Vous êtes sur la page 1sur 7

Research Article

Received: 25 September 2016 Revised: 27 November 2016 Accepted article published: 13 February 2017 Published online in Wiley Online Library: 10 March 2017

(wileyonlinelibrary.com) DOI 10.1002/jsfa.8269

Positive and negative aspects of green coee


consumption antioxidant activity versus
mycotoxins
Magdalena Jeszka-Skowron,a* Agnieszka Zgoa-Grzeskowiak,a Agnieszka
Waskiewicz,b ukasz Stepienc and Ewa Stanisza

Abstract
BACKGROUND: The quality of coee depends not only on the contents of healthy compounds but also on its contamination with
microorganisms that can produce mycotoxins during development, harvesting, preparation, transport and storage.

RESULTS: The antioxidant activity of green coee brews measured in this study by ABTS, DPPH and FolinCiocalteu assays
showed that coee extracts from Robusta beans possessed higher activity in all assays than extracts from Arabica beans. The
occurrence of ochratoxin A and aflatoxins (B1, B2, G1 and G2) in green coee beans was studied using liquid chromatogra-
phy/mass spectrometry. Apart from mycotoxins, the content of ergosterol as a marker indicating fungal occurrence was also
determined. Among aflatoxins, aflatoxin B1 was the dominant mycotoxin in coee bean samples, with the highest level at
17.45 ng g1 . Ochratoxin A was detected in four samples at levels ranging from 1.27 to 4.34 ng g1 , and fungi potentially pro-
ducing this toxin, namely Aspergillus oryzae, Alternaria sp., Aspergillus foetidus, Aspergillus tamarii and Penicillium citrinum, were
isolated.

CONCLUSION: Steaming and decaeination of coee beans increased antioxidant activities of brews in comparison with those
prepared from unprocessed beans. Although toxins can be quantified in green coee beans and novel fungi were isolated, their
concentrations are acceptable according to legal limits.
2017 Society of Chemical Industry

Keywords: green coee beans; antioxidant activity; ergosterol; mycotoxins; fungal pathogens

INTRODUCTION humidity.15 Approximately 25% of global food and feed crop


Coee is one of the most popular beverages in the world. Coea production is aected by fungi.14 They are associated with many
arabica and Coea canephora var. robusta are the most important diseases such as nephropathy and immunosuppression, which
species of coee. Arabica species come mainly from South Amer- are caused by ochratoxins.16 Aflatoxins are hepatotoxic and show
ica (Brazil) and the uplands and mountain areas of East Africa, carcinogenic eects.17 In the European Union, the legal limit for
while Robusta species come mainly from South Asia (India and ochratoxin A in roasted coee beans is 5 ng g1 .18
Vietnam) and the lowlands of Central and West Africa.1,2 Coee Therefore there is a need for a quick and cheap method of esti-
can be considered as a functional food owing to the presence mation of the quality of coee. Ergosterol could be one indicator
of some active compounds. The most important are chloro- of potential fungal and mycotoxin contamination. Saxena et al.19
genic acids and caeine, which reduce the incidence of cancer, revealed that HPTLC measurement of ergosterol on long-grain
diabetes and liver diseases, protect against Parkinsons disease white rice can be a useful method to detect fungal activity and
and thus decrease mortality risk.3,4 Green coee bean extract ochratoxin production. However, another study showed a lack of
shows health-promoting eects by reducing visceral fat and body correlation between ergosterol content and mycotoxin produc-
weight.5,6 These eects are observed owing to the content of tion in cereal products.20
bioactive compounds7 11 that show antioxidant activity.8,9,12
The quality of coee depends not only on the contents of
Correspondence to: M Jeszka-Skowron, Institute of Chemistry and Technical
healthy compounds but also on its contamination with toxic
Electrochemistry, Poznan University of Technology, Berdychowo 4, PL-60-965
substances.13 Coee cherries (and then beans) can be aected
Poznan, Poland. E-mail: magdalena.jeszka-skowron@put.poznan.pl
by microorganisms (e.g. fungi) that can produce toxins during
development, harvesting, preparation, transport and storage.14 a Institute of Chemistry and Technical Electrochemistry, Poznan University of
Mycotoxins are produced by fungi of various genera, including Technology, Poznan, Poland
Fusarium, Trichoderma, Alternaria, Penicillium, Aspergillus and b Department of Chemistry, Poznan University of Life Sciences, Poznan, Poland
Stachybotrys, under dierent growth conditions, with the level
4022

of production depending on the substrate, temperature and c Institute of Plant Genetics, Polish Academy of Sciences, Poznan, Poland

J Sci Food Agric 2017; 97: 40224028 www.soci.org 2017 Society of Chemical Industry
Positive and negative aspects of green coee consumption www.soci.org

The first aim of this study was to measure the antioxidant (mmol L1 ) using the following linear equation (r2 = 0.998) based
activity of Arabica and Robusta green coee extracts in relation on the calibration curve:
to their species, origin and preparation of beans (decaeination,
steaming and roasting) by three dierent assays. The second A = 36.8660C 0.3932
aim was the determination of ergosterol as a fungal marker, the
isolation of fungi and the quantitation of mycotoxins as potential The results were expressed as mmol Trolox per 100 g green
toxic products of their growth. As a result, both positive and coee beans.
negative aspects of green coee consumption could be assessed.
DPPH assay
The ability of green coee extracts to scavenge DPPH radicals was
MATERIALS AND METHODS estimated according to the method of Blois22 and Jeszka-Skowron
Chemicals and reagents
and Zgoa-Grzeskowiak.23 Briefly, 1 mL of a 0.5 mmol L1 methano-
Ergosterol, aflatoxin (B1, B2, G1 and G2) and ochratoxin A certi-
lic solution of DPPH was mixed with 3 mL of extract (earlier diluted
fied standards were purchased from Sigma-Aldrich (Steinheim,
in methanol). The mixture was left for 30 min at room temperature
Germany). 2,2 -Azinobis(3-ethylbenzothiazoline-6-sulfonic acid)
in the dark. The absorbance of samples was measured at 516 nm
diammonium salt (ABTS), 2,2-diphenyl-1-picrylhydrazyl (DPPH),
using the Beckman spectrophotometer. Antioxidant activity was
potassium persulfate, gallic acid, FolinCiocalteu reagent and
expressed as percentage of DPPH scavenging relative to the con-
all other chemicals were also acquired from Sigma-Aldrich. Mass
trol using the equation
spectrometry (MS)-grade acetonitrile and methanol were sup-
plied by POCH (Gliwice, Poland) and MS-grade formic acid was [(
DPPH scavenging activity (%) = absorbance of control
obtained from Sigma-Aldrich. High-purity deionized (DEMIWA ) ]
5 rosa, Watek, Ledec nad Szavou, Czech Republic) and doubly absorbance of sample absorbance of control 100
distilled (quartz apparatus, Bi18, Heraeus, Hanau, Germany) water
(resistivity 18 M cm) was used throughout the study. Trolox was used as a standard for the calibration curve
(r2 = 0.997). The results were expressed as mmol Trolox per
Green coee beans and brews 100 g green coee beans.
Green coee beans of dierent origin C. arabica: Brazil, Rwanda,
China, Laos and Peru; C. robusta: Vietnam, Vietnam decaeinated FolinCiocalteu assay
(dichloromethane), Vietnam steamed (3 bar pressure for 30 min), The reducing ability of coee extracts was analyzed using
Vietnam roasted (180 C temperature for 8 min), India, Indonesia, FolinCiocalteu reagent.24 Gallic acid was used as an external
Laos, Uganda Sc and Uganda Bugishu were obtained from a standard (r2 = 0.999). The absorbance of samples was measured at
coee roasting plant. The moisture content of the coee beans was 754 nm using the Beckman spectrophotometer. The results were
above 120 g kg1 . expressed as mg gallic acid equivalent (GAE) g1 extract.

Extraction process for determination of antioxidant activity Isolation of fungal strains, DNA extraction, species
Ground beans (0.5 g) were treated with distilled water (20 mL) at identification, polymerase chain reaction (PCR) primers,
94 C for 10 min. The resulting solution was cooled to room tem- cycling profiles, DNA sequencing and analysis
perature, centrifuged (2086 g, 5 min) and decanted. All extracts Coee beans were ground to fine powder which was later used
were prepared immediately before analysis and filtered through a for isolation of fungal strains. Each sample was distributed onto
0.45 m polytetrafluoroethylene (PTFE) syringe filter (Agilent Tech- two 90 mm sterile potato dextrose agar (PDA) plates with strep-
nologies, Santa Clara, CA, USA). tomycin (0.3 g L1 ) and incubated at room temperature for 3 days.
Then individual fungal strains were subcultured via several pas-
sages on PDA medium. Pure mycelia were harvested to Eppen-
ABTS assay
dorf tubes and used for genomic DNA extraction. Genomic DNAs
The ability of green coee extracts to scavenge ABTS radicals was
of all strains were extracted using a modified cetyltrimethylam-
estimated according to the method of Re et al.21 with slight mod-
monium bromide (CTAB) method.25,26 Mycelia of pure cultures
ification. Briefly, 3 mL of ABTS+ solution (prepared 24 h earlier,
grown on solid PDA medium were harvested and homogenized
containing 7 mmol L1 ABTS and 2.45 mmol L1 potassium persul-
using liquid nitrogen. Subsequently, 800 L of CTAB buer with
fate and diluted to an absorbance of 0.70 0.01) was mixed with
4 mL L1 -mercaptoethanol was added, followed by 150 L of
0.03 mL of coee extract. The mixture was left for 6 min at room
chloroform/isoamyl alcohol (24:1 v/v). Samples were incubated at
temperature in the dark. The absorbance of samples against a
65 C for 25 min. After addition of 150 L of chloroform/isoamyl
reagent blank was measured at 734 nm using a UVvisible spec-
alcohol (24:1 v/v), samples were shaken vigorously, left at room
trophotometer (7500DU, Beckman, Brea, CA, USA). Antioxidant
temperature for 10 min and then centrifuged at 12 000 g for
activity was expressed as percentage ABTS scavenging relative to
15 min. The aqueous upper phase was recovered and the DNA was
the control using the equation
precipitated with 1 mL of ice-cold ethanol at 20 C for 20 min. The
[( precipitate was centrifuged at 12 000 g for 15 min, the DNA was
ABTS scavenging activity (%) = absorbance of control
washed with ethanol (700 mL L1 ) and centrifuged at 12 000 g for
) ]
absorbance of sample absorbance of control 100 5 min. Samples were dried and re-dissolved in 200 L of TE buer
(pH 8.0). The primers used are listed in Table 1. ITS4/5 primers
Trolox was used as a standard for the calibration curve. The were chosen to amplify the genomic region encoding the rRNA
4023

ABTS scavenging activity was presented as Trolox equivalents subunits.27 The primers have been validated previously.28 The PCR

J Sci Food Agric 2017; 97: 40224028 2017 Society of Chemical Industry wileyonlinelibrary.com/jsfa
www.soci.org M Jeszka-Skowron et al.

45 V and the confirmatory transition was from m/z 379.3 to 145.1


Table 1. Sequences of PCR primers used in this study
at collision energy set to 22 V.
Amplified Amplicon
Marker 5 3 sequence sequence size (bp) Extraction of ochratoxin A and aflatoxins and determination
of mycotoxins
ITS4 TCCTCCGCTTATTGATATGC ITS1/2 rDNA 600
ITS5 GGAAGTAAAAGTCGTAACAAGG
Ground samples (10 g) were soaked overnight in 30 mL of ace-
tonitrile/water (60:40 v/v) to extract ochratoxin A. After filtration
through Filtrak 389 paper, 10 mL of extract was collected, diluted
was done in 20 L volume using a C-1000 thermal cycler (Bio- with 40 mL of water and filtered through Whatman No. 5 paper.
Rad, Hercules, CA, USA). Each sample contained 1 unit of Phire II The filtrate was applied on top of an Ochra Test immunoanity
HotStart Taq DNA polymerase (Thermo Scientific, Waltham, MA, column (IAC; Vicam, Milford, MA, USA) at a rate of 12 drops s1 .
USA), 4 L of 5 PCR buer, 12.5 pmol of forward/reverse primers, Subsequently, the IAC was washed at the same rate with 10 mL
2.5 mmol L1 of each dNTP and about 1020 ng of fungal DNA. of 0.1 g L1 washing solution (25 g NaCl + 5 g NaHCO3 + 0.1 mL
PCR conditions were as follows: 30 s at 98 C, 35 cycles of 5 s at Tween-20 + 1 L water) followed by 10 mL of distilled water. Ochra-
98 C, 5 s at 58 C and 10 s at 72 C, then 1 min at 72 C. Amplicons toxin A was eluted from the column with 1.5 mL of methanol.
were electrophoresed in 15 g L1 agarose gels (Invitrogen, Thermo Ground samples (10 g) were soaked overnight in 30 mL of
Scientific, Waltham, MA, USA) with GelRed staining (Biotium, Fre- methanol/water (80:20 v/v) to extract aflotoxins. After filtration
mont, CA, USA). through Whatman No. 5 paper, 10 mL of extract was diluted with
For sequence analysis, PCR-amplified DNA fragments were puri- 40 mL of phosphate-buered saline (PBS; pH 7.4). An Afla Test IAC
fied with exonuclease I (Epicentre, Madison, WI, USA) and shrimp (Vicam) was conditioned with 10 mL of PBS. In the next step, the
alkaline phosphatase (Promega, Madison, WI, USA) using the fol- filtrate was applied on top of the IAC at a rate of 12 drops s1 .
lowing program: 30 min at 37 C, followed by 15 min at 80 C. Subsequently, the IAC was washed at the same rate with 10 mL of
Both DNA strands were labeled using BigDyeTerminator 3.1 kit distilled water. Aflatoxins were eluted from the column with 2 mL
(Applied Biosystems, Foster City, CA, USA) according to Stepien of methanol.
et al.28 and the manufacturers instructions. Subsequently, the frag- The solutions were evaporated and the residues reconstituted
ments were precipitated with ethanol to remove the remains of with 1 mL of methanol for chromatographic analysis of mycotox-
the reagents. Sequence reading was performed using Applied ins. The occurrence of aflatoxins and ochratoxin A was analyzed
Biosystems equipment. The sequences of PCR products were com- using a Waters Acquity ultra-performance liquid chromatography
pared with reference accessions obtained from the NCBI GenBank (UPLC) system coupled to a tandem quadrupole (TQ) mass detec-
database using the BLASTn algorithm. tor (Milford, MA, USA) and equipped with Empower 2 software for
data processing. The mass spectrometer was operated in positive
electrospray ionization (ESI+ ) mode using multiple reaction moni-
Analysis of ergosterol and mycotoxin contents using liquid toring (MRM). High-purity nitrogen was used as the ESI nebulizing
chromatography/tandem mass spectrometry (LC/MS/MS)
gas, and argon was used as the collision gas for collision-induced
Ergosterol was determined according to the procedure described dissociation. The mycotoxins were separated on a Waters Acquity
previously.20,29 Briefly, 200 mg of sample was placed in a 12 mL BEH C18 column (2.1 mm 100 mm, 1.7 m particle size) using a
glass culture tube, then 2 mL of methanol was added, fol- mobile phase consisting of 1 mL L1 formic acid in deionized water
lowed by 0.5 mL of 2 mol L1 NaOH. The tube was sealed with (line A) and 1 mL L1 formic acid in acetonitrile (line B) at a flow
a rubber-lined screw cap and placed in a 200 mL screw-capped rate of 0.3 mL min1 . The volume of each injected sample was 3 L.
high-density polyethylene bottle (Bel-Art Products, Wayne, NJ, The ESI+ source had the following parameters: capillary voltage
USA). A Microchef 460 microwave oven (Moulinex, Caen, France) 3.5 kV, cone voltage 3550 V, source temperature 150 C, desolva-
operating at 2450 MHz and 300 W for 20 s was used for irradiation tion temperature 350 C, cone gas flow rate 50 L h1 and desolva-
of the sample. The cooled sample was extracted four times with tion gas flow rate 800 L h1 with nitrogen. The parameters (m/z and
1 mL portions of pentane. Pentane extractions were separated by collision energy) of precursor ions and product ions selected for
centrifugation at 2086 g for 1 min. The combined extracts were the analysis of mycotoxins are shown in Table 2.
evaporated with a gentle stream of nitrogen, reconstituted in
0.5 mL of methanol and filtered through a 0.2 m PTFE syringe fil-
ter. The samples were analyzed using the chromatographic system Statistical analysis
UltiMate 3000 RSLC from Dionex (Sunnyvale, CA, USA) coupled Results are expressed as mean standard deviation (of at least
with the API 4000 QTRAP triple quadrupole mass spectrometer three replicates). Analysis of variance and significant dierences
from AB Sciex (Foster City, CA, USA) using the atmospheric pres- among means and correlation analysis were performed with
sure chemical ionization (APCI) interface. Samples (10 L) were one-way ANOVA. The significance level was based on a confidence
injected into a Gemini-NX C18 column (100 mm 2.0 mm i.d., level of 95.0%. The experimental data were analyzed using Statis-
3 m) from Phenomenex (Torrance, CA, USA) maintained at 35 C. tica 12.0 (StatSoft Inc., Tulsa, OK, USA).
The isocratic mobile phase employed during the analysis con-
sisted of methanol/water (95:5 v/v) at a flow rate of 0.4 mL min1 .
The APCI source operated in positive ion mode. The following set- RESULTS AND DISCUSSION
tings for the ion source and mass spectrometer were used: curtain Antioxidant activity of coee extracts
gas 10 psi, nebulizer gas 20 psi, temperature 400 C, nebulizing More than one method should be used to determine the antiox-
current 3 A and collision gas 10 psi. The de-clustering potential idant activity of food samples, because each individual method
was 65 V and the dwell time was set to 200 ms. The quantitative shows antioxidant activity in a dierent way. Therefore the use of at
4024

transition was from m/z 379.3 to 69.1 at collision energy set to least two methods based on dierent reaction mechanisms leads

wileyonlinelibrary.com/jsfa 2017 Society of Chemical Industry J Sci Food Agric 2017; 97: 40224028
Positive and negative aspects of green coee consumption www.soci.org

Table 2. Optimized parameters of LC/MS/MS for each mycotoxin

Quantification ion Confirmation ion


Mycotoxin Precursor ion (m/z) Production (m/z) Collision energy (eV) Product ion (m/z) Collision energy(eV)

Ochratoxin A 404.0 358.0 45 239.0 43


Aflatoxin B1 313.1 241.1 40 285.1 35
Aflatoxin B2 315.1 287.0 36 259.0 38
Aflatoxin G1 329.0 243.0 28 311.1 30
Aflatoxin G2 331.1 313.3 30 245.2 32

to a better estimation of the antioxidant activity of a sample.30 ochratoxigenic fungi, but only 22% of these were contaminated
In this study, the antioxidant activity of green coee extracts was with ochratoxin A.37 Leong et al.38 reported that Robusta beans
measured by the three most common assays: FolinCiocalteu from Vietnam were more contaminated with fungi than Arabica
(F-C), ABTS and DPPH. These methods are widely used to deter- beans. Aspergillus niger infected 89% of Robusta beans, whereas
mine the in vitro antioxidant activity of foods and beverages.30 32 Aspergillus carbonarius and other yellow aspergilli infected
Coee extracts from green beans of Robusta showed higher 1214% of beans. Other researchers reported that one-third
antioxidant activities than Arabica extracts (Table 3). Coee of 18 isolates of Aspergillus flavus produced aflatoxins B1 and B2.37
extracts from Vietnam (particularly the processed Robusta beans)
possessed the highest activity in all assays. Arabica from Laos Ergosterol
showed the lowest antioxidant activity in all tests. Similar results Ergosterol is present in fungal cell walls and therefore can be used
were reported by Naidu et al.33 as an indicator of fungal contamination at the present time or in
Arabica coee is believed to be of better quality than Robusta the past.20 Thus, even in dead fungal biomass, it can be isolated
especially regarding flavor and taste. Steaming of Robusta coee from a sample and quantified. In this study, ergosterol was found in
is a method used to improve the flavor and create the specific every tested sample (Table 4), which clearly shows the widespread
acidic taste unique to Arabica. Decaeination of coee resulted presence of fungi in coee beans. It also proves that proper storage
in higher activities in ABTS and DPPH assays in comparison with and transport conditions of green coee are of great importance
Vietnamese coee by 11 and 15% respectively. Similarly, steaming for reducing the possibility of fungal growth.
of coee increased antioxidant activities in ABTS and DPPH assays It was found that the ergosterol content of Arabica coee sam-
in comparison with Vietnamese coee by 7 and 23% respectively. ples was on average lower than that of Robusta samples (Table 4).
Roasting of coee decreased antioxidant activities, especially in Two Arabica samples contained about 200 ng g1 of ergosterol and
ABTS and F-C assays, by 28.4 and 31.6% respectively. Literature three other samples contained between 1000 and 2000 ng g1 .
data showed that roasting can reduce the antioxidant activity of Most Robusta samples contained between 1000 and 3000 ng g1
coee extract by as much as 3-fold.31 of ergosterol. The only Robusta sample containing a low amount
Using Pearsons linear correlation coecient, a high positive cor- of ergosterol was the Vietnamese roasted coee, which shows
relation (r2 = 0.956) was found between ABTS and DPPH antioxi- that roasting of coee beans greatly reduces their ergosterol con-
dant activity assays of green coee extracts. Positive correlations tent. For the tested Vietnamese coee beans, a 90% reduction
between ABTS and F-C assays (r2 = 0.867) and DPPH and F-C assays in ergosterol was noted after coee roasting. On the other hand,
(r2 = 0.823) were also recorded. Similar correlations between total one of the analyzed Robusta samples contained a very high level
phenolic compounds and F-C and DPPH assays were reported in of ergosterol. Indonesian green coee beans contained about 14
the literature.32,34 The F-C assay is used to determine the amount 000 ng g1 of ergosterol, which clearly shows that this sample was
of phenolic compounds in plant extracts.34 The method is not ded- heavily contaminated with fungi. High contamination with fungal
icated to these compounds and detects also other reducing com- toxins could also be expected in this sample (Table 4).
pounds such as Maillard condensation products and tryptophan.35
However, regardless of the dierences between the methods used
Mycotoxins
in this study, high correlation between results has been obtained.
The most harmful mycotoxins aflatoxins B1, B2, G1 and G2 and
ochratoxin A were identified and quantified using LC/MS/MS
Fungi analysis. These fungal compounds cause low performance, sick-
It has been shown that mycotoxigenic Aspergillus, Penicillium and ness or even death when ingested, inhaled or absorbed.
Fusarium spp. are present throughout the entire coee production Aflatoxin B1 was the dominant mycotoxin in coee samples,
process chain.14 The molecular identification of isolated fungal and the highest level was detected in Laotian green beans of
strains originating from coee samples is summarized in Table 4. Robusta coee (Table 4). All determined mycotoxins were found
Pathogens from the Aspergillus genus were the most common con- in Chinese green Arabica beans. Aflatoxins B1 and G1 were the
taminating agents in the green coee beans tested. Ochratoxin most abundant among the four selected aflatoxins in green coee
A was detected in green coee beans, and potentially ochratox- beans, a finding consistent with Bokhari39 and Bokhari and Aly.40
igenic fungi Aspergillus oryzae, Alternaria sp., Aspergillus foetidus, The level of these aflatoxins was lower in comparison with previous
Aspergillus tamarii and Penicillium citrinum were isolated. reports.39,40
Dierent results for isolated fungi have been reported by Ochratoxin A was found in two samples of Robusta green coee
Rezende et al.36 and Batista et al.37 The latter authors found beans originating from Vietnam and Uganda. Arabica green coee
4025

that 58% of green coee beans were infected with potentially beans from Laos and China contained ochratoxin A at high levels,

J Sci Food Agric 2017; 97: 40224028 2017 Society of Chemical Industry wileyonlinelibrary.com/jsfa
www.soci.org M Jeszka-Skowron et al.

Table 3. Antioxidant activity of coee extracts measured by ABTS, DPPH and F-C assays

Coee type Coee origin ABTS (mmol Trolox per 100 g) DPPH (mmol Trolox per 100 g) F-C (mg GAE g1 )

Robusta Vietnam GR2 green 228.26 8.24e 37.60 0.46d 68.01 0.27 g
Vietnam decaeinated green 253.56 1.17 g 43.24 2.02e 64.90 0.68f
Vietnam steamed 270.36 8.98 h 46.23 0.54f 66.60 0.27f
Vietnam GR2 roasted 163.51 9.15c 34.26 1.23c 46.49 0.78c
India Cherry green 239.70 3.21f 42.70 0.72e 62.08 0.95e
Laos green 220.23 2.27e 32.14 2.62c 59.47 1.60e
Indonesia green 194.70 3.96d 32.55 0.29c 57.27 1.31de
Uganda Sc12 green 244.17 4.32f 44.40 1.26ef 56.93 0.52d
Uganda Bugishu green 193.25 2.40d 33.56 0.67c 50.81 0.88b
Arabica Brazil green 151.48 2.79c 22.24 0.82b 15.41 1.39a
Rwanda green 140.21 8.05a 21.39 1.40b 17.37 1.96a
China green 154.89 4.07c 23.41 1.81b 15.76 0.95a
Laos green 134.87 0.85a 16.13 2.57a 14.17 0.52a

Mean values with dierent letters in a column are significantly dierent by Tukeys test (P 0.05).

Table 4. Concentrations of ergosterol and mycotoxins in selected coees and molecular identification of fungal species isolated from tested products

Coee type Coee origin Ergosterol (ng g1 ) Aflatoxins B1/B2/G1/G2 (ng g1 ) Ochratoxin A (ng g1 ) Identified fungal strains

Robusta Vietnam 833 85 4.87 1.02/ND/4.52 0.91/ND ND Aspergillus foetidus Aspergillus


decaeinated tamarii Cladosporium
green macrocarpum
Vietnam GR2 2218 156 9.28 1.07/ND/ND/ND 2.06 0.13 Alternaria sp.
green
Vietnam GR2 229 8 ND/ND/ND/ND ND Aspergillus tamarii Aspergillus
roasted foetidus
Laos green 3407 104 17.45 2.11/ND/5.23 0.62/ND ND Alternaria sp.
Uganda Bugishu 934 44 ND/ND/ND/ND ND
green
Uganda Sc12 2385 139 5.53 0.66/ND/ND/ND 1.27 0.08 Aspergillus oryzae
green
Indonesia green 13954 324 10.35 0.93/ND/9.28 1.84/ND ND Aspergillus foetidus Aspergillus
tamarii Phoma sp.
India Cherry 900 22 7.11 1.45/ND/ND/ND ND Aspergillus foetidus Aspergillus
green tamarii
Arabica Laos green 194 11 6.16 0.74/1.11 0.09/5.73 4.34 0.32 Aspergillus tamarii Penicillium
0.49/ND citrinum
Brazil green 979 71 1.83 0.42/ND/0.82 0.07/ND ND
Rwanda green 1202 64 2.91 0.40/ND/ND/ND ND
China green 173 14 11.03 0.64/1.44 0.08/8.14 1.57 0.33 Aspergillus foetidus Aspergillus
0.71/0.95 0.05 oryzae
Peru green 2088 175 3.55 0.21/ND/2.41 0.33/ND ND Fusarium proliferatum Fusarium
verticillioides Aspergillus
tamarii Penicillium
chrysogenum Aspergillus
foetidus

ND, not detected.

but not exceeding the legal limits. Leong et al.38 reported that the 14 000 ng g1 of ergosterol was expected to be highly contami-
level of ochratoxin A in coee beans from Vietnam was very similar, nated with mycotoxins. Indeed, it contained 10 ng g1 of aflatoxin
1.8 ng g1 , while Bokhari39 observed that the ochratoxin A content B1 and 9 ng g1 of aflatoxin G1. Nevertheless, there were two
ranged between 3.77 and 25.9 ng g1 in Saudi Arabian coee. other samples (Laotian green Robusta and Chinese green Arabica)
Apart from reducing ergosterol, roasting of coee beans containing more aflatoxins despite the fact that they contained
also decreased mycotoxin contents (Table 4). A reduction in smaller amounts of ergosterol. Laotian green Robusta contained
mycotoxin contamination during roasting was also observed 3400 ng g1 of ergosterol and Chinese green Arabica contained
previously.39,41 43 only 173 ng g1 of the metabolite. Shantha44 reported that some
No correlation was found between ergosterol and mycotoxin fungal species reduce aflatoxin B1 contamination. One such
4026

contents. The Indonesian green coee sample containing about species, i.e. Phoma sp., was identified in the Indonesia green

wileyonlinelibrary.com/jsfa 2017 Society of Chemical Industry J Sci Food Agric 2017; 97: 40224028
Positive and negative aspects of green coee consumption www.soci.org

coee beans (Table 4), and its presence could lead to a reduction 8 Jeszka-Skowron M, Sentkowska A, Pyrzynska K and De Pea MP,
in aflatoxin contamination, while ergosterol content was still high. Chlorogenic acids, caeine content and antioxidant properties of
green coee extracts: influence of green coee bean preparation.
The presence of a fungal strain not producing aflatoxins could be Eur Food Res Technol 242:14031409 (2016).
another reason for high ergosterol content in this sample, even 9 Jeszka-Skowron M, Stanisz E and De Pea MP, Relationship between
if this strain was not viable anymore (and therefore not isolated). antioxidant activity, chlorogenic acids and elemental composition
This clearly shows that ergosterol cannot be used to predict of green coee. LWT Food Sci Technol 73:243250 (2016).
10 Jeszka-Skowron M, Zgoa-Grzeskowiak A and Grzeskowiak T, Analytical
high contamination of green coee samples with aflatoxins and methods applied for the characterization and the determination
ochratoxin A. of bioactive compounds in coee. Eur Food Res Technol 240:1931
(2015).
11 Perrone D, Farah A, Donangelo CM, de Paulis T and Martin PR, Com-
prehensive analysis of major and minor chlorogenic acids and lac-
CONCLUSIONS tones in economically relevant Brazilian coee cultivars. Food Chem
Robusta coee brews possessed higher antioxidant activity in 106:859867 (2008).
12 Shahidi F and Chandrasekara A, Hydroxycinnamates and their in vitro
ABTS, DPPH and F-C assays than Arabica. Steaming of Robusta cof- and in vivo antioxidant activities. Phytochem Rev 9:147170 (2010).
fee increased antioxidant activities in ABTS and DPPH assays in 13 Graca-Moraleja A, Font G, Maes J and Ferrer E, Development of a new
comparison with coee extracts from unprocessed beans. Simi- method for the simultaneous determination of 21 mycotoxins in cof-
larly, decaeination of coee resulted in higher activity in ABTS and fee beverages by liquid chromatography tandem mass spectrome-
try. Food Res Int 72:247255 (2015).
DPPH assays in comparison with unprocessed coee. On the other 14 Paterson RRM, Lima N and Taniwaki MH, Coee, mycotoxins and
hand, roasting of beans decreased the antioxidant activity. climate change. Food Res Int 61:115 (2014).
Potentially ochratoxigenic fungi A. oryzae, Alternaria sp., A. 15 Paterson RRM and Lima N, How will climate change aect mycotoxins
foetidus, A. tamarii and P. citrinum were isolated. Aspergillus niger in food? Food Res Int 43:19021914 (2010).
infected 89% of Robusta beans. Robusta green coee beans origi- 16 Nuru AA, Occurrence, harmful eects and analytical determination of
ochratoxin A in coee. J Appl Pharmaceut Sci 5:120127 (2015).
nating from Vietnam and Uganda as well as Arabica green coee 17 Wild CP and Gong YY, Mycotoxins and human disease: a largely ignored
beans from Laos and China contained ochratoxin A at higher global health issue. Carcinogenesis 31:7182 (2010).
levels than other analyzed samples, but not exceeding the legal 18 European Commission, Commission Regulation (EC) No 1881/2006 of
limits. 19 December 2006 setting maximum levels for certain contaminants
in foodstus [Online]. Available: http://eur-lex.europa.eu/ [28 Febru-
The content of ergosterol as a fungal marker was studied. Its ary 2017].
presence was confirmed in every tested sample. Nevertheless, 19 Saxena J, Munimbazi C and Bullerman LB, Relationship of mould
no correlation between ergosterol and mycotoxin contents was count, ergosterol and ochratoxin A production. Int J Food Microbiol
found. Thus this marker cannot be used to assess contamination 71:2934 (2001).
20 Stanisz E, Zgoa-Grzeskowiak A, Waskiewicz A, Stepien and Beszterda
with mycotoxins. M, Can ergosterol be an indicator of Fusarium fungi and mycotoxins
To summarize, antioxidant activity of green coee brews, espe- in cereal products? J Braz Chem Soc 26:705712 (2015).
cially from C. canephora var. robusta, makes these products valu- 21 Re R, Pellegrini N, Proteggente A, Pannala A, Yang M and Rice-Evans
able for consumers. Although toxins can be quantified in green C, Antioxidant activity applying an improved ABTS radical cation
decolorization assay. Free Radic Biol Med 26:12311237 (1999).
coee beans, their concentrations are acceptable according to cur- 22 Blois MS, Antioxidant determinations by the use of a stable free radical.
rent legal limits. Nature 26:11991200 (1958).
23 Jeszka-Skowron M and Zgoa-Grzeskowiak A, Analysis of antioxidant
activity, chlorogenic acid, and rutin content of Camellia sinensis
infusions using response surface methodology optimization. Food
ACKNOWLEDGEMENTS Anal Meth 7:20332041 (2014).
This work was supported by grant 03/31/DSMK/0326 from the 24 Singleton VL and Rossi JA, Colorimetry of total phenolics with
phosphomolybdic-phosphotungstic acid reagents. Am J Enol Vitic
Polish Ministry of Science and Higher Education. The authors 16:144158 (1965).
would also like to thank Strauss Cafe Poland for coee samples. 25 Doohan FM, Parry DW, Jenkinson P and Nicholson P, The use of
species-specific PCR-based assays to analyse Fusarium ear blight of
wheat. Plant Pathol 47:197205 (1998).
26 Waskiewicz A and Stepien , Mycotoxins biosynthesized by
REFERENCES plant-derived Fusarium isolates. Arch Ind Hyg Toxicol 63:437446
1 ICO, International Coee Organization Annual Review 2011/12. ICO, Lon- (2012).
don (2013). 27 White TJ, Bruns T, Lee S and Taylor J, Amplification and direct sequenc-
2 FAO/WHO, Discussion Paper on Ochratoxin A in Coee [Online]. (2008). ing of fungal ribosomal RNA genes for phylogenetics, in PCR Proto-
Available: ftp://ftp.fao.org/codex/meetings/CCCF/CCCF2/cf02_14e cols: a Guide to Methods and Applications, ed. by Innis MA, Gelfand
.pdf [28 February 2017]. DH, Sninsky JJ and White TJ. Academic Press, San Diego, CA, pp.
3 Bhupathiraju SN, Pan A, Malik VS, Manson JE, Willett WC, van Dam 315322 (1990).
RM et al., Caeinated and caeine-free beverages and risk of type 28 Stepien , Koczyk G and Waskiewicz A, FUM cluster divergence in
2 diabetes. Am J Clin Nutr 97:155166 (2013). fumonisins-producing Fusarium species. Fungal Biol 115:112123
4 Cano-Marquina A, Tarn JJ and Cano A, The impact of coee on health. (2011).
Maturitas 75:721 (2013). 29 Horbik D, owinska-Kluge A, Grski Z, Stanisz E and Zgoa-Grzeskowiak
5 Onakpoya I, Terry R and Ernst E, The use of green coee extract as A, Microwave-assisted extraction combined with HPLC-MS/MS for
a weight loss supplement: a systematic review and meta-analysis diagnosis of fungal contamination in building materials. J Braz Chem
of randomised clinical trials. Gastroenterol Res Pract 2011:382852 Soc 24:14781486 (2013).
(2011). 30 Pekal A, Drzd z P, Biesaga M and Pyrzynska K, Screening of the
6 Shimoda H, Seki E and Aitani M, Inhibitory eect of green coee bean antioxidant properties and polyphenol composition of aromatised
extract on fat accumulation and body weight gain in mice. BMC green tea infusions. J Sci Food Agric 92:22442249 (2012).
Compl Altern Med 17:69 (2006). 31 Nebesny E and Budryn G, Antioxidative activity of green and roasted
7 Gmez-Ruiz JA, Leake DS and Ames JM, In vitro antioxidant activity coee beans as influenced by convection and microwave roasting
of coee compounds and their metabolites. J Agric Food Chem methods and content of certain compounds. Eur Food Res Technol
4027

55:69626929 (2007). 217:157163 (2003).

J Sci Food Agric 2017; 97: 40224028 2017 Society of Chemical Industry wileyonlinelibrary.com/jsfa
www.soci.org M Jeszka-Skowron et al.

32 Stelmach E, Pohl P and Szymczycha-Madeja A, The content of Ca, Cu, 38 Leong SL, Hien LT, An TV, Trang NT, Hocking AD and Scott ES, Ochra-
Fe, Mg and Mn and antioxidant activity of green coee brews. Food toxin A-producing Aspergilli in Vietnamese green coee beans. Lett
Chem 182:302308 (2015). Appl Microbiol 45:301306 (2007).
33 Naidu M, Sulochanamma G, Sampathu SR and Srinivas P, Studies on 39 Bokhari FM, Mycotoxins and toxigenic fungi in Arabic coee beans in
extraction and antioxidant potential of green coee. Food Chem Saudi Arabia. Adv Biol Res 1:5666 (2007).
107:377384 (2008). 40 Bokhari FM and Aly MM, Evolution of traditional means of roasting and
34 Prez-Hernndez LM, Chvez-Quiroz K, Medina-Jurez L and mycotoxins contaminated coee beans in Saudi Arabia. Adv Biol Res
Gmez Meza N, Phenolic characterization, melanoidins, and 3:7178 (2009).
antioxidant activity of some commercial coees from Coea 41 Napolitano A, Fogliano V, Tafuri A and Ritieni A, Natural occurrence of
arabica and Coea canephora. J Mex Chem Soc 56:430435 ochratoxin A and antioxidant activities of green and roasted coees
(2012). and corresponding byproducts. J Agric Food Chem 55:1049910504
35 Everette JD, Bryant QM, Green AM, Abbey YA, Wangila GW and (2007).
Walker RB, Thorough study of reactivity of various compound 42 Prez De Obanos A, Gonzlez-Peas E and Lpez De Cerain A, Influence
classes toward the FolinCiocalteu reagent. J Agric Food Chem of roasting and brew preparation on the ochratoxin A content in
58:81398144 (2010). coee infusion. Food Addit Contam 22:463471 (2005).
36 Rezende EF, Borges JG, Cirillo M, Prado G, Paiva LC and Batista LR, 43 Drunday V and Pacin A, Occurrence of Ochratoxin A in coee beans,
Ochratoxigenic fungi associated with green coee beans (Coea ground roasted coee and soluble coee and method validation.
arabica L.) in conventional and organic cultivation in Brazil. Braz J Food Control 30:675678 (2013).
Microbiol 44:377384 (2013). 44 Shantha T, Fungal degradation of aflatoxin B1. Nat Toxins 7:175178
37 Batista LR, Chalfoun SM, Prado G, Schwan RF and Wheals AE, Toxigenic (1999).
fungi associated with processed (green) coee beans (Coea arabica
L.). Int J Food Microbiol 25:293300 (2003).
4028

wileyonlinelibrary.com/jsfa 2017 Society of Chemical Industry J Sci Food Agric 2017; 97: 40224028

Vous aimerez peut-être aussi