundance of classes, rather than the number of
identications, the phenolic acids class was the most abundant in all
samples in both forms (free and bound). Although these samples have a
greater diversity of molecules belonging to the avonoid class, the
content of phenolic acids present is greater.
In Fig. 2B, the relative total abundance of phenolic compounds
(our, dough and bread) in free and bound forms can be observed. The
bread sample B2 (4.5% FYM) presented the greatest abundance of
phenolic compounds of all samples, followed by B3 (2.5% MYM). In
contrast, for the dough samples D2 and D4 (4.5% ne and MYM) these
compounds were more abundant compared with the other dough sam-
ples. The our sample F2 (4.5% ne YM) showed a greater relative total
abundance than the other ours.
The differences between samples in terms of the abundance of
phenolic compounds and the notable presence of these compounds in
bread even after baking can be attributed to the breakage of the matrix
during the thermal process and the formation of new compounds due to
the Maillard reaction (Angelino et al., 2017; Xu, Guo, Romana, Pico, &
Martinez, 2020). It has been reported by Domingu´
andez et al.
(2021) that during baking the chlorogenic acids (or 5-caffeoylquinic
acids) may link to the melanoidins structures via covalent bonds. This
reaction can allow the phenolic compounds in bread to be quantied,
and in our study they are mostly present in free forms (compounds
number 55, 60, 62, 75 and 76 in Table S1 of the Supplementary Mate-
rial). In addition, during bread-making, the heat transfer into the center
of the product is relatively slow. Therefore, the compounds may not be
degraded (Bureˇ
a et al., 2021). Moreover, free phenolic compounds
can be linked to proteins leading to the formation of new compounds
through condensation reactions. For example, quinine in YM leaves and
stems can react with the S-H and amino groups of proteins, favoring its
identication in bread (Ozdal, Capanoglu, & Altay, 2013).
3.3. Multivariate analysis for discrimination of phenolic prole of our,
dough and bread samples
To obtain a better understanding of the behavior of the phenolic
compounds during the bread-making process, the phenolic proles of
the our, dough and bread with YM added and the control samples were
analyzed by multivariate statistical analysis (PCA and HCA).
The degree of similarity of the extracts (free and bound) was evi-
denced through the PCA biplot analysis (scores: samples; loadings:
phenolic compounds) of the samples (Fig. 3A). The two main compo-
nents (PC1 and PC2) explained 86% of the total variance, with a clear
distinction between the proles of phenolic compounds present in the
free and bound forms on comparing all samples. PC1 explained most of
the data (70%) and PC2 explained 16% of the total dispersion.
Furthermore, for samples B1 and B3 the proles for the bound phenolic
compounds were more distinct compared with the other samples.
In the second PCA, the samples were grouped into our, dough and
bread to perform a PLS analysis (Fig. 3B). In this case, the two main
components (PC1 and PC2) explained 81% of the total variance. The
approximation of the dough and bread samples can be noted due to the
great similarity of the phenolic compounds proles, while the our
samples had a more distinct prole. This could be associated with the
transformation of phenolic compounds during the bread-making pro-
cess, notably in the fermentation and baking steps, which can help to
differentiate between these two groups (Angelino et al., 2017; Santetti
et al., 2021a).
In addition, the compounds that could most strongly inuence the
differentiation of the samples were estimated by the variable importance
in the projection (VIP) method, obtained from the partial least squares-
discriminant analysis (PLS-DA) (Fig. 3C). Some of these compounds
were noted in the samples containing YM, namely caffeic acid, caf-
feoylquinic acid isomer I, dicaffeoylquinic acid isomer II, 5-caffeoyl-
quinic acid and caffeoylquinic acid isomer II. These have been
reported as the major compounds in YM (Mateos et al., 2018; Kaltbach,
Ballert, Kabrodt, Schellenberg, 2020; Butiuk et al., 2021; Cheminet
et al., 2021), which explains the difference compared with the samples
without or with smaller proportions of YM. These compounds and the
isomers may be present due to the transformations that occur during
bread-dough fermentation. Leonard et al. (2021) reported that the
biotransformation of hydroxycinnamic acids, such as caffeic acid, can
occur due to decarboxylase or reductase enzymes during fermentation
with yeast.
On the other hand, isoferulic, ferulic and p-coumaric acids, consid-
ered the most abundant compounds in wheat (Santos et al., 2019), were
identied in the our, dough and bread control samples. In all of the free
extracts of samples with YM incorporated it was possible to note the
presence of different structures of chlorogenic acids. In contrast, the
most abundant compounds in bound extracts were simple phenolic
acids, such as ferulic acid, p-coumaric acid and caffeic acid.
Fig. 4A shows the HCA and heat map for selected groups in free and
bound extracts. The total number of compounds in common for the free
and bound extracts was 28 and included isoferulic acid, vanillin and
sinapic acid. In the distribution and variation of the relative abundance
of these compounds in each sample according to the color intensity, a
wide variation between free and bound phenolics in the extracts was
observed. The highest prevalence of the selected compounds was found
in the bound form. Some phenolic compounds were identied in greater
abundance in the bread samples and in the bound extract, including
rosmarinic acid, caffeic acid, esculetin and apigenin 7-O-glucoside.
During bread-making, many phenolic compounds can be involved in the
formation of Maillard reaction products, which leads to an increase in
the phenolic content (Domínguez-Fern´
andez et al., 2021). In addition,
this increase may be due to the baking temperature resulting in the
release of phenolics bound to the cell wall or the conversion of free
phenolics to bound phenolics via condensation reactions with proteins
and sugars through hydrogen and water bonds (Li et al., 2020). Ac-
cording to Alves and Perrone (2015), the bound phenolic compounds are
easily incorporated into melanoidins structures. This is consistent with
the results of this study, in which the major fractions of phenolics were
present as bound forms and may be linked through ester or glycosidic
In the B4 sample (free phenolics extract), vanillin was present in the
greatest abundance, which can be attributed to the bread baking tem-
perature. According to C˘
alinoiu and Vodnar (2020), an increase in
vanillin after baking is probably due to particular bonds and the position
between parts of the cell wall and the phenolic acids. In addition, caffeic
acid was predominant in dough and bread samples with YM added, since
this is a major compound in YM leaves (Bracesco et al., 2011).
In summary, the phenolic proles were distinct for all samples
studied, as the addition of YM increased the abundance of phenolic
compounds. Moreover, the bread-making process, particularly certain
steps such as baking, can be an important factor when seeking to
improve the phenolic content, as many of these compounds are released
with the rupture of the cell wall caused by the increase in temperature.
4. Conclusions
This study elucidated and evaluated for the rst time the prole of
phenolic compounds in whole wheat our, dough and bread samples
prepared with the incorporation of YM. The results demonstrated the
effect of the bread-making process on the phenolic prole using robust
metabolomics tools. Samples with YM incorporated showed a greater
diversity of phenolic compounds than the control samples. The dough
and bread samples with the highest YM content (4.5%) presented a
greater abundance of phenolic compounds than the our samples.
The fermentation and baking processes can help release phenolic
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
Fig. 3. (A), (B) PCA and PLS of the metabolites identied in the samples, the score (samples represented by symbol) is distributed according to the relative ion
abundance of phenolic compounds identied (grey circles); and (C) compounds discriminated by variable importance projection analysis in all samples (our, dough
and bread). Legend: D: dough whole wheat our. B: bread whole wheat our. D1 and D2: Dough elaborated with 2.5 and 4.5% of ne YM (w/w). D3 and D4: Dough
elaborated with 2.5 and 4.5% of medium YM (w/w). B1 and B2: Bread elaborated with 2.5 and 4.5% of ne YM (w/w). B3 and B4: Bread elaborated with 2.5 and
4.5% of medium YM (w/w).
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
compounds. In addition, multivariate statistical analysis demonstrated
that most phenolic compounds were present in free form, possibly due to
the action of the oven temperature and the fermentation process,
causing the breaking of chemical bonds between the compounds and the
matrix. In conclusion, the addition of YM to whole wheat our in bread
making offers potential benets due to the presence of several phenolic
compounds of different classes. Even with the incorporation of a low
content (<5%; w/w) of yerba mate leaves in the bread, the increase was
pronounced and the technological properties were maintained. Thus,
this is a feasible strategy for the enrichment of bakery products, pro-
moting possible health benets for the consumers and enhancing the use
of a regional product. However, more research is needed to investigate
the storage of bread, since unwanted sensory changes can occur along
with losses of phenolic compounds. In addition, further research should
be aimed at evaluating the bioaccessibility of the phenolic compounds
during in vitro gastrointestinal simulation, since several biochemical
reactions can occur during bread ingestion, which can change the bio-
logical action of phenolic compounds in the body.
CRediT authorship contribution statement
Gabriela Soster Santetti: Conceptualization, Investigation, Data
curation, Writing original draft, Writing review & editing. Luciana
Ribeiro da Silva Lima: Methodology, Investigation, Data curation,
Writing original draft, Writing review & editing. Barbara Biduski:
Conceptualization, Data curation, Writing original draft, Writing
review & editing. Millena Cristina Barros Santos: Methodology,
Investigation, Data curation, Writing original draft, Writing review &
editing. Carolina Thomaz dos Santos DAlmeida: Methodology,
Investigation, Data curation, Writing original draft, Writing review &
editing. Luiz Claudio Cameron: Funding acquisition, Resources. Luiz
Carlos Gutkoski: Conceptualization, Supervision, Funding acquisition,
Resources. Mariana Sim˜
oes Larraz Ferreira: Methodology, Investiga-
tion, Data curation, Resources, Writing review & editing. Renata Dias
de Mello Castanho Amboni: Conceptualization, Supervision, Writing
original draft, Writing review & editing.
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.
Acknowledgments and fundings
This study was nanced in part by the Coordenaç˜
ao de Aperfeiçoa-
mento de Pessoal de Nível Superior - Brasil (CAPES) (Financial Code
001); Fundaç˜
ao de Amparo `
a Pesquisa do Estado do Rio de Janeiro
(FAPERJ) (26/202.709/2018), Conselho Nacional de Desenvolvimento
Cientíco e Tecnol´
ogico (CNPq) (Grant numbers 162-16.00/16-2;
427.116/2018; 401053/2019-9). We would like to thank Inovamate
for providing the YM leaves, Biotrigo Gen´
etica for providing the wheat,
and the Federal University of the State of Rio de Janeiro (UNIRIO) and
University of Passo Fundo (UPF) for its assistance. Renata D. M. C.
Amboni and Mariana S. L. Ferreira were granted a fellowship (Pq1D and
Pq2, respectively) from CNPq.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
oa, A., ˇ
Cesla, P., ˇ
a, P., Calabr`
o, M. L., & ˇ
a, L. (2018). Development of
extraction method for characterization of free and bonded polyphenols in barley
(Hordeum vulgare L.) grown in Czech Republic using liquid chromatography-tandem
mass spectrometry. Food Chemistry, 245, 829837.
Acosta-Estrada, B. A., Guti´
errez-Uribe, J. A., & Serna-Saldivar, S. O. (2014). Bound
phenolics in foods, a review. Food Chemistry, 152, 4655.
Angelino, D., Cossu, M., Marti, A., Zanoletti, M., Chiavaroli, L., Brighenti, F., et al.
(2017). Bioaccessibility and bioavailability of phenolic compounds in bread: A
review. Food & Fuction, 8, 23682393.
Fig. 4. Hierarchical cluster analysis (HCA) and heat map of free and bound phenolic compounds in our, dough and bread samples prepared with YM added. Legend:
D: dough whole wheat our. B: bread whole wheat our. D1 and D2: Dough elaborated with 2.5 and 4.5% of ne YM (w/w). D3 and D4: Dough elaborated with 2.5
and 4.5% of medium YM (w/w). B1 and B2: Bread elaborated with 2.5 and 4.5% of ne YM (w/w). B3 and B4: Bread elaborated with 2.5 and 4.5% of medium YM
G. Soster Santetti et al.
Food Research International 159 (2022) 111635
Antognoni, F., Mandrioli, R., Potente, G., Saa, D. L. T., & Gianotti, A. (2019). Changes in
carotenoids, phenolic acids and antioxidant capacity in bread wheat doughs
fermented with different lactic acid bacteria strains. Food Chemistry, 292, 211216.
Alves, G., & Perrone, D. (2015). Breads enriched with guava our as a tool for studying
the incorporation of phenolic compounds in bread melanoidins. Food Chemistry, 185,
Butiuk, A. P., Maidana, A., Adachi, O., Akakabe, Y., Roque, A. M., & Hours, A. (2021).
Optimization and modeling of the chlorogenic acid extraction from a residue of
yerba mate processing. Journal of Applied Research on Medicinal and Aromatic Plants,
25, Article 100329.
a, B., Paznocht, L., Kotíkov´
a, Z., Giampaglia, B., Martinek, P., & Lachman, J.
(2021). Changes in carotenoids and tocols of colored-grain wheat during unleavened
bread preparation. Journal of Food Composition and Analysis, 103, Article 104108.
Bressiani, J., Oro, T., Santetti, G. S., Almeida, J. L., Bertolin, T. E., Gomez, M., et al.
(2017). Properties of whole grain wheat our and performance in bakery products as
a function of particle size. Journal of Cereal Science, 75, 269277.
Bracesco, N., Sanchez, A., Contreras, V., Menini, T., Gugliucci, A. (2011). Recent ad-
vances on Ilex paraguariensis research: Minireview. Journal of Ethnopharmacology,
136, 378384. / j.jep.2010.06.032.
Brites, L. T. G. F., Rebellato, A. P., Meinhart, A. D., Godoy, H. T., Pallone, J. A. L., &
Steel, C. J. (2022). Technological, sensory, nutritional and bioactive potential of pan
breads produced with rened and whole grain buckwheat ours. Food Chemistry, 13,
Article 100243.
alinoiu, L. F., & Vodnar, D. C. (2020). Thermal processing for the release of phenolic
compounds from wheat and oat bran. Biomolecules, 10, 114.
Çelik, E. E., & G¨
okmen, V. (2020). Effects of fermentation and heat treatments on bound-
ferulic acid content and total antioxidant capacity of bread crust-like system made of
different whole grain ours. Journal of Cereal Science, 93, Article 102978. https://
Cheminet, G., Baroni, M. V., & Wunderlin, D. A. (2021). Antioxidant properties and
phenolic composition of Composed Yerba Mate. Journal of Food Science and
Technology, 58, 47114721. 020-04961-x
andez, M., Ilrigoyen, A., Vargas-Alvarez, M. A., Ludwig, I. A.,
na, M. P., & Cid, C. (2021). Inuence of culinary process on free and bound (poly)
phenolic compounds and antioxidant capacity of artichokes. International Journal of
Gastronomy and Food Science, 25, Article 100389.
Drakula, S., Novotni, D., Mustaˇ
c, N.ˇ
C., Vouˇ
cko, B., Krpan, M., Hruˇ
skar, M., et al. (2021).
Alteration of phenolics and antioxidant capacity of gluten-free bread by yellow pea
our addition and sourdough fermentation. Food Bioscience, 44, Article 101424.
Gan, R., Zhang, D., Wang, M., & Corke, H. (2018). Health benets of bioactive
compounds from the genus Ilex a source of traditional caffeinated beverages.
Nutrients, 10, 1682.
Geng, P., Harnly, J. M., & Chen, P. (2016). Differentiation of bread made with whole
grain and rened wheat (T. aestivum) our using LC/MS-based chromatographic
ngerprinting and chemometric approaches. Journal of Food Composition and
Analysis, 47, 92100.
Germ, M., ´
Arvy, J., Vollmannov´
a, A., T´
oth, T., Golob, A., Luthar, Z., et al. (2019). The
temperature treshold for the transformation of rutin to quercetin in tartary
buckwheat dough. Food Chemistry, 283, 2831.
Kaltbach, P., Ballert, S., Kabrodt, K., & Schellenberg, I. (2020). New HPTLC methods for
analysis of major bioactive compounds in mate (Ilex paraguariensis) tea. Journal of
Food Composition and Analysis, 92, Article 103568.
Leonard, W., Zhang, P., Ying, D., Adhikari, B., & Fang, Z. (2021). Fermentation
transforms the phenolic proles and bioactivities of plant-based foods. Biotechnology
Advances, 49, Article 107763.
Li, M., Chen, X., Deng, J., Ouyang, D., Wang, D., Liang, Y., et al. (2020). Effect of thermal
processing on free and bound phenolic compounds and antioxidant activities of
hawthorn. Food Chemistry, 332, Article 127429.
Martins, Z. E., Pinho, O., & Ferreira, I. M. P. L. V. O. (2017). Food industry by-products
used as functional ingredients of bakery products. Trends in Food Science &
Technology, 67, 106128.
Mateos, R., Baeza, G., Sarri´
a, B., & Bravo, L. (2018). Improved LC-MS
of hydroxycinnamic acid derivatives and avonols in different commercial mate (Ilex
paraguariensis) brands. Quantication of polyphenols, methylxanthines, and
antioxidant activity. Food Chemistry, 241, 232241.
Ning, J., Hou, G. G., Sun, J., Wan, X., & Dubat, A. (2017). Effect of green tea power on
the quality attributes and antioxidant activity of whole-wheat our pan bread. LWT
Food Science and Technology, 79, 342348.
Nooshkam, M., Varidi, M., & Bashash, M. (2019). The Maillard reaction products as food
born antioxidant and antibrowning agentes model and real food systems. Food
Chemistry, 275, 644660.
Ozdal, T., Capanoglu, E., & Altay, F. (2013). A review on protein-phenolic interactions
and associated changes. Food Research International, 51, 954970.
Ou, J., Wang, M., Zheng, J., & Ou, S. (2019). Positive and negative effects of polyphenol
incorporation in baked foods. Food Chemistry, 284, 9099.
Palermo, M., Pellegrini, N., & Fogliano, V. (2014). The effect of cooking on the
phytochemical content of vegetables. Journal of the Science of Food and Agriculture,
94, 10571070.
Santetti, G. S., Dacoreggio, M. V., Silva, A. C. M., Biduski, B., Bressiani, J., Oro, T., et al.
(2021b). Effect of yerba mate (Ilex paraguariensis) leaves on dough properties,
antioxidant activity, and bread quality using whole wheat our. Journal of Food
Science, 111.
Santos, M. C. B., Lima, L. R. S., Nascimento, F. R., Nascimento, T. P., Cameron, L. C., &
Ferreira, M. S. L. (2019). Metabolomic approach for characterization of phenolic
compounds in different wheat genotypes during grain development. Food Research
International, 124, 118128.
Santetti, G. S., Dacoreggio, M. V., In´
acio, H. P., Biduski, B., Hoff, R. B., Fritzen-
Freire, C. B., et al. (2021a). The addition of yerba mate leaves on bread dough has
inuences on fermentation time and the availability of phenolic compounds? LWT
Food Science and Technology, 146, Article 111442.
Singleton, V. L., Orthofer, R., & Lamuela-Ravent´
os, R. M. (1999). Analysis of total
phenols and other oxidation substrates and antioxidants by means of folin-ciocalteu
reagent. Methods in Enzymology, 299, 152178.
Sumner, L. W., et al. (2007). Proposed minimum reporting standards for chemical
analysis. Metabolomics, 3, 211221.
Shahidi, F., & Yeo, J. (2016). Insoluble-Bound Phenolics in Food. Molecules, 21, 1216.
Tian, W., Chen, G., Tilley, M., & Li, Y. (2021). Changes in phenolic proles and
antioxidant activities during the whole wheat our bread-making process. Food
Chemistry, 345, Article 128851.
Valli, V., Taccari, A., Di Nunzio, M., Danesi, F., & Bordoni, A. (2018). Health benets of
ancient grains. Comparison among bread made with ancient, heritage and modern
grain ours in human cultured cells. Food Research International, 107, 206215.
Verardo, V., Glicerina, V., Cocci, E., Frenich, A. G., Romani, S., & Caboni, M. F. (2018).
Determination of free and bound phenolic compounds and their antioxidant activity
in buckwheat bread loaf, crust and crumb. LWT - Food Science and Technology, 87,
Yuan, W., Fan, W., Mua, Y., Meng, D., Yan, Z., Li, Y., et al. (2021). Baking intervention
for the interaction behaviours between bamboo (Phyllostachys heterocycla) leaf
avonoids and gliadin. Industrial Crops & Products, 164, Article 113385. https://doi.
Xu, J., Wang, W., & Li, Y. (2019). Dough properties, bread quality, and associated
interactions with added phenolic compounds: A review. Journal of Functional Foods,
52, 629639.
Xu, K., Guo, M., Romana, L., Pico, J., & Martinez, M. M. (2020). Okra seed and seedless
pod: Comparative study of their phenolics and carbohydrate fractions and their
impact on bread-making. Food Chemistry, 317, Article 126387.
c, S., Mogol, B. A., Akıllıo˘
glu, G., Serpen, A., Babi´
c, M., & G¨
okmen, V. (2013). Effects
of infrared heating on phenolic compounds and Maillard reaction products in maize
our. Journal of Cereal Science, 58, 17.
G. Soster Santetti et al.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
The nutritional quality and bioactive potential of breads made with partial replacement of refined wheat flour (RWF) with 30% or 45% refined buckwheat flour (RBF) or whole buckwheat flour (WGBF) was assessed through mineral bioaccessibility, starch digestibility, dietary fiber content and bioactive potential by determining rutin and quercetin levels during processing. Moreover, technological quality and sensory acceptance were also evaluated. Breads made with 30% or 45% WGBF showed higher mineral and fiber contents compared to the control, while the formulations with RBF showed higher bioaccessibility. No changes were observed in the rutin levels of the dough before and after fermentation, but after baking, rutin and quercetin levels increased. The highest starch hydrolysis was found in the formulation containing 45% RBF. The formulations made with 30% RBF or 30% WGBF were well accepted by consumers. Our study shows interesting results, as few studies report the effect of processing on bioactive compounds.
Gluten-free bread is often characterised by poor nutritional value and bioactive profile. Legume flours have potential for its enrichment, but the effect of sourdough fermentation of legume matrices on the phenolics and antioxidant capacity of bread is scarcely investigated. Thus, this study aimed was to determine the effect of partial replacement (25%) of wholemeal rice flour with yellow pea flour on phenolics and antioxidant capacity of gluten-free sourdough and bread. Sourdough was fermented with Lactobacillus reuteri DSM 20016, Lactobacillus fermentum DSM 20052 or Lactobacillus brevis DSM 20054. Free phenolic acids (protocatechuic, 4-hydroxybenzoic, vanillic and ferulic), lactic and acetic acid content was determined by HPLC, and free total phenolic content (TPC), DPPH and FRAP antioxidant capacity by spectrophotometric methods. After 16 h of fermentation, total titrable acidity of the sourdough ranged from 11.85 to 18.97 mL of 0.1 M sodium hydroxide, with lactic/acetic acid content of 2.65–9.41. Yellow pea flour addition substantially increased protocatechuic acid and 4-hydroxybenzoic acid content, but decreased the antioxidant capacity of unfermented dough and bread. Depending on the starter, the sourdough fermentation of pea-rice flour blend and its addition to bread increased the phenolic acid content, TPC and antioxidant capacity. Bread with yellow pea flour and L. brevis sourdough showed the highest improvement in phenolic acid content (40%), TPC (44%) and antioxidant capacity (30% DPPH, 50% FRAP) compared to bread without added sourdough. The study demonstrates the importance of using sourdough fermentation with a carefully selected starter when adding pea flour to gluten-free bread to ensure high antioxidant potential.
This study investigated the effects of different yerba mate (YM) proportions (1.5, 2.5, and 4.5 g YM/100 g whole wheat flour (WWF) and particle sizes (245, 415.5, and 623.9 µm) on dough rheological properties, antioxidant activity, and bread characteristics. The addition of YM leaves led to a possible interaction between its phenolic compounds and the gluten network within the dough, without negative effects on dough formation. However, the larger YM particle size (623.9 µm) caused a weakening of the protein network, resulting in lower quality product compared to the other samples. Improved bread quality was found when the YM leaves were added at 2.5 g YM/100 g WWF. The total amount of phenolic compounds and the antioxidant activity increased as the proportion of YM increased in both flour and bread. Moreover, the phenolic compounds in 2.5 g YM/100 g WWF breads were stable during baking, showing no significant losses in the amount of phenolic compounds and antioxidant activity. These results suggest the YM can be successfully incorporated into baked product, improving its functional characteristics. Practical Application This study evaluates the technological quality of bakery product made by incorporating yerba mate leaves in whole wheat flour. The results will contribute to the production of a bread with greater functional properties due to the presence of polyphenols and phytochemicals.
The content of carotenoids and tocols in the individual steps of technological processing was determined using HPLC-DAD and HPLC-FLD. The dough preparation resulted in a more pronounced loss of the investigated antioxidants (51.5 and 33.0%) compared to baking (22.5 and 9.1%). Carotenoids appear to be markedly less stable than tocols, as only 26.0% compared to 57.9% of the initial flour content was preserved in the final product, accounting for 0.81 and 29.4 μg/g. In the dough preparation phase, a strong correlation (R = 0.693, p ≤ 0.001) was noted between the tocol content in the flour and the share of retained carotenoids in the dough, indicating a protective effect of tocols on carotenoids. In total, a similar overall retention of carotenoids, as in the leavened bun preparation (27.4%), was found, since the differences in shape of the final product turn out to cause increased carotenoid losses due to high baking temperatures in flatbread.
Artichokes are an important source of (poly)phenolic compounds, mainly caffeoylquinic acids, which consumption has been associated with health benefits. However, heat treatments have shown to affect the amounts of these bioactive food compounds. In the present study the influence of culinary techniques (boiling, griddling, and frying) on the total (poly)phenolic content of artichokes (Cynara Scolymus cv. Blanca de Tudela) was evaluated by LC-MS/MS. Additionally, the antioxidant capacity of cooked artichokes was evaluated by spectrophotometric methods. A total of 31 (poly)phenols were identified and quantified, being caffeoylquinic acids the most abundant compounds in raw artichokes accounting for more than 95% of total (poly)phenolic compounds. With the different culinary techniques, these compounds suffered degradation but also redistribution, probably due to isomerization and hydrolysis reactions. Frying and griddling showed the lowest content of (poly)phenolic compounds and antioxidant capacity suggesting thermal degradation. Boiling also provoked losses, which were mainly due to leaching of phenolic compounds into the water. However, it was the heat treatment that best preserved (poly)phenolic compounds in artichokes.
Chlorogenic acid (CGA) and its hydrolysis products, quinic and caffeic acids are considered as fine chemicals and have a high commercial value. Yerba mate stems, useless residue from yerba mate processing, contains significant amounts of CGA. The goal of this study was to determine the best conditions for maximizing the CGA aqueous extraction from residues of yerba mate processing and to fit the extraction kinetics to empirical models. The effect of single factors such as solid-liquid ratio, numbers of extraction, temperature and time were studied for two particle sizes, using one factor at a time method and Response Surface Methodology (RSM). The highest and almost instantaneous CGA extraction (1.051 ± 0.015 gCGA L⁻¹) was obtained using Ø particle size < 500 μm, at 65 °C, solid-liquid ratio of 1:20 and double stage extraction. For the largest particle size (2.5 × 5 mm > particle size >1 × 5 mm) the optimized conditions were 85 ± 5 °C and 25 ± 5 min, solid-liquid ratio of 1:20 and double stage extraction. For these last particles, the applied models of Pilosof et al. and Spiro and Jago showed a good agreement with the experimental and model calculated data. In the present study, it was possible to optimize the CGA extraction conditions for two particle sizes of yerba mate stems, in order to obtain extracts with high content of CGA that can be used in the production of others fine chemicals of interest in pharmaceutical industries.
Phenolics are a group of compounds derived from plants that have displayed potent biological activities and health-promoting effects. Fermentation is one of the most conventional but still prevalent bioprocessing methods in the food industry, with the potential to increase phenolic content and enhance its nutritive value. This review details the biotransformation of different classes of phenolics (hydroxycinnamic and hydroxybenzoic acids, flavonoids, tannins, stilbenoids, lignans, alkylresorcinols) by various microorganisms (lactic acid bacteria, yeast, filamentous fungi) throughout the fermentation process in plant-based foods. Several researchers have commenced the use of metabolic engineering, as in recombinant Saccharomyces cerevisiae yeast and Escherichia coli, to enhance the production of this transformation. The impact of phenolics on the metabolism of microorganisms and fermentation process, although complex, is reviewed for the first time. Moreover, this paper highlights the general effect of fermentation on the food's phenolic content, and its bioaccessibility, bioavailability and bioactivities including antioxidant capacity, anti-cancer, anti-diabetic, anti-inflammation, anti-obesity properties. Phenolics of different classes are converted into compounds that are often more bioactive than the parent compounds, and fermentation generally leads to a higher phenolic content and antioxidant activity in most studies. However, biotransformation of several phenolic classes is less studied due to its low concentration and apparent insignificance to the food system. Therefore, there is potential for application of metabolic engineering to further enhance the content of different phenolic classes and bioactivities in food.
This study investigated the effect of different fermentation times of dough made with whole wheat flour (WWF) added with yerba mate (YM), and the effect on rheological behavior, release of phenolic compounds, and antioxidant activity. Three treatments were evaluated: YM1 (2.5%; w/w), YM2 (4.5%; w/w), and the control sample (0%). The dough fermentative behavior showed an alteration in CO2 production, and dough volume was reduced in the presence of YM. The addition of YM, together with the fermentation process, increased the free, bound, and total phenolic content, as well as the antioxidant activity of the dough. The addition of 4.5% of YM showed higher content for free phenolic compounds on the dough after the fermentation. 26 phenolic compounds were identified in the dough samples. Rutin was the main compound in doughs with YM, comprising 79% and 61% of total phenolic content of YM2 and YM1, respectively. The phenolic profile demonstrated that fermentation times of 30 and 60 min released greater amounts of compounds in the three studied doughs, mostly caffeic acid, ferulic acid, chlorogenic acid, p-coumaric acid, isoquercetin, and rutin. These results suggest that the dough fermentation process may be a determining factor for the release of phenolic compounds.
With the growing interest in formulating nutritionally fortified baked products with added flavonoid-rich ingredients for health benefits, interactions between these ingredient classes have become critical, as they may affect product quality. In this study, the interaction behaviours that are caused by baking intervention between bamboo leaf flavonoids and gliadin were investigated. The flavonoids and gliadin were profiled via Fourier transform infrared spectroscopy (FTIR), nuclear magnetic resonance (NMR) and high-performance liquid chromatography coupled with diode array detection and electrospray ionization hybrid ion trap and time-of-flight mass spectrometry (HPLC-DAD-ESI/IT-TOF-MS/MSⁿ) to obtain molecular structural, qualitative and quantitative information. The content of flavonoids was reduced, the number of −OH groups of the flavonoids was decreased and the chemical shift of phenolic −OH was changed in the baking process. Gliadin exhibited structural changes upon sulfhydryl exposure and α-helix reduction in the baking process. Molecular docking was further performed to explore the binding mechanism in the complex of bamboo leaf flavonoids and gliadin, and the results demonstrated that hydrogen bonding and hydrophobic interaction were the main binding types. This study identified the mechanism of interaction between bamboo leaf flavonoids and gliadin in the baking process, thereby providing data support for the application of bamboo leaf flavonoids in the baking industry.
Yerba mate contains bioactive compounds, and is widely consumed as a decoction beverage in several Southern American countries. At present, the consumption of mate with added herbal blends and flavors, called "composed yerba mate", has increased; however, no studies on the antioxidant characteristics of these products have been published. In this sense, the main objective was to assess the antioxidant characteristics of "composed yerba mate" compared to "traditional yerba mate", in the form it is traditionally consumed. Total polyphenols content ranged from 15 to 45 mg/g GAE in all decoctions analyzed. Seventeen phenolic compounds were identified and quantified by HPLC-DAD-MS/MS, mainly belonging to the caffeoylquinic acids group. The antioxidant capacity was measured using in vitro assays, Ferric reducing ability of plasma (FRAP) and Trolox equivalent antioxidant capacity (TEAC), and with Saccharomyces cerevisiae as the in vivo model organism. All decoctions displayed antioxidant activity and were capable of rescuing yeast cells between 10.68 and 18.38% from oxidative stress. Multiple regression analysis showed a high correlation between phenolic composition and activity of samples, where different compounds indicate a significant contribution to the observed activity. Significant differences were found in the content, profile and antioxidant activity of polyphenols when "traditional yerba mate" and "composed yerba mate" were compared. In some cases, the antioxidant capacity was similar or higher in composed yerba mate; while the rest displayed lower biological activity. Based on these findings, it would be possible to assume that the addition of herb mixtures modifies the antioxidant and biological properties of mate. Supplementary information: The online version contains supplementary material available at (10.1007/s13197-020-04961-x).