- Email: [email protected]
Dynamics of microbial ecology during cocoa fermentation and drying: Towards the identification of molecular markers
Accepted Manuscript Dynamics of microbial ecology during cocoa fermentation and drying: towards the identification of molecular markers Yasmine Hamdouche , Guehi Tagro , Noel Durand , Kra Brou Didier Kedjebo , Didier Montet , Jean Christophe Meile PII:
S0956-7135(14)00286-2
DOI:
10.1016/j.foodcont.2014.05.031
Reference:
JFCO 3864
To appear in:
Food Control
Received Date: 29 November 2013 Revised Date:
15 March 2014
Accepted Date: 7 May 2014
Please cite this article as: HamdoucheY., TagroG., DurandN., Didier KedjeboK.B., MontetD. & MeileJ.C., Dynamics of microbial ecology during cocoa fermentation and drying: towards the identification of molecular markers, Food Control (2014), doi: 10.1016/j.foodcont.2014.05.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights •
Microbial communities exhibited dynamic changes during cocoa fermentation & drying. Bacterial ecology of cocoa beans is highly altered during post-harvest processing.
•
Gluconobacter xylinum and Clostridium sp were only detected in drying step.
•
Gluconobacter xylinum and Clostridium sp could be molecular markers of quality.
AC C
EP
TE D
M AN U
SC
RI PT
•
ACCEPTED MANUSCRIPT Title:
2
Dynamics of microbial ecology during cocoa fermentation and drying: towards the
3
identification of molecular markers
4
Authors:
5
Yasmine Hamdouche a*, Guehi Tagro b, Noel Durand a, Kra Brou Didier Kedjebo b, Didier
6
Montet a, Jean Christophe Meile a
7
a
8
5, France.
9
b
RI PT
1
SC
CIRAD-UMR Qualisud, TA B-95/16, 73, Jean-François Breton, 34398 Montpellier Cedex
Laboratory of Microbiology and Molecular Biology. Department of Food Science and
Technology. University of Nangui Abrogoua, Abidjan, Ivory Coast
11
*
12
E-mail address: [email protected]
13
Postal address: CIRAD-UMR Qualisud, TA B-95/16, 73, rue Jean-François Breton, 34398
14
Montpellier Cedex 5, France
15
Tel: (33) 4 67 61 57 28, Fax: (33) 4 67 61 44 33
16
Abstract:
17
This study aimed to evaluate molecular microbial markers associated with cocoa fermentation
18
and drying steps. The dynamics of microbial populations were analyzed during postharvest
19
processing (fermentation and sun drying) of cocoa from Ivory Coast. The variations in
20
bacterial and yeast communities structure were monitored by a molecular technique (PCR-
21
DGGE) using 16S rRNA and 26S rRNA profiles. Dominant microbial species were
22
determined by sequencing. Our results showed that treatments of cocoa generated great
23
variations in the microbial ecology in terms of number of species detected. The bacterial
24
ecology quickly underwent dramatical changes during the fermentation step after two days.
25
The dominance of Enterobacteria was first observed and then accompanied by Lactic acid
26
bacteria and at the latest stages by acetic acid bacteria. Furthermore, Hanseniaspora opuntiae
27
yeast species predominated during cocoa bean fermentation and drying processes. In addition,
28
two bacterial species were only identified during the drying step of fermented cocoa beans
29
and are potential candidates for biological markers of this process.
M AN U
10
AC C
EP
TE D
Corresponding Author: Hamdouche Yasmine
1
ACCEPTED MANUSCRIPT 30 31
Key words: PCR-DGGE, Cocoa Fermentation, Cocoa drying, Post-harvest, Bacterial, Yeast
32
ecology.
33
1. Introduction
RI PT
34
Post harvest cocoa processing is a crucial aspect of cocoa production as it mostly determines
36
the quality of cocoa beans. The key processing steps in producing the most desirably
37
flavoured cocoa beans are fermentation, drying and roasting. Fermentation is generally carried
38
out in a traditional practice following three processes in boxes, heaps and trays through a
39
spontaneous four to six days fermentation process (Wood & Lass, 2001; Schwan & Wheals,
40
2004; De Vuyst et al., 2010). Fermentation step begins with the growth of microorganisms
41
which activities contribute to the degradation of the mucilaginous pulp surrounding the beans,
42
and produces a range of metabolic end-products (alcohols, organic acids). The main phases of
43
cocoa beans fermentation process are alcoholic fermentation by yeasts which convert the
44
sugars from the pulp into ethanol. Lactic fermentation, which produces lactic acid from
45
sugars, is mainly driven by lactic acid bacteria (LAB) and has to be avoided to obtain high
46
quality cocoa. In the last phase of fermentation, acetic acid bacteria (AAB) transform alcohol
47
into acetic acid (Schwan et al., 1995; Camu et al., 2007; Nielsen et al., 2007a).
48
Following fermentation, the beans are either sundried or artificially dried (in regions where
49
constant rainfall occurs), in order to reduce the moisture content from about 60% to about
50
7.5%. Drying must be carried out carefully to ensure that off-flavour did not develop. Drying
51
should take place slowly. If the beans are dried too quickly some of the chemical reactions
52
started in the fermentation process are not allowed to complete and the beans remain acidic,
53
with a bitter flavour. However, if the drying is too slow, moulds and off flavours can develop
54
(Fowler et al., 1998; Wood & Lass, 2001; Thompson et al., 2001).
55
Microorganisms play a crucial role in the quality of cocoa during fermentation step (Schwan
56
et al., 1995; 1997, Nielsen et al., 2007b). The influence of cocoa origin and fermentation
57
practices in
58
investigated extensively and multiphasically (Camu et al., 2007, 2008; Nielsen et al., 2007b,
59
2008; Garcia-Armisen et al., 2010; Lefeber et al., 2010; Papalexandratou & De Vuyst, 2011;
60
Papalexandratou et al., 2011a,b,c, 2013). These studies helped to understand the impact of
AC C
EP
TE D
M AN U
SC
35
the farm on the evolution of cocoa bean fermentation processes has been
2
ACCEPTED MANUSCRIPT fermentation practices on the microbial succession during cocoa bean fermentations.
62
However, little is known about the microbial communities, their dynamics and role in cocoa
63
quality during the drying step.
64
In this present study, we analyzed both cocoa fermentation and drying steps in terms of
65
microbial ecology (bacteria, yeast) in order to get an overview of the post-harvest processes
66
and their impact on the microbial ecology of cocoa. To this end, we used the denaturing
67
gradient gel electrophoresis (DGGE) method that provides a rapid and overall view of
68
community structures (Ampe et al., 1999, 2001; Ben Omar & Ampe, 2000; Cocolin et al.,
69
2001; Ercolini et al., 2001). This method was successfully employed for monitoring cocoa
70
fermentation and to discriminate geographical origin of foods products such as fish (Le
71
Nguyen et al., 2008), fruits (El Sheikha et al., 2009; 2011) and coffee (Nganou Donkeng et
72
al., 2012). In this way, we aim at define molecular markers that could be linked to cocoa
73
quality at different steps during post-harvest processing.
74 75 76
2. Materials and methods 2.1. Sampling
M AN U
SC
RI PT
61
Two hundred cocoa pods were harvested by traditional methods in Ivory Coast in March
78
2013. After opening of the pods, approximately 80 kg of beans surrounded by mucilage were
79
fermented in a wooden box during 6 days (one batch was followed), mixed by hand at 48 and
80
96 hours of fermentation, and then dried for 4 days. Cocoa beans samples (800 g) were
81
collected during fermentation at 3 time intervals (2, 4 and 6 days). 6-days fermented cocoa
82
beans were sun dried during 4 days. 200 grams of coca beans were sampled at 3 time intervals
83
(2, 3 and 4 days). Samples were collected with gloves and placed in sterile bags, put in a
84
cooler box and sent immediately by plane to Montpellier (France) and analyzed immediately
85
at Cirad laboratories. The transportation time from Ivory Coast to Cirad laboratories was
86
about 24 hours.
87
AC C
EP
TE D
77
2.2. Total DNA extraction
88
DNA extraction from bacteria and yeast is based on the method of Ampe et al. (1999) and
89
Leesing (2005). For each sample, DNA extraction was performed in duplicate. Ten grams of
90
cocoa beans were put in Falcon tubes containing 10 mL of peptone water (Biomérieux) and
91
mixed at room temperature for 30 min on a rotating wheel. Two milliliters of the resulting 3
ACCEPTED MANUSCRIPT suspensions were mixed with 0.3 g of 0.5 mm diameter acid washed glass beads (Sigma-
93
Aldrich) in Eppendorf tubes. The mixture was vortexed vigorously for 2 min in a bead beater
94
instrument (Vortex Genie 2 SI-A256) then centrifuged at 12000 g for 10 min. The pellet
95
(containing microbial cells) was resuspended in 300 µL of breaking buffer [2% Triton X-100
96
(Prolabo); 1% sodium dodecyl sulphate (SDS, Sigma); 100 mM NaCl (Sigma); 100 mM Tris–
97
HCl pH 8.0; 1 mM EDTA; pH 8.0 (Promega)]. Then, 100 µL TE (10 mM Tris-HCl; pH 8.0,
98
Promega) were added. 100 µL of lysosyme solution (25 mg. mL¯1, Eurobio) were added and
99
incubated at room temperature for 5 min and 100 µL of proteinase K solution (20 mg.mL¯1,
100
Eurobio) were added and incubated at 42°C for 20 min. Then 50 µL of 20% SDS were added
101
and incubated at 42°C for 10 min, 400 µL of mixed alkyltrimethyl ammonium bromide
102
(MATAB, Sigma) were added followed by an incubation at 65°C for 10 min. 700 µL of
103
phenol/chloroform/isoamyl alcohol (25/24/1 v/v/v, Carlo Erba) were added and the tubes
104
were centrifuged at 12000 g for 15 min. The aqueous layer was transferred to a new tube, and
105
a second phenol-chloroform extraction was performed. 600 µL of Chloroform/isoamyl
106
alcohol (24/1) was added followed by a 12000 g centrifugation for 10 min. The aqueous phase
107
was collected and 0.1 volume of sodium acetate (3M, pH 5) and an equal volume of ice-cold
108
isopropanol were added and further stored at -20°C overnight. The next day, the tubes were
109
centrifuged at 12000 g for 30 min, then the supernatant was discarded, DNA pellets were
110
washed with 500 µL 70% ethanol, after 5 min of centrifugation, the DNA pellets were air
111
dried for 1 to 2 hours. Finally, the DNA was re-suspended in 100 µL of molecular water and
112
stored at -20°C until analyses.
113
The quantity and purity of extracted DNA were verified by UV spectrophotometer (Biospec-
114
Nano), and by electrophoresis through a 0.8% agarose gel in 1x TAE buffer (containing 40
115
mM Tris-HCl, pH 7.4, 20 mM sodium acetate, 1.0 mM Na2-EDTA; Eppendorf,) with a 1 Kb
116
molecular weight ladder (Promega). After running at 100 V for 30 min, the gels were stained
117
for 15 min in an ethidium bromide solution (50 µg.mL¯1; Promega), rinsed for 5 min in
118
distilled water, then observed and photographed on a UV transilluminator, using a black and
119
white CCD camera (Scion Company) and Gel Smart 7.3 system software (Clara Vision).
AC C
EP
TE D
M AN U
SC
RI PT
92
120 121
2.3. PCR amplification of extracted DNA
122
A 160 bp fragment of The V3 variable region of bacterial 16S rDNA was amplified using gc-
123
338f (5΄-GCG CCG CCG CGC GCG GCG GGC GGG GCG GGG GCA CGG GGG GCA 4
ACCEPTED MANUSCRIPT TCC TAC GGG AGG CAG CAG-3΄, Sigma) and 518r (5΄-ATT ACC GCG GCT GCT GG-
125
3΄, Sigma) primers (Ampe et al., 1999; Leesing., 2005; ∅vreas et al., 1997). A 40-pb GC-
126
clamp was added to the forward primer in order to ensure that the fragment of DNA will
127
remain partially double stranded and that the region screened is in the lowest melting domain
128
(Sheffield et al., 1989). Each mixture (final volume 50 µL) contained about 100 ng of
129
extracted DNA, all primers at 0.2 µM, deoxyribonucleotide triphosphate (dNTPs) at 200 µM,
130
1.5 mM MgCl2, 5µL of 10x of reaction MgCl2-free Taq buffer and 1.25 U of a-Taq
131
polymerase (Promega). In order to increase the specificity of amplification and to reduce the
132
formation of spurious by-products, DNA was amplified in a “touchdown” PCR as follow:
133
94°C for 3 min, followed by 10 touchdown cycles of denaturation at 94°C for 1 min annealing
134
at 65°C (1st cycle) to 55°C (10th cycle) for 1 min and extension at 72°C for 1 min, then 25
135
cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 1 min. The extension step was
136
increased to 10 min in the last cycle.
137
For Yeasts DNA PCR amplification: a 250 bp fragment of D1/D2 region of the 26S rDNA
138
gene was amplified using eukaryotic universal primers gc-NL1 (5΄-CGC CCG CCG CGC
139
GCG GCG GGC GGG GCG GGG GCC ATA TCA ATA AGC GGA GGA AAA G-3΄) and
140
the reverse primer LS2 (5΄-ATT CCC AAA CAA CTC GAC TC-3΄, Sigma) (Kurtzman &
141
Robnett, 1998; Cocolin et al., 2000). A 30-pb GC-clamp was added to the forward primer.
142
The mixture final volume is adjusted to 50 µL, it contained about 100 ng of DNA, and all
143
reagents as described above. The amplification started by one cycle at 94°C for 3 min,
144
followed by 30 cycles of 95°C for 1 min, 52°C for 1 min and 72°C for 1 min, and final
145
extension at 72°C for 10 min.
SC
M AN U
TE D
EP
2.4. Denaturing Gradient Gel Electrophoresis (DGGE) analysis
AC C
146
RI PT
124
147
PCR products were analyzed by DGGE, by using a Bio-Rad Dcode universal mutation
148
detection system (Bio-) using the procedure described by Muyzer et al. (1993) and improved
149
by Leesing (2005). PCR amplicons were loaded into 8% (w/v) polyacrylamide gels
150
(acrylamide/N,N΄- methylene bisacrylamide, 37.5/1, Promega) in 1x TAE buffer.
151
Electrophoresis experiments were performed at 60°C using a denaturing gradient ranging
152
from 30 to 60% (100% corresponded to 7 M urea (Promega) and 40% (v/v) formamide,
153
Promega). The gel was electrophoresed at 20 V for 10 min and then at 80 V for 12 h.
5
ACCEPTED MANUSCRIPT After electrophoresis the gel was stained for 30 min in an ethidium bromide solution (50µg
155
mL¯¹ in 1x TAE buffer) and rinsed for 20 min in distilled water and then photographed as
156
described above.
157
Purification and identification of DGGE bands
158
Detected bands were cut from the DGGE gel with a sterile scalpel. DNA of each band was
159
then eluted in 100 µL TE buffer at 4°C overnight, followed by one hour at 37°C. 100µL of
160
DNA eluted from each band was purified by adding 300 µL of TE, 40 µL of sodium acetate
161
(3M, pH 5) and 1 mL of absolute ethanol and incubated at -20°C for 30 min, then centrifuged
162
12000 g for 15 min. The supernatant was discarded, DNA pellets were washed with 500 µL
163
70% ethanol and after 5 min of centrifugation, the DNA pellets were air dried for 1 to 2 hours.
164
Finally, the DNA was re-suspended in 50 µL of molecular water and stored at 4°C. Bacterial
165
and yeast purified DNA were PCR-amplified using 338f and NL-1 (without GC clamp) and
166
518r and LS-2 primers respectively. PCR amplicons were sent for sequencing at GATC
167
Biotech (Germany).
168
Sequences were compared to the GenBank database using the BLAST program (Altschul et
169
al., 1997) (http://www.ncbi.nlm.nih.gov/BLAST/) and the Ribosomal Database Project
170
(http:rdp.cme.msu.edu/index.jsp). Sequences with a percentage of identity of 97% or greater
171
were considered to belong to the same species (Stackebrandt and Goebel, 1994; Palys et al.,
172
1997).
173
3. Results
SC
M AN U
TE D
EP
175
3.1.
Analysis of bacterial communities associated to cocoa beans during wooden boxes fermentation and sun drying
AC C
174
RI PT
154
176
DGGE profiles were generated from bacterial DNA extracted at different steps during cocoa
177
bean fermentation and drying. Two bands were detected on fresh cocoa beans, one of them
178
was identified as Chloroplast DNA (Q), and a second one was not identified (P) (Figure 1).
179
Both bands could be detected throughout the process. In all cases, these two bands were not
180
informative. After 2 days of fermentation, we observed the presence of a high number (13) of
181
bacterial species. Most of them belong to the Enterobacteriaceae family (9 species); Shigella
182
dysenteriae (V), Kluyevera cryocrescens (N), Escherichia albertii (I), Escherichia sp (U),
183
Escherichia fergusonii (L), Enterobacter sp (R), Panteoa vagans (T), Tatumella terrea (O)
184
and Budvicia sp (M). Lactobacillus plantarum (A and M1) was present during all the
6
ACCEPTED MANUSCRIPT fermentation process. Two bands were revealed at 2 days and identified as AAB of the
186
Acetobacter gender (Acetobacter pasteurianus, (W) and Acetobacter sp (X), (Figure 1)
187
After 4 days of fermentation, several bands present at 2 days were detected such as Kluyevera
188
cryocrescens (N), Tatumella terrea (O) and Budvicia sp (M). The other enterobacterial
189
species found on 2 days fermented cocoa beans were no longer detected at 4 days of
190
fermentation. 2 new bands corresponding to AAB species were detected (Acetobacter
191
cerivisiae (J) and Acetobacter nitrogenifigens (D) together with the two above mentioned
192
AAB species identified at day 2. In addition to the bacterial species detected on 4-day
193
fermented beans, Lactobacillus fermentum (C), Weeksella sp (E), Enterobacter cloacae (F)
194
and Acetobacter lovaniensis (H) were detected in 6-day fermented beans. All of the bands
195
corresponding to AAB species remained detectable through the whole fermentation process.
196
During the drying step (2, 3 and 4 days), some bands that were present on (6-day) fermented
197
cocoa beans were still detectable such as Lactobacillus plantarum (A), Kluyevera
198
cryocrescens (N) and Weeksella sp (E). Two bands corresponding to Acetobacter lovaniensis
199
(H) and Lactobacillus fermentum (C) were no longer detected after 2 and 3 days of drying,
200
respectively. During the drying step, the remaining detected bands identified as AAB as well
201
as two bands (not previously detected on fermented cocoa beans) were detected and
202
corresponded to Gluconobacter xylinum (G) and Clostridium sp (K) species.
SC
M AN U
TE D
203
RI PT
185
3.2. Analysis of yeast communities of cocoa beans during wooden boxes fermentation and sun drying
204
One single band was isolated on fresh cocoa beans which corresponded to Hanseniaspora
206
opuntiae (A) that was the dominant species through the whole fermentation step (Figure 2).
207
Four bands were detected and identified as Candida insectorum (C), Pichia kudriavezii (D),
208
Pichia galeiformis (F) and Pichia species group 1 (E) after 2 days of fermentation. After 4
209
days of fermentation, the major bands (A), (C) and (D) remained detectable, and 3 bands
210
appeared corresponding to Candida sp (B), yeast species group 2 (H) and Pichia manshurica
211
(G).
212
At 6 days of fermentation, the bands identified as Candida sp (B) and Pichia species group 1
213
(E) were no longer detectable, while Pichia manshurica (G), Candida insectorum (C), Pichia
214
galeiformis (F) and Pichia kudriavezii (D), and yeast species group 2 (H) and (A) were still
215
present. Yeast species associated to 6-day fermented cocoa beans were Hanseniaspora
216
opuntiae (A), Candida insectorum (C), Pichia kudriavezii (D), Pichia galeiformis (F), Pichia
AC C
EP
205
7
ACCEPTED MANUSCRIPT 217
manshurica (G) and yeast species group 2 (H) stayed present during 3 days of sun drying.
218
Pichia species group 1 (E) was not detected during the drying step. At the last drying day (4
219
days), only half of the number of bands found at 3 days (3 bands) were detected as
220
Hanseniaspora opuntiae (A), Pichia manshurica (G) and yeast species group 2 (H).
RI PT
221
4. Discussion
223
Analyses of the bacterial and yeast ecologies of cocoa showed that post-harvest steps of cocoa
224
processes are associated to great variations in microbial ecology in terms of number of species
225
that could be detected at the molecular level.
226
During fermentation step, bacterial ecology quickly underwent dramatical changes after two
227
days and most of the identified bands were Enterobacterial species which may contribute to
228
glucose conversion into lactic acid and citric acid (Grimont & Grimont, 2006). Previous study
229
showed that Enterobacteriaceae were involved during the initial phase of cocoa beans
230
fermentation (Garcia-Amisen et al., 2010; Papalexandratou, 2011a, b, c, and 2013). Tatumella
231
saanichensis could be involved in pectinolytic degradation of cocoa pulp and in citric acid
232
assimilation (Marin-Cevada et al., 2010). Lefeber et al. (2010) found genus of Pantoea and
233
Escherichia during the early stages of both of heap and box fermentations in Ivory Coast.
234
Pantoea could be derived from environmental soil or plant material (Kageyama et al., 1992;
235
Pujol & Kado, 2002).
236
Two Lactic Acid Bacterial (LAB) species such as Lactobacillus plantarum and Lactobacillus
237
fermentum were detected on fermented cocoa beans with the dominance of L. plantarum
238
during the whole fermentation step while L. fermentum was detected in the last stages. L.
239
plantarum was already isolated during fermentation of Malaysian, Brazilian, Ivorian and
240
Ghanaian cocoa beans (Gracia-Amisen et al., 2010; Papalexandratou et al., 2011a, b, c, and
241
2013 and Camu et al., 2007). L. fermentum was isolated and detected from cocoa beans
242
fermentation, often as dominant species (Nielsen et al., 2007b; Camu et al., 2007, 2008b;
243
Kostinek et al., 2008, Papalexandratou et al., 2011a). In our case, L. fermentum was detected
244
until 3 days of drying.
245
Two Acetic acid bacteria (AAB) species were detected during the first stages of fermentation
246
(Acetobacter pasteurianus and Acetobacter sp) whereas the other AAB species (identified as
247
Acetobacter
AC C
EP
TE D
M AN U
SC
222
lovaniensis,
Acetobacter
cerivisiae,
8
Acetobacter
nitrogenifigens
and
ACCEPTED MANUSCRIPT Gluconobacter xylinum) were detected later in the process at 4 and 6 days and during drying.
249
The presence of another representative of Acetobacter and Gluconobacter species were
250
previously reported in cocoa bean fermentations (Schawan & Wheals, 2004, Camu et al.,
251
2007; De Melo Periera et al., 2013). Nielsen et al. (2007b) showed that succession of different
252
AAB species took place during fermentation as Acetobacter syzygii and Acetobacter
253
pasterianus were the dominant AAB during the first half of all fermentation, and the later
254
stages of fermentation were dominated by Acetobacter tropicalis.
255
In our study, we showed that AAB dominated in the last stages of fermentation and all drying
256
stages. The dominance of this species during the last stages of fermentation and drying could
257
be explained by their resistance to acidity and heat (Ndoye et al., 2006). These species could
258
potentially be used as good markers of complete cocoa fermentation and drying.
259
Gluconobacter xylinum and Clostridium sp were only detected during sun drying step, which
260
might indicate that they could be present in environment (soils, materials) as Clostridium
261
species generally are considered as soil contaminant (Jeong-Dong et al., 2004).
262
In the last day of cocoa beans drying the microbiota still consisted of several bacterial species,
263
which made bacterial ecology totally different from the initial fresh cocoa beans.
264
During the whole cocoa post-harvest processing, including fermentation and drying, 23
265
bacterial species were detected among which 21 were identified. All bacterial species were
266
found during the whole fermentation step and 18 bacterial species during the drying step. We
267
observed 14 common species between fermentation and drying steps. Seven species (among
268
which six Enterobacteriaceae) were only detected during fermentation step and could be
269
potential specific markers of fermentation step (Escherichia sp, Escherichia fergusonii,
270
Escherichia albertii, Enterobacter sp, Panteoa vagans, Shigella dysenteriae). During sun
271
drying, 2 species were detected which were Gluconobacter xylinum and Clostridium sp and
272
could be specific markers of the drying step.
273
Hanseniaspora opuntiae was the unique yeast species detected on fresh cocoa beans,
274
confirming the quality of cocoa pods and the relative sterility of beans inside the pod. This
275
species could be present at the surface of the pod as it was involved in all studied fermentation
276
until now and was usually dominant in cocoa fermentation (Camu et al., 2007, Lefeber et al.,
277
2011, Papalexandratou & De Vuyst, 2011, Papalexandratou et al., 2011a, b, c, 2013).
278
Hanseniaspora opuntiae was dominant during the whole fermentation and drying steps
279
followed by Pichia manshurica and yeast species group 2.
AC C
EP
TE D
M AN U
SC
RI PT
248
9
ACCEPTED MANUSCRIPT During drying step, the bands identified as Hanseniaspora opuntiae, Candida insectorum,
281
Pichia kudriavezii and yeast species group 2 (Pichia sporocuriosa and Issatchenkia
282
hanoiensis) decreased in time and, at the end of drying, these three species were the only ones
283
detected.
284
Most of the yeast species present on fermented cocoa beans disappeared during the drying
285
process, and no new species could be detected. Yeast species appeared less resistant to (sun)
286
drying than bacterial species. Dynamics of yeast communities associated to cocoa beans
287
fermentation were more important than dynamics of yeast communities associated to cocoa
288
beans during the drying step.
289
Eight yeast species were detected in the whole process, all of these species were present
290
during the fermentation step and seven during sun drying.
291
One species (Candida sp) was only detected at a specific step during fermentation (4 days)
292
before completion. These species could be specific marker for incomplete fermentation.
293
Both yeast and bacterial underwent dramatical changes during both cocoa fermentation and
294
drying. While yeast and bacterial ecologies undergo changes during both treatments such as
295
fermentation and sun drying, a much reduced number of yeast species was recorded at the
296
fourth day of drying.
TE D
297
M AN U
SC
RI PT
280
Aknowledgements
299
This work was supported by DESI grant from CIRAD France (to support PhD students) and
300
Y. Hamdouche was supported by a PhD fellowship from HUBBARD.
AC C
301
EP
298
302
References
303
Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller, W., & Lipman,
304
D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein databases search
305
programs. Nucleic Acids Research, 25, 3389-3402.
306
Ampe, F., Ben Omar, N., Moizan, C., Wacher, C., & Guyot, J. P. (1999). Polyphasic study of
307
the spatial distribution of microorganisms in Mexican pozol, a fermented maize dough,
10
ACCEPTED MANUSCRIPT demonstrates the need for cultivation-independent methods to investigate traditional
309
fermentations. Applied and Environmental Microbiology, 65, 5464–5473.
310
Ampe, F., Audrey, S., & Nadine, Z. (2001). Dynamics of the microbial community
311
responsible for traditional sour cassava starch fermentation studied by denaturing gradient gel
312
electrophoresis and quantitative rRNA hybridization. International Journal of Food
313
Microbiology, 65, 45–54.
314
Ben Omar, N., & Ampe, F. (2000). Microbial community dynamics during production of the
315
Mexican fermented maize dough pozol. Applied and Environmental Microbiology, 47, 3664–
316
3673.
317
Camu, N., De Winter, T., Verbrugghe, K., Cleenwerck, I., Vandamme, P., Takrama, J. S.,
318
Vancanneyt, M., & De Vuyst, L. (2007). Dynamics and biodiversity of populations of lactic
319
acid bacteria and acetic acid bacteria involved in spontaneous heap fermentation of cocoa
320
beans in Ghana. Applied and Environmental Microbiology, 73, 1809–1824.
321
Camu, N., González, Á., De Winter, T., Van Schoor, A., De Bruyne, K., Vandamme, P.,
322
Takrama, J. S., Addo, S. K., & De Vuyst, L.(2008). Influence of turning and environmental
323
contamination on the dynamics of populations of lactic acid and acetic acid bacteria involved
324
in spontaneous cocoa bean heap fermentation in Ghana. Applied and Environmental
325
Microbiology, 74, 86–98.
326
Cocolin, L., Bisson, L. F., & Mills, D. A. (2000). Direct profiling of the yeast dynamics in
327
wine fermentations. FEMS Microbiology Letters, 189, 81-87.
328
Cocolin, L., Manzano, M., Cantoni, C., & Comi, G. (2001). Denaturing gradient gel
329
electrophoresis analysis of the 16S rRNA gene V1 region to monitor dynamic changes in the
330
bacterial population during fermentation of Italian sausages. Applied and Environmental
331
Microbiology, 67, 5113–5121.
332
Daniel, H. M., Vrancken, G., Takrama, J. F., Camu, N., De Vos, P., & De Vuyst, L. (2009).
333
Yeast diversity of Ghanaian cocoa bean heap fermentations. FEMS Yeast Research, 9, 774–
334
783.
335
De Melo Pereira, G. F., Magalhães-Guedes, K. T., Schwan, R. S. (2013). rDNA-based DGGE
336
analysis and electron microscopic observation of cocoa beans to monitor microbial diversity
337
and distribution during the fermentation process. Food Research International 53, 482–486.
AC C
EP
TE D
M AN U
SC
RI PT
308
11
ACCEPTED MANUSCRIPT De Vuyst, L., Lefeber, T., Papalexandratou, Z., & Camu, N. (2010). The functional role of
339
lactic acid bacteria in cocoa bean fermentation. In: Mozzi, F., Raya, R.R., &Vignolo, G.M.
340
(Eds.), Biotechnology of Lactic Acid Bacteria (pp. 301-326). Ames, IA: Novel Applications.
341
Wiley-Blackwell.
342
Dias, D. R., Schwan, R. F., Carlos, L & Lima, O. (2003). Metodologia para elaboração de
343
fermentado de Cajá (Spondias mombin L.). Ciência e Technologia de Alimentos, 23(3), 342-
344
350.
345
El Sheikha, A. F., Condur, A., Métayer, I., Le Nguyen, D. D., Loiseau, G., & Montet, D.
346
(2009). Determination of fruit origin by using 26S rDNA fingerprinting of yeast communities
347
by PCR-DGGE: preliminary application to Physalis fruits from Egypt. Yeast, 26, 567-573.
348
El Sheikha, A. F., Bouvet, J. M., & Montet, D. (2011). Biological bar code for determining
349
the geographical origin of fruits using 28S rDNA fingerprinting of fungal communities by
350
PCR-DGGE: an application to Shea tree fruits. Crops &Food, 3, 40-47.
351
Ercolini, D., Moschetti, G., Blaiotta, G., & Coppola, S. (2001). Behavior of variable V3
352
region from 16S rDNA of lactic acid bacteria in denaturing gradient gel electrophoresis.
353
Current Microbiology, 42, 199–202.
354
Fowler, M. S., Leheup, P., & Cordier, J. L. (1998). Cocoa, coffee and tea. In: Wood, B.J.B
355
(Ed.), Microbiology of Fermented Foods (pp. 128-147). London: Blackie Academic and
356
Professional.
357
Garcia-Armisen, T., Papalexandratou, Z., Hendryckx, H., Camu, N., Vrancken, G., & De
358
Vuyst, L., Cornelis, P. (2010). Diversity of the total bacterial community associated with
359
Ghanaian and Brazilian cocoa bean fermentation samples as revealed by a 16 S rRNA gene
360
clone library. Applied Microbiology and Biotechnology, 87, 2281–2292.
361
Grimont, F., & Grimont, P.A.D. (2006). The genus Enterobacter, In: Dworkin, M., Falkow,
362
S., Rosenberg, E., Schleifer, K. M., & Stackebrandt, E., (Eds.), The Prokaryotes (pp. 163–
363
200). New York: Springer.
364
Jeong-Dong , K., Dae-Weon, L.,
365
(2004). Distribution and antimicrobial susceptibility of Clostridium species in soil
366
contaminated with domestic livestock feces of Korea. Journal of microbiology and
367
biotechnology, 14, 1017-7825.
AC C
EP
TE D
M AN U
SC
RI PT
338
Kyou-Seung, L., Chang-Hyun, C., & Kook-Hee , K.
12
ACCEPTED MANUSCRIPT Kageyama, B., Nakae, M., Yagi, S., & Sonoyama, T. (1992). Pantoea punctata sp. nov.,
369
Pantoea citrea sp. nov., and Pantoea terrea sp. nov. isolated from fruit and soil samples.
370
International Journal of Systematic Bacteriology, 42, 203-210.
371
Kurtzman, C. P., & Robnett, C. J. (1998). Identification and phylogeny of ascomycetous
372
yeasts from analysis of nuclear large subunit (26S) ribosomal DNA partial sequences. Antonie
373
Van Leeuwenhoek, 73, 331-371.
374
Leesing, R. (2005). Identification and validation of specific markers for traceability of
375
aquaculture fish for import/export. PhD thesis. University of Montpellier 2, France.
376
Lefeber, T., Gobert, W., Vrancken, G., Camu, N., & De Vuyst, L. (2010). Dynamics and
377
species diversity of communities of lactic acid bacteria and acetic acid bacteria during
378
spontaneouscocoa bean fermentations in vessels. Food Microbiology, 28, 457–464.
379
Muyzer, G., De Waal, E. C., & Uitterlinden, A. G. (1993). Profiling of complex microbial
380
populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-
381
amplified genes coding for 16S rRNA. Applied Environmental Microbiology, 59, 695-700.
382
Nielsen, D. S., Schillinger, U., Franz, C. A. M. P., Bresciani, J., Amoa-Awua, W., Halzapel,
383
W. H., & Jakobsen, M. (2007a). Lactobacillus ghanensis sp. nov., a motile lactic acid
384
bacterium isolated from Ghanaian cocoa fermentations. International Journal of Systematic
385
and Evolutionary Microbiology, 57, 1468-1472.
386
Nganou, D. N., Durand, N., Tatsadjieu, N. L., Meile, J. C., El Sheikha, A. F., Montet, D., et
387
al. (2012). Determination of coffee origin by using 28S rDNA fingerprinting of fungal
388
communities by PCR-DGGE: Application to the Cameroonian coffee. International Journal
389
of Biosciences, 2(5), 18-30.
390
Nielsen, D. S., Teniola, O. D., Ban-Koffi, L., Owusu, M., Andersson, T. S., & Holzapfel, W.
391
H. (2007b). The microbiology of Ghanaian cocoa fermentations analysed using culture-
392
dependent and culture-independent methods. Inernational Journal of Food Microbiology,
393
114, 168–186.
394
Nielsen, D. S., Nielsen P. S., & Van den Berg, F. (2008). Investigating the fermentation of
395
cocoa by correlating denaturing gradient gel electrophoresis profiles and near infrared spectra.
396
International Journal of Food Microbiology, 125, 133-40.
AC C
EP
TE D
M AN U
SC
RI PT
368
13
ACCEPTED MANUSCRIPT Le Nguyen, D. D, Ha, H, Dijoux, D, et al. (2008). Determination of fish origin by using 16S
398
rDNA fingerprinting of bacterial communities by PCR–DGGE: an application on Pangasius
399
fish from Vietnam. Food Control, 19, 454–460.
400
Øvreas, L., Forney, L., Dae, F. L., & Torsvik, V. (1997). Distribution of Bacterioplankton in
401
Meromictic Lake Sælenvannet, as Determined by Denaturing Gradient Gel Electrophoresis of
402
PCR-Amplified Gene Fragment Coding for 16S rRNA. Applied and Environmental
403
Microbiology, 63, 3367-3373.
404
Palys, T., Nakamura, L. K., & Coha, F. M. (1997). Discovery and classification of ecological
405
diversity in the bacterial world: the role of DNA sequence data. International Journal of
406
Systematic Bacteriology, 47, 1145-1156.
407
Papalexandratou, Z., & De Vuyst, L. (2011). Assessment of the yeast species diversity of
408
cocoa bean fermentations in different cocoa-producing regions using denaturing gradient gel
409
electrophoresis. FEMS. Yeast Research, 11, 564–574.
410
Papalexandratou, Z., Camu, N., Falony, G., & De Vuyst, L. (2011a). Comparison of the
411
bacterial species diversity of spontaneous cocoa bean fermentations carried out at selected
412
farms in Ivory Coast and Brazil. Food Microbiology, 28, 964–973.
413
Papalexandratou, Z., Falony, G., Romanens, E., Jimenez, J. C., Amores, F., Daniel, H. M., &
414
De Vuyst, L., (2011b). Species diversity, community dynamics, and metabolite kinetics of the
415
microbiota associated with traditional Ecuadorian spontaneous cocoa bean fermentations.
416
Applied and Environmental Microbiology, 77, 7698–7714.
417
Papalexandratou, Z., Vrancken, G., De Bruyne, K., Vandamme, P., & De Vuyst, L. (2011c).
418
Spontaneous organic cocoa bean fermentations in Brazil are characterized by a restricted
419
species diversity of lactic acid bacteria and acetic acid bacteria. Food Microbiology, 28,
420
1326– 1338.
421
Papalexandratou, Z., Lefeber,T., Bahrim, B., Lee, O. S., Daniel, H. M., & De Vuyst, L.
422
(2013). Hanseniaspora opuntiae, Saccharomyces cerevisiae, Lactobacillus fermentum, and
423
Acetobacter pasteurianus predominate during well-performed Malaysian cocoa bean box
424
fermentations, underlining the importance of these microbial species for a successful cocoa
425
bean fermentation process. Food Microbiology, 28, 1326-1338.
426
Pujol, C. J., & Kado, C. I. (2002). Genetic and biochemical characterization of the pathway in
427
Pantoea citrea leading to pink disease of pineapple. Journal of Bacteriology, 182, 2230-2237.
AC C
EP
TE D
M AN U
SC
RI PT
397
14
ACCEPTED MANUSCRIPT Schwan, R. F., Rose, A. H., & Board, R. G. (1995). Microbial fermentation of cocoa beans,
429
with emphasis on enzymatic degradation of the pulp. Journal of Applied Bacteriology
430
Symposium Supplement, 79, 96-107.
431
Schwan, R. F., Cooper, R.M., & Wheals, A.E., (1997). Endopolygalacturonase secretion by
432
Kluyveramyces marxianus and other cocoa pulp-degrading yeasts. Enzyme and Microbiology
433
Technology 21, 234-244.
434
Schwan, R. F., & Wheals, A. E. (2004). The microbiology of cocoa fermentation and its role
435
in chocolate quality. Critical Reviews in Food Science and Nutrition, 44(4), 205–221.
436
Sheffield, V. C., Beck, J. S., Stone, E. M., & Myers, R. M. (1989). Attachment of a 40 bp
437
G+C rich sequence (GC-clamp) to genomic DNA fragments by polymerase chain reaction
438
results in improved detection of single-base changes. Proceeding of the National Academy of
439
Sciences of the United States of American, 86, 232-236.
440
Thompson, S. S., Miller, K. B., & Lopez, A. S. (2001). Cocoa and coffee. In: Doyle, M. P.,
441
Beuchat, L. R., Montville, T. J. (Eds.), Food Microbiology Fundamentals and Frontiers (pp.
442
649–661). Washington: ASM Press.
443
Wood, G. A. R., & Lass, R. A. (2001). Cocoa. (4th ed). Oxford: Blackwell Science.
AC C
EP
TE D
M AN U
SC
RI PT
428
15
ACCEPTED MANUSCRIPT Table 1: Identification of DGGE bands for bacterial species (A to X)
AC C
EP
TE D
M AN U
SC
RI PT
Table 2: Identification of DGGE bands for yeast species (A to H)
ACCEPTED MANUSCRIPT Table 1: Letter
Genus/ Species
A
Lactobacillus
Percent identity
Query cover
E value
References
100%
100%
1e-33
HM218742
99%
100%
100%
100%
100%
B
Not identified
C
Lactobacillus
D
Acetobacter nitrogenifigens
1e-44
CP005958
2e-22
KC763628
3e-47
JX628863
SC
fermentum
RI PT
plantarum
Weeksella sp
97%
F
Enterobacter
100%
100%
4e-50
HG421017
97%
100%
2e-47
NR074338
98%
100%
8e-47
HM217974
cloacae G
Gluconobacter xylinum Acetobacter lovaniensis
TE D
H
M AN U
E
Escherichia albertii
97%
100%
2e-58
HM194885
J
Acetobacter
100%
100%
3e-51
KC485819
100%
100%
4e-55
JX575132
99%
100%
1e-34
JQ319655
Budvicia sp
96%
100%
1e-45
JQ080726
Kluyvera
98%
100%
2e-42
AM933754
98%
100%
3e-41
JQ726629
EP
I
cerivisiae
L
Clostridium sp
AC C
K
Escherichia fergusonii
M N
cryocrescens O
Tatumella terrea
P
Not identified
ACCEPTED MANUSCRIPT Q
Chloroplast DNA
100%
100%
4e-60
JX912466
R
Enterobacter sp
100%
100%
2e-68
JX267074
S
Tatumella
98%
100%
3e-41
JQ726633
2e-48
HG421010
T
Pantoea vagans
100%
100%
U
Escherichia sp
100%
100%
V
Shigella dysenteriae
100%
100%
W
Acetobacter
2e-68
HM346189
8e-62
KC768788
100%
95%
AC C
EP
TE D
Acetobacter sp
100%
M AN U
Acetobacter sp
SC
KC478491
pasteurianus
X
RI PT
saanichensis
98%
9e-57
3e-68
KF668642 AB853266
ACCEPTED MANUSCRIPT Table 2: Letter
Genus/ Species
Percent identity
Query cover
E value
References
A
Hanseniaspora
100%
100%
7e-90
KC544476
Candida sp
96%
100%
C
Candida insectorum
99%
100%
D
Pichia kudriavezii
99%
100%
96%
100%
strain YM25991 E
Species group of
Pichia klyvera Pichia fermentans
Pichia galeiformis
G
Pichia manshurica
H
Species group 2:
100%
EF042050
2e-37
JN544058
2e-47
KC494718
2e-49
JX103191 EU647835
100%
4e-49
JX880398
100%
96%
1e-87
JX183960
99%
100%
1e-32
TE D
F
M AN U
Pichia:
1e-40
SC
B
RI PT
opuntiae
JF749216
Issatchenkia
FJ153148
EP
Pichia sporocuriosa
AC C
hanoiensis
ACCEPTED MANUSCRIPT
Figure 1: Picture of DGGE gel showing the evolution of bacterial ecology structure on fresh, fermented and sun dried cocoa beans at different time intervals. 1: Fresh Cocoa beans (T0 time of fermentation), 2: Cocoa beans fermented for 2 days, 3:
RI PT
Cocoa beans fermented for 4 days, 4: Cocoa beans fermented for 6 days and not dried, 5: Cocoa beans fermented for 6 days and dried for 2 days, 6: Cocoa beans fermented for 6 days and dried for 3days, 7: Cocoa beans fermented for 6 days and dried for 4 days.
M1: Lactobacillus plantarum DNA, and M2: Escherichia coli DNA are pure bacterial strain
SC
markers.
M AN U
Identification of DGGE bands for bacterial species (A to X) (Table 1)
Figure 2: Picture of DGGE gel showing the evolution of yeast ecology structure from on fresh, fermented and sun dried cocoa beans at different time intervals. 1: Fresh Cocoa beans (T0 time of fermentation), 2: Cocoa beans fermented for 2 days, 3:
TE D
Cocoa beans fermented for 4 days, 4: Cocoa beans fermented for 6 days and not dried, 5: Cocoa beans fermented for 6 days and dried for 2 days, 6: Cocoa beans fermented for 6 days and dried for 3 days, 7: Cocoa beans fermented for 6 days and dried for 4 days. M2: Pichia sorbitophila DNA are pure yeast strain markers.
EP
M1: Candida apicola DNA
AC C
Identification of DGGE bands for yeast species (A to H) (Table 2)
1
3
2
ACCEPTED MANUSCRIPT 4 7 5 6
M1 B C
D E F G H
I
J K
P Q
M N O
R
S T U V
AC C
Figure 1
X
EP
W
TE D
M2
M AN U
L
SC
RI PT
A
ACCEPTED MANUSCRIPT
1
3
2
4
5
M1
B
SC
C D
7
RI PT
A
6
M AN U
M2
TE D
E F
AC C
EP
G H
Figure 2
ACCEPTED MANUSCRIPT 1. Images and statistical analysis The photographs of DGGE gels were analyzed using Image Quant TL software v. 2003 (Amersham Biosciences, USA). In each lane, banding patterns were standardized with two reference patterns for bacterial DNA (Escherichia coli and Lactobacillus plantarum) and for yeast DNA (Candida apicola and Pichia sorbitophila). The band relative positions were
RI PT
compared with standard patterns.
In DGGE analysis, the generated banding pattern is considered as an image of all the major microbial populations. An individual band generally refers to a unique ‘sequence type’ or phylotype (Van Hannen et al., 1999; Muyzer et al., 1995; Kowalchuk et al., 1997). The
SC
DGGE fingerprinting was manually scored by presence and absence of co-migrating bands, independently of intensity. Pair wise community similarities were quantified using Dice
M AN U
similarity coefficient (SD ): SD =2 Nc/ Na + Nb (Heyndrickx et al., 1996)
Na represented the number of bands detected in sample A, Nb represented the number of bands in sample B, and Nc represented the number of bands common to both samples. The similarity index was expressed within a range of 0 (completely dissimilar) to 100 (perfect similarity). Statistical analysis were performed using the similarity matrix to group samples
TE D
according to their similarity index by Principal Component Analysis (PCA) using Statistica (version 7) software (StatSoft, USA).
EP
2. Results of principal component analysis (PCA) Principal component analysis (PCA) was used to compare bacterial communities during the
AC C
post-harvest process of cocoa beans (supplementary Figure S1). About 95% of the variability of the sample set could be described using this method and four distinct groups could be discriminated: the first group contained only fresh cocoa beans, a second group included only cocoa beans fermented for 2 days, a third group included 2 types of samples (cocoa beans fermented for 4 and 6 days) and the last group dried cocoa beans samples (2, 3 and 4 days). Principal component analysis (PCA) was used to compare yeast communities; the statistical analysis separates three sample groups (see supplementary Figure S2) and represents more than 98% of the variation within the sample data set. A first group contained initial fresh cocoa beans and final (4 days drying) cocoa bean samples, a second group contained cocoa
ACCEPTED MANUSCRIPT beans fermented for 2 days, and a third group included samples of 4- and 6-day fermented as
TE D
M AN U
SC
RI PT
well as 2- and 3-day dried cocoa beans).
Figure S1: Factorial variance analysis of PCR-DGGE gel of 16S rDNA band profiles of
AC C
EP
Bacteria from fermentation and drying of cocoa beans
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S2: Factorial variance analysis of PCR-DGGE gel of 26S rDNA band profiles of
References
EP
Yeasts from fermentation and drying of cocoa beans.
AC C
Heyndrickx, M., Vauterin, L., Vandamme, P., Kersters, K., & De Vos, P. (1996). Applicability of combined amplified ribosomal DNA restriction analysis (ARDRA) patterns in bacterial phylogeny and taxonomy. Journal of Microbiological Methods, 26, 247-259. Muyzer, G., Teske, A., Wirsen, C. O., & Jannasch, H. W. (1995). Phylogenetic relationships of Thiomicrospira species and their identification in deep-sea hydrothermal vent sample by denaturing gradient gel electrophoresis of 16S rDNA fragment. Archives of Microbiology, 164, 165-172. Kowalchuk, G. A., Stephen, J. R., de Boer, W., Prosser, J. I., Embley, T. M., & Woldendorp, J. W. (1997). Analysis of ammonia-oxydizing bacteria of the beta subdivision of the class
ACCEPTED MANUSCRIPT Proteobacteria in coastal sand dunes by denaturing gradient gel electrophoresis and sequencing of PCR amplified 16S ribosomal DNA fragments. Applied and Environmental Microbiology, 63, 1489-1497. Van Hannen, E. J., Zwart, G., Van Agterveld, M. P., Gons, H. J., Ebert, J., & Laanbroek, H. J. (1999). Changes in bacterial and eukaryotic community structure after mass lysis of
RI PT
filamentous cyanobacteria associated with viruses. Applied and Environmental Microbiology
AC C
EP
TE D
M AN U
SC
65, 795-801.
S0956-7135(14)00286-2
DOI:
10.1016/j.foodcont.2014.05.031
Reference:
JFCO 3864
To appear in:
Food Control
Received Date: 29 November 2013 Revised Date:
15 March 2014
Accepted Date: 7 May 2014
Please cite this article as: HamdoucheY., TagroG., DurandN., Didier KedjeboK.B., MontetD. & MeileJ.C., Dynamics of microbial ecology during cocoa fermentation and drying: towards the identification of molecular markers, Food Control (2014), doi: 10.1016/j.foodcont.2014.05.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights •
Microbial communities exhibited dynamic changes during cocoa fermentation & drying. Bacterial ecology of cocoa beans is highly altered during post-harvest processing.
•
Gluconobacter xylinum and Clostridium sp were only detected in drying step.
•
Gluconobacter xylinum and Clostridium sp could be molecular markers of quality.
AC C
EP
TE D
M AN U
SC
RI PT
•
ACCEPTED MANUSCRIPT Title:
2
Dynamics of microbial ecology during cocoa fermentation and drying: towards the
3
identification of molecular markers
4
Authors:
5
Yasmine Hamdouche a*, Guehi Tagro b, Noel Durand a, Kra Brou Didier Kedjebo b, Didier
6
Montet a, Jean Christophe Meile a
7
a
8
5, France.
9
b
RI PT
1
SC
CIRAD-UMR Qualisud, TA B-95/16, 73, Jean-François Breton, 34398 Montpellier Cedex
Laboratory of Microbiology and Molecular Biology. Department of Food Science and
Technology. University of Nangui Abrogoua, Abidjan, Ivory Coast
11
*
12
E-mail address: [email protected]
13
Postal address: CIRAD-UMR Qualisud, TA B-95/16, 73, rue Jean-François Breton, 34398
14
Montpellier Cedex 5, France
15
Tel: (33) 4 67 61 57 28, Fax: (33) 4 67 61 44 33
16
Abstract:
17
This study aimed to evaluate molecular microbial markers associated with cocoa fermentation
18
and drying steps. The dynamics of microbial populations were analyzed during postharvest
19
processing (fermentation and sun drying) of cocoa from Ivory Coast. The variations in
20
bacterial and yeast communities structure were monitored by a molecular technique (PCR-
21
DGGE) using 16S rRNA and 26S rRNA profiles. Dominant microbial species were
22
determined by sequencing. Our results showed that treatments of cocoa generated great
23
variations in the microbial ecology in terms of number of species detected. The bacterial
24
ecology quickly underwent dramatical changes during the fermentation step after two days.
25
The dominance of Enterobacteria was first observed and then accompanied by Lactic acid
26
bacteria and at the latest stages by acetic acid bacteria. Furthermore, Hanseniaspora opuntiae
27
yeast species predominated during cocoa bean fermentation and drying processes. In addition,
28
two bacterial species were only identified during the drying step of fermented cocoa beans
29
and are potential candidates for biological markers of this process.
M AN U
10
AC C
EP
TE D
Corresponding Author: Hamdouche Yasmine
1
ACCEPTED MANUSCRIPT 30 31
Key words: PCR-DGGE, Cocoa Fermentation, Cocoa drying, Post-harvest, Bacterial, Yeast
32
ecology.
33
1. Introduction
RI PT
34
Post harvest cocoa processing is a crucial aspect of cocoa production as it mostly determines
36
the quality of cocoa beans. The key processing steps in producing the most desirably
37
flavoured cocoa beans are fermentation, drying and roasting. Fermentation is generally carried
38
out in a traditional practice following three processes in boxes, heaps and trays through a
39
spontaneous four to six days fermentation process (Wood & Lass, 2001; Schwan & Wheals,
40
2004; De Vuyst et al., 2010). Fermentation step begins with the growth of microorganisms
41
which activities contribute to the degradation of the mucilaginous pulp surrounding the beans,
42
and produces a range of metabolic end-products (alcohols, organic acids). The main phases of
43
cocoa beans fermentation process are alcoholic fermentation by yeasts which convert the
44
sugars from the pulp into ethanol. Lactic fermentation, which produces lactic acid from
45
sugars, is mainly driven by lactic acid bacteria (LAB) and has to be avoided to obtain high
46
quality cocoa. In the last phase of fermentation, acetic acid bacteria (AAB) transform alcohol
47
into acetic acid (Schwan et al., 1995; Camu et al., 2007; Nielsen et al., 2007a).
48
Following fermentation, the beans are either sundried or artificially dried (in regions where
49
constant rainfall occurs), in order to reduce the moisture content from about 60% to about
50
7.5%. Drying must be carried out carefully to ensure that off-flavour did not develop. Drying
51
should take place slowly. If the beans are dried too quickly some of the chemical reactions
52
started in the fermentation process are not allowed to complete and the beans remain acidic,
53
with a bitter flavour. However, if the drying is too slow, moulds and off flavours can develop
54
(Fowler et al., 1998; Wood & Lass, 2001; Thompson et al., 2001).
55
Microorganisms play a crucial role in the quality of cocoa during fermentation step (Schwan
56
et al., 1995; 1997, Nielsen et al., 2007b). The influence of cocoa origin and fermentation
57
practices in
58
investigated extensively and multiphasically (Camu et al., 2007, 2008; Nielsen et al., 2007b,
59
2008; Garcia-Armisen et al., 2010; Lefeber et al., 2010; Papalexandratou & De Vuyst, 2011;
60
Papalexandratou et al., 2011a,b,c, 2013). These studies helped to understand the impact of
AC C
EP
TE D
M AN U
SC
35
the farm on the evolution of cocoa bean fermentation processes has been
2
ACCEPTED MANUSCRIPT fermentation practices on the microbial succession during cocoa bean fermentations.
62
However, little is known about the microbial communities, their dynamics and role in cocoa
63
quality during the drying step.
64
In this present study, we analyzed both cocoa fermentation and drying steps in terms of
65
microbial ecology (bacteria, yeast) in order to get an overview of the post-harvest processes
66
and their impact on the microbial ecology of cocoa. To this end, we used the denaturing
67
gradient gel electrophoresis (DGGE) method that provides a rapid and overall view of
68
community structures (Ampe et al., 1999, 2001; Ben Omar & Ampe, 2000; Cocolin et al.,
69
2001; Ercolini et al., 2001). This method was successfully employed for monitoring cocoa
70
fermentation and to discriminate geographical origin of foods products such as fish (Le
71
Nguyen et al., 2008), fruits (El Sheikha et al., 2009; 2011) and coffee (Nganou Donkeng et
72
al., 2012). In this way, we aim at define molecular markers that could be linked to cocoa
73
quality at different steps during post-harvest processing.
74 75 76
2. Materials and methods 2.1. Sampling
M AN U
SC
RI PT
61
Two hundred cocoa pods were harvested by traditional methods in Ivory Coast in March
78
2013. After opening of the pods, approximately 80 kg of beans surrounded by mucilage were
79
fermented in a wooden box during 6 days (one batch was followed), mixed by hand at 48 and
80
96 hours of fermentation, and then dried for 4 days. Cocoa beans samples (800 g) were
81
collected during fermentation at 3 time intervals (2, 4 and 6 days). 6-days fermented cocoa
82
beans were sun dried during 4 days. 200 grams of coca beans were sampled at 3 time intervals
83
(2, 3 and 4 days). Samples were collected with gloves and placed in sterile bags, put in a
84
cooler box and sent immediately by plane to Montpellier (France) and analyzed immediately
85
at Cirad laboratories. The transportation time from Ivory Coast to Cirad laboratories was
86
about 24 hours.
87
AC C
EP
TE D
77
2.2. Total DNA extraction
88
DNA extraction from bacteria and yeast is based on the method of Ampe et al. (1999) and
89
Leesing (2005). For each sample, DNA extraction was performed in duplicate. Ten grams of
90
cocoa beans were put in Falcon tubes containing 10 mL of peptone water (Biomérieux) and
91
mixed at room temperature for 30 min on a rotating wheel. Two milliliters of the resulting 3
ACCEPTED MANUSCRIPT suspensions were mixed with 0.3 g of 0.5 mm diameter acid washed glass beads (Sigma-
93
Aldrich) in Eppendorf tubes. The mixture was vortexed vigorously for 2 min in a bead beater
94
instrument (Vortex Genie 2 SI-A256) then centrifuged at 12000 g for 10 min. The pellet
95
(containing microbial cells) was resuspended in 300 µL of breaking buffer [2% Triton X-100
96
(Prolabo); 1% sodium dodecyl sulphate (SDS, Sigma); 100 mM NaCl (Sigma); 100 mM Tris–
97
HCl pH 8.0; 1 mM EDTA; pH 8.0 (Promega)]. Then, 100 µL TE (10 mM Tris-HCl; pH 8.0,
98
Promega) were added. 100 µL of lysosyme solution (25 mg. mL¯1, Eurobio) were added and
99
incubated at room temperature for 5 min and 100 µL of proteinase K solution (20 mg.mL¯1,
100
Eurobio) were added and incubated at 42°C for 20 min. Then 50 µL of 20% SDS were added
101
and incubated at 42°C for 10 min, 400 µL of mixed alkyltrimethyl ammonium bromide
102
(MATAB, Sigma) were added followed by an incubation at 65°C for 10 min. 700 µL of
103
phenol/chloroform/isoamyl alcohol (25/24/1 v/v/v, Carlo Erba) were added and the tubes
104
were centrifuged at 12000 g for 15 min. The aqueous layer was transferred to a new tube, and
105
a second phenol-chloroform extraction was performed. 600 µL of Chloroform/isoamyl
106
alcohol (24/1) was added followed by a 12000 g centrifugation for 10 min. The aqueous phase
107
was collected and 0.1 volume of sodium acetate (3M, pH 5) and an equal volume of ice-cold
108
isopropanol were added and further stored at -20°C overnight. The next day, the tubes were
109
centrifuged at 12000 g for 30 min, then the supernatant was discarded, DNA pellets were
110
washed with 500 µL 70% ethanol, after 5 min of centrifugation, the DNA pellets were air
111
dried for 1 to 2 hours. Finally, the DNA was re-suspended in 100 µL of molecular water and
112
stored at -20°C until analyses.
113
The quantity and purity of extracted DNA were verified by UV spectrophotometer (Biospec-
114
Nano), and by electrophoresis through a 0.8% agarose gel in 1x TAE buffer (containing 40
115
mM Tris-HCl, pH 7.4, 20 mM sodium acetate, 1.0 mM Na2-EDTA; Eppendorf,) with a 1 Kb
116
molecular weight ladder (Promega). After running at 100 V for 30 min, the gels were stained
117
for 15 min in an ethidium bromide solution (50 µg.mL¯1; Promega), rinsed for 5 min in
118
distilled water, then observed and photographed on a UV transilluminator, using a black and
119
white CCD camera (Scion Company) and Gel Smart 7.3 system software (Clara Vision).
AC C
EP
TE D
M AN U
SC
RI PT
92
120 121
2.3. PCR amplification of extracted DNA
122
A 160 bp fragment of The V3 variable region of bacterial 16S rDNA was amplified using gc-
123
338f (5΄-GCG CCG CCG CGC GCG GCG GGC GGG GCG GGG GCA CGG GGG GCA 4
ACCEPTED MANUSCRIPT TCC TAC GGG AGG CAG CAG-3΄, Sigma) and 518r (5΄-ATT ACC GCG GCT GCT GG-
125
3΄, Sigma) primers (Ampe et al., 1999; Leesing., 2005; ∅vreas et al., 1997). A 40-pb GC-
126
clamp was added to the forward primer in order to ensure that the fragment of DNA will
127
remain partially double stranded and that the region screened is in the lowest melting domain
128
(Sheffield et al., 1989). Each mixture (final volume 50 µL) contained about 100 ng of
129
extracted DNA, all primers at 0.2 µM, deoxyribonucleotide triphosphate (dNTPs) at 200 µM,
130
1.5 mM MgCl2, 5µL of 10x of reaction MgCl2-free Taq buffer and 1.25 U of a-Taq
131
polymerase (Promega). In order to increase the specificity of amplification and to reduce the
132
formation of spurious by-products, DNA was amplified in a “touchdown” PCR as follow:
133
94°C for 3 min, followed by 10 touchdown cycles of denaturation at 94°C for 1 min annealing
134
at 65°C (1st cycle) to 55°C (10th cycle) for 1 min and extension at 72°C for 1 min, then 25
135
cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 1 min. The extension step was
136
increased to 10 min in the last cycle.
137
For Yeasts DNA PCR amplification: a 250 bp fragment of D1/D2 region of the 26S rDNA
138
gene was amplified using eukaryotic universal primers gc-NL1 (5΄-CGC CCG CCG CGC
139
GCG GCG GGC GGG GCG GGG GCC ATA TCA ATA AGC GGA GGA AAA G-3΄) and
140
the reverse primer LS2 (5΄-ATT CCC AAA CAA CTC GAC TC-3΄, Sigma) (Kurtzman &
141
Robnett, 1998; Cocolin et al., 2000). A 30-pb GC-clamp was added to the forward primer.
142
The mixture final volume is adjusted to 50 µL, it contained about 100 ng of DNA, and all
143
reagents as described above. The amplification started by one cycle at 94°C for 3 min,
144
followed by 30 cycles of 95°C for 1 min, 52°C for 1 min and 72°C for 1 min, and final
145
extension at 72°C for 10 min.
SC
M AN U
TE D
EP
2.4. Denaturing Gradient Gel Electrophoresis (DGGE) analysis
AC C
146
RI PT
124
147
PCR products were analyzed by DGGE, by using a Bio-Rad Dcode universal mutation
148
detection system (Bio-) using the procedure described by Muyzer et al. (1993) and improved
149
by Leesing (2005). PCR amplicons were loaded into 8% (w/v) polyacrylamide gels
150
(acrylamide/N,N΄- methylene bisacrylamide, 37.5/1, Promega) in 1x TAE buffer.
151
Electrophoresis experiments were performed at 60°C using a denaturing gradient ranging
152
from 30 to 60% (100% corresponded to 7 M urea (Promega) and 40% (v/v) formamide,
153
Promega). The gel was electrophoresed at 20 V for 10 min and then at 80 V for 12 h.
5
ACCEPTED MANUSCRIPT After electrophoresis the gel was stained for 30 min in an ethidium bromide solution (50µg
155
mL¯¹ in 1x TAE buffer) and rinsed for 20 min in distilled water and then photographed as
156
described above.
157
Purification and identification of DGGE bands
158
Detected bands were cut from the DGGE gel with a sterile scalpel. DNA of each band was
159
then eluted in 100 µL TE buffer at 4°C overnight, followed by one hour at 37°C. 100µL of
160
DNA eluted from each band was purified by adding 300 µL of TE, 40 µL of sodium acetate
161
(3M, pH 5) and 1 mL of absolute ethanol and incubated at -20°C for 30 min, then centrifuged
162
12000 g for 15 min. The supernatant was discarded, DNA pellets were washed with 500 µL
163
70% ethanol and after 5 min of centrifugation, the DNA pellets were air dried for 1 to 2 hours.
164
Finally, the DNA was re-suspended in 50 µL of molecular water and stored at 4°C. Bacterial
165
and yeast purified DNA were PCR-amplified using 338f and NL-1 (without GC clamp) and
166
518r and LS-2 primers respectively. PCR amplicons were sent for sequencing at GATC
167
Biotech (Germany).
168
Sequences were compared to the GenBank database using the BLAST program (Altschul et
169
al., 1997) (http://www.ncbi.nlm.nih.gov/BLAST/) and the Ribosomal Database Project
170
(http:rdp.cme.msu.edu/index.jsp). Sequences with a percentage of identity of 97% or greater
171
were considered to belong to the same species (Stackebrandt and Goebel, 1994; Palys et al.,
172
1997).
173
3. Results
SC
M AN U
TE D
EP
175
3.1.
Analysis of bacterial communities associated to cocoa beans during wooden boxes fermentation and sun drying
AC C
174
RI PT
154
176
DGGE profiles were generated from bacterial DNA extracted at different steps during cocoa
177
bean fermentation and drying. Two bands were detected on fresh cocoa beans, one of them
178
was identified as Chloroplast DNA (Q), and a second one was not identified (P) (Figure 1).
179
Both bands could be detected throughout the process. In all cases, these two bands were not
180
informative. After 2 days of fermentation, we observed the presence of a high number (13) of
181
bacterial species. Most of them belong to the Enterobacteriaceae family (9 species); Shigella
182
dysenteriae (V), Kluyevera cryocrescens (N), Escherichia albertii (I), Escherichia sp (U),
183
Escherichia fergusonii (L), Enterobacter sp (R), Panteoa vagans (T), Tatumella terrea (O)
184
and Budvicia sp (M). Lactobacillus plantarum (A and M1) was present during all the
6
ACCEPTED MANUSCRIPT fermentation process. Two bands were revealed at 2 days and identified as AAB of the
186
Acetobacter gender (Acetobacter pasteurianus, (W) and Acetobacter sp (X), (Figure 1)
187
After 4 days of fermentation, several bands present at 2 days were detected such as Kluyevera
188
cryocrescens (N), Tatumella terrea (O) and Budvicia sp (M). The other enterobacterial
189
species found on 2 days fermented cocoa beans were no longer detected at 4 days of
190
fermentation. 2 new bands corresponding to AAB species were detected (Acetobacter
191
cerivisiae (J) and Acetobacter nitrogenifigens (D) together with the two above mentioned
192
AAB species identified at day 2. In addition to the bacterial species detected on 4-day
193
fermented beans, Lactobacillus fermentum (C), Weeksella sp (E), Enterobacter cloacae (F)
194
and Acetobacter lovaniensis (H) were detected in 6-day fermented beans. All of the bands
195
corresponding to AAB species remained detectable through the whole fermentation process.
196
During the drying step (2, 3 and 4 days), some bands that were present on (6-day) fermented
197
cocoa beans were still detectable such as Lactobacillus plantarum (A), Kluyevera
198
cryocrescens (N) and Weeksella sp (E). Two bands corresponding to Acetobacter lovaniensis
199
(H) and Lactobacillus fermentum (C) were no longer detected after 2 and 3 days of drying,
200
respectively. During the drying step, the remaining detected bands identified as AAB as well
201
as two bands (not previously detected on fermented cocoa beans) were detected and
202
corresponded to Gluconobacter xylinum (G) and Clostridium sp (K) species.
SC
M AN U
TE D
203
RI PT
185
3.2. Analysis of yeast communities of cocoa beans during wooden boxes fermentation and sun drying
204
One single band was isolated on fresh cocoa beans which corresponded to Hanseniaspora
206
opuntiae (A) that was the dominant species through the whole fermentation step (Figure 2).
207
Four bands were detected and identified as Candida insectorum (C), Pichia kudriavezii (D),
208
Pichia galeiformis (F) and Pichia species group 1 (E) after 2 days of fermentation. After 4
209
days of fermentation, the major bands (A), (C) and (D) remained detectable, and 3 bands
210
appeared corresponding to Candida sp (B), yeast species group 2 (H) and Pichia manshurica
211
(G).
212
At 6 days of fermentation, the bands identified as Candida sp (B) and Pichia species group 1
213
(E) were no longer detectable, while Pichia manshurica (G), Candida insectorum (C), Pichia
214
galeiformis (F) and Pichia kudriavezii (D), and yeast species group 2 (H) and (A) were still
215
present. Yeast species associated to 6-day fermented cocoa beans were Hanseniaspora
216
opuntiae (A), Candida insectorum (C), Pichia kudriavezii (D), Pichia galeiformis (F), Pichia
AC C
EP
205
7
ACCEPTED MANUSCRIPT 217
manshurica (G) and yeast species group 2 (H) stayed present during 3 days of sun drying.
218
Pichia species group 1 (E) was not detected during the drying step. At the last drying day (4
219
days), only half of the number of bands found at 3 days (3 bands) were detected as
220
Hanseniaspora opuntiae (A), Pichia manshurica (G) and yeast species group 2 (H).
RI PT
221
4. Discussion
223
Analyses of the bacterial and yeast ecologies of cocoa showed that post-harvest steps of cocoa
224
processes are associated to great variations in microbial ecology in terms of number of species
225
that could be detected at the molecular level.
226
During fermentation step, bacterial ecology quickly underwent dramatical changes after two
227
days and most of the identified bands were Enterobacterial species which may contribute to
228
glucose conversion into lactic acid and citric acid (Grimont & Grimont, 2006). Previous study
229
showed that Enterobacteriaceae were involved during the initial phase of cocoa beans
230
fermentation (Garcia-Amisen et al., 2010; Papalexandratou, 2011a, b, c, and 2013). Tatumella
231
saanichensis could be involved in pectinolytic degradation of cocoa pulp and in citric acid
232
assimilation (Marin-Cevada et al., 2010). Lefeber et al. (2010) found genus of Pantoea and
233
Escherichia during the early stages of both of heap and box fermentations in Ivory Coast.
234
Pantoea could be derived from environmental soil or plant material (Kageyama et al., 1992;
235
Pujol & Kado, 2002).
236
Two Lactic Acid Bacterial (LAB) species such as Lactobacillus plantarum and Lactobacillus
237
fermentum were detected on fermented cocoa beans with the dominance of L. plantarum
238
during the whole fermentation step while L. fermentum was detected in the last stages. L.
239
plantarum was already isolated during fermentation of Malaysian, Brazilian, Ivorian and
240
Ghanaian cocoa beans (Gracia-Amisen et al., 2010; Papalexandratou et al., 2011a, b, c, and
241
2013 and Camu et al., 2007). L. fermentum was isolated and detected from cocoa beans
242
fermentation, often as dominant species (Nielsen et al., 2007b; Camu et al., 2007, 2008b;
243
Kostinek et al., 2008, Papalexandratou et al., 2011a). In our case, L. fermentum was detected
244
until 3 days of drying.
245
Two Acetic acid bacteria (AAB) species were detected during the first stages of fermentation
246
(Acetobacter pasteurianus and Acetobacter sp) whereas the other AAB species (identified as
247
Acetobacter
AC C
EP
TE D
M AN U
SC
222
lovaniensis,
Acetobacter
cerivisiae,
8
Acetobacter
nitrogenifigens
and
ACCEPTED MANUSCRIPT Gluconobacter xylinum) were detected later in the process at 4 and 6 days and during drying.
249
The presence of another representative of Acetobacter and Gluconobacter species were
250
previously reported in cocoa bean fermentations (Schawan & Wheals, 2004, Camu et al.,
251
2007; De Melo Periera et al., 2013). Nielsen et al. (2007b) showed that succession of different
252
AAB species took place during fermentation as Acetobacter syzygii and Acetobacter
253
pasterianus were the dominant AAB during the first half of all fermentation, and the later
254
stages of fermentation were dominated by Acetobacter tropicalis.
255
In our study, we showed that AAB dominated in the last stages of fermentation and all drying
256
stages. The dominance of this species during the last stages of fermentation and drying could
257
be explained by their resistance to acidity and heat (Ndoye et al., 2006). These species could
258
potentially be used as good markers of complete cocoa fermentation and drying.
259
Gluconobacter xylinum and Clostridium sp were only detected during sun drying step, which
260
might indicate that they could be present in environment (soils, materials) as Clostridium
261
species generally are considered as soil contaminant (Jeong-Dong et al., 2004).
262
In the last day of cocoa beans drying the microbiota still consisted of several bacterial species,
263
which made bacterial ecology totally different from the initial fresh cocoa beans.
264
During the whole cocoa post-harvest processing, including fermentation and drying, 23
265
bacterial species were detected among which 21 were identified. All bacterial species were
266
found during the whole fermentation step and 18 bacterial species during the drying step. We
267
observed 14 common species between fermentation and drying steps. Seven species (among
268
which six Enterobacteriaceae) were only detected during fermentation step and could be
269
potential specific markers of fermentation step (Escherichia sp, Escherichia fergusonii,
270
Escherichia albertii, Enterobacter sp, Panteoa vagans, Shigella dysenteriae). During sun
271
drying, 2 species were detected which were Gluconobacter xylinum and Clostridium sp and
272
could be specific markers of the drying step.
273
Hanseniaspora opuntiae was the unique yeast species detected on fresh cocoa beans,
274
confirming the quality of cocoa pods and the relative sterility of beans inside the pod. This
275
species could be present at the surface of the pod as it was involved in all studied fermentation
276
until now and was usually dominant in cocoa fermentation (Camu et al., 2007, Lefeber et al.,
277
2011, Papalexandratou & De Vuyst, 2011, Papalexandratou et al., 2011a, b, c, 2013).
278
Hanseniaspora opuntiae was dominant during the whole fermentation and drying steps
279
followed by Pichia manshurica and yeast species group 2.
AC C
EP
TE D
M AN U
SC
RI PT
248
9
ACCEPTED MANUSCRIPT During drying step, the bands identified as Hanseniaspora opuntiae, Candida insectorum,
281
Pichia kudriavezii and yeast species group 2 (Pichia sporocuriosa and Issatchenkia
282
hanoiensis) decreased in time and, at the end of drying, these three species were the only ones
283
detected.
284
Most of the yeast species present on fermented cocoa beans disappeared during the drying
285
process, and no new species could be detected. Yeast species appeared less resistant to (sun)
286
drying than bacterial species. Dynamics of yeast communities associated to cocoa beans
287
fermentation were more important than dynamics of yeast communities associated to cocoa
288
beans during the drying step.
289
Eight yeast species were detected in the whole process, all of these species were present
290
during the fermentation step and seven during sun drying.
291
One species (Candida sp) was only detected at a specific step during fermentation (4 days)
292
before completion. These species could be specific marker for incomplete fermentation.
293
Both yeast and bacterial underwent dramatical changes during both cocoa fermentation and
294
drying. While yeast and bacterial ecologies undergo changes during both treatments such as
295
fermentation and sun drying, a much reduced number of yeast species was recorded at the
296
fourth day of drying.
TE D
297
M AN U
SC
RI PT
280
Aknowledgements
299
This work was supported by DESI grant from CIRAD France (to support PhD students) and
300
Y. Hamdouche was supported by a PhD fellowship from HUBBARD.
AC C
301
EP
298
302
References
303
Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller, W., & Lipman,
304
D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein databases search
305
programs. Nucleic Acids Research, 25, 3389-3402.
306
Ampe, F., Ben Omar, N., Moizan, C., Wacher, C., & Guyot, J. P. (1999). Polyphasic study of
307
the spatial distribution of microorganisms in Mexican pozol, a fermented maize dough,
10
ACCEPTED MANUSCRIPT demonstrates the need for cultivation-independent methods to investigate traditional
309
fermentations. Applied and Environmental Microbiology, 65, 5464–5473.
310
Ampe, F., Audrey, S., & Nadine, Z. (2001). Dynamics of the microbial community
311
responsible for traditional sour cassava starch fermentation studied by denaturing gradient gel
312
electrophoresis and quantitative rRNA hybridization. International Journal of Food
313
Microbiology, 65, 45–54.
314
Ben Omar, N., & Ampe, F. (2000). Microbial community dynamics during production of the
315
Mexican fermented maize dough pozol. Applied and Environmental Microbiology, 47, 3664–
316
3673.
317
Camu, N., De Winter, T., Verbrugghe, K., Cleenwerck, I., Vandamme, P., Takrama, J. S.,
318
Vancanneyt, M., & De Vuyst, L. (2007). Dynamics and biodiversity of populations of lactic
319
acid bacteria and acetic acid bacteria involved in spontaneous heap fermentation of cocoa
320
beans in Ghana. Applied and Environmental Microbiology, 73, 1809–1824.
321
Camu, N., González, Á., De Winter, T., Van Schoor, A., De Bruyne, K., Vandamme, P.,
322
Takrama, J. S., Addo, S. K., & De Vuyst, L.(2008). Influence of turning and environmental
323
contamination on the dynamics of populations of lactic acid and acetic acid bacteria involved
324
in spontaneous cocoa bean heap fermentation in Ghana. Applied and Environmental
325
Microbiology, 74, 86–98.
326
Cocolin, L., Bisson, L. F., & Mills, D. A. (2000). Direct profiling of the yeast dynamics in
327
wine fermentations. FEMS Microbiology Letters, 189, 81-87.
328
Cocolin, L., Manzano, M., Cantoni, C., & Comi, G. (2001). Denaturing gradient gel
329
electrophoresis analysis of the 16S rRNA gene V1 region to monitor dynamic changes in the
330
bacterial population during fermentation of Italian sausages. Applied and Environmental
331
Microbiology, 67, 5113–5121.
332
Daniel, H. M., Vrancken, G., Takrama, J. F., Camu, N., De Vos, P., & De Vuyst, L. (2009).
333
Yeast diversity of Ghanaian cocoa bean heap fermentations. FEMS Yeast Research, 9, 774–
334
783.
335
De Melo Pereira, G. F., Magalhães-Guedes, K. T., Schwan, R. S. (2013). rDNA-based DGGE
336
analysis and electron microscopic observation of cocoa beans to monitor microbial diversity
337
and distribution during the fermentation process. Food Research International 53, 482–486.
AC C
EP
TE D
M AN U
SC
RI PT
308
11
ACCEPTED MANUSCRIPT De Vuyst, L., Lefeber, T., Papalexandratou, Z., & Camu, N. (2010). The functional role of
339
lactic acid bacteria in cocoa bean fermentation. In: Mozzi, F., Raya, R.R., &Vignolo, G.M.
340
(Eds.), Biotechnology of Lactic Acid Bacteria (pp. 301-326). Ames, IA: Novel Applications.
341
Wiley-Blackwell.
342
Dias, D. R., Schwan, R. F., Carlos, L & Lima, O. (2003). Metodologia para elaboração de
343
fermentado de Cajá (Spondias mombin L.). Ciência e Technologia de Alimentos, 23(3), 342-
344
350.
345
El Sheikha, A. F., Condur, A., Métayer, I., Le Nguyen, D. D., Loiseau, G., & Montet, D.
346
(2009). Determination of fruit origin by using 26S rDNA fingerprinting of yeast communities
347
by PCR-DGGE: preliminary application to Physalis fruits from Egypt. Yeast, 26, 567-573.
348
El Sheikha, A. F., Bouvet, J. M., & Montet, D. (2011). Biological bar code for determining
349
the geographical origin of fruits using 28S rDNA fingerprinting of fungal communities by
350
PCR-DGGE: an application to Shea tree fruits. Crops &Food, 3, 40-47.
351
Ercolini, D., Moschetti, G., Blaiotta, G., & Coppola, S. (2001). Behavior of variable V3
352
region from 16S rDNA of lactic acid bacteria in denaturing gradient gel electrophoresis.
353
Current Microbiology, 42, 199–202.
354
Fowler, M. S., Leheup, P., & Cordier, J. L. (1998). Cocoa, coffee and tea. In: Wood, B.J.B
355
(Ed.), Microbiology of Fermented Foods (pp. 128-147). London: Blackie Academic and
356
Professional.
357
Garcia-Armisen, T., Papalexandratou, Z., Hendryckx, H., Camu, N., Vrancken, G., & De
358
Vuyst, L., Cornelis, P. (2010). Diversity of the total bacterial community associated with
359
Ghanaian and Brazilian cocoa bean fermentation samples as revealed by a 16 S rRNA gene
360
clone library. Applied Microbiology and Biotechnology, 87, 2281–2292.
361
Grimont, F., & Grimont, P.A.D. (2006). The genus Enterobacter, In: Dworkin, M., Falkow,
362
S., Rosenberg, E., Schleifer, K. M., & Stackebrandt, E., (Eds.), The Prokaryotes (pp. 163–
363
200). New York: Springer.
364
Jeong-Dong , K., Dae-Weon, L.,
365
(2004). Distribution and antimicrobial susceptibility of Clostridium species in soil
366
contaminated with domestic livestock feces of Korea. Journal of microbiology and
367
biotechnology, 14, 1017-7825.
AC C
EP
TE D
M AN U
SC
RI PT
338
Kyou-Seung, L., Chang-Hyun, C., & Kook-Hee , K.
12
ACCEPTED MANUSCRIPT Kageyama, B., Nakae, M., Yagi, S., & Sonoyama, T. (1992). Pantoea punctata sp. nov.,
369
Pantoea citrea sp. nov., and Pantoea terrea sp. nov. isolated from fruit and soil samples.
370
International Journal of Systematic Bacteriology, 42, 203-210.
371
Kurtzman, C. P., & Robnett, C. J. (1998). Identification and phylogeny of ascomycetous
372
yeasts from analysis of nuclear large subunit (26S) ribosomal DNA partial sequences. Antonie
373
Van Leeuwenhoek, 73, 331-371.
374
Leesing, R. (2005). Identification and validation of specific markers for traceability of
375
aquaculture fish for import/export. PhD thesis. University of Montpellier 2, France.
376
Lefeber, T., Gobert, W., Vrancken, G., Camu, N., & De Vuyst, L. (2010). Dynamics and
377
species diversity of communities of lactic acid bacteria and acetic acid bacteria during
378
spontaneouscocoa bean fermentations in vessels. Food Microbiology, 28, 457–464.
379
Muyzer, G., De Waal, E. C., & Uitterlinden, A. G. (1993). Profiling of complex microbial
380
populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-
381
amplified genes coding for 16S rRNA. Applied Environmental Microbiology, 59, 695-700.
382
Nielsen, D. S., Schillinger, U., Franz, C. A. M. P., Bresciani, J., Amoa-Awua, W., Halzapel,
383
W. H., & Jakobsen, M. (2007a). Lactobacillus ghanensis sp. nov., a motile lactic acid
384
bacterium isolated from Ghanaian cocoa fermentations. International Journal of Systematic
385
and Evolutionary Microbiology, 57, 1468-1472.
386
Nganou, D. N., Durand, N., Tatsadjieu, N. L., Meile, J. C., El Sheikha, A. F., Montet, D., et
387
al. (2012). Determination of coffee origin by using 28S rDNA fingerprinting of fungal
388
communities by PCR-DGGE: Application to the Cameroonian coffee. International Journal
389
of Biosciences, 2(5), 18-30.
390
Nielsen, D. S., Teniola, O. D., Ban-Koffi, L., Owusu, M., Andersson, T. S., & Holzapfel, W.
391
H. (2007b). The microbiology of Ghanaian cocoa fermentations analysed using culture-
392
dependent and culture-independent methods. Inernational Journal of Food Microbiology,
393
114, 168–186.
394
Nielsen, D. S., Nielsen P. S., & Van den Berg, F. (2008). Investigating the fermentation of
395
cocoa by correlating denaturing gradient gel electrophoresis profiles and near infrared spectra.
396
International Journal of Food Microbiology, 125, 133-40.
AC C
EP
TE D
M AN U
SC
RI PT
368
13
ACCEPTED MANUSCRIPT Le Nguyen, D. D, Ha, H, Dijoux, D, et al. (2008). Determination of fish origin by using 16S
398
rDNA fingerprinting of bacterial communities by PCR–DGGE: an application on Pangasius
399
fish from Vietnam. Food Control, 19, 454–460.
400
Øvreas, L., Forney, L., Dae, F. L., & Torsvik, V. (1997). Distribution of Bacterioplankton in
401
Meromictic Lake Sælenvannet, as Determined by Denaturing Gradient Gel Electrophoresis of
402
PCR-Amplified Gene Fragment Coding for 16S rRNA. Applied and Environmental
403
Microbiology, 63, 3367-3373.
404
Palys, T., Nakamura, L. K., & Coha, F. M. (1997). Discovery and classification of ecological
405
diversity in the bacterial world: the role of DNA sequence data. International Journal of
406
Systematic Bacteriology, 47, 1145-1156.
407
Papalexandratou, Z., & De Vuyst, L. (2011). Assessment of the yeast species diversity of
408
cocoa bean fermentations in different cocoa-producing regions using denaturing gradient gel
409
electrophoresis. FEMS. Yeast Research, 11, 564–574.
410
Papalexandratou, Z., Camu, N., Falony, G., & De Vuyst, L. (2011a). Comparison of the
411
bacterial species diversity of spontaneous cocoa bean fermentations carried out at selected
412
farms in Ivory Coast and Brazil. Food Microbiology, 28, 964–973.
413
Papalexandratou, Z., Falony, G., Romanens, E., Jimenez, J. C., Amores, F., Daniel, H. M., &
414
De Vuyst, L., (2011b). Species diversity, community dynamics, and metabolite kinetics of the
415
microbiota associated with traditional Ecuadorian spontaneous cocoa bean fermentations.
416
Applied and Environmental Microbiology, 77, 7698–7714.
417
Papalexandratou, Z., Vrancken, G., De Bruyne, K., Vandamme, P., & De Vuyst, L. (2011c).
418
Spontaneous organic cocoa bean fermentations in Brazil are characterized by a restricted
419
species diversity of lactic acid bacteria and acetic acid bacteria. Food Microbiology, 28,
420
1326– 1338.
421
Papalexandratou, Z., Lefeber,T., Bahrim, B., Lee, O. S., Daniel, H. M., & De Vuyst, L.
422
(2013). Hanseniaspora opuntiae, Saccharomyces cerevisiae, Lactobacillus fermentum, and
423
Acetobacter pasteurianus predominate during well-performed Malaysian cocoa bean box
424
fermentations, underlining the importance of these microbial species for a successful cocoa
425
bean fermentation process. Food Microbiology, 28, 1326-1338.
426
Pujol, C. J., & Kado, C. I. (2002). Genetic and biochemical characterization of the pathway in
427
Pantoea citrea leading to pink disease of pineapple. Journal of Bacteriology, 182, 2230-2237.
AC C
EP
TE D
M AN U
SC
RI PT
397
14
ACCEPTED MANUSCRIPT Schwan, R. F., Rose, A. H., & Board, R. G. (1995). Microbial fermentation of cocoa beans,
429
with emphasis on enzymatic degradation of the pulp. Journal of Applied Bacteriology
430
Symposium Supplement, 79, 96-107.
431
Schwan, R. F., Cooper, R.M., & Wheals, A.E., (1997). Endopolygalacturonase secretion by
432
Kluyveramyces marxianus and other cocoa pulp-degrading yeasts. Enzyme and Microbiology
433
Technology 21, 234-244.
434
Schwan, R. F., & Wheals, A. E. (2004). The microbiology of cocoa fermentation and its role
435
in chocolate quality. Critical Reviews in Food Science and Nutrition, 44(4), 205–221.
436
Sheffield, V. C., Beck, J. S., Stone, E. M., & Myers, R. M. (1989). Attachment of a 40 bp
437
G+C rich sequence (GC-clamp) to genomic DNA fragments by polymerase chain reaction
438
results in improved detection of single-base changes. Proceeding of the National Academy of
439
Sciences of the United States of American, 86, 232-236.
440
Thompson, S. S., Miller, K. B., & Lopez, A. S. (2001). Cocoa and coffee. In: Doyle, M. P.,
441
Beuchat, L. R., Montville, T. J. (Eds.), Food Microbiology Fundamentals and Frontiers (pp.
442
649–661). Washington: ASM Press.
443
Wood, G. A. R., & Lass, R. A. (2001). Cocoa. (4th ed). Oxford: Blackwell Science.
AC C
EP
TE D
M AN U
SC
RI PT
428
15
ACCEPTED MANUSCRIPT Table 1: Identification of DGGE bands for bacterial species (A to X)
AC C
EP
TE D
M AN U
SC
RI PT
Table 2: Identification of DGGE bands for yeast species (A to H)
ACCEPTED MANUSCRIPT Table 1: Letter
Genus/ Species
A
Lactobacillus
Percent identity
Query cover
E value
References
100%
100%
1e-33
HM218742
99%
100%
100%
100%
100%
B
Not identified
C
Lactobacillus
D
Acetobacter nitrogenifigens
1e-44
CP005958
2e-22
KC763628
3e-47
JX628863
SC
fermentum
RI PT
plantarum
Weeksella sp
97%
F
Enterobacter
100%
100%
4e-50
HG421017
97%
100%
2e-47
NR074338
98%
100%
8e-47
HM217974
cloacae G
Gluconobacter xylinum Acetobacter lovaniensis
TE D
H
M AN U
E
Escherichia albertii
97%
100%
2e-58
HM194885
J
Acetobacter
100%
100%
3e-51
KC485819
100%
100%
4e-55
JX575132
99%
100%
1e-34
JQ319655
Budvicia sp
96%
100%
1e-45
JQ080726
Kluyvera
98%
100%
2e-42
AM933754
98%
100%
3e-41
JQ726629
EP
I
cerivisiae
L
Clostridium sp
AC C
K
Escherichia fergusonii
M N
cryocrescens O
Tatumella terrea
P
Not identified
ACCEPTED MANUSCRIPT Q
Chloroplast DNA
100%
100%
4e-60
JX912466
R
Enterobacter sp
100%
100%
2e-68
JX267074
S
Tatumella
98%
100%
3e-41
JQ726633
2e-48
HG421010
T
Pantoea vagans
100%
100%
U
Escherichia sp
100%
100%
V
Shigella dysenteriae
100%
100%
W
Acetobacter
2e-68
HM346189
8e-62
KC768788
100%
95%
AC C
EP
TE D
Acetobacter sp
100%
M AN U
Acetobacter sp
SC
KC478491
pasteurianus
X
RI PT
saanichensis
98%
9e-57
3e-68
KF668642 AB853266
ACCEPTED MANUSCRIPT Table 2: Letter
Genus/ Species
Percent identity
Query cover
E value
References
A
Hanseniaspora
100%
100%
7e-90
KC544476
Candida sp
96%
100%
C
Candida insectorum
99%
100%
D
Pichia kudriavezii
99%
100%
96%
100%
strain YM25991 E
Species group of
Pichia klyvera Pichia fermentans
Pichia galeiformis
G
Pichia manshurica
H
Species group 2:
100%
EF042050
2e-37
JN544058
2e-47
KC494718
2e-49
JX103191 EU647835
100%
4e-49
JX880398
100%
96%
1e-87
JX183960
99%
100%
1e-32
TE D
F
M AN U
Pichia:
1e-40
SC
B
RI PT
opuntiae
JF749216
Issatchenkia
FJ153148
EP
Pichia sporocuriosa
AC C
hanoiensis
ACCEPTED MANUSCRIPT
Figure 1: Picture of DGGE gel showing the evolution of bacterial ecology structure on fresh, fermented and sun dried cocoa beans at different time intervals. 1: Fresh Cocoa beans (T0 time of fermentation), 2: Cocoa beans fermented for 2 days, 3:
RI PT
Cocoa beans fermented for 4 days, 4: Cocoa beans fermented for 6 days and not dried, 5: Cocoa beans fermented for 6 days and dried for 2 days, 6: Cocoa beans fermented for 6 days and dried for 3days, 7: Cocoa beans fermented for 6 days and dried for 4 days.
M1: Lactobacillus plantarum DNA, and M2: Escherichia coli DNA are pure bacterial strain
SC
markers.
M AN U
Identification of DGGE bands for bacterial species (A to X) (Table 1)
Figure 2: Picture of DGGE gel showing the evolution of yeast ecology structure from on fresh, fermented and sun dried cocoa beans at different time intervals. 1: Fresh Cocoa beans (T0 time of fermentation), 2: Cocoa beans fermented for 2 days, 3:
TE D
Cocoa beans fermented for 4 days, 4: Cocoa beans fermented for 6 days and not dried, 5: Cocoa beans fermented for 6 days and dried for 2 days, 6: Cocoa beans fermented for 6 days and dried for 3 days, 7: Cocoa beans fermented for 6 days and dried for 4 days. M2: Pichia sorbitophila DNA are pure yeast strain markers.
EP
M1: Candida apicola DNA
AC C
Identification of DGGE bands for yeast species (A to H) (Table 2)
1
3
2
ACCEPTED MANUSCRIPT 4 7 5 6
M1 B C
D E F G H
I
J K
P Q
M N O
R
S T U V
AC C
Figure 1
X
EP
W
TE D
M2
M AN U
L
SC
RI PT
A
ACCEPTED MANUSCRIPT
1
3
2
4
5
M1
B
SC
C D
7
RI PT
A
6
M AN U
M2
TE D
E F
AC C
EP
G H
Figure 2
ACCEPTED MANUSCRIPT 1. Images and statistical analysis The photographs of DGGE gels were analyzed using Image Quant TL software v. 2003 (Amersham Biosciences, USA). In each lane, banding patterns were standardized with two reference patterns for bacterial DNA (Escherichia coli and Lactobacillus plantarum) and for yeast DNA (Candida apicola and Pichia sorbitophila). The band relative positions were
RI PT
compared with standard patterns.
In DGGE analysis, the generated banding pattern is considered as an image of all the major microbial populations. An individual band generally refers to a unique ‘sequence type’ or phylotype (Van Hannen et al., 1999; Muyzer et al., 1995; Kowalchuk et al., 1997). The
SC
DGGE fingerprinting was manually scored by presence and absence of co-migrating bands, independently of intensity. Pair wise community similarities were quantified using Dice
M AN U
similarity coefficient (SD ): SD =2 Nc/ Na + Nb (Heyndrickx et al., 1996)
Na represented the number of bands detected in sample A, Nb represented the number of bands in sample B, and Nc represented the number of bands common to both samples. The similarity index was expressed within a range of 0 (completely dissimilar) to 100 (perfect similarity). Statistical analysis were performed using the similarity matrix to group samples
TE D
according to their similarity index by Principal Component Analysis (PCA) using Statistica (version 7) software (StatSoft, USA).
EP
2. Results of principal component analysis (PCA) Principal component analysis (PCA) was used to compare bacterial communities during the
AC C
post-harvest process of cocoa beans (supplementary Figure S1). About 95% of the variability of the sample set could be described using this method and four distinct groups could be discriminated: the first group contained only fresh cocoa beans, a second group included only cocoa beans fermented for 2 days, a third group included 2 types of samples (cocoa beans fermented for 4 and 6 days) and the last group dried cocoa beans samples (2, 3 and 4 days). Principal component analysis (PCA) was used to compare yeast communities; the statistical analysis separates three sample groups (see supplementary Figure S2) and represents more than 98% of the variation within the sample data set. A first group contained initial fresh cocoa beans and final (4 days drying) cocoa bean samples, a second group contained cocoa
ACCEPTED MANUSCRIPT beans fermented for 2 days, and a third group included samples of 4- and 6-day fermented as
TE D
M AN U
SC
RI PT
well as 2- and 3-day dried cocoa beans).
Figure S1: Factorial variance analysis of PCR-DGGE gel of 16S rDNA band profiles of
AC C
EP
Bacteria from fermentation and drying of cocoa beans
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S2: Factorial variance analysis of PCR-DGGE gel of 26S rDNA band profiles of
References
EP
Yeasts from fermentation and drying of cocoa beans.
AC C
Heyndrickx, M., Vauterin, L., Vandamme, P., Kersters, K., & De Vos, P. (1996). Applicability of combined amplified ribosomal DNA restriction analysis (ARDRA) patterns in bacterial phylogeny and taxonomy. Journal of Microbiological Methods, 26, 247-259. Muyzer, G., Teske, A., Wirsen, C. O., & Jannasch, H. W. (1995). Phylogenetic relationships of Thiomicrospira species and their identification in deep-sea hydrothermal vent sample by denaturing gradient gel electrophoresis of 16S rDNA fragment. Archives of Microbiology, 164, 165-172. Kowalchuk, G. A., Stephen, J. R., de Boer, W., Prosser, J. I., Embley, T. M., & Woldendorp, J. W. (1997). Analysis of ammonia-oxydizing bacteria of the beta subdivision of the class
ACCEPTED MANUSCRIPT Proteobacteria in coastal sand dunes by denaturing gradient gel electrophoresis and sequencing of PCR amplified 16S ribosomal DNA fragments. Applied and Environmental Microbiology, 63, 1489-1497. Van Hannen, E. J., Zwart, G., Van Agterveld, M. P., Gons, H. J., Ebert, J., & Laanbroek, H. J. (1999). Changes in bacterial and eukaryotic community structure after mass lysis of
RI PT
filamentous cyanobacteria associated with viruses. Applied and Environmental Microbiology
AC C
EP
TE D
M AN U
SC
65, 795-801.