Food Chemistry 282 (2019) 109–119

Contents lists available at ScienceDirect

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

Identification of new bioactive peptides from Kefir milk through T

proteopeptidomics: Bioprospection of antihypertensive molecules
⁎
Fernanda Gobbi Amorima, , Luciana Barbosa Coitinhob, Ananda Tissianel Diasc,
Andreia Gomes Freitas Friquesa,c, Brenna Lepaus Monteiroc, Lucas Cunha Dias de Rezended,
Thiago de Melo Costa Pereiraa, Bianca Prandi Campagnaroa, Edwin De Pauwe,
Elisardo Corral Vasqueza, Loïc Quintone
a
Pharmaceutical Sciences Graduate Program, Vila Velha University, Vila Velha, ES, Brazil
b
School of Pharmaceutical Sciences of Ribeirão Preto, University of São Paulo – USP, Ribeirão Preto, SP, Brazil
c
Laboratory of Translational Physiology, Federal University of Espirito Santo, Vitoria, ES, Brazil
d
Departament of Natural Science, Federal University of Espirito Santo, CEUNES, São Mateus, ES, Brazil
e
Laboratory of Mass Spectrometry, Department of Chemistry, University of Liège, Liège, Belgium

A R T I C LE I N FO A B S T R A C T

Keywords: Kefir, a probiotic beverage prepared from fermented milk, has been associated with antihypertensive activity.
Kefir However, the bioactive molecules responsible for this activity still remain unclear. Therefore, in this study we
Proteomics aim to evaluate the mechanisms of the antihypertensive effects of Kefir in the two-kidney one-clip hypertension
Antihypertension model, and to bioprospect for bioactive peptides identified by proteomic methodologies. Treatment with Kefir
Bioactive peptides
was performed via gavage, and resulted in a 37 mmHg reduction in systolic arterial pressure and 19% inhibition
ACE
of angiotensin converting enzyme (ACE) activity. For the proteopeptidomic study, the protein extract of Kefir
Inhibition
beverage and non-fermented bovine milk were analysed by MALDI-TOF mass spectrometry, and their tryptic
digestion products sequenced via Shotgun proteomics (Q-Exactive mass spectrometer). A list of 35 peptides with
potential hypertensive activity due to ACE inhibition were identified. These results demonstrate the benefits of
Kefir products, and may guide the design of new antihypertensive drugs.

1. Introduction sheep's or buffalo's milk (Prado et al., 2015).

Kefir grains are composed of microbial species, such as lactic acid
Kefir is a probiotic fermented milk originating in the Northern bacteria, acetic acid bacteria, yeasts, and other fungi, which are in-
Caucasus Mountains. It represents an ancient tradition in human nu- corporated in a matrix of viscous polysaccharides, called kefiran (Prado
trition, its consumption dating as far back as 2000 B.C. Its name ori- et al., 2015). The microorganisms that coexist in kefiran show a com-
ginates from the Slavic Keif, meaning “well-being”, owing to the ben- plex symbiotic relationship, and several species have been identified,
eficial effects associated with its consumption (Ebner et al., 2015; Rosa including Lactobacillus brevis, Lactobacillus helveticus, Lactobacillus kefir,
et al., 2017). Leuconostoc mesenteroides, Kluyveromyces lactis, Kluyveromyces marx-
In appearance, Kefir grains are small irregular masses with a gela- ianus and Pichia fermentans (Ferreira et al., 2010). Interestingly, the
tinous texture and yellowish colour (Friques et al., 2015). The bio- microbial composition varies according to the origin of the Kefir, the
chemical composition of the grains includes fat, mucopolysaccharides, substrate used in the fermentation process, and the culture maintenance
proteins (27% insoluble, 6% soluble and 5.6% free amino acids), vita- methods used (Prado et al., 2015; Yaman, Elmali, Karadagoglu, &
mins B and K, tryptophan, calcium, phosphorus and magnesium (Rosa Cetinkaya, 2006).
et al., 2017). Traditionally, Kefir can be produced using cow's, goat's, This symbiotic dairy product has attracted great interest in the

⁎
Corresponding author at: Programa de Ciências Farmacêuticas, Universidade de Vila Velha (UVV), Av. Comissário José Dantas de Melo, n 21 – Prédio de
Biopráticas – Laboratório 33. Boa Vista-Vila Velha ES, CEP 29102-920, Brazil.
E-mail addresses: [email protected] (F.G. Amorim), [email protected] (L.B. Coitinho), [email protected] (A.T. Dias),
[email protected] (A.G.F. Friques), [email protected] (B.L. Monteiro), [email protected] (L.C.D.d. Rezende),
[email protected] (T.d.M.C. Pereira), [email protected] (B.P. Campagnaro), [email protected] (E. De Pauw),
[email protected] (E.C. Vasquez), [email protected] (L. Quinton).

https://doi.org/10.1016/j.foodchem.2019.01.010
Received 10 September 2018; Received in revised form 21 December 2018; Accepted 2 January 2019
Available online 08 January 2019
0308-8146/ © 2019 Elsevier Ltd. All rights reserved.
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

scientific community, because of its benefits, including antibacterial After confirmation of hypertension in these animals (systolic
effects (Rodrigues, Caputo, Carvalho, Evangelista, & Schneedorf, 2005), pressure > 160 mmHg after seven days of surgery), they received Kefir
control of glucose and cholesterol plasma levels (Hadisaputro, beverage or a vehicle composed of non-fermented bovine milk with pH
Djokomoeljanto, Judiono, & Soesatyo, 2012; Santanna et al., 2017), and corrected to 4, which is similar to the pH of Kefir beverage. The
anti-hypertensive and anti-inflammatory activities (Friques et al., 2015; treatment was administered by gavage over 21 days, at doses of 0.3 mL/
Klippel et al., 2016), among other activities. 100 g. The animals were divided into three groups: Sham animals that
These beneficial effects are due the constituents present in this received vehicle; 2K1C animals that received vehicle; and 2K1C animals
symbiotic dairy and may represent an important source of a variety of that received Kefir beverage. The animals were supplied by the Central
bioactive compounds (Capriotti, Cavaliere, Piovesana, Samperi, & Vivarium of the Federal University of Espírito Santo (UFES). All the
Laganà, 2016; Piovesana et al., 2015; Zenezini Chiozzi et al., 2016) experiments were performed in accordance with the guidelines on
Therefore, characterization of these bioactive compounds in food is of ethical principles of animal research, established by the National
crucial importance. Some of these potentially beneficial compounds are Technical Biosafety Commission (CTNBio) and approved by the
small peptides derived from proteins, which are often cryptic and latent Committee for Ethics in the Use of Experimental Animals (CEUA-UFES,
until proteolytic release of the active peptides. The release can occur by protocol no. 040/2014). The animals were kept in a light-dark cycle of
enzymatic activity during gastrointestinal digestion or produced during 12 h, temperature of 20–25 °C, humidity of 70% and free access to
ripening and fermentation (Piovesana et al., 2015; Zenezini Chiozzi water and feed.
et al., 2016).
Although Kefir has many beneficial effects, the composition of the 2.2.2. Plethysmography
bioactive peptides formed during the fermentation of the milk remains For indirect measurement of systolic arterial pressure (SAP) by
unclear. Peptides derived from milk proteins are a rich source of plethysmography (IITC Life Science − 23,924 Victory Blvd, Woodland
functional ingredients in foods, and are at least partially responsible for Hills, CA), the rats were conditioned in a cylindrical acrylic tube, which
many of the aforementioned effects of Kefir (Ebner et al., 2015). was kept in a heated and ventilated environment. The animal’s tail was
Among the most recognized bioactive peptides from dairy products, fitted into a rubber cuff attached to a sphygmomanometer to inflate and
we highlight those with inhibitory activity towards the angiotensin deflate. Inside the cuff there is a pulse transducer (photo sensor) cou-
converting enzyme (ACE). ACE is involved in a cascade of metabolic pled to a computer. The sensor picked up the signals and recorded them
reactions that trigger increased blood pressure (de Oliveira et al., on the computer. During the cuff insufflation, there is a momentary loss
2018). Accordingly, the inhibition of ACE represents putative anti-hy- of pulse signals, which return as soon as the cuff is deflated. Therefore,
pertensive action, and has great therapeutic potential (de Oliveira et al., SAP was considered the first pulse signal of this process.
2018). Therefore, this study aimed to perform proteopeptidomic ana- The animals were previously adapted to their environment for three
lysis of kefir, before and after milk fermentation, in order to bioprospect days before any SAP measurement was taken, to prevent blood pressure
new bioactive peptides, especially antihypertensive peptides, in this changes due to the stress of the animal. SAP measurement was per-
complex drink. formed on day 7 after surgery, to confirm whether the animal had
developed hypertension, and on day 28, to analyse the effects of the
2. Material and methods treatment with Kefir on blood pressure. During the SAP measurements,
at least 5 records were taken for each animal.
2.1. Kefir preparation
2.2.3. Evaluation of angiotensin converting enzyme (ACE) activity in blood
The process of preparing synbiotic the kefir beverage and non-fer- samples
mented milk (control) for this study is described in Fig. 1. The kefir The effect of Kefir on ACE activity was measured by the fluorimetric
beverage was prepared by adding 4% (w/v) of kefir grains to pasteur- method described by Friedland and Silverstein (1976) adapted to mi-
ized whole bovine milk, and leaving at room temperature for 24 h. This croplate and with modifications such as using ZPhe-His-Leu as substrate
mixture was filtered through a plastic mesh, and the resulting product (Aragão et al., 2015; Friedland & Silverstein, 1976; Junod, 2012).
was refrigerated (average temperature of 10 °C) to permit micro- Plasma samples (3 μL) were incubated for 15 min at 37 °C with 40 μL of
organism growth for further 24 h. At the end of this process, a 10 mL buffer (5 mM Zphe-His-Leu, 0.4 M sodium borate buffer and 0.9 M NaCl
sample of fermented kefir milk was lyophilized and stored at −20 °C. pH 8.3) containing the ACE substrate. The reaction was stopped by the
As a control group, the same batch of bovine milk was treated with 4% addition of 190 μL of 0.34 M NaOH. The reaction resulted in a product,
(w/v) kefir grains and immediately lyophilized before any fermentation His-Leu, which was incubated with 17 μL of 2% o-phthaldialdehyde
could take place. A 10 mL sample of the non-fermented milk was also (OPA) (Sigma, U.S.A.) in methanol for 10 min. Then, 31 μL of 3 N HCl
lyophilized and kept refrigerated in similar conditions. was added to the solution in order to precipitate the protein-OPA
complex, which was removed by centrifugation for 10 min at 1,000 xg.
2.2. In vivo analysis The dipeptide was measured fluorimetrically (Synergy 2, Biotek,
U.S.A.). The fluorescence measurements were performed at 37 °C on a
2.2.1. Hypertensive animals and Kefir beverage treatment 96-well black reading plate (Black polystyrene, Corning, U.S.A.) with
With the aim of analysing the effects of Kefir beverage over the excitation filters of 365 nm and 495 nm emission, and analysed by the
Renin Angiotensin system and hypertension, the well-established model software Gen5. A calibration curve with ACE substrate was included on
of hypertensive animals was used, based on the two-kidney one-clip each plate. For the blank, 40 μL buffer (5 mM Zphe-His-Leu, 0.4 M so-
(2K1C) procedure. Male Wistar rats weighing 140–160 g were an- dium borate buffer and 0.9 M NaCl pH 8.3), 190 μL of 0.34 M NaOH,
esthetized with Ketamine (90 mg/kg, i.p.) and Xylazine (10 mg/kg, i.p.) 17 μL of OPA and ACE substrate were added. The assays were per-
and subjected to laparotomy to expose the left renal artery, where a formed in triplicate.
silver clip was placed. The clip was pre-calibrated with an internal
diameter of 0.2 mm, leading to a decrease in blood flow and a reduction 2.2.4. Statistical analyses
in renal perfusion pressure. This kidney stenosis was used to mimic The experimental data are presented as mean ± SE. The data were
secondary renovascular hypertension. The animals of the Sham group analysed using the software GraphPad Prism, version 6.0 for Windows
underwent the same surgical procedure, but without the placement of (GraphPad Software, La Jolla, CA, USA, 2012). Data from the ACE ac-
the clip. After the procedure, the incision was sutured, and the animals tivity and plethysmography were analysed using one-way analysis of
were kept in individual cages. variance (ANOVA) followed by the Tukey’s test. Values of p < 0.05

110
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

Fig. 1. Sample preparation of synbiotic Kefir. The samples selected to be studied herein are marked with a gray dotted box.

were considered statistically significant. mass spectra were processed and analysed using DataAnalysis 4.0
software (Bruker Daltonics, Bremen, Germany). To compare the mole-
2.3. Proteomics analysis cular masses (m/z) identified between the groups, we selected ion peaks
with relative intensity over 5% and ions with a higher resolution of the
2.3.1. MALDI-TOF/MS monoisotopic pattern (reflective mode).
Lyophilized samples of Kefir beverage and non-fermented bovine
milk were desalted in a 10 µL pipette tip containing C18 reversed-phase 2.3.2. Shotgun proteomics
media (ZipTip® C18 tips, Millipore) with elution in a solution of acet- Grain/milk and fermented milk samples (20 µg) were suspended in
onitrile/water/formic acid (49.8/50/0.2). Sample were spotted with 159 µL of 50 mM NH4HCO3 pH 7.8 and reduced with 3.2 µL of 500 mM
1 µL of 2,5-dihydroxybenzoic acid matrix (DHB; 10 mg/mL in 0.2% dithiothreitol for 40 min, at 56 °C, under shaking at 300 rpm. The
formic acid and 50% acetonitrile) for analysis on a MALDI-TOF samples were then alkylated with 6.4 µL of 500 mM iodoacetamide for
UltrafleXtreme mass spectrometer (Bruker Daltonics, Bremen, 30 min, at room temperature, in the dark, followed by a second re-
Germany) operated in reflected positive mode, equipped with duction with 3.6 µL of 500 mM dithiothreitol for a further 30 min, at
SmartBeam Laser. The analyser was previously calibrated with Peptide room temperature. Next, the grain/milk and fermented milk samples
Calibration Standard II in a mass range of 700–4000 m/z. For each were submitted to two trypsin digestions in 50 mM NH4HCO3 pH 7.8: in
sample, we accumulated at least 5000 records of mass spectra. The the first, the protein:trypsin mixture at a ratio of 1:50 was incubated

111
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

overnight, at 37 °C, under 300 rpm; in the second, the protein:trypsin the human angiotensin converting enzyme was retrieved from PDB
mixture at a ratio of 1:100 supplemented with 640 µL of 100% acet- under the code 1O86, and prepared for docking by removing ligands
onitrile was incubated for 3 h, at 37 °C, under 300 rpm. The reactions and solvents in the software Chimera. Molecular docking was per-
were stopped by adding 10% TFA to the reaction mixtures, and the formed in AutoDock Vina built in PyRx 0.8, search space was centred at
samples were dried in a speed vacuum. X:39.3 Y:34.5 Z:43.4 and the dimensions were set at X:39,9A Y:40.4A Z:
For the shotgun proteomics analyses, the samples were suspended in 40.9A. For each ligand peptide, the nine poses with the lowest energy
20 µL of 0.1% TFA for desalting on ZipTip™ pipette tips with C18 resin were retrieved and saved in .pdb format. LigPlot + 1.4.5 was used to
and eluted with acetonitrile/water/TFA (49.8/50/0.2 V/V) solution. prepare bidimensional schematic diagrams of the predicted hydrogen
The digested material was analysed in the Acquity UPLC® M-Class bonding and hydrophobic interactions between ligand and enzyme.
(Waters, Milford, MA, USA) coupled to a Q-Exactive™ Plus Hybrid
Quadrupole-Orbitrap™ Mass Spectrometer (Thermo Scientific, Bremen, 3. Results and discussion
Germany). Peptides were eluted using a gradient of 2–90% of solution B
in 150 min (A: water/0.1% formic acid; B: acetonitrile), at a flow rate of 3.1. Kefir reduces SAP in hypertensive mice and shows ACE inhibitory
0.6 mL/min, and data were acquired in positive-ion mode. activity
Protein identification by automated de novo sequencing was per-
formed using the software Peaks Studio 8.5, with database created by The beneficial effects of Kefir consumption are well-documented in
the search of the keywords “Milk AND Bovine” in the UniProt re- the literature (Friques et al., 2015; Hsu et al., 2018; Klippel et al.,
pository, downloaded in February 2018 (25,029 sequences). 2016). Among the bioactive peptides derived from dairy, those with
Carbamidomethylation was set as fixed modification, while amidation blood pressure-lowering effects have attracted special attention, due to
(C-terminus) and oxidation (M) were set as variable modifications, with the prevalence and importance of hypertension in the population
maximum missed cleavages at 3. Parent mass and fragment mass error (López-Fandiño, Otte, & Van Camp, 2006). They are mainly classified as
tolerance were set at 5 ppm and 0.2 Da, respectively. A false discovery ACE inhibitors, which are responsible for the effects of probiotic bev-
rate (FDR) of 1% and unique peptide ≥ 2 were used to filter out in- erage consumption (Ebner et al., 2015; Quirós, Hernández-Ledesma,
accurate proteins for the PEAKS search and “De novo only” analysis. Ramos, Amigo, & Recio, 2005).
A − 10lgP > 20 indicated that the detected proteins were relatively The anti-hypertensive effects of Kefir were demonstrated in a study
high in confidence, as it targeted very few decoy matches above that in which treatment with Kefir over 60 days reduced mean arterial
threshold. These data can be found in jPOSTrepo repository (https:// pressure (MAP) in Spontaneously Hypertensive Rats (SHR) at 24 mmHg
repository.jpostdb.org/) under the accession numbers of JPST000532 (Friques et al., 2015). Another study showed that the treatment with
and PXD012074 for ProteomeXchange. the soluble non-bacterial kefir fraction for 8 weeks reduced MAP at
42 mmHg (Brasil et al., 2018). However, this is the first time Kefir
2.4. In silico analysis consumption has been evaluated in the 2K1C renovascular hyperten-
sion model.
2.4.1. Prediction of bioactive molecules Renovascular hypertension by the 2K1C model is the result of the
After performing peptide sequencing with mass spectrometry, in release of renin by the stenotic kidney, and the consequent production
silico analyses were carried out with Milk Bioactive Peptide Database of Angiotensin II (Ang II) due to the higher activity of ACE. 2K1C leads
(MBPDB; http://mbpdb.nws.oregonstate.edu). Thus, the peptides were to a gradual and chronic increase in blood pressure, which plateaus
classified according to the milk protein from which they originated, after 2 weeks (Lerman, Chade, Sica, & Napoli, 2005).
their putative biological function, and references. Local ACE in the kidneys converts Ang I to Ang II, an important
vasoconstrictor. In addition, Ang II causes a reduction in urinary ex-
2.4.2. Prediction of anti-hypertensive peptides and their physic-chemical cretion of water and sodium, which plays an important role in long-
features term stabilization of hypertension. In addition, ACE acts in the de-
The peptides identified as potential antihypertensive agents by the gradation of bradykinin, a vasodilator (Dzau, Burt, & Pratt, 1988).
MBPDB database were selected and submitted to various computational Therefore, inhibition of ACE results in a lowering of blood pressure and
analyses, in order to predict their physicochemical properties. is an important therapeutic target in the treatment of hypertension. In
Molecular mass, isoelectric point (pI), water solubility and net charge at this study, the renovascular 2K1C model of hypertension was used to
neutral pH were calculated using Innovagen (http://www.innovagen. mimic renal artery constriction in humans.
com/) and Protparam Expasy (https://web.expasy.org/cgi-bin/ In vitro ACE-inhibitory activity and in vivo antihypertensive effects
protparam/protparam). The amino acid sequence of each peptide was have been reported for different fermented milks cultivated in the la-
submitted to the Peptide Ranker server (http://bioware.ucd.ie/ boratory with different strains of lactic acid bacteria (Gobbetti,
~compass/biowareweb/Server_pages/peptideranker.php), which uses Ferranti, Smacchi, Goffredi, & Addeo, 2000; Quirós et al., 2005). Sur-
neural networks to predict the likelihood of a peptide being bioactive. prisingly, before this study, there were no reports evaluating the ACE
For each peptide, scores from 0 to 1 were assigned, with 1 being the inhibition effects of Kefir using the 2K1C model. In our study, we ob-
most likely to be bioactive. The toxicity of the peptides was predicted served that the 2K1C model lead to the development of hypertension in
using ToxinPred (http://crdd.osdd.net/raghava/toxinpred/), which mice, in which the animals presented systolic arterial pressure (SAP) of
uses a dataset of 1805 toxic peptides to train a machine learning al- 228.5 mmHg (SE ± 6.8) compared to the Sham animals
gorithm. (124.6 mmHg ± 6.2) 28 days after clipping. The treatment with Kefir
reduced the SAP in 36.7 mmHg (p < 0.001) (Fig. 2A). In addition,
2.4.3. Molecular docking Kefir inhibited ACE activity in 19.3% compared to the 2K1C group
The MBPDB database was used to predict the ACE-inhibition capa- (p < 0.05) (Fig. 2B), which confirms the presence of ACE inhibitors in
city of the peptides, based on the similarity of previously published the Kefir, and supports the thesis that they are involved in the anti-
sequences. A blind molecular docking strategy was applied to access the hypertensive effects of this dairy beverage.
lowest energy poses of each compound when bound to the ACE enzyme.
For the docking analyses, the chemical structure of each peptide (li- 3.2. Proteomics studies of Kefir reveal bioactive peptides
gands) was prepared in MDL SD file format from the amino acid se-
quence using the “expand label” tool in ChemDraw 16.0, and energy Milk from different animals is a valuable source of proteins and
minimized using Open Babel built in PyRX 0.8. The crystal structure of nutrients for human consumption. This is due the dynamic range of

112
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

Fig. 2. In vivo effects of the 14 days treatment with Kefir fermented milk in renovascular hypertensive mices. (A) SAP of Sham and 2K1C animals after receiving Kefir
for 14 days; **p < 0.001. (B) Inhibitory activity of Kefir over ACE in blood samples of Sham, 2K1C and 2K1C Kefir animals; *p < 0.05. Statistical analyses were
performed with one-way ANOVA followed Turkey post-hoc.

proteoforms among animal species, which provide different nutritional in biological discoveries. In this study, shotgun proteomics was per-
profiles (Soggiu, Roncada, & Piras, 2018). formed on a Q-Exactive Orbitrap mass spectrometer in order to evaluate
Milk peptidome is considered a valuable tool to identify biologically the total milk proteins content which was digested by trypsin in vitro.
active peptides. The growing application of peptidomic techniques, Mass spectrometry analyses resulted in similar MS and MS/MS
such as high-resolution mass spectrometry, has led to an increase in our identification between both groups: 17,289 and 17,839 MS and 65,664
knowledge of the health benefits of dairy products (Giacometti & and 64,025 MS/MS scans for before and after fermentation, respec-
Buretić-Tomljanović, 2017). Therefore, in order to evaluate how fer- tively. After applying the filtering parameters, the samples were nar-
mentation by Kefir grains affects milk composition, we performed a rowed to 9110 (before fermentation) and 7146 sequenced peptide in
proteomic study integrating two mass spectrometry techniques: MALDI- the mass spectras. These sequences were compared to a database, as
TOF and Q-Exactive LC-MS/MS. described in the Methods section, aiming the annotation proteins and
In the MALDI-TOF analysis, it was observed that new ions appeared peptides. The analysis of the non-fermented milk retrieved 965 Peptide
after the fermentation, probably due the endogenous proteolysis of milk Spectrum Matches compared to 4592 for the fermented Kefir beverage,
protein with high molecular masses (Fig. 3A). Most of the signals were representing a 4.75-fold increase. These peptides enabled the identifi-
detected 800–1900 m/z for the non-fermented bovine milk and cation of 18 proteins in the non-fermented bovine group and 28 pro-
700–2500 m/z for Kefir milk. After the fermentation, it is possible to teins in the Kefir beverage.
note some signals out of this mass range. Similar results were found by In a comparative analysis, it was observed that > 50% of the pep-
Ebner et al. (2015) which suggests that this signal pattern represents tides identified in the fermented Kefir milk originated from serum al-
peptides with lengths of approximately 7–20 amino acid residues bumin, while 17% of the peptides originated from caseins, 13% from β-
(Ebner et al., 2015). lactoglobulin, 10% from lactotransferrin, and 5% from α-lactoalbumin
In regard to MALDI-TOF mass spectra, it is noted that peptides at m/ (Fig. 4A). Both groups shared 110 peptides, while 49 peptides were
z 1718.0 and 1881.1 have already been reported in a previous pro- exclusively identified before fermentation compared to 624 peptides
teomic study of Kefir by Ebner et al. (2015). The authors attributed after fermentation (Fig. 4B). The complete list of peptides identified in
these signals to the peptide sequences of QEPVLGPVRGPFPIIV and this analysis is given in the supplementary material (S2).
YQEPVLGPVRGPFPIIV, fragments from the C-terminal region of β- Usually, caseins (α-S1-casein, α-S2-casein, β-casein and κ-casein)
casein f194–209 and f193–209, respectively. In our study, the signal represent the major fractions of milk proteins (80%), followed by the
1881.1 m/z appears only after fermentation, which indicates that this whey proteins (α-lactalbumin, β-lactoglobulin, lactotransferrin and
peptide could be indeed a product of β-casein produced by Kefir pro- serum albumin), which represents a total of 20% of the milk beverages
biotic. (Ebner et al., 2015; Rai, Sanjukta, & Jeyaram, 2017). Although it was
In general, fermentation increased the diversity of detected ions, as expected that the microbial content of Kefir would act randomly in all
can be seen in Fig. 3. Even if we take into account the suppressing milk proteins, this did not, in fact, happen. Whey proteins are more
effects that could occur in MALDI-TOF MS, an initial observation is that resistant to the hydrolytic activity of proteases when compared to
a total of 183 ions were exclusively detected after fermentation, com- caseins, which is explained by the globular structure of the former
pared to 51 ions before fermentation. Both groups share 43 ions, and (Swaisgood, 1993). The fact that we induced in vitro digestion with
they are mostly distributed in the mass range of > 1200–2200 (Fig. 3B trypsin may explain why we detected several peptides derived from
and C). The complete list of the detected ions is shown in the serum albumin. The sample processing methodology was selected to
Supplementary material (S1). obtain an extensive overview of kefir products. Thus, the in vitro di-
In order to perform an in-depth analysis of how Kefir fermentation gestion enabled the detection of high molecular weight proteins. Other
affects the protein content of the beverages, we also applied shotgun Kefir proteomic studies did not process their samples as we did, and
proteomics in this study. Shotgun proteomics allows the analysis of a therefore, excluded the high molecular protein present in Kefir milk
mixture of peptides from complex samples that have undergone in vitro (Ebner et al., 2015; Quirós et al., 2005). A complete list with the parent
tryptic digestion. Therefore, complex mixtures of peptides are separated protein of each peptide is provided in supplementary material (S3).
by liquid chromatography and analysed in a mass spectrometer coupled
to liquid chromatography (LC-MS/MS) in an online system (Zhang,
Fonslow, Shan, Baek, & Yates, 2013). The use of shotgun proteomics to
carry out a wide range of research experiments has promoted advances

113
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

Fig. 3. Distribution of ions between the groups after MALDI-TOF analysis. (A) MALDI-TOF mass spectra before and after milk fermentation by Kefir. (B) Number of
detected ions according to the mass range before and after fermentation. (C) Venn diagram of ions identified in MALDI-TOF analysis before and after the Kefir
fermentation.

3.3. Bioprospection of anti-hypertensive peptides from Kefir and their non-fermented bovine milk. The peptide sequences selected in this
physicochemical features process are listed in the supplementary material (S4). The main source
of peptides are caseins, which was to be expected (Yamamoto, Maeno,
Aiming to explore new peptides that appeared exclusively after & Takano, 1999).
fermentation, we selected a range of peptides that were sequenced by Therefore, those peptides that showed reports on the MBPDB da-
shotgun proteomics/Peaks Studio software and for which the m/z ion tabase for putative anti-hypertensive peptides, and those with no re-
signal was detected in MALDI-TOF. We considered an error of 0.09 Da cords (ND and classified as milk proteins) were selected for in silico and
for the data matching. By comparing the results from the shotgun docking analysis. The physicochemical features were predicted for a
proteomics with those from MALDI-TOF, we were able to identify total of 35 peptides, and the most favourable interactions with the ACE
peptides that are potentially present in the Kefir beverage and not in the enzyme were evaluated (Table 1). None of the 35 peptides were

114
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

identified 14 peptides with known anti-hypertensive activity from milk

proteins fermented with Lactobacillus helveticus NCC 2765 (Robert,
Razaname, Mutter, & Juillerat, 2004). Milk fermented with L. helveticus
CP790 showed anti-hypertensive activity in SHR rats with 5 mL/kg of
body weight (15 mg of peptide/kg), which suggests that peptides lib-
erated from the hydrolysis of casein by the proteinase in the culture
medium have antihypertensive effect in SHR rats (Yamamoto, Akino, &
Takano, 1994).
Milk is a complex mixture in which the bioactive peptides can be
released by the action of various unspecific and specific endogenous
proteases. In addition, the commercial milk used for the kefir produc-
tion is subjected to a previous food-processing for its commercializa-
tion, such as pasteurization. These processing can increase the com-
plexity of the peptidomic profile (Piovesana et al., 2015; Zenezini
Chiozzi et al., 2016). Some of the ACE inhibiting peptides can also be
released in the process of enzymatic hydrolysis in vivo (in the gastro-
intestinal tract) by the action of pepsin, trypsin and chymotrypsin (Rai
et al., 2017). The ACE-inhibitory activity after simulated digestion was
similar to, or slightly lower than for non-hydrolysed peptides, except for
peptide β–casein (DKIHPF), which exhibited 8 times greater activity
after hydrolysis (Quirós et al., 2005).
ACE inhibitor peptides usually contain 2–12 amino acids with two
active sites of ACE: in the N- and C-domain. The C-domain plays a major
role in blood pressure regulation and appears to be a promising phar-
Fig. 4. Distribution of protein classes in before and after fermentation de-
macological target for reducing blood pressure (Lunow, Kaiser,
termined by shotgun-proteomics. (A) Relative distribution of peptides origi-
nated from Milk proteins after fermentation. (B) Venn diagram with the pep-
Rückriemen, Pohl, & Henle, 2015). It has been suggested that hydro-
tides identified by mass spectrometry before and after the Kefir fermentation. phobic amino acids, such as Try, Phe, Trp, Ala, Ile, Val, and Met, or
positively charged amino acids such as Arg and Lys, as well as Pro at the
C terminal position of the peptides, may be related to a better affinity
considered toxic, according to the software ToxinPred. All the 35 mass
with ACE (Rai et al., 2017).
spectra from the antihypertensive peptides is represented in supple-
Our docking studies suggests that the peptides with most favourable
mentary material (S5).
ACE interaction (most negative values for Kcal/mol), around −10
It is important to emphasise that the search for those peptides in the
Kcal/mol, had hydrophobic amino acids or positively charged amino
MBPDB database also returns similar peptides, and only three of the 35
acids, with exceptions of LVYPFPGPIHNSLPQ, YQEPVLGPVRGPFPII
peptides shown in Table 1 were exactly the same as those previously
and YPFPGPIPN. Although these three peptides do not possess those
published. The other retrieved sequences showed some modifications in
particular amino acid residues in their structures, their similar peptides
the amino acid sequence, and could be seen as similar peptides (un-
have already studied, and their ACE inhibitory capacity demonstrated
derlined in the “Literature Sequence” column). In addition, none of the
(Pihlanto et al., 2010; Saito et al., 2000; Yamamoto et al., 1994).
peptides reported in Table 1 have previously been isolated from Kefir
Computational approaches, such as docking, allow the identification of
beverage. Quiros et al. (2015), studied Kefir from goat’s milk and
intermolecular potential interactions between ACE and peptides, as
identified and tested in vitro two ACE inhibitor peptides (PYVRYL and
well as measurement of the distances and angles in atoms involved. In
LVYPFTGPIPN) that differ from those of our study (Quirós et al., 2005).
addition, molecular modelling methodologies are strategic tools to
In a proteomic study of Kefir from bovine milk, 12 putative ACE in-
guide researchers in the de novo design of new peptides with high af-
hibitors peptides were described, one of which was similar to our study
finity by the catalytic cleft of this enzyme, and their subsequent cus-
(LLYQEPVLGPVRGPFPIIV) (Ebner et al., 2015). However, the Kefir
tomized production (de Oliveira et al., 2018).
used in this study was obtained from the region of Espirito Santo
We also predicted in silico the products of enzymatic hydrolysis from
(Brazil), while that used in Ebner’s study originated from Olztyn (Po-
gastric digestion, considering the cleavage site for trypsin, chymo-
land). The microbial composition may vary according to the origin of
trypsin and pepsin (Table 1). Therefore, this analysis can help guide
the Kefir, which may explain the differences in peptide composition
bench-top experiments to identify the most bioactive peptide frag-
found in our study compared to those of Quiros and Ebner. Ad-
ments. However, in silico proteolysis does not guarantee that computer-
ditionally, the substrate used in the fermentation process, and the cul-
generated peptides can be experimentally produced, as the proteins
ture maintenance methods, may have differed among different groups
may not undergo complete protease hydrolysis {Udenigwe, 2014,
(Prado et al., 2015).
Bioinformatics approaches‘, prospects and challenges of food bioactive
The ACE inhibitory activity of different types of fermented dairy
peptide research}(Udenigwe, 2014). Also, theoretical enzyme specifi-
products have been studied, including some varieties of cheese (Ong,
cities are not always as described, and actual proteolytic process of the
Henriksson, & Shah, 2007; Saito, Nakamura, Kitazawa, Kawai, & Itoh,
enzymes is more complicated than in silico proteolysis (Lin et al., 2018).
2000), fermented milk and yogurt from different sources (Moslehishad
In our study, only the peptide AVPYPQR did not undergo hydrolysis,
et al., 2013; Papadimitriou et al., 2007), and Kefir milk (Brasil et al.,
and it is important to note that in most of the studies that have been
2018; Quirós et al., 2005). Therefore, fermented milk products are
evaluated the in vivo effects of bioactive peptides, they were orally
natural sources of ACE inhibitor peptides, which are produced during
administered.
the hydrolysis of milk protein by microbial proteases presented in the
A bidimensional schematic model showing the predicted interac-
starter culture.
tions between peptide AVPYPQR and ACE with a relatively high affinity
In a recent study, ACE inhibitory peptides from β-casein were iso-
(-10.1 Kcal/mol) is given in Fig. 5. According to this model, there are
lated by reversed phase HPLC from Lactobacillus jensenii fermented
two points of coordination with the zinc atom in the active site of ACE:
milk, and identified by spectrometry analysis and amino acid sequen-
the carboxyl group from the C-terminal arginine residue, and the amide
cing (Pihlanto, Virtanen, & Korhonen, 2010). Robert et al. (2000)
forming the peptide bond between glutamine and C-terminal arginine.

115
 Table 1
Physicochemical features predicted and the most favourable interaction with ACE enzyme for each AH peptide.
Peptides MM (Da) Peptide pI Net charge Water Sequence literature References IC50 Docking Digestion simulation (Trypsin,
F.G. Amorim et al.

Ranker at pH7 solubility (Kcal/mol) Chymotrypsin, Pepsin)

AVPYPQR 830.47 0.57 9.1 1 Good Same (Anne Pihlanto-Leppälä, 274 µM −10.1 non
Rokka, & Korhonen, 1998)
CGLVPVLAENR 1170.64 0.4 6.32 −0.1 Poor −8.6 CG↓L↓VPVL↓AENR
DFGHIQYVAAYR 1440.73 0.3 7.09 0.1 Poor −10.1 D↓F↓GH↓IQY↓VAA↓Y↓R
DMPIQAFLLY 1210.63 0.71 3.8 −1 Poor −9.1 DMPIQA↓F↓L↓L↓Y
EINQFYQK 1069.54 0.21 6.35 0 Good −8.1 EINQ↓F↓Y↓QK
EMPFPKYPVEPF 1480.74 0.7 4.54 −1 Good EMPFPK (Perpetuo, Juliano, & 423 µg/mL −10.1 EMPFPK↓Y↓PVEP↓F
Lebrun, 2003)
FLLYQEPVLGPVRGPFPIIV 2254.31 0.8 6.35 0 Poor LLYQQPVLGPVRGPFPIIV and (Yamamoto et al., 1994) 21 µg/mL – −10.1 F↓L↓L↓Y↓QEPVL↓GPVR↓GPFPIIV
YQQPVLGPVRGPFPIIV 101 µg/mL
FVAPFPEVFG 1109.58 0.83 4 −1 Poor Same (Robert et al., 2004) 650 µM −9.8 F↓VAPFPEV↓F↓G↓
GLFQINNK 933.53 0.32 9.11 1 Poor −7.8 G↓L↓F↓QINNK
GTWYSLAMAASDISLLDAQSAPLR 2538.29 0.47 4.21 −1 Poor −9.5 G↓T↓W↓Y↓S↓L↓AMAASDIS↓L↓L↓
DAQSAP↓L↓R
IHPFAQTQSLVYPFPGP 1898.99 0.44 7.09 0.1 Poor −10.6 IHP↓F↓AQTQS↓L↓VY↓PFPGP
IHPFAQTQSLVYPFPGPIPNSLPQN 2763.45 0.14 7.09 0.1 Poor −8.9 IHP↓F↓AQTQS↓L↓VY↓
PFPGPIPNSL↓PQN
IPPLTQTPVVVPPFLQPEVMGVSK 2574.46 0.03 6.35 0 Poor LPQNIPPLTQTPVVVPPFLQPEVMGVSK (Yamamoto et al., 1994) 144 µg/mL −9.7 IPPL↓TQTPVVVPPF↓L↓
QPEVM↓GVSK
KLDQWLCEK 1162.61 0.47 6.38 −0.1 Good −7.8 K↓L↓DQ↓W↓L↓CEK
KVGINYWLAHK 1329.77 0.49 9.72 2.1 Poor VGINYWLAHK (A. Pihlanto-Leppälä, 327 µM −9.7 K↓VGIN↓Y↓W↓L↓AH↓K
Koskinen, Piilola,
Tupasela, & Korhonen,

116
2000)
LACQCLVR 905.49 0.55 8.39 0.9 Poor −8.3 L↓ACQC↓L↓VR
LLDDDLTDDIMCVK 1608.76 0.27 3.66 −4 Good −7.9 L↓L↓DDD↓L↓TDDIM↓CVK
LLYQEPVLGPVRGPFPIIV 2107.24 0.72 6.35 0 Poor LLYQQPVLGPVRGPFPIIV and (Yamamoto et al., 1994) 21 µg/mL – −10 L↓L↓Y↓QEPVL↓GPVR↓GPFPIIV
YQQPVLGPVRGPFPIIV 101 µg/mL
LVYPFPGPIHNSLPQ 1678.91 0.54 7.09 0.1 Poor LVYPFPGPIHNSLPQN (Pihlanto et al., 2010) 71 µM −12 L↓VY↓PFPGPIH↓NSL↓PQ
LVYPFPGPIPN 1213.68 0.65 5.88 0 Poor LVYPFPGPIH and YPFPGPIPN (Pihlanto et al., 2010; 89 µM – −9.7 L↓VY↓PFPGPIPN
Saito et al., 2000) 14.8 µM
LYQEPVLGPVRGPFPII 1895.09 0.71 6.35 0 Poor LLYQQPVLGPVRGPFPIIV and (Yamamoto et al., 1994) 21 µg/mL – −10.2 L↓Y↓QEPVL↓GPVR↓GPFPII
LYQEPVLGPVRGPFPIIV 1994.16 0.73 6.35 0 Poor YQQPVLGPVRGPFPIIV 101 µg/mL −10.6 L↓Y↓QEPVL↓GPVR↓GPFPIIV
NLHLPLPLLQ 1157.72 0.54 7.1 0.1 Poor NLHLPLPLL (Robert et al., 2004) 15 µM −9.6 N↓L↓H↓L↓PLP↓L↓L↓Q
NLHLPLPLLQS 1244.75 0.38 7.1 0.1 Poor −8.7 N↓L↓H↓L↓PLP↓L↓L↓Q
PFPEVFGK 920.50 0.72 6.35 0 Good FFVAPFPEVFGK and FPEVFGK (Maruyama, Mitachi, 77 µM and −9 PFPEV↓F↓GK
Tanaka, Tomizuka, & 140 µM
Suzuki, 1987)
QFLPYPYYAKPA 1457.76 0.78 8.76 1 Poor −10.5 Q↓F↓L↓PYP↓Y↓Y↓AKPA
RFFVAPFPEVFGK 1541.85 0.75 9.1 1 Poor FFVAPFPEVFGK (Maruyama et al., 1987) 77 µM −10.1 R↓F↓F↓VAPFPEV↓F↓GK
SLVYPFPGPIH 1226.67 0.76 7.09 0.1 Poor LVYPFPGPIH and LVYPFPGPIHNSLPQN (Pihlanto et al., 2010) 89 µM – −9.7 S↓L↓VY↓PFPGPIH
SLVYPFPGPIHNSLPQ 1765.94 0.67 7.09 0.1 Poor 71 µM −9.8 S↓L↓VY↓PFPGPIH↓NSL↓PQ
SQSLTLTDVENLHLPLP 1877.01 0.29 4.36 −1.9 Poor −9.6 SQS↓L↓T↓L↓TDVEN↓L↓H↓L↓PLP
TDLLNVCMDAK 1222.60 0.15 4.21 −1.1 Good −8.2 T↓D↓L↓LNVCM↓DAK
VAPFPEVFGK 1090.61 0.77 6.35 0 Poor FVAPFPEVFGKEKVNELSKDIGS and (Robert et al., 2004) ND and −9.8 VAPFPEV↓F↓GK
FVAPFPEVFG 650 µM
WCTISQPEWFK 1424.68 0.72 6.32 −0.1 Poor −9.1 W↓CTISQPEW↓F↓K
YPFPGPIPN 1001.52 0.82 5.88 0 Poor Same (Saito et al., 2000) 14.8 µM −10.7 Y↓PFPGPIPN
YPVEPFTESQSL 1396.68 0.19 3.8 −2 Poor KYPVQPFTESQSLTL (Yamamoto et al., 1994) 93 µg/mL −9.2 Y↓PVEP↓F↓TESQS↓L

pI: isoelectric point; IC50: half maximal inhibitory concentration over ACE activity in vitro.
Food Chemistry 282 (2019) 109–119
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

Fig. 5. Bidimensional schematic model of the predicted interactions between peptide AVPYPQR and ACE. The chemical structure of the peptide and hydrogen-
bonding amino acids of ACE are represented in different colours (purple and dark yellow respectively), with coloured spheres representing different atoms (black for
carbon, red for oxygen and blue for nitrogen). The amino acid residues predicted to be involved in hydrophobic contacts are shown in a stylized red semicircle. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

It is interesting to note that enalapril and lisinopril are two widely gastrointestinal digestion by pepsin, trypsin and chymotrypsin
applied ACE inhibitors that also contain a carboxylate group co- (Goodman, 2010). Qureshi, Vegarud, Abrahamsen, and Skeie (2013)
ordinating zinc (Wei, Clauser, Alhenc-Gelas, & Corvol, 1992). This showed that in vitro ACE activity inhibition of a peptide does not always
model also shows hydrogen bonding or hydrophobic interactions be- imply that the peptide will exhibit in vivo antihypertensive effects
tween AVPYPQR peptide and other important amino acid residues at (Qureshi et al., 2013). Thus, it is very difficult to establish a direct re-
the ACE binding sites. These interactions include hydrogen bonding lationship between in vitro and in vivo inhibition of ACE activity of a
between a peptide bond in AVPYPQR and the imidazolic group in peptide (Rai et al., 2017). ACE inhibitor peptides can be classified into
amino acid residue His-353; and between a primary amide in AVPYPQR three groups according to Fujita, Yokoyama, and Yoshikawa (2000): a)
with another amide at Gln-281. In previous crystallographic studies, true inhibitors (Type I), in which the IC50 of the peptide are not af-
similar interactions involving the same amino acid residues have been fected by gastrointestinal proteases or preincubation with ACE; (b)
shown to be important for the interaction of lisinopril and enalapril to Substrate type (Type II): represented by ACE inhibitors peptides that
ACE. Additionally, hydrophobic interactions with Ala-354, Tyr-523, are converted to low activity peptides on digestion by gastrointestinal
Ser-355, Val-518 are also known to be important for enalapril and li- proteases; and (c) pro-drug (Type III), which is peptides that are con-
sinopril binding, and were also present in the AVPYPQR peptide verted to potential ACE inhibitors by the action of gastrointestinal
(Natesh, Schwager, Evans, Sturrock, & Acharya, 2004; Natesh, proteases (Fujita et al., 2000; Rai et al., 2017).
Schwager, Sturrock, & Acharya, 2003).
Some bioactive peptides seem to be resistant to gastrointestinal di-
gestion, so that they reach the cardiovascular system in active form. 4. Conclusion
These peptides are inactive after fermentation, and are activated during
In this work, we demonstrate that bovine milk fermented by the

117
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

probiotic Kefir has anti-hypertensive effects, and is capable of inhibiting 100–104.

ACE activity in vivo. The novelty of the present study is that we identify Hsu, Y. J., Huang, W. C., Lin, J. S., Chen, Y. M., Ho, S. T., Huang, C. C., & Tung, Y. T.
(2018). Kefir Supplementation Modifies Gut Microbiota Composition, Reduces
some bioactive peptides through two different proteomic approaches, Physical Fatigue, and Improves Exercise Performance in Mice. Nutrients, 10(7).
and characterize those with potential anti-hypertensive actions, ac- Junod, A. (2012). Lung Metabolism: Proteolysis and Antioproteolysis Biochemical
cording to their physicochemical characteristics and molecular docking Pharmacology Handling of Bioactive Substances. Elsevier.
Klippel, B. F., Duemke, L. B., Leal, M. A., Friques, A. G., Dantas, E. M., Dalvi, R. F., ...
in the ACE. These findings on bioactive peptides in kefir may open new Vasquez, E. C. (2016). Effects of kefir on the cardiac autonomic tones and baroreflex
opportunities to the dairy industry for their expansion and promotion of sensitivity in spontaneously hypertensive rats. Frontiers in Physiology, 7, 211.
milk products with potential health benefits. Furthermore, the work- Lerman, L. O., Chade, A. R., Sica, V., & Napoli, C. (2005). Animal models of hypertension:
An overview. Journal of Laboratory and Clinical Medicine, 146(3), 160–173.
flow used for the discovery of novel milk-derived bioactive molecules Lin, K., Zhang, L.-W., Han, X., Xin, L., Meng, Z.-X., Gong, P.-M., & Cheng, D.-Y. (2018).
has significantly evolved over the years, and could become crucial for Yak milk casein as potential precursor of angiotensin I-converting enzyme inhibitory
the drug development. These results may guide future research in the peptides based on in silico proteolysis. Food Chemistry, 254, 340–347.
Lunow, D., Kaiser, S., Rückriemen, J., Pohl, C., & Henle, T. (2015). Tryptophan-con-
design of new drugs to treat hypertension.
taining dipeptides are C-domain selective inhibitors of angiotensin converting en-
zyme. Food Chemistry, 166, 596–602.
Acknowledgments López-Fandiño, R., Otte, J., & Van Camp, J. (2006). Physiological, chemical and tech-
nological aspects of milk-protein-derived peptides with antihypertensive and ACE-
inhibitory activity. International Dairy Journal, 16(11), 17.
This study was supported by the National Council for Scientific and Maruyama, S., Mitachi, H., Tanaka, H., Tomizuka, N., & Suzuki, H. (1987). Studies on the
Technological Development (CNPq, scholarship to FGA n. 150037/ active site and antihypertensive activity of angiotensin I-converting enzyme in-
2018-0; grant to BPC no. 445736/2014-3 and grant to ECV no. 303001/ hibitors derived from casein. Agricultural and Biological Chemistry, 51(6), 5.
Moslehishad, M., Ehsani, M. R., Salami, M., Mirdamadi, S., Ezzatpanah, H., Naslaji, A. N.,
2015-1), Espirito Santo Research Foundation (grant to LVDR – EDITAL & Moosavi-Movahedi, A. A. (2013). The comparative assessment of ACE-inhibitory
FAPES PPP 22/2018; grant to ECV – EDITAL CNPq/FAPES N.24/2018 – and antioxidant activities of peptide fractions obtained from fermented camel and
PRONEX 569/2018; scholarship to BPC – EDITAL Pesquisador Capixaba bovine milk by Lactobacillus rhamnosus PTCC 1637. International Dairy Journal, 29(2),
82–87.
18/2018 – 552/2018 and grant to BPC – EDITAL UNIVERSAL 21/2018) Natesh, R., Schwager, S. L., Evans, H. R., Sturrock, E. D., & Acharya, K. R. (2004).
and the State of São Paulo Research Foundation (FAPESP, scholarship Structural details on the binding of antihypertensive drugs captopril and enalaprilat
to LBC, no. 2014/01497-9 and no. 2016/12516-0). to human testicular angiotensin I-converting enzyme. Biochemistry, 43(27),
8718–8724.
Natesh, R., Schwager, S. L., Sturrock, E. D., & Acharya, K. R. (2003). Crystal structure of
Appendix A. Supplementary data the human angiotensin-converting enzyme-lisinopril complex. Nature, 421(6922),
551–554.
Ong, L., Henriksson, A., & Shah, N. P. (2007). Angiotensin converting enzyme-inhibitory
Supplementary data to this article can be found online at https://
activity in Cheddar cheeses made with the addition of probiotic Lactobacillus casei sp.
doi.org/10.1016/j.foodchem.2019.01.010. Le Lait, 87(2), 16.
Papadimitriou, C. G., Vafopoulou-Mastrojiannaki, A., Silva, S. V., Gomes, A.-M., Malcata,
References F. X., & Alichanidis, E. (2007). Identification of peptides in traditional and probiotic
sheep milk yoghurt with angiotensin I-converting enzyme (ACE)-inhibitory activity.
Food Chemistry, 105(2), 647–656.
Aragão, D. S., de Andrade, M. C., Ebihara, F., Watanabe, I. K., Magalhães, D. C., Juliano, Perpetuo, E. A., Juliano, L., & Lebrun, I. (2003). Biochemical and pharmacological aspects
M. A., ... Casarini, D. E. (2015). Serine proteases as candidates for proteolytic pro- of two bradykinin-potentiating peptides obtained from tryptic hydrolysis of casein.
cessing of angiotensin-I converting enzyme. International Journal of Biological Journal of Protein Chemistry, 22(7–8), 601–606.
Macromolecules, 72, 673–679. Pihlanto, A., Virtanen, T., & Korhonen, H. (2010). Angiotensin I converting enzyme (ACE)
Brasil, G. A., Silva-Cutini, M. A., Moraes, F. S. A., Pereira, T. M. C., Vasquez, E. C., Lenz, inhibitory activity and antihypertensive effect of fermented milk. International Dairy
D., ... de Andrade, T. U. (2018). The benefits of soluble non-bacterial fraction of kefir Journal, 20(1), 7.
on blood pressure and cardiac hypertrophy in hypertensive rats are mediated by an Pihlanto-Leppälä, A., Koskinen, P., Piilola, K., Tupasela, T., & Korhonen, H. (2000).
increase in baroreflex sensitivity and decrease in angiotensin-converting enzyme Angiotensin I-converting enzyme inhibitory properties of whey protein digests:
activity. Nutrition, 51–52, 66–72. Concentration and characterization of active peptides. Journal of Dairy Research,
Capriotti, A. L., Cavaliere, C., Piovesana, S., Samperi, R., & Laganà, A. (2016). Recent 67(1), 53–64.
trends in the analysis of bioactive peptides in milk and dairy products. Analytical and Pihlanto-Leppälä, A., Rokka, T., & Korhonen, H. (1998). Angiotensin I converting enzyme
Bioanalytical Chemistry, 408(11), 2677–2685. inhibitory peptides derived from bovine milk proteins. International Dairy Journal,
de Oliveira, M. R., Silva, T. J., Barros, E., Guimarães, V. M., Baracat-Pereira, M. C., Eller, 8(4), 6.
M. R., ... de Oliveira, E. B. (2018). Anti-hypertensive peptides derived from caseins: Piovesana, S., Capriotti, A. L., Cavaliere, C., La Barbera, G., Samperi, R., Zenezini Chiozzi,
mechanism of physiological action, production bioprocesses, and challenges for food R., & Laganà, A. (2015). Peptidome characterization and bioactivity analysis of
applications. Applied Biochemistry and Biotechnology. donkey milk. Journal of Proteomics, 119, 21–29.
Dzau, V. J., Burt, D. W., & Pratt, R. E. (1988). Molecular biology of the renin-angiotensin Prado, M. R., Blandón, L. M., Vandenberghe, L. P., Rodrigues, C., Castro, G. R., Thomaz-
system. American Journal of Physiology, 255(4 Pt 2), F563–F573. Soccol, V., & Soccol, C. R. (2015). Milk kefir: Composition, microbial cultures, bio-
Ebner, J., Aşçı Arslan, A., Fedorova, M., Hoffmann, R., Küçükçetin, A., & Pischetsrieder, logical activities, and related products. Frontiers in Microbiology, 6, 1177.
M. (2015). Peptide profiling of bovine kefir reveals 236 unique peptides released Quirós, A., Hernández-Ledesma, B., Ramos, M., Amigo, L., & Recio, I. (2005).
from caseins during its production by starter culture or kefir grains. Journal of Angiotensin-converting enzyme inhibitory activity of peptides derived from caprine
Proteomics, 117, 41–57. kefir. Journal of Dairy Science, 88(10), 3480–3487.
Ferreira, I. M., Pinho, O., Monteiro, D., Faria, S., Cruz, S., Perreira, A., ... Tavares, P. Qureshi, T. M., Vegarud, G. E., Abrahamsen, R. K., & Skeie, S. (2013). Angiotensin I-
(2010). Short communication: Effect of kefir grains on proteolysis of major milk converting enzyme-inhibitory activity of the Norwegian autochthonous cheeses
proteins. Journal of Dairy Science, 93(1), 27–31. Gamalost and Norvegia after in vitro human gastrointestinal digestion. Journal of
Friedland, J., & Silverstein, E. (1976). A sensitive fluorimetric assay for serum angio- Dairy Science, 96(2), 838–853.
tensin-converting enzyme. American Journal of Clinical Pathology, 66(2), 416–424. Rai, A. K., Sanjukta, S., & Jeyaram, K. (2017). Production of angiotensin I converting
Friques, A. G., Arpini, C. M., Kalil, I. C., Gava, A. L., Leal, M. A., Porto, M. L., ... Vasquez, enzyme inhibitory (ACE-I) peptides during milk fermentation and their role in re-
E. C. (2015). Chronic administration of the probiotic kefir improves the endothelial ducing hypertension. Critical Reviews in Food Science and Nutrition, 57(13),
function in spontaneously hypertensive rats. J Transl Med, 13, 390. 2789–2800.
Fujita, H., Yokoyama, K., & Yoshikawa, M. (2000). Classification and Antihypertensive Robert, M. C., Razaname, A., Mutter, M., & Juillerat, M. A. (2004). Identification of an-
Activity of Angiotensin I-Converting Enzyme Inhibitory Peptides Derived from Food giotensin-I-converting enzyme inhibitory peptides derived from sodium caseinate
Proteins. Journal of Food Science, 65(4), 564–569. hydrolysates produced by Lactobacillus helveticus NCC 2765. Journal of Agriculture
Giacometti, J., & Buretić-Tomljanović, A. (2017). Peptidomics as a tool for characterizing and Food Chemistry, 52(23), 6923–6931.
bioactive milk peptides. Food Chemistry, 230, 91–98. Rodrigues, K. L., Caputo, L. R., Carvalho, J. C., Evangelista, J., & Schneedorf, J. M. (2005).
Gobbetti, M., Ferranti, P., Smacchi, E., Goffredi, F., & Addeo, F. (2000). Production of Antimicrobial and healing activity of kefir and kefiran extract. International Journal of
angiotensin-I-converting-enzyme-inhibitory peptides in fermented milks started by Antimicrobial Agents, 25(5), 404–408.
Lactobacillus delbrueckii subsp. bulgaricus SS1 and Lactococcus lactis subsp. cremoris Rosa, D. D., Dias, M. M. S., Grześkowiak, Ł., Reis, S. A., Conceição, L. L., & Peluzio, M. D.
FT4. Applied and Environment Microbiology, 66(9), 3898–3904. C. G. (2017). Milk kefir: Nutritional, microbiological and health benefits. Nutrition
Goodman, B. E. (2010). Insights into digestion and absorption of major nutrients in hu- Research Reviews, 30(1), 82–96.
mans. Advances in Physiology Education, 34(2), 44–53. Saito, T., Nakamura, T., Kitazawa, H., Kawai, Y., & Itoh, T. (2000). Isolation and struc-
Hadisaputro, S., Djokomoeljanto, R. R., Judiono, & Soesatyo, M. H. (2012). The effects of tural analysis of antihypertensive peptides that exist naturally in Gouda cheese.
oral plain kefir supplementation on proinflammatory cytokine properties of the hy- Journal of Dairy Science, 83(7), 1434–1440.
perglycemia Wistar rats induced by streptozotocin. Acta Med Indones, 44(2), Santanna, A. F., Filete, P. F., Lima, E. M., Porto, M. L., Meyrelles, S. S., Vasquez, E. C., ...

118
F.G. Amorim et al. Food Chemistry 282 (2019) 109–119

Andrade, T. U. (2017). Chronic administration of the soluble, nonbacterial fraction of CP790. Journal of Dairy Science, 77(4), 917–922.
kefir attenuates lipid deposition in LDLr. Nutrition, 35, 100–105. Yamamoto, N., Maeno, M., & Takano, T. (1999). Purification and characterization of an
Soggiu, A., Roncada, P., & Piras, C. (2018). Proteomics in milk and dairy products. In E. D. antihypertensive peptide from a yogurt-like product fermented by Lactobacillus hel-
de Almeida, A., & I. Miller (Eds.). Proteomics in domestic animals: from farm to systems veticus CPN4. Journal of Dairy Science, 82(7), 1388–1393.
biology.). Cham: Springer. Yaman, H., Elmali, M., Karadagoglu, G., & Cetinkaya, A. (2006). Observations of kefir
Swaisgood, H. E. (1993). Review and update of casein chemistry. Journal of Dairy Science, grains and their structure from different geographical regions: Turkey and germany.
76(10), 3054–3061. Atatürk University Journal of Veterinary Sciences, 1(1–2), 5.
Udenigwe, C. C. (2014). Bioinformatics approaches, prospects and challenges of food Zenezini Chiozzi, R., Capriotti, A. L., Cavaliere, C., La Barbera, G., Piovesana, S., Samperi,
bioactive peptide research. Trends in Food Science & Technology, 36(2), 137–143. R., & Laganà, A. (2016). Purification and identification of endogenous antioxidant
Wei, L., Clauser, E., Alhenc-Gelas, F., & Corvol, P. (1992). The two homologous domains and ACE-inhibitory peptides from donkey milk by multidimensional liquid chroma-
of human angiotensin I-converting enzyme interact differently with competitive in- tography and nanoHPLC-high resolution mass spectrometry. Analytical and
hibitors. Journal of Biological Chemistry, 267(19), 13398–13405. Bioanalytical Chemistry, 408(20), 5657–5666.
Yamamoto, N., Akino, A., & Takano, T. (1994). Antihypertensive effect of the peptides Zhang, Y., Fonslow, B. R., Shan, B., Baek, M. C., & Yates, J. R. (2013). Protein analysis by
derived from casein by an extracellular proteinase from Lactobacillus helveticus shotgun/bottom-up proteomics. Chemical Reviews, 113(4), 2343–2394.

119