Postharvest Biology and Technology 86 (2013) 294–299

Contents lists available at ScienceDirect

Postharvest Biology and Technology

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

Spontaneous postharvest fermentation of açai (Euterpe oleracea) fruit

Fagner Aguiar, Verena Menezes, Hervé Rogez ∗
Faculdade de Engenharia de Alimentos & Centre for Agro-food Valorization of Amazonian Bioactive Compounds (CVACBA), Universidade Federal do Pará,
Av. Perimetral s/n, CEP: 66.095-780, Belém, PA, Brazil1

a r t i c l e i n f o a b s t r a c t

Article history: Açai (Euterpe oleracea) fruit (EOF) are widely commercialized in the Brazilian Amazon. These fruit contain
Received 11 September 2012 a high bacterial load and are transported on boards stowed inside or outside the holds of small boats. In
Accepted 6 July 2013 this context, postharvest parameters were assessed under conditions that simulated these two methods
of EOF transport: stowage in closed polystyrene boxes, simulating the inside of cargo holds, i.e., transport
Keywords: in a closed system; and open baskets, simulating transport in an open environment, i.e., transport in the
Euterpe oleracea
prow or bow of the boat. EOF suffered spontaneous fermentation of alcoholic, acetic, and lactic types in
Spontaneous fermentation
the closed system, which is the most common type of transportation of this fruit. In the closed system,
Transport
Respiration rate
there was a predominance of lactic acid bacteria over acetic acid bacteria, with 82% and 95% of the initial
SCFA content of d-glucose and d-fructose being consumed, respectively, after 27 h of experiment. The weight
Phenolic compounds loss reached 1.7% and there was a logarithmic decrease of the major phenolic compounds of the fruit
in the closed system, with losses of 78% of cyanidin-3-rutinoside, 88% of cyanidin-3-glucoside, 78% of
homorientin, and 72% of orientin after 27 h, which was higher than in the open system (58%, 66%, 73%
and 62%, respectively). Analyses on EOF stowed in a closed system indicated that the respiratory rate
was characteristic of a non-climacteric fruit, i.e., it showed a logarithmic decay in the production of CO2
(R2 = 0.995; P < 0.05). Thus, transport in a closed system results in more drastic nutritional and functional
changes on EOF than when transport is carried out in an open system, suggesting that transportation in
continuous aerobic conditions and a short period of time between picking and processing are preferable.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction level in the United States, Japan and Europe, for use in açai drinks,
because of the fruit’s nutritional value and antioxidant quality
Euterpe oleracea is a palm tree that is widespread in northern (Bichara and Rogez, 2011). The antioxidant quality of the açai
South America, with its greatest occurrence and economic impor- drink is due to high concentrations of phenolic compounds, mainly
tance in the floodplains of the Amazonian delta. E. oleracea fruit cyanidin-3-rutinoside (C-3-R) and cyanidin-3-glucoside (C-3-G),
(EOF), known as açai, are produced by the palm tree in bunches which come to represent 30% of phenolic compounds and are
beginning in the tree’s third year. Each fruit is a sessile stone fruit responsible by the purple colour, and orientin and homorientin,
with a woody endocarp, round in shape, with a diameter of 1–2 cm, secondarily (Rogez et al., 2011).
and mass varying from 0.8 to 2.3 g. The green and tinga varieties are The local supply chain of EOF (Fig. 1) has potential for micro-
green before ripening and turn a pale green colour after ripening. biological hazards, because the fruit undergo much handling
In the case of the black variety, the most common one, the fruit throughout the supply chain (Rogez et al., 2012). The data report
turns a purple/violet colour (Bichara and Rogez, 2011). The produc- that the fruit of this palm have a basal level of microorgan-
tion of EOF in Brazil in 2011 was 215,000 tons, with 109,000 tons isms. The critical microbiological points are during threshing and
originating in the state of Pará (IBGE, 2011), and this produce was transport (up to 30 h) in the holds of boats, and during sales in
transported to the main commercial centres of the Brazilian Ama- the markets of Amazonian cities (commercial centres), such as
zon mainly in small boats. Belém, the capital of the state of Pará (Rogez, 2000). These condi-
In the past 15 years, a boom on marketing this fruit has occurred, tions favour spontaneous fermentation processes in tropical fruits
not only on the Brazilian market, but also at the international because the pH (between 3 and 5) and the sugar and water content
are favourable to microbiological action. The respiratory activity of
tropical fruits also favours fermentation, because they exhibit high
∗ Corresponding author.
oxygen consumption and carbonic gas dissipation, progressively
E-mail addresses: [email protected], [email protected] (H. Rogez).
making the environment anaerobic and prone to fermentation
1
Tel.: +55 91 3201 7456; fax: +55 91 3201 7456. (Montet et al., 2004). Therefore, the aim of this paper was to

0925-5214/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.postharvbio.2013.07.015
 F. Aguiar et al. / Postharvest Biology and Technology 86 (2013) 294–299 295

evaluate the metabolic parameters and the type(s) of spontaneous 1.80

fermentation(s) occurring in EOF during postharvest, particularly 1.60
during transportation to commercial centres.
1.40
1.20 Fruit in closed system
2. Materials and methods

%Weight loss
Fruit in open system
1.00
2.1. Fruit 0.80
0.60
EOF (126 kg – with 1 kg corresponding to about 770 fruit) were
collected in Abaetetuba, close to Belém (Eastern Brazilian Ama- 0.40
zon) and immediately transported to the laboratory; the total time 0.20
between picking and arrival was 10 h. The fruit were harvested at
0.00
the fully mature stage, as recommended by Rogez et al. (2011). 0 5 10 15 20 25 30
Time (h)
2.2. Fermentation conditions of EOF
Fig. 2. Weight loss of Euterpe oleracea fruit over 27 h. The vertical bar is standard
The two most common conditions of transport of the EOF deviation (n = 3).
on boats were simulated as follows: in closed polystyrene boxes
(closed system, such as in boat cargo holds), and in open baskets other samples were stored at −20 ◦ C for the analysis of sugars, and
(open system, such as on the open decks of boats). Ten kilogram for chromatographic analysis. Thermodynamic, weight loss, and
portions of EOF were stacked in each of eight polystyrene boxes, respiration determinations were directly performed on EOF.
and ten additional 1.8-kg portions of EOF were stacked in an open
basket, for the simulation of the first and second conditions of trans- 2.4. Weight loss
portation, respectively, for 27 h. In the open basket, the upper and
lower portions were not analysed, as they were not representative This parameter was monitored by the gravimetric method. Ini-
of all fruit. tially, the fruit were weighed into synthetic nets in 1.8-kg portions,
which were then placed inside polystyrene boxes and open bas-
2.3. Fermentation kinetics and sample preparation kets among free EOF. Then, at each sampling time, the nets were
retrieved and weighed on a balance with an accuracy of 0.01 g.
From both open and closed conditions, 1.8-kg samples of EOF The temperature of the EOF was monitored using thermometers
were collected after 0, 3, 7 (in triplicate), 13, 20, and 27 h. Each fruit (±0.2 ◦ C; Model 2292; Incoterm, São Paulo, Brazil) located at the
sample was softened in water at 45 ◦ C for 1 h, and then mechani- centre of the stack of fruit under both experimental conditions.
cally pulped in a 1:1 ratio of water: fruit according to the method
of Pompeu et al. (2009b) (Fig. 1). The pulping machine was washed 2.5. Estimation of the fruit’s specific heat and enthalpy of
and sterilized routinely using 70% alcohol. Microbiological anal- fermentation
yses were immediately performed on the fresh juice. Meanwhile
To estimate specific heat, the method of calorimetric mixtures
of Shrivastava and Datta (1999) was used, as follows: 150 g EOF
was mixed with 250 g water contained in an adiabatic calorimeter.
Picking
At equilibrium, the temperature of the mixture was measured. The
specific heat of the EOF was estimated using Eq. (1). This determi-
nation was carried out in triplicate.
Threshing
(Cc + Mw ∗ cw ) ∗ (Te − Tw )
cp = (1)
Mf ∗ (Tf − Te )
Transport of the
fruit where cp is the specific heat of EOF; Cc is the heat capacity of
calorimeter; Mw is the mass of water; cw is the specific heat of
water; Mf is the exact mass of EOF; Te , Tw and Tf are the tem-
Distribution peratures of equilibrium, water and fruit before of the mixture,
respectively.
The enthalpy (H) of spontaneous fermentation was cal-
Washing of the fruit culated using Eq. (2), from the temperature variation of EOF
(T = Tfinal − Tinitial ) during the fermentation at constant pressure.

H = cp ∗ T (2)
Pulp Softening
2.6. Determination of respiration rate

Pulp crushing and The respiration rate of the EOF was only measured in the closed
filtration system (Bhande et al., 2008). The percentages of O2 and CO2 inside
the boxes were obtained through a gas analyser (Eurotron/Green
Line 8000; Milan, Italy). Eqs. (3) and (4) were used to obtain the
respiration rate.
Açai juice  
(GO2 )t − (GO2 )t+1 V
RO2 = ∗ (3)
t M
Fig. 1. Flow chart of the postharvest steps and of the processing of açai juice.
296 F. Aguiar et al. / Postharvest Biology and Technology 86 (2013) 294–299

 
(GCO2 )t+1 − (GCO2 )t V Helium was used as the carrier gas. The temperature ramp of the
RCO2 = ∗ (4) column was set as follows: 100–150 ◦ C at a rate of 0.60 ◦ C s−1 ,
t M
increasing to 200 ◦ C at a rate of 0.16 ◦ C s−1 , and maintained at this
where R is the respiration rate of O2 or CO2 (mL kg−1 h−1 ), G is the temperature for 2 min for SCFA and lactic acid; and 50–130 ◦ C with
percentage fraction of gases, t is the time (h), V is the free volume a rate of 0.60 ◦ C s−1 for ethanol. In both methods, the tempera-
inside polystyrene boxes (mL), and M is the mass of the fruit (kg). tures of the injector and the detector were set to 240 ◦ C and 260 ◦ C,
Then, respiration rates were converted to mmol kg−1 s−1 . This was respectively (Brunetto et al., 2009; Sabatini et al., 2008).
done using the formula V = (nRT)/P, where P is the pressure (kPa), n
is the number of moles, R = 8.314 (the ideal gas constant), T is the 2.10. Quantification of the major phenolic compounds by
absolute temperature (K) and V is the volume (L). high-performance liquid chromatography (HPLC)

2.7. Microbiology The major phenolic compounds of açai (C-3-R, C-3-G, orientin,
and homorientin) were assessed using a HPLC system (LC-10Avp
Açai juice from closed system samples was subjected to 10× series; Shimadzu, Tokyo, Japan), with the conditions of the column
dilutions in peptone and subsequently plated in plate count and the separation and quantification of each compound conducted
agar (PCA; Difco, Detroit, MI) at 36 ◦ C for 48 h (Vanderzant and as prescribed by Pompeu et al. (2009b) and Rogez et al. (2011).
Splittstoesser, 1992), in Man Rogosa Sharp (MRS) agar (Dfico, The flow rate was 16.66 ␮L s−1 , with the gradient established as
Detroit, MI) (Downes and Ito, 2001; Wehr and Frank, 2004), both follows: 5–15% B in 10 min, 15–50% B in 3 min, 50–70% B in 4 min,
using the pour-plate technique, or in Glucose-Yeast (GY) agar and 70–5% B in 3 min to return to the initial column equilibrium
(Torija et al., 2010) using the spread-plate technique, to count condition. Compounds were identified based on the retention times
the mesophiles, lactic acid bacteria (LAB), and acetic acid bacte- of the standard compounds acquired from Extrasynthèse and on
ria (AAB), respectively. The incubation of the samples in MRS agar UV–Visible spectra (Genay, France).
and GY agar was conducted at 30 ◦ C for 72 h with the addition of
cycloheximide (Fluka, Steinheim, Germany) to avoid fungal con- 2.11. Validations
tamination (100 mg L−1 ). The GY environment was created in our
laboratory using 1% yeast extract, 1% glucose, and 1.5% agar (w/v, 2.11.1. Selection assessment of the culture media MRS and GY
Difco) (Torija et al., 2010), with a final pH of 5.43. Microbiological agar
kinetics were only tracked on fruit stowed in the closed system, Three commercial samples of açai juice, purchased at commer-
because spontaneous fermentation would be more evident at low cial establishments in Belém (Brazil), were diluted in peptone and
oxygen concentrations. then plated on GY agar (spread-plate) and MRS agar (pour-plate).
Because the GY agar medium was formulated in our laboratory, a
2.8. Sugars positive control (vinegar containing strains of Acetobacter spp.) was
used. Cycloheximide was added to the culture media (100 mg L−1 )
Sucrose, d-glucose and d-fructose were determined using the to avoid fungal contamination. After 72 h at 30 ◦ C, 21 colonies of
enzymatic kit Megazyme K-SUFRG (Megazyme, Bray, Ireland). Açai MRS agar and 30 colonies of agar GY were isolated and submitted
juice was diluted in water at 90 ◦ C in a 1:4 (w/w) ratio, immediately for biochemical analyses of gram colouration by the Huker method
placed for 30 min in a refrigerator, and then filtered in Whatmann (Bier et al., 2001), and catalysis (Fung and Petrishko, 1973).
4 qualitative paper to remove the oil from the samples, as recom-
mended by the manufacturer of the enzymatic kit. The samples 2.11.2. Recovery test for ethanol, lactic acid and SCFA
were then subjected to enzymatic reactions and the absorbance Three commercial samples of açai juice were purchased in com-
was measured at 340 nm on a UV/Vis Biopharma Ultrospec spec- mercial establishments in Belém (Brazil). Known masses of ethanol,
trophotometer (Manchester, UK). lactic acid, and SCFA were added to these three samples. The sam-
ples were divided into two parts, one to determine the percent
2.9. Identification and quantification of ethanol, lactic acid and recovery of lactic acid and another one to determine the percent
short chain fatty acid (SCFA) recovery of ethanol and SCFA because they require different sam-
ple treatments. Subsequently, the protocol described in Section 2.9
Açai juice samples were diluted directly in acidified acetone was followed.
with hydrochloric acid (0.37%, v/v) (Merck) to identify and quan-
tify SCFA and ethanol according to the method of Pinto et al. (2006). 3. Results

To determine lactic acid content, the samples were centrifuged at 3.1. Weight loss, specific heat and enthalpy of fermentation
1700 × g for 15 min, frozen at −20 ◦ C, and lyophilised at −70 ◦ C
and 0.027 kPa for 48 h (Södergârd and Stolt, 2002). Subsequently, The loss of mass in the EOF stowed in the closed system (0.28%) was much
the samples were diluted in acidified acetone for identification less than for those stowed in the open system after 27 h (1.7%). As to the thermo-
dynamic parameters, EOF had an average cp value of 3.15 ± 0.10 kJ kg−1 ◦ C−1 . H
and quantification. All samples were filtered through 0.45 ␮m
values of 13.41 kJ kg−1 and 9.67 kJ kg−1 were found for the closed and open systems,
Vertical Chromatography filters (Bangkok, Thailand) for manual respectively. For both conditions, the minimum temperature was 29.0 ◦ C, and the
injection (1 ␮L) into a gas chromatograph (Shimadzu GC 2010; maximum temperatures reached were 32.1 ◦ C and 33.3 ◦ C in the open and closed
Kyoto, Japan). To identify the compounds, the retention times were systems, respectively, after 7 h of experimentation.
compared and the samples were diluted in standard commer-
3.2. Respiration rate
cial compounds (ethanol from Tedia, Fairfield, OH; lactic, acetic,
propionic, butyric, valeric, and caproic acids from Sigma–Aldrich, The measured respiration rate (CO2 evolution) in the closed system e after 3 h
Steinheim, Germany). Quantitative results were obtained by inte- was 335 nmol kg−1 s−1 (Fig. 3) at a temperature within the fruit of 30.0 ◦ C, and
grating to obtain area under the peaks using calibration curves with decreased to around 80 nmol kg−1 s−1 , after 27 h when the fruit was at 32.2 ◦ C.
The influence of fruit respiration on fermentation is perceptible as the environ-
GCSolution Release 2.30 Software (Tokyo, Japan). The gas chro-
ment becomes microaerobic due to production of CO2 by the fruit. There was
matograph was equipped with a ZB-Wax column (60 m × 0.25 mm a rapid decrease of oxygen absorption during the 3 h (402 nmol kg−1 s−1 down
in internal diameter × 0.25 ␮m width of the glycol polyethylene to 88 nmol kg−1 s−1 ). The logarithmic first-order model fit the data well (P < 0.05)
film), a detector of flame ionization, and split mode injection (10:1). (Fig. 3).
 F. Aguiar et al. / Postharvest Biology and Technology 86 (2013) 294–299 297

12 (a)

10

Ethanol
8
D-Glucose

mmol kg-1
6 D-Frutose

Acetic Acid
4
Lactic Acid

0
0 5 10 15 20 25 30
time (h)

12
Ethanol (b)
Fig. 3. Respiration rate of Euterpe oleracea fruit over 27 h of postharvest in closed
system. (t) = time in hours. Vertical bar is standard deviation (n = 3). D-Glucose
10
D-Frutose

8 Acetic Acid
3.3. Microbiology
Lactic Acid

mmol kg-1
3.3.1. Selection assessment of the culture media MRS and GY agar 6
The results indicated that 64% (n = 21) of the colonies plated on MRS agar were
Gram-positive and catalase-negative, characterizing them as belonging to LAB. In GY
4
agar, 57% (n = 30) of the plated colonies were Gram-negative and catalase-positive,
characterizing them as belonging to AAB. Both percentages were applied to the
counts of LAB and AAB and experimental samples. 2

3.3.2. O2 and CO2 concentrations and experimental bacterial kinetics 0

In the closed system, LAB showed very similar growth to that of mesophiles 0 5 10 15 20 25 30
during the entire experimental period (about 1011 CFU kg−1 ). In this condition, EOF time (h)
presented an average temperature equal to 31.7 ◦ C. The load of AAB was two base-ten
logarithmic orders of magnitude lower than LAB during the entire experiment, vary- Fig. 5. Kinetics of some organic compounds produced in Euterpe oleracea fruit over
ing from 7.1 up to 8.8 log CFU kg−1 (Fig. 4). After 27 h in the closed system, bacterial 27 h of postharvest storage in a closed system (a) and in an open system (b). Vertical
growth and gas concentrations reached about 4% O2 and 14% CO2 . bar is standard deviation (n = 3).

3.4. Sugars and some products of fermentation

3.4.3. Identification and quantification of some fermentation products
3.4.1. Sugars content During the postharvest period, ethanol, lactic acid, and acetic acid were iden-
The simple sugars d-glucose and d-fructose were the only ones quantifiable. tified in EOF after only 3 h, indicating the occurrence of alcoholic, lactic, and acetic
The sucrose content is very low in açai (<0.1% dry matter), as previously reported by fermentation processes in both simulated transport conditions. Alcoholic fermenta-
Bichara and Rogez (2011) and Schauss et al. (2006). High consumption of d-glucose tion was predominant in both conditions studied (Fig. 5a and b). Meanwhile, lactic
and d-fructose by microorganisms in the closed system is inferred from the 73% and acetic acid fermentations were secondary, occurring more vigorously in the
reduction in d-glucose and 95% in d-fructose after 27 h (Fig. 5a). Meanwhile, in the closed system. Interestingly, acetic acid was produced at a greater quantity than
open system, this consumption was more moderate (Fig. 5b), with a 44% reduction lactic acid, the former increasing from 0.28 to 1.50 mmol kg−1 , whereas the latter
in d-glucose content and 45% in d-fructose content. from 0.26 to 0.97 mmol kg−1 . Both organic acids were modeled with the logarithmic
first-order model (P < 0.05) (Fig. 6).
3.4.2. Recovery test for products of fermentation
The recovery of target compounds was high (96–103%) with low coefficients of
3.5. Quantification of the main phenolic compounds by HPLC
variation (CV < 9%) (Table 1). This fact indicates significant accuracy in the method-
ology used to quantify compounds arising from the spontaneous fermentation of
In the closed system, there was a 78% decrease in the C-3-R content and an 88%
EOF.
decrease in the C-3-G content of the EOF, after 27 h. Homorientin and orientin were
decreased by 78% and 72%, respectively. The logarithmic first-order model fit the
12 25 data well (P < 0.05) (Fig. 7a). In the open system the losses of phenolic compounds

11
20
Gas Concentration (%)

10
Log CFU kg-1

15
9
10
8

5
7

6 0
0 5 10 15 20 25 30
Time (h)

Fig. 4. Kinetics of growth of () mesophiles bacteria, () lactic acid bacteria and ()
acetic acid bacteria in Euterpe oleracea fruit. () Oxygen and () carbonic gas con-
centration inside polystyrene boxes over 27 h of postharvest. Vertical bar is standard Fig. 6. Production of lactic and acetic acid by Euterpe oleracea fruit over 27 h in a
deviation (n = 3). closed system. (t) = time in hours. Vertical bar is standard deviation (n = 3).
298 F. Aguiar et al. / Postharvest Biology and Technology 86 (2013) 294–299

Table 1
Recovery test for ethanol, lactic acid and short chain fatty acids in 3 açai juice samples.

Compound Blank (␮g)a Amount added (␮g) Amount found (␮g) Recovery (%) Coefficient of variation (%)

Ethanol 29.7 785.0 783.4 96 8.7

Lactic acid 300 2411.0 2756.7 102 8.7
Acetic acid 0 524.5 535.8 102 8.2
Propionic acid 0 49.7 50.3 101 5.5
Butyric acid 0 48.2 47.6 99 2.5
Valeric acid 0 47.0 47.6 101 5.9
Caproic acid 0 46.4 47.8 103 2.8
a
Ethanol, acetic, propionic, butyric, valeric and caproic acids in 1.5 mL of açai juice; lactic acid in 20 mL of açai juice.

The cp value was relevant (3.15 ± 0.10 kJ kg−1 ◦ C−1 ), and is

in the same order of magnitude as other tropical fruits such
as orange (3.91 kJ kg−1 ◦ C−1 ), pineapple (3.80 kJ kg−1 ◦ C−1 ),
lemon (3.69 kJ kg−1 ◦ C−1 ), cashew (3.88 kJ kg−1 ◦ C−1 ), guava
(3.80 kJ kg−1 ◦ C−1 ), mango (3.78 kJ kg−1 ◦ C−1 ), and banana
(3.45 kJ kg−1 ◦ C−1 ) (Ikegwu and Ekwu, 2009). The enthalpy of
fermentation indicated that the quantity of dissipated energy is
very high, considering that approximately 5 tons of EOF are trans-
ported per boat every day in the Brazilian Amazon. This fact draws
attention to the refrigeration of boat holds in order to decrease
the speed of metabolic reactions and preserve EOF quality. This
technology also is important for reducing the respiration rate of
EOF, which was characteristic of non-climacteric fruits and strong
after 3 h in the closed system.
With regard to microbiology, the selectivity of the media MRS
and GY agar was good. Inside polystyrene boxes, the microaerobic
conditions favoured the growth of LAB, since the concentration of
O2 decreased to 4% after 27 h (Fig. 4). This favours sugar catabolism
by the fermentation media of the bacteria, producing organic com-
pounds such as ethanol and carboxylic acids.
LAB produces bacteriocins that inhibit the growth of other
groups of bacteria (Helander et al., 1997), such as AAB and aero-
bic bacteria. Furthermore, the microaerobic condition was adverse
to the development of AAB because the concentration of O2 was
already low after 7 h (≤6.87 ± 0.68%) (Fig. 4). In fact, Massaguer
(2005) reported that this group of bacteria is usually found in lower
levels than LAB in the fermentation of fruits and fruit juices, due to
the lack of oxygen. In addition, LAB adapt well to intrinsic charac-
teristics of fresh food material. Spontaneous fermentation typically
arises from competition between a variety of indigenous and con-
taminating microorganisms and those that adapt better prevail in
this type of process (Rodriguez et al., 2009), which explains the
Fig. 7. Phenolic content of Euterpe oleracea fruit over 27 h (a) in a closed system and similar counts of AAB and mesophiles (Fig. 4).
(b) in an open system. Legend: cyanidin-3-glucoside (C-3-G), cyanidin-3-rutinoside The sugar content quickly decreased due to microbial activ-
(C-3-R). (t) = time in hours. Vertical bar is standard deviation (n = 3). ity in the microaerobic environment (Fig. 5a). It has been shown
that simple sugars are the first substrates consumed during the
spontaneous fermentation process in EOF. This microaerobic envi-
were smaller: C-3-R, C-3-G, homorientin, and orientin sustained decreases of 58%,
ronment also favoured the production of ethanol by yeasts, which
66%, 73%, and 62%, respectively (Fig. 7b).
are found in large quantities in the açai fruit (2.2 × 109 CFU kg−1 )
(Rogez, 2000). In the open system, alcoholic fermentation resulted
4. Discussion in decay over time (Fig. 5b), probably due to the low boiling point
of ethanol, and may have been carried by water vapour.
EOF showed greater weight loss in the open system (Fig. 2), Field reports indicate that EOF transported in boat holds result
being in greater contact with the atmosphere and its currents, facil- in juice with a more accentuated acidity (acrimony). The palatabil-
itating increased evaporation of water generated by transpiration ity threshold of acetic acid (pungent) is 0.9 mmol L−1 (Schoenberger
from the fruit. In the closed system there was condensation of the et al., 2002); this threshold was reached after 7 h in the closed sys-
water vapour on the fruit, causing a wet appearance. In addition, tem (0.92 ± 0.17 mmol kg−1 ) (Fig. 6). This fact explains the sour
complete oxidation of sugars by bacteria into water and carbon taste of açai juice, when EOF is transported for a long time in
dioxide takes place in an aerobic environment; therefore, there is a microaerobic conditions (in boat cargo holds).
greater loss of biomass under these conditions (Solieri and Giudici, In the open system the degradation of phenolics was lower than
2009). Pompeu et al. (2009a) observed a 3.2% loss of mass in EOF that in the closed system for the four compounds studied (C-3-G,
after 27 h at 30 ◦ C; however, in the referenced experiment, they C-3-R, orientin, and homorientin) (Fig. 7a and b). The degrada-
were stowed in a forced air circulation oven. tion in an aerobic environment could be explained to acetification
 F. Aguiar et al. / Postharvest Biology and Technology 86 (2013) 294–299 299

by AAB (Cerezo et al., 2010). Su and Silva (2006) also reported Fung, D.Y.C., Petrishko, D.T., 1973. Capillary tube catalase test. Applied Microbiology
drastic decreases in anthocyanins, total polyphenols, and antiox- 26, 631–632.
Helander, I.M., Wright, A.V., Mattila-Sandhlon, T.M., 1997. Potential of lactic acid
idant activity during the acetification of blackberry wines for 96 h bacteria and novel antimicrobials against Gram-negative. Trends in Food Science
at 15 ◦ C in a closed system with a supply of oxygen. & Technology 8, 146–150.
In microaerobic environments, LAB are also able to use phe- IBGE, 2011. Instituto Brasileiro de Geografia e Estatística. In: Tabela 1 – Quantidade
e valor dos produtos de extração vegetal e silvicultura, segundo os principais
nolic compounds as carbon and energy sources (Othman et al., produtos. http://www.ibge.gov.br/home (02.25.13).
2009). In addition, the degradation is catalysed by polyphenoloxi- Ikegwu, O.J., Ekwu, F.C., 2009. Thermal and physical properties of some tropical fruits
dase (Bichara and Rogez, 2011), and the average temperature in the and their juices in Nigeria. Journal of Food Technology 7, 38–42.
Massaguer, P.R., 2005. Microbiologia dos Processos Fermentativos. Varela, São Paulo.
open system (30.1 ± 1.2 ◦ C) and in the closed system (31.7 ± 1.8 ◦ C)
Montet, D., Loiseau, G., Zakhia, N., 2004. Fermented vegetables. In: Pandey Ashok
favoured the enzymatic activity. (Ed.), Concise Encyclopedia of Bioresource Technology. Food Products Press,
Londres, pp. 211–223.
Othman, N.B., Roblain, D., Chammen, N., Thonart, P., Hamdi, M., 2009. Antioxidant
5. Conclusion
phenolic compounds loss during the fermentation of Chétoui olives. Food Chem-
istry 116, 662–669.
Due to the long transport times, açai fruit undergo three types Pinto, A.B., Guedes, C.M., Moreira, R.F.A., De Maria, C.A.B., 2006. Volatile constituents
from headspace and aqueous solution of genipap (Genipa americana) fruit iso-
of spontaneous fermentation: alcoholic, lactic, and acetic, which
lated by solid-phase extraction method. Flavour and Fragrance Journal 21,
are much stronger in closed conditions than in open systems, and 488–491.
result in a higher degradation rate of phenolic compounds. The Pompeu, D.R., Barata, V.P., Rogez, H., 2009a. Impacto da refrigeração sobre variáveis
higher quality loss is the bioactive compounds, since more than de qualidade dos frutos do açaizeiro (Euterpe oleracea). Alimentos e Nutrição 20,
141–148.
50% of each phenolic compound has degraded after only 13 h in Pompeu, D.R., Silva, E.M., Rogez, H., 2009b. Optimization of the solvent extraction
closed environments where low oxygen and high microbial load are of phenolic antioxidants from fruits of Euterpe oleracea using Response Surface
present, at an average temperature of 31.7 ± 1.8 ◦ C; in other words, Methodology. Bioresource Technology 100, 6076–6082.
Rodriguez, H., Curiel, J.A., Landete, J.M., Rivas, B., Felipe, F.L., Gómez-Cordovés, C.,
transporting açai fruit in aerobic conditions mitigates the negative Mancheño, J.M., Muñoz, R., 2009. Food phenolics and lactic acid bacteria. Inter-
effects of spontaneous fermentation. Therefore, it can be suggested national Journal of Food Microbiology 132, 79–90.
that abbreviating the time between picking and processing açai Rogez, H., 2000. Açaí: Preparo, Composição e Melhoramento da Conservação.
EDUFPA, Belém, PA, Brazil.
fruit and non-closed shipping is of great importance for limiting Rogez, H., Akwie, S.N.L.T., Moura, F.G., Larondelle, Y., 2012. Kinetic modeling
nutritional and functional losses. of anthocyanin degradation and microorganism growth during posthar-
vest storage of açai fruits (Euterpe oleracea). Journal of Food Science 77,
C1299–C1305.
Acknowledgements Rogez, H., Pompeu, D.R., Akwie, S.N.T., Larondelle, Y., 2011. Sigmoidal kinetics of
anthocyanin accumulation during fruit ripening: a comparison between açai
The authors are grateful to Fundação de Amparo à Pesquisa do fruits (Euterpe oleracea) and other anthocyanin-rich fruits. Journal of Food Com-
position and Analysis 24, 796–800.
Estado do Pará (FAPESPA), Pró-Reitoria de Pesquisa e Pós-Graduação
Sabatini, N., Mucciarella, M.R., Marsilio, V., 2008. Volatile compounds in
(PROPESP/UFPA), Fundação de Amparo e Desenvolvimento da uninoculated and inoculated table olives with Lactobacillus plantarum (Olea
Pesquisa (FADESP) and Conselho Nacional de Desenvolvimento Cientí- europaea L., cv. Moresca and Kalamata). Food Science and Technology 41,
2017–2022.
fico e Tecnológico (CNPq), all from Brazil, for their financial support.
Schauss, A.G., Wu, X., Prior, R.L., Ou, B., Patel, D., Huang, D., Kababick, J.P., 2006.
Phytochemical and nutrient composition of the freezed-dried Amazonian palm
References berry, Euterpe oleracea Mart. (Acai). Journal of Agricultural and Food Chemistry
54, 8598–8603.
Schoenberger, C., Krottenthaler, M., Back, W., 2002. Sensory and analytical charac-
Bhande, S.D., Ravindra, M.R., Goswami, T.K., 2008. Respiration rate of banana fruit
terization of nonvolatile taste-active compounds in bottom fermented beers.
under aerobic conditions at different storage temperatures. Journal of Food Engi-
Technical Quarterly – Master Brewers Association of the Americas 39, 210–217.
neering 87, 116–123.
Shrivastava, M., Datta, A.K., 1999. Determination of specific heat and thermal con-
Bichara, C.M.G., Rogez, H., 2011. Chapter 1: Açai (Euterpe oleracea Martius). In:
ductivity of mushrooms (Pleurotus florida). Journal of Food Engineering 39,
Yahia, E.M. (Ed.), Postharvest Biology and Technology of Tropical and Subtropical
255–260.
Foods: Açai to Citrus, vol. 2. Woodhead Publishing, Oxford, England, pp. 1–23.
Södergârd, A., Stolt, M., 2002. Properties of lactic acid based polymers and their
Bier, J.W., Splittsoesser, D.F., Tortorello, M.L., 2001. Microscopic methods. In:
correlation with composition. Progress in Polymer Science 27, 1123–1163.
Downes, F.P., Ito, K. (Eds.), Compendium of Methods for the Microbiological
Solieri, L., Giudici, P., 2009. Vinegars of the World. Springer, Milan.
Examination of Foods. , fourth ed. American Public Health Association, Wash-
Su, M., Silva, J.L., 2006. Antioxidant activity, anthocyanins, and phenolics of rabbit-
ington, DC, pp. 37–44 (Cap. 4).
eye blueberry (Vaccinium ashei) by-products as affected by fermentation. Food
Brunetto, M.R., Cayama, Y.D., Gutiérrez, L., Roa, S.C., Méndrez, Y.C., Gallignani, M.,
Chemistry 97, 447–451.
Zambrano, A., Gómez, A., Ramos, G., 2009. Headspace gas chromatography–mass
Torija, M.J., Mateo, E., Guillamón, J.M., Mas, A., 2010. Identification and quantifica-
spectrometry determination of alkylpyrazines in cocoa liquor samples. Food
tion of acetic acid bacteria in wine and vinegar by TaqMan-MGB probes. Food
Chemistry 112, 253–257.
Microbiology 27, 257–265.
Cerezo, A.B., Cuevas, E., Winterhalter, P., Garcia-Parrilla, M.C., Troncoso, A.M., 2010.
Vanderzant, C., Splittstoesser, D.F., 1992. Compendium of Methods for Microbi-
Anthocyanin composition in Cabernet Sauvignon red wine vinegar obtained by
ological Examination of Foods, third ed. American Public Health Association,
submerged acetification. Food Research International 43, 1577–1584.
Washington, DC.
Downes, F.P., Ito, K., 2001. Compendium of Methods for the Microbiological Exam-
Wehr, H.M., Frank, J.F., 2004. Standard Methods for the Examination of Dairy Prod-
ination of Foods, fourth ed. American Public Health Association, Washington,
ucts, seventeenth ed. American Public Health Association, Washington, DC.
DC.