Journal of Food Composition and Analysis 45 (2016) 113–120

Contents lists available at ScienceDirect

Journal of Food Composition and Analysis

journal homepage: www.elsevier.com/locate/jfca

Original Research Article

Characterization of concentrated agave saps and storage effects

on browning, antioxidant capacity and amino acid content
Liliana Santos-Zea a, Ana M. Leal-Dı́az a, Daniel A. Jacobo-Velázquez a,
José Rodrı́guez-Rodrı́guez b, Silverio Garcı́a-Lara a, Janet A. Gutiérrez-Uribe a,*
a
Tecnologico de Monterrey, Campus Monterrey, Centro de Biotecnologı´a-FEMSA, Escuela de Ingenierı´a y Ciencias, Av. Eugenio Garza Sada 2501 Sur,
C.P. 64849 Monterrey, N.L., Mexico
b
Tecnologico de Monterrey, Campus Monterrey, Centro de Calidad Ambiental, Escuela de Ingenierı´a y Ciencias, Av. Eugenio Garza Sada 2501 Sur,
C.P. 64849 Monterrey, N.L., Mexico

A R T I C L E I N F O A B S T R A C T

Article history:
Received 14 May 2015
Received in revised form 17 October 2015
Accepted 19 October 2015
Available online 22 October 2015
Chemical compounds studied in this article:
2,3-Dihydro-3,5-dihydroxy-6-methyl-4H-
pyran-4-one (PubChem CID: 16038313)
Hecogenin (PubChem CID: 91453)
Kammogenin (PubChem CID: 12305426)
5-Hydroxymethylfurfural (PubChem CID:
237332)
Lysine (PubChem CID: 5962)
Phenylalanine (PubChem CID: 6140)
Serine (PubChem CID: 5951)
Fructose (PubChem CID: 5984)
Glucose (PubChem CID: 5793)
Sucrose (PubChem CID: 5988).
Sap from agave plants (‘‘aguamiel’’) is traditionally consumed in Mexico as a fresh beverage, fermented
or concentrated. Concentrated agave sap (CAS) is used as a sweetener but, due to heating, a brown color
develops and intensifies during storage. Browning varies among CAS batches and this work was focused
on its correlation with the chemical composition changes observed during 20 weeks of storage. The
browning index (BI), measured as the optical density at 490 nm (OD490 nm) per gram of sample, increased
54.4% in the batch that initially had 57 OD490 nm/g but in the other two batches that had a lower BI, the
increase was less than 26.1%. Antioxidant capacity only increased in the batch with the highest BI going
from 18 to 23 Trolox equivalent mmol/g dry weight. Saponin content was different in the three batches
(224.2–434.7 protodioscin equivalents/gram dry weight) but did not change after 20 weeks of storage.
Browning index and antioxidant capacity were negatively correlated with free amino acid concentration,
particularly serine, phenylalanine and lysine decreased 29.4, 50 and 30%, respectively. Browning was
positively correlated to furosine, an early Maillard reaction derivative of lysine previously reported as a
free radical scavenger.
ß 2015 Elsevier Inc. All rights reserved.

Keywords:
Agave sap
Storage
Saponins
Browning
Antioxidant
Furosine
Aguamiel
Food analysis
Food composition

1. Introduction

Abbreviations: ACE, acetonic extract; ACN, acetonitrile; AOXC, antioxidant capacity;

ASX, asparagine/aspartic acid; AUC, area under curve; B1, batch 1; B2, batch 2; B3,
batch 3; CAS, concentrated agave sap; DDMP, 2,3-dihydro-3,5-dihydroxy-6-
methyl-4H-pyran-4-one; dw, dry weight; EI, electron impact ionization; ESI,
electrospray ionization; fw, fresh weight; GLX, glutamine/glutamic acid; HMF, 5-
hydroxymethylfurfural; LYS, lysine; OD, optical density; ORAC, oxygen raidcal
absorbance capacity; PHE, phenylalanine; PRO, proline; SER, serine; TE, Trolox
equivalents; TP, total phenols; TSS, total soluble solids; u, units.
* Corresponding author. Tel.: +52 81 83284322; fax: +52 81 83284262.
E-mail address: jagu@itesm.mx (J.A. Gutiérrez-Uribe).
Agave (Agave spp.) is a genus of monocotyledon, monocarpic
plants, distributed throughout the American continent (Ortiz-
Basurto et al., 2008) and among the 293 species, 75% may be found
in Mexico (Good-Avila et al., 2006). This plant can be used as a raw
material for the manufacture of traditional beverages, such as
tequila and mezcal (Narváez-Zapata and Sánchez-Teyer, 2009). It is
also considered as a source of functional food ingredients, such as
prebiotic fructans, and as a medicinal plant. Agave leaves and

http://dx.doi.org/10.1016/j.jfca.2015.10.005
0889-1575/ß 2015 Elsevier Inc. All rights reserved.
114 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120
stems contain bioactive metabolites conferring a diversity of
effects on health, such as anticancer, anti-inflammatory and
immunomodulatory (Santos-Zea et al., 2012). Saponins, in
particular, have been previously reported in fresh agave sap
(Leal-Dı́az et al., 2015), and these compounds have shown
pharmacological properties, such as anti-inflammatory and
ulceroprotective effects (Mina et al., 2014).
The plant sap, called ‘‘aguamiel’’ in Mexico, is a liquid composed
mainly of sugars, including fructo-oligosaccharides (Ortiz-Basurto
et al., 2008). This fluid is harvested exclusively from certain agave
species known as ‘‘agaves pulqueros’’, including Agave americana,
A. atrovirens, A. ferox, A. mapisaga and A. salmiana (Escalante et al.,
2008). When the plant reaches maturity (8–12 years), a cavity is
made in the center by cutting the floral stem, where sap is
manually collected twice a day (Gutierrez-Uribe and Serna-
Salvdivar, 2013). It may be directly consumed as a beverage but
due to its composition and the presence of endophytic micro-
organisms, it is not suitable for long-term storage. In fact, more
than 300 endophytic bacteria strains have been isolated from
Agave tequilana leaf base (Martı́nez-Rodrı́guez et al., 2015). Agave
sap is also traditionally fermented to produce a mild ethanolic
drink named ‘‘pulque’’ (Narváez-Zapata and Sánchez-Teyer, 2009)
or submitted to a thermal process to yield a product denominated
concentrated agave sap (CAS) (Gutierrez-Uribe and Serna-Salvdi-
var, 2013). CAS production is an artisanal process and we consider
that the information included in this manuscript would be very
helpful to understand the importance of setting time and
temperature set points and study the agave sap characteristics
that must be analyzed for its use as raw material.
As in other foods with a high concentration of sugars and
phytochemicals, including honey, agave sap concentrate may show
changes in the concentration or transformation of antioxidants and
other phytochemicals with beneficial effects on health. Moreover,
a relationship between antioxidant capacity and a diversity of
compounds generated or released during honey storage has been
demonstrated (Brudzynski and Miotto, 2011). Browning due to
formation of Maillard compounds, deteriorative reactions and
complexing of polyphenols with amino acids and proteins has been
observed in honey. Citrus honey physicochemical and organoleptic
properties changed during one-year storage because of the loss of
volatile compounds and sugars and Maillard derivatives formation
(Castro-Vázquez et al., 2008). Maillard reaction is known to occur
during storage in products containing sugar and free amino acids,
such as honey, juice concentrates and jams (Aslanova et al., 2010;
Bulut and Kilic, 2009; Selen Burdurlu and Karadeniz, 2003). Some
of the markers used to analyze this complex system of reactions
in food products include furosine, carboxymethyllysine and
hydroxymethylfurfural (Bastos and Gugliucci, 2015).
The current study was carried out to determine the effect of
storage on the main characteristics of concentrated agave sap.
Physicochemical parameters (water activity, dry matter, pH,
soluble solids, browning) and chemical composition (content of
sugars, protein, amino acids, saponins, phenolic acids, Maillard
reaction products and antioxidant capacity) were evaluated over
the course of 20 week storage. Correlations among the analyzed
parameters were established.
2. Materials and methods
2.1. Concentrated agave sap preparation and storage conditions
Three different batches of concentrated agave sap (CAS),
elaborated by artisanal means, were obtained from a commercial
supplier (AGMEL, Monterrey, Mexico). Each batch was produced as
follows: agave sap (aguamiel) was collected from mature agaves
(over 7 years of age) from the species Agave salmiana by traditional
means. It was then pooled and concentrated to nearly 10% of its
original volume by heating in a stove until boiling point was
reached (Gutierrez-Uribe and Serna-Salvdivar, 2013). The brown-
colored syrup obtained by this process was then stored in plastic
containers until delivery to the laboratory. First batch (B1) was
produced in the state of Hidalgo, Mexico on July 2011. Second (B2)
and third (B3) batches were acquired from the same region on
November and December 2013, respectively.
The storage study started 14 days after CAS was produced. Each
batch was separated into 50 mL aliquots and stored at 25  2 8C
and 85% relative humidity in darkness. Every two weeks, an aliquot
was collected until 20 weeks were completed. Aliquots were stored
at 20 8C until analysis.
2.2. Physicochemical tests
Dry matter was determined by gravimetric means (AOAC
425.45) drying at 60 8C in vacuum. Liquid sample (2 g) was mixed
with 0.5 g dry silicon dioxide (Celite 545, Desarrollo de Especia-
lidades Quı́micas, Monterrey, Mexico) and left to dry during 24 h.
Weight was recorded before and after drying to obtain dry matter
percentage. Water activity was measured using the instrument
Aqualab with dew point sensor (4TEV, Decagon Devices, Pullman,
WA, USA) at 25 8C, pH with a digital pH meter (Mettler-Toledo AG,
Schwerzenbach, Switzerland) and total soluble solids (TSS) with a
manual refractometer in scale 58–90 8Brix (ATAGO, Tokyo, Japan).
Browning index was determined according to a procedure used to
evaluate color for tequila ‘‘cooking honey’’ (Waleckx et al., 2008),
with some modifications. Concentrated agave sap was diluted 50-
fold (w/v) with distilled water, vortexed and 200 mL were plated in
96-well microplates. Optical density at 490 nm (OD490) was
recorded in a microplate reader (Synergy HT, Bio-Tek, Winooski,
VT, USA). Results were reported as OD490 per gram of CAS fresh
weight (OD490/g fw). Protein content was determined according
the AOAC 978.02 micro Kjeldahl standardized method and results
were reported as % dry weight (% dw), using conversion factor
N  6.25.
2.3. Sugars and free amino acid determination
To quantitate sugars and amino acids, CAS was diluted 10-fold
(7 8Brix) with distilled water, and centrifuged at 13,800 g to
eliminate suspended matter (Centrifuge 5804 R, Eppendorf,
Hamburg, Germany). The supernatant was then filtered through
0.22 mm nylon membranes (VWR International LLC, Radnor, PA,
USA) and kept at 20 8C until determination was carried out.
Sugar concentrations were quantitated using liquid chroma-
tography with evaporative light scattering (HPLC-ELSD) detection
(Agilent Technologies, 1200 Series, Santa Clara, CA, USA) according
to application XBridge Amide WA64101 (Waters Corporation,
Milford, MA, USA). Separation was carried out in XBridge Amide
column, particle size 3.5 mm, 250 mm  4.6 mm i.d. (Waters
Corporation, Milford, MA, USA) with mobile phase A: 80%
acetonitrile (ACN) (BDH, Poole, UK) and 20% water (BDH, Poole,
UK) with 0.2% triethylamine (Sigma–Aldrich, St. Louis, MO, USA)
(v/v/v) and B: 30% ACN and 70% water with 0.2% triethylamine
(v/v/v) at a flow rate of 1 mL/min, using the following gradient:
0!16 min, 10!70% B; 16!17 min, 70!10% B; 17!30 min,
isocratic elution at 10% B. Conditions for ELSD were: gain 2,
pressure 2 bar, temperature 50 8C, filter: 0.5 s and sampling
time 100–10 Hz. D-Glucose (Sigma–Aldrich, St. Louis, MO, USA),
D-fructose (Merck KgaA, Darmstadt, Germany), and sucrose (Fluka,
St. Louis, MO, USA) were used as standards, in a range from 0.5 to
10 mg/mL for each sugar. Before injection (10 mL), samples were
further diluted with HPLC grade water to 1 8Brix. Results were
reported as % dw.
 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120
Free amino acid concentration was evaluated using the Waters
AccQ*Tag Chemistry Package (Waters Corporation, Milford, MA,
USA) by high performance liquid chromatography coupled to a
fluorescence detector (HPLC-FLD). Separation was carried out on a
Nova-Pak AccQ*Tag AminoAcid Analysis Column (Waters Corpo-
ration, Milford, MA, USA) at 37 8C. Mobile phase A consisted of a
AccQ*Tag Eluent A (Waters Corporation, Milford, MA, USA), diluted
at a 1:10 ratio in HPLC grade water and phase B contained 60% ACN
and 40% water (v/v). Elution was carried out with the following
gradient: 0!0.5 min, 0!2% B; 0.5!15 min, 2!7% B; 15!19 min,
7!10% B; 19!32 min, 10!33% B; 32!34 min, 33!100% B;
34!37 min, 100% isocratic elution; 37!38 min, 100!0% B;
38!45 min, 0% B isocratic elution. Flow was constant at 1.0 mL/
min and injection volume was 10 mL. Detector parameters were set
at lex/em 250/395 nm. Tryptophan was not detected by this
method, while aspartic acid and asparagine (ASX, Rt = 13.5 min)
and glutamic acid and glutamine (GLX, Rt = 16.1) were quantitated
as mixtures at the retention times specified. Calibration was
carried out with a mix of amino acid standards (12.5–150 pM).
Results were reported as milligrams per gram of CAS dry weight
(mg/g dw).
2.4. Analysis of total phenols, antioxidant capacity and 5-
hydroxymethylfurfural
For total phenols (TP) and antioxidant capacity (AOXC)
evaluation, a CAS acetonic extract (ACE) was produced to
minimize interference by reducing sugars. In preliminary
experiments, acetone and methanol were used to extract
antioxidants from 17 CAS samples demonstrating that acetone
was a better solvent for the extraction of antioxidants (data not
shown). Extraction was done using 5 g of CAS and 10 mL of a
solution of 80% distilled acetone (Desarrollo de Especialidades
Quı́micas, Monterrey, Mexico) and 20% distilled water (v/v) in a
shaking incubator (Innova 4000, New Brunswick Scientific,
Edison, NJ, USA) at 250 rpm and 30 8C during 20 min. It is
important to point out that, based on preliminary experiments,
these extraction conditions allowed the recovery of 50–70% TP,
depending on the CAS TP concentration. But for comparison
purposes, these conditions were selected to find out if TP and
AOXC were affected during the storage of CAS. A 1 mL aliquot was
taken for TP and AOXC. The remaining ACE (9 mL) was evaporated
to dryness under vacuum at 50 8C (Rocket Evaporator, Genevac
Ltd, Ipswich, England) and used for 5-hydroxymethylfurfural
(HMF) extraction.
Total phenols were evaluated by the Folin–Ciocalteau method
as previously reported (Zheng and Wang, 2001). Absorbance was
recorded at 765 nm in a microplate reader (Epoch, Biotek,
Winooski, VT, USA). Gallic acid (Sigma–Aldrich, St. Louis, MO,
USA) was used as standard (25–150 mg/mL) and results were
expressed as gallic acid equivalent mg per gram dry weight CAS
(GAE mg/g dw CAS).
Antioxidant capacity was determined according to the hydro-
philic oxygen radical absorbance capacity (ORAC) assay as
previously reported (Huang et al., 2002). Fluorescence was
recorded at lex/em = 485/530 nm in a microplate reader (Synergy
HT, Biotek, Winooski, VT, USA). ()-6-Hydroxy-2,5,7,8-tetramethyl-
chromane-2-carboxylic acid (Trolox) purchased at Sigma–Aldrich, St.
Louis, MO, USA (25–100 mM) was used as standard, and results were
expressed as Trolox equivalent mmol per gram dry weight CAS (TE
mmol/g dw).
5-Hydroxymethylfurfural (HMF) was extracted from ACE
(9 mL), following a modified procedure established by Mancilla-
Margalli and López (2002). Dry ACE was dissolved in 5 mL distilled
water and partitioned with 5 mL HPLC grade dichloromethane
(Honeywell Burdick and Jackson, Muskegon, MI, USA). Organic
115
phase was collected and aqueous phase was partitioned twice
more with fresh solvent. Sodium sulfate (Desarrollo de Especia-
lidades Quı́micas, Monterrey, Mexico) was used to eliminate water
from the organic phase. Pooled dichloromethane extract for each
sample was dried under nitrogen gas flow and stored at 20 8C
until analysis.
Quantitation of HMF was determined by gas chromatography
coupled to mass spectroscopy detector (GC–MS), modifying a
method reported by Mancilla-Margalli and López (2002). Dry
extracts were diluted in 200–1000 mL dichloromethane and 1 mL
was injected in 1:5 split mode on a gas chromatograph (Model
6890N, Agilent Technologies, Santa Clara, CA, USA) coupled to a
mass spectroscopy detector (5973N, Agilent Technologies, Santa
Clara, CA, USA). Separation was carried out on a 25 m  0.32 mm
i.d., 0.5 mm FFAP column (Agilent Technologies, Santa Clara, CA,
USA) with helium (AOC, Saltillo, Mexico) as carrier gas at a
flow rate of 1 mL/min. Temperature ramp was initiated at 50 8C
during 2 min, followed by an increase to 100 8C at a rate of
20 8C/min, 1 min hold at 100 8C and increase to 230 8C was
reached at 7 8C/min and maintained for 30 min. Injector and
detector temperatures were 250 8C and 150 8C, respectively.
Ionization was done by an Electron Impact (EI) source at
70 eV. A HMF (Sigma–Aldrich, St. Louis, MO, USA) standard
curve (10–250 mg/mL) was used for quantitation and results
were reported as HMF mg/g dw CAS. The peak corresponding to
HMF was identified by the standard retention time and
confirmed by the mass spectra library.
2.5. Saponin quantitation
Saponin extraction was modified from Leal-Dı́az et al. (2015)
using 2.5 g CAS and extracting with 50 mL of 50% n-butanol
(Desarrollo de Especialidades Quı́micas, Monterrey, Mexico) and
50% water (v/v), under agitation at 750 rpm during 25 min in a
shaking incubator (VorTemp1550 Labnet, Edison, NJ, USA). The
extract was centrifuged (SL16R Thermo Fisher Scientific Inc.,
Waltham, MA, USA) at 3000 rpm, 4 8C during 5 min and the upper
phase was collected. The process was carried out twice and the
extracts were pooled and vacuum concentrated at 40 8C (Rocket
Evaporator, Genevac Ltd, Ipswich, England). Dry extract was
resuspended in 2 mL 50% methanol (BDH, Poole, UK) and 50%
water (v/v) and filtered through 0.45 mm GHP membranes (Pall
Life Sciences, Port Washington, NY, USA).
Quantitation was performed by HPLC-ELSD (Agilent Technolo-
gies, 1200 Series, Santa Clara, CA, USA). Chromatographic
conditions were the same previously reported (Leal-Dı́az et al.,
2015). Separation was performed at 25 8C by reversed phase
chromatography on a 150 mm  4.6 mm i.d., 5 mm, Zorbax Eclipse
XDB-C18 column with a 12.5 mm  4.6 mm i.d. guard column of
the same material (Agilent Technologies, Santa Clara, CA, USA),
using A: water with 0.1% formic acid (Sigma–Aldrich, St. Louis, MO,
USA) and B: ACN with 0.1% formic acid. Elution was done as
follows: 0!8 min, isocratic elution at 28% B; 8!28 min, 28!55%
B; 28!31 min, 55!100% B; 31!38 min, isocratic at 100% B,
38!40 min, 100!28% B; 40!45 min, isocratic at 28% B. Detection
was carried out using nitrogen gas; pressure, 3.4 bar; temperature
40 8C; gain, 9; sampling time, 33 Hz. Chromatograms were
obtained with Chemstation for LC (Agilent Technologies, Santa
Clara, CA, USA). Hecogenin (Santa Cruz Biotechnology Inc., Santa
Cruz, CA, USA) was used as quantitation standard (10–250 ppm)
for sapogenins and protodioscin (Sigma–Aldrich, St. Louis, MO,
USA) (10–250 ppm) for saponins. Results were reported as
hecogenin and protodioscin equivalent mg per gram of CAS dry
weight (HE mg/g dw CAS and PE mg/g dw CAS). Methodology
used for saponin identification can be found in Supplementary
information S3.
116 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120

2.6. Maillard reaction products determination Table 1

Characterization of physicochemical parameters of three independent batches
(B1–3) of concentrated agave sap.
Maillard reaction products determination was carried out by
HPLC-DAD (Agilent Technologies, 1260 Series, Santa Clara, CA, Parameter B1 B2 B3
USA) on 500-fold diluted CAS in 50% methanol and 50% water (v/v) Water activity (aw) 0.73  0.01 b 0.83  0.01 a 0.83  0.01 a
and filtered through 0.45 mm GHP membranes (Pall Life Sciences, Moisture (%) 25.2  0.2 b 32.3  0.2 a 32.5  0.1 a
Port Washington, NY, USA). Separation was carried out on a pH 4.4  0.0 b 5.0  0.1 a 4.5  0.0 b
Total soluble solids (8Brix) 75.5  0.3 a 65.9  0.1 b 64.8  0.3 c
250 mm  4.6 mm i.d., 5 mm, Luna C18 column (Phenomenex Inc.,
Browning index (OD490/g fw) 57.7  3.9 a 23.5  0.4 b 23.7  0.2 b
Torrance CA, USA), using as mobile phase A: water pH 2.4 adjusted Fructose (% dw) 20.7  2.4 a 12.4  3.3 b 17.5  0.2 ab
with ortho-phosphoric acid (Desarrollo de Especialidades Quı́mi- Glucose (% dw) 18.6  2.1 a 13.9  0.1 b 19.9  0.2 a
cas, Monterrey, Mexico) and B: 60% methanol and 40% water (v/v) Sucrose (% dw) 13.8  1.8 c 41.7  0.2 a 19.8  0.2 b
Protein (% dw) 3.9  0.1 a 4.2  0.1 a 3.4  0.1 b
pH 2.4 adjusted with ortho-phosphoric acid. According to a
Amino acids (mg/g dw)a 5.2  2.3 b 14.2  1.5 a 16.3  2.5 a
method for phenolic acids, chromatograms were obtained at Total phenols (GAE mg/g dw) 1.14  0.10 a 0.56  0.02 b 0.56  0.03 b
280 nm and UV–visible absorption spectra were recorded (Torres- Antioxidant capacity 18.9  0.9 a 12.7  0.7 b 13.2  1.3 b
Contreras et al., 2014). A mix of vanillic, gallic, chlorogenic, p- (TE mmol/g dw)
coumaric and ferulic acids (Sigma–Aldrich, St. Louis, MO, USA) Different letters in each row mean significant statistical differences between results
were used to identify by retention time. Compound identification (n = 3, p < 0.05).
a
was carried out by HPLC coupled to a mass spectroscopy time of Tryptophan was not quantitated by this method.
flight (HPLC-MS-TOF) detector (Model G1969A Agilent 1100 Santa
Clara, CA, USA). Mass spectra for unknown peaks were obtained to that of high fructose syrup obtained from cooking agave heads for
characterize CAS fingerprint. To compare the three batches and the tequila production (pH = 4.5–5.0) (Mancilla-Margalli and López,
changes during storage, relative quantitation was done based on 2002) and was within the range (pH = 4.1–5.5) reported in
area under the curve (AUC) for each peak obtained at 280 nm commercial agave nectar (Mellado-Mojica and López, 2015). It
according to the following formula: has also been reported that A. salmiana syrups have a moisture
content ranging from 20 to 30%, TSS from 70 to 80 8Brix and pH
AUCdilution factor=mg CAS dw 3.5 to 5.2 (Mellado-Mojica and López, 2015).
Another important difference among batches was observed in
Mass spectroscopy was carried out with the same elution the browning index, with B1 presenting values 2.4 times greater
method using an electrospray source in positive mode (ESI+) under than B2 and B3. Agave syrup, as well as the sugary liquid used for
the following conditions: m/z range, 120–600; nitrogen gas, 8 L/ tequila production, has been reported to be dark brown due to the
min; temperature, 300 8C; nebulizer pressure, 20 psi; capillary formation of Maillard compounds (Montañez-Soto et al., 2011;
voltage; 4000 V; fragment voltage, 70 V; and skimmer, 40 V. Waleckx et al., 2008).
Extracted ion chromatograms were obtained using Analyst QS Fructose (1) and glucose (2) were more abundant than sucrose
1.1 software (Applied Biosystems, Carlsbad, CA, USA). (3) only in B1 (Fig. 1). In the three batches analyzed, the fructose
content was lower than 20.7% (Table 1) in accordance to previous
2.7. Statistical analysis reports that showed that A. salmiana syrups had a sucrose content
of up to 32% in contrast to those obtained from A. tequilana syrups
Measurements were done in triplicate, and results were that had more than 60% fructose (Mellado-Mojica and López,
expressed as the mean  standard deviation. Statistical analysis 2015). The sucrose concentration in B2 and B3 was higher than the
was conducted by one-way ANOVA and differences among means observed for B1 (Table 1), indicating a possible hydrolysis during
were calculated by Tukey’s test. Spearman’s correlation analysis was extended heating that also produced a darker concentrated
carried out to determine relationship between variables. A level of agave sap.
significance of 0.05 was used in all tests, using JMP1 Version The protein content was higher in B2 than in B1 or B3 (Table 1).
12 software (SAS Institute Inc., Cary, NC, USA). B1 had the lowest free amino acid content (Table 1). Amino acid
profile was different for the three batches studied (Supplementary
3. Results and discussion Table S1), but cysteine was consistently not present in any of them,
comparable to results reported for non-concentrated agave sap
3.1. Characterization of concentrated agave saps (Ortiz-Basurto et al., 2008). Proline was the most abundant free

This work reports for the first time the changes that occurred in
concentrated agave sap (CAS) during storage. There were impor- mV
3
tant differences among the three batches tested. Particularly, the
first batch had the lowest water activity and dry matter, and 160
logically the highest concentration of soluble solids (Table 1).
These results demonstrate that batch 1 (B1) had a higher exposure
120
to thermal process since the only indicator used to stop artisanal 1
concentration is when the liquid starts boiling. Agave sap 2
concentrate from batch 2 (B2) and batch 3 (B3) were similar in 80
terms of aw and dry matter content, but soluble solids were higher
for B2. According to these results, aw in CAS was higher than in 40
agave syrup produced from Agave tequilana fructan hydrolysis
(aw = 0.699), while TSS (70 8Brix) and dry matter content (70%)
were similar (Montañez-Soto et al., 2011). Batches 1 and 3 had a
0 2 4 6 8 10 12 14 min
similar pH, while B2 was different (pH = 5.0). In all cases pH values
were lower than the reported pH for fresh agave sap or ‘‘aguamiel’’ Fig. 1. HPLC-ELSD chromatogram showing (1) fructose, (2) glucose and (3) sucrose
(pH = 6.0) (Escalante et al., 2008). The CAS pH was very similar to for three batches, B1 (____), B2 (_._._._), B3 (.........), of concentrated agave sap.
 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120
Fig. 2. Changes during the storage of concentrated agave sap related to (a) browning, (b) antioxidant capacity, (c) 5-hydroxymethylfurfural concentration, and (d) fructose,
glucose and sucrose concentration at 2, 10 and 20 week storage for three batches of concentrated agave sap (n = 3, statistical significance was 0.05).
amino acid in B1 and aspartic acid in B2 and B3 (Supplementary
Table S1). Total phenols (TP) content and antioxidant capacity
(AOXC) of ACE were two-fold higher for B1 than for B2 and B3
(Table 1). Concentration of TP in B1 was similar to the reported for
Agave atrovirens leaves extracts (1.072–1.338 mg GAE/g dw)
(Olvera-Garcia et al., 2015).
3.2. Changes observed during storage of concentrated agave saps
Water activity, dry matter, total soluble solids and pH did not
show significant changes during storage (Supplementary Fig. S2).
Browning increased linearly from 57 to 88 OD490/g fw after
20 weeks of storage of B1 (Fig. 2a). For B2 and B3, even after
20 weeks of storage, browning did not reach the initial value
observed for B1. A similar phenomenon was observed in a study of
honey stored during 90 days at 37 8C, in which the authors
concluded that initial color was correlated to the degree of
browning occurring during storage (Gonzales et al., 1999).
Additionally, a relationship between moisture content and color
changes in honey has been observed, demonstrating that a lower
moisture content accelerated the darkening reactions (Bulut and
Kilic, 2009), and it is worthy of note that B1 had a lower moisture
and water activity than B2 and B3. The increase in browning has
also been observed in products with similar TSS content
(70 8Brix), such as apple juice concentrates (Selen Burdurlu
and Karadeniz, 2003) as well as diverse fruit jams (Aslanova et al.,
2010; Rababah et al., 2014).
Interestingly, AOXC of B1 increased from 18 to 23 TE mmol/g dw
CAS after 20 weeks (Fig. 2b), a change not observed for B2 and B3.
Contrary to these results, antioxidant capacity of honey diminished
after 6 months of storage even if total phenolic content remained
stable (Wang et al., 2004). TP did not change significantly during
20 weeks storage in any of three batches analyzed in this study
(Supplementary Fig. S2).
5-Hydroxymethyl-furfural (HMF) has been reported in agave
products that undergo thermal hydrolysis of inulin and fructoo-
ligosaccharides to produce sugary syrups. This compound was
found in higher concentrations in thermally hydrolyzed Agave
tequilana juice (Mancilla-Margalli and López, 2002; Waleckx et al.,
2008) than the 82.6 mg/g reached in the present study (Fig. 2c).
Throughout storage, HMF in B1 decreased during the first 8 weeks,
then increased significantly from weeks 10 to 14 and reduced
117
towards week 20. A general declining tendency of HMF concen-
tration with relation to time occurred for B2. By contrast, HMF in
B3 decreased during the first 10 weeks and increased towards the
end of the study to levels similar to the initial value. Interestingly, a
similar behavior had been observed for compounds related to
Maillard reaction produced during inulin hydrolysis of Agave
tequilana exudates for tequila production (Mancilla-Margalli and
López, 2002).
Changes in sugar concentration during storage were different
among batches. B2 had no significant changes in any of the three
sugars analyzed (Fig. 2d), as previously reported for Nigerian
honey stored during 24 months at room temperature (Fasasi,
2012). For B1, sugars concentration incremented from 53 to 65%,
indicating a possible fructan hydrolysis, though a further study is
required to validate this hypothesis. In B3 only glucose concentra-
tion changed significantly (Fig. 2d).
Significant (p-value < 0.0001) negative correlations (r defined
as Spearman correlation coefficient) between browning index and
the amino acids serine (SER, r = 0.91), phenylalanine (PHE,
r = 0.89) and lysine (LYS, r = 0.88) were observed for B1. These
three amino acids decreased markedly during 20 weeks, reaching
29.4%, 50% and 30% initial concentration (Fig. 3a). This decrease
was not observed in B2 and B3 (Fig. 3b and c). Other amino acids,
such as glutamic acid/glutamine, threonine, methionine, histidine,
tyrosine, glycine and leucine (Supplementary Table S1) also had a
correlation with the browning observed in B1 (p-value < 0.001).
Loss of free amino acids during storage has been detected in honey
(Iglesias et al., 2006), possibly due to their conversion to non-
enzymatic browning derivatives (Sanz et al., 2003).
3.3. Determination of saponin content
Based on the mass spectra obtained (Supplementary informa-
tion S3), two kammogenin (4, 5), three manogenin (6, 7, 8) and one
hecogenin (9) glycosides were detected at different concentrations
among CAS batches (Fig. 4). Compound 4 corresponded to
magueyoside A and 6 to magueyoside B, both isolated from agave
flowers (Perez et al., 2013) and found in fresh agave sap (Leal-Dı́az
et al., 2015). Compound 9, agavoside C0 , had been identified in the
leaves of various agave species (Bodeiko and Kintya, 1975; Chen
et al., 2011), and also found in flowers (Perez et al., 2013), and fresh
agave sap (Leal-Dı́az et al., 2015).
118 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120
Fig. 3. Changes in concentration of amino acids related to browning of concentrated agave sap (CAS) of (a) batch 1, (b) batch 2, (c) batch 3 (n = 3, statistical significance was
0.05).
The most abundant glycosides in B2 and B3 were 5 and 6,
quantitated as the sum of both peaks and representing half the
total amount of saponins (Table 2). On the other hand, B1 showed a
different profile, with 4 as the most plentiful compound for this
batch, 40% of the sum of saponins (Table 2). Out of the sapogenins,
only kammogenin (mass spectrum on Supplementary information
S3) was found in free form in the three batches, and greater
amounts were found in B2 and B3 than in B1 (Table 2). Meanwhile,
manogenin was only quantitated in B2 (8 HE mg/g dw).
Saponin concentration was different among batches, with B2
(>400 PE mg/g dw) containing the highest amount followed by B3
(330 mg/g dw) and B1 (>200 mg/g dw) (Table 2). Saponin content
was not significantly altered due to storage after 20 weeks in all
batches studied. CAS contained from 3.5 to 7.5-fold the amount of
saponins when compared to fresh agave sap from immature Agave
salmiana (Leal-Dı́az et al., 2015). In this study, saponin concentra-
tion correlated positively with water activity (p-value = 0.02,
r = 0.73) and negatively with browning (p-value = 0.03, r = 0.71).
3.4. Analysis of Maillard reaction products
Chromatogram at 280 nm (Fig. 5a) showed predominant peaks
in the first 15 min of analysis. The most plentiful in the three
batches were the ones eluting at 4.1 (10), 9.2 (11) and 10.1 min
(12). Retention times and UV-absorption spectra did not
Fig. 4. Saponin profile for three concentrated agave sap (CAS) batches at 2 weeks of
storage. B1 (____), B2 (.........), B3 (_._._._). (4) Magueyoside A; (5) kammogenin
tetraglycoside; (6) magueyoside D; (7) manogenin tetraglycoside; (8) manogenin
triglycoside; (9) agavoside C0 .
correspond to the phenolic acid standards used for comparison,
or to the cinnamic acid and benzoic acid derivatives previously
reported in agave products (Almaraz-Abarca et al., 2013).
Therefore, the compounds measured through the Folin–Ciocalteau
method (Table 1) do not necessarily correspond to phenolic
compounds, but rather to other reducing compounds. Reports in
literature have indicated that Maillard reaction products may
cause an overestimation in of total phenols by Folin–Ciocalteau
(Ghedolf et al., 2002).
Compound 10 (Fig. 5b) had maximum absorption at 224 and
288 nm and a pseudomolecular ion m/z 381.10 [M+H]+. Though no
exact identity was obtained, the UV–visible absorption spectra was
similar to those observed for Maillard reaction products obtained
in sugar-amino acid model systems (Yu et al., 2012). Compound 11
(Fig. 5c) presented a UV-maximum at 296 nm and a molecular ion
m/z 144.07 [M]+, and thus identified as 2,3-dihydro-3,5-dihydroxy-
6-methyl-4H-pyran-4-one (DDMP). According to reports in
literature, DDMP has a UV-maxima at 296 nm and a characteristic
m/z 144 molecular ion. Moreover, this compound was also
detected in the dichloromethane extract analyzed by GC–MS,
where the characteristic ions m/z 144, 115, 72, 55 and 43 were
observed (Yu et al., 2013). Compound 12 (Fig. 5d) absorption
spectra had two maxima at 228 and 284 nm and a pseudomole-
cular ion m/z 255.1 [M+H]+, corresponding to the Amadori product
known as furosine (Baptista and Carvalho, 2004). Furosine has
been reported in honey as an indicator of quality (Sanz et al., 2003).
The three peaks were found in B1, B2 and B3, and changes were
observed with storage time. Compound 10 only increased with
respect to time in B1, reaching its maximum at weeks 16 and 18
(Fig. 5e). Concentration of DDMP decreased notably during storage
of the three batches analyzed (Fig. 5f). In the case of B1, furosine
doubled its concentration during 20 weeks of storage but in B2 was
found in lower concentrations and it decreased during storage
(Fig. 5g). The same decreasing trend was observed for B3 but with a
higher initial concentration than the observed for B1 or B2,
reaching nearly half of its initial value. Furosine has been detected
as an early Maillard reaction product in foods that undergo thermal
processing (Silván et al., 2006), and it correlated to antioxidant
capacity in a model system generated from hydrolyzed cereal
proteins and glucose (Zilić et al., 2012). In this study, AOXC
correlated positively with furosine concentration in B1 (r = 0.60,
p-value = 0.0014) but it was not the case for B2 or B3 since no
significant increase in AOXC or furosine was observed.
 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120 119

Table 2
Saponin and sapogenin concentration for three batches (B1–3) of concentrated agave sap at 2- and 20-week storage.

Compound IDa B1 B2 B3

2w 20w 2w 20w 2w 20w

Saponins (PEb mg/g dw) 4 90.4  11.8 ab 80.7  6.7 ab 97.3  4.2 a 97.6  4.1 a 78.0  5.5 b 76.2  2.0 b
5+6 55.2  5.6 c 51.3  2.8 c 207.0  23.7 a 230.0  12.5 a 148.3  20.9 b 140.8  4.7 b
7+8 41.7  1.5 c 39.3  1.5 c 83.8  5.1 a 87.5  8.6 a 63.7  2.8 b 64.3  1.2 b
9 36.9  1.1 b 36.4  1.2 b 46.6  02.3 a 46.7  1.0 a 49.3  2.7 a 49.2  0.6 a

Sum 224.2  19.4 c 207.7  11.7 c 434.7  30.6 a 462.4  10.8 a 339.4  31.6 b 330.5  7.8 b

Sapogenins (HEc mg/g dw) Kamogenin 7.2  0.2 c 7.3  0.4 c 22.5  1.6 b 21.6  1.3 b 24.9  1.7 ab 26.2  1.2 a
Manogenin <5 <5 8.1  0.4 8.3  0.2 <5 <5

Sum 7.2  0.2 c 7.3  0.4 c 30.6  1.6 a 30.0  1.4 a 24.9  1.7 b 26.2  1.2 b

Different letters in each row mean significant statistical differences between results (n = 3, p < 0.05).
a
Compound ID: 4 – magueyoside A; 5 – kammogenin tetraglycoside; 6 – magueyoside D; 7 – manogenin tetraglycoside; 8 – manogenin triglycoside; 9 – agavoside C0 .
b
PE, protodioscin equivalents.
c
HE, hecogenin equivalents (LOD 5 mg/g dw).

In general, one of the most important differences among the
batches was saponin concentration. When CAS had a higher
saponin content, browning index was lower. Browning index
increased more rapidly during the 20 weeks of storage for B1. This
batch had the highest initial browning index compared to B2 and
B3. Browning was positively correlated with antioxidant capacity
and negatively with amino acid concentration, particularly with
LYS, PHE and SER. No phenolic compounds were detected in this
study, but furosine concentration in B1 correlated positively with
antioxidant capacity and browning index.

Fig. 5. (a) Maillard reaction products (compounds 11, 12 and 13) chromatograms obtained at 280 nm after 2 (____) and 20 weeks (.........) of storage, UV–vis spectra of (b)
compound 10, (c) compound 11 – 2,3-dihydro-3,5-dihydroxy-6-methyl-4H-pyran-4-one, and (d) compound 12 – furosine, and AUC changes of compounds (e) 10 (f), 11 (f)
and 12 during 20 weeks of storage (n = 3, statistical significance was 0.05).
120 L. Santos-Zea et al. / Journal of Food Composition and Analysis 45 (2016) 113–120
Conflict of interest
The authors declare that they have no competing financial
interests.
Acknowledgements
The authors offer gratitude and acknowledgements to the
NutriOmics Research Chair Fund from Tecnológico de Monterrey,
Nutrigenomics Research Chair Fund from Fundacion FEMSA and
CONACYT (CVU 270166 and 388427) for financial support for
graduate studies and for this research. We are also grateful to
AGMEL S.A. de C.V. for providing the concentrated agave sap
samples and to Research Center for Protein Development (CIDPRO)
from Tecnológico de Monterrey, Campus Monterrey for technical
aid in protein determination.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jfca.2015.10.005.
References
Almaraz-Abarca, N., González-Elizondo, M.d.S., Campos, M.d.G., Ávila-Sevilla,
Z.E., Delgado-Alvarado, E.A., Ávila-Reyes, J.A., 2013. Variability of the foliar
phenol profiles of the Agave victoriae-reginae complex (Agavaceae). Bot. Sci.
91 (3), 295–306.
Aslanova, D., Bakkalbasi, E., Artik, N., 2010. Effect of storage on 5-
hydroxymethylfurfural (HMF) formation and color change in jams. Int. J.
Food Prop. 13 (4), 904–912, http://dx.doi.org/10.1080/10942910902908896.
Baptista, J.A., Carvalho, R.C., 2004. Indirect determination of Amadori compounds
in milk-based products by HPLC/ELSD/UV as an index of protein
deterioration. Food Res. Int. 37 (8), 739–747, http://dx.doi.org/10.1016/
j.foodres.2004.02.006.
Bastos, D.H.M., Gugliucci, A., 2015. Contemporary and controversial aspects of
the Maillard reaction products. Curr. Opin. Food Sci. 1, 13–20, http://
dx.doi.org/10.1016/j.cofs.2014.08.001.
Bodeiko, V.A., Kintya, P.K., 1975. Steroid saponins VIII. The structure of agave
saponin C0 and D from the leaves of Agave americana. Chem. Nat. Compd. 11
(6), 775–777.
Brudzynski, K., Miotto, D., 2011. Honey melanoidins: analysis of the
compositions of the high molecular weight melanoidins exhibiting radical-
scavenging activity. Food Chem. 127 (3), 1023–1030, http://dx.doi.org/
10.1016/j.foodchem.2011.01.075.
Bulut, L., Kilic, M., 2009. Kinetics of hydroxymethylfurfural accumulation and
color change in honey during storage in relation to moisture content. J. Food
Process. Preserv. 33 (1), 22–32.
Castro-Vázquez, L., Dı́az-Maroto, M.C., González-Viñas, M.A., de la Fuente, E.,
Pérez-Coello, M.S., 2008. Influence of storage conditions on chemical
composition and sensory properties of citrus honey. J. Agric. Food Chem. 56
(6), 1999–2006, http://dx.doi.org/10.1021/jf072227k.
Chen, P.-Y., Chen, C.-H., Kuo, C.-C., Lee, T.-H., Kuo, Y.-H., Lee, C.-K., 2011.
Cytotoxic steroidal saponins from Agave sisalana. Planta Med. 77 (9),
929–933.
Escalante, A., Giles Gomez, M., Hernandez, G., Cordova Aguilar, M.S., Lopez
Munguia, A., Gosset, G., Bolivar, F., 2008. Analysis of bacterial community
during the fermentation of pulque, a traditional Mexican alcoholic beverage,
using a polyphasic approach. Int. J. Food Microbiol. 124 (2), 126–134, http://
dx.doi.org/10.1016/j.ijfoodmicro.2008.03.003.
Fasasi, K.A., 2012. Physicochemical attributes of Nigerian natural honey from
honeybees (Apis mellifera adansonii) (Hymenoptera: Apidae) and its shelf life
in storage at room temperature. Pak. J. Biol. Sci. 15 (21), 1027–1033.
Ghedolf, N., Wang, X.-H., Engeseth, N.J., 2002. Identification and quantification of
antioxidant components of honeys from various floral sources. J. Agric. Food
Chem. 50 (21), 5870–5877.
Gonzales, A.P., Burin, L., Buera, M.D.P., 1999. Color changes during storage of
honeys in relation to their composition and initial color. Food Res. Int. 32
(3), 185–191.
Good-Avila, S.V., Souza, V., Gaut, B.S., Eguiarte, L.E., 2006. Timing and rate of
speciation in Agave (Agavaceae). Proc. Natl. Acad. Sci. U. S. A. 103 (24),
9124–9129.
Gutierrez-Uribe, J.-A., Serna-Salvdivar, S., 2013. Agave syrup having anticancer
activity. USA Patent 8,470,858.
Huang, D., Ou, B., Hampsch-Woodill, M., Flanagan, J.A., Prior, R.L., 2002. High-
throughput assay of oxygen radical absorbance capacity (ORAC) using a
multichannel liquid handling system coupled with a microplate fluorescence
reader in 96-well format. J. Agric. Food Chem. 50 (16), 4437–4444, http://
dx.doi.org/10.1021/jf0201529.
Iglesias, M.T., Martı́n-Álvarez, P.J., Polo, M.C., de Lorenzo, C., González, M., Pueyo,
E., 2006. Changes in the free amino acid contents of honeys during storage
at ambient temperature. J. Agric. Food Chem. 54 (24), 9099–9104, http://
dx.doi.org/10.1021/jf061712x.
Leal-Dı́az, A.M., Santos-Zea, L., Martı́nez-Escobedo, H.C., Guajardo-Flores, D.,
Gutierrez-Uribe, J.A., Serna-Saldivar, S.O., 2015. Effect of Agave americana
and A. salmiana ripeness on saponin content from aguamiel (agave sap). J.
Agric. Food Chem. 63 (15), 3924–3930, http://dx.doi.org/10.1021/acs.jafc.
5b00883.
Martı́nez-Rodrı́guez, J.C., de la Mora-Amutio, M., Plascencia-Correa, L.A., Audelo-
Regalado, E., Guardado, F.R., Hernández-Sánchez, E., Peña-Ramı́rez, Y.J.,
Escalante, A., Beltrán-Garcı́a, M.J., Ogura, T., 2015. Cultivable endophytic
bacteria from leaf bases of Agave tequilana and their role as plant growth
promoters. Braz. J. Microbiol. 45 (4), 1333–1339.
Mancilla-Margalli, N.A., López, M.G., 2002. Generation of Maillard compounds
from inulin during the thermal processing of Agave tequilana Weber var.
Azul. J. Agric. Food Chem. 50 (4), 806–812, http://dx.doi.org/10.1021/
jf0110295.
Mellado-Mojica, E., López, M.G., 2015. Identification, classification, and
discrimination of agave syrups from natural sweeteners by infrared
spectroscopy and HPAEC-PAD. Food Chem. 167, 349–357, http://dx.doi.org/
10.1016/j.foodchem.2014.06.111.
Mina, S.A., Melek, F.R., Abdel-Khalik, S., El-Shaarawy, F.S., Eskander, J., 2014.
Pharmacological activities of Agave seemanniana and isolation of three
steroidal saponins. Eur. J. Med. Plants 4 (3), 271–283.
Montañez-Soto, J.L., Venegas-González, J., Bernardino-Nicanor, A., Ramos-
Ramı́rez, E.G., 2011. Enzymatic production of high fructose syrup from Agave
tequilana fructans and its physicochemical characterization. Afr. J.
Biotechnol. 10 (82), 19137–19143, http://dx.doi.org/10.5897/AJB11.2704.
Narváez-Zapata, J.A., Sánchez-Teyer, L.F., 2009. Agaves as a raw material: recent
technologies and applications. Recent Pat. Biotechnol. 3 (3), 185–191.
Olvera-Garcia, V., Cardador-Martinez, A., Martin del Campo, S.T., 2015. Effect of
thermal process and filtration on the antioxidant activity and
physicochemical parameters of Agave atrovirens extracts. J. Food Res. 4 (1),
155–167, http://dx.doi.org/10.5539/jfr.v4n1p155.
Ortiz-Basurto, R.I., Pourcelly, G., Doco, T., Williams, P., Dornier, M., Belleville, M.-
P., 2008. Analysis of the main components of the aguamiel produced by the
maguey-pulquero (Agave mapisaga) throughout the harvest period. J. Agric.
Food Chem. 56 (10), 3682–3687, http://dx.doi.org/10.1021/jf072767h.
Perez, A.J., Calle, J.M., Simonet, A.M., Guerra, J.O., Stochmal, A., Macı́as, F.A., 2013.
Bioactive steroidal saponins from Agave offoyana flowers. Phytochemistry 95,
298–307.
Rababah, T.M., Al-U’Datt, M., Al-Mahasneh, M., Yang, W., Feng, H., Ereifej, K.,
Kilani, I., Ishmais, M.A., 2014. Effect of jam processing and storage on
phytochemicals and physiochemical properties of cherry at different
temperatures: effect of jam processing and storage. J. Food Process. Preserv.
38 (1), 247–254, http://dx.doi.org/10.1111/j.1745-4549.2012.00770.x.
Santos-Zea, L., Leal-Diaz, A.M., Cortes-Ceballos, E., Gutierrez-Uribe, J.A., 2012.
Agave (Agave spp.) and its traditional products as a source of bioactive
compounds. Curr. Bioact. Compd. 8 (3), 218–231.
Sanz, M.L., del Castillo, M.D., Corzo, N., Olano, A., 2003. 2-Furoylmethyl amino
acids and hydroxymethylfurfural as indicators of honey quality. J. Agric.
Food Chem. 51 (15), 4278–4283, http://dx.doi.org/10.1021/jf021235s.
Selen Burdurlu, H., Karadeniz, F., 2003. Effect of storage on nonenzymatic
browning of apple juice concentrates. Food Chem. 80 (1), 91–97.
Silván, J.M., van de Lagemaat, J., Olano, A., del Castillo, M.D., 2006. Analysis and
biological properties of amino acid derivates formed by Maillard reaction in
foods. J. Pharm. Biomed. Anal. 41 (5), 1543–1551, http://dx.doi.org/10.1016/
j.jpba.2006.04.004.
Torres-Contreras, A.M., Nair, V., Cisneros-Zevallos, L., Jacobo-Velázquez, D.A.,
2014. Plants as biofactories: stress-induced production of chlorogenic acid
isomers in potato tubers as affected by wounding intensity and storage
time. Ind. Crop Prod. 62, 61–66.
Waleckx, E., Gschaedler, A., Colonna-Ceccaldi, B., Monsan, P., 2008. Hydrolysis of
fructans from Agave tequilana Weber var. Azul during the cooking step in a
traditional tequila elaboration process. Food Chem. 108 (1), 40–48, http://
dx.doi.org/10.1016/j.foodchem.2007.10.028.
Wang, X.H., Gheldof, N., Engeseth, N.J., 2004. Effect of processing and storage on
antioxidant capacity of honey. J. Food Sci. 69 (2), fct96–fct101.
Yu, X., Zhao, M., Hu, J., Zeng, S., Bai, X., 2012. Correspondence analysis of
antioxidant activity and UV–vis absorbance of Maillard reaction products as
related to reactants. LWT – Food Sci. Technol. 46 (1), 1–9.
Yu, X., Zhao, M., Liu, F., Zeng, S., Hu, J., 2013. Identification of 2,3-dihydro-3,5-
dihydroxy-6-methyl-4H-pyran-4-one as a strong antioxidant in glucose–
histidine Maillard reaction products. Food Res. Int. 51, 397–403, http://
dx.doi.org/10.1016/j.foodres.2012.12.044.
Zheng, W., Wang, S.Y., 2001. Antioxidant activity and phenolic compounds in
selected herbs. J. Agric. Food Chem. 49 (11), 5165–5170, http://dx.doi.org/
10.1021/jf010697n.
Žilić, S., Akıllıoğlu, G., Serpen, A., Barać, M., Gökmen, V., 2012. Effects of
isolation, enzymatic hydrolysis, heating, hydratation and Maillard reaction
on the antioxidant capacity of cereal and legume proteins. Food Res. Int. 49
(1), 1–6, http://dx.doi.org/10.1016/j.foodres.2012.06.031.