- Email: [email protected]
Comparison study of the volatile profiles and microbial communities of Wuyi Qu and Gutian Qu, two major types of traditional fermentation starters of Hong Qu glutinous rice wine
Accepted Manuscript Comparison study of the volatile profiles and microbial communities of Wuyi Qu and Gutian Qu, two major types of traditional fermentation starters of Hong Qu glutinous rice wine Zhibin Liu, Zhiyao Wang, Xucong Lv, Xiaoping Zhu, Liling Chen, Li Ni PII:
S0740-0020(17)30093-X
DOI:
10.1016/j.fm.2017.07.019
Reference:
YFMIC 2834
To appear in:
Food Microbiology
Received Date: 31 January 2017 Revised Date:
17 July 2017
Accepted Date: 25 July 2017
Please cite this article as: Liu, Z., Wang, Z., Lv, X., Zhu, X., Chen, L., Ni, L., Comparison study of the volatile profiles and microbial communities of Wuyi Qu and Gutian Qu, two major types of traditional fermentation starters of Hong Qu glutinous rice wine, Food Microbiology (2017), doi: 10.1016/ j.fm.2017.07.019. 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 1
Comparison study of the volatile profiles and microbial communities of Wuyi Qu
2
and Gutian Qu, two major types of traditional fermentation starters of Hong Qu
3
glutinous rice wine
RI PT
4
Zhibin Liu1, Zhiyao Wang1, Xucong Lv2, Xiaoping Zhu1, Liling Chen1, Li Ni1*
5 6 1
Institute of Food Science & Technology, Fuzhou University, Fuzhou 350108, P.R. China
8
2
National Engineering Research Center of JUNCAO Technology, Fujian Agriculture and
M AN U
9
SC
7
Forestry University, Fuzhou, Fujian 350002, China
10
Corresponding author: Dr. Li Ni, Institute of Food Science & Technology, Fuzhou
12
University, No. 2 Xueyuan Road, Fuzhou 350108, P.R. China
13
Tel/ Fax: +86-591-22866378
14
Email: [email protected]
EP
16
AC C
15
TE D
11
1
ACCEPTED MANUSCRIPT 17
Abstract Hong Qu, which mainly contains Monascus sp. and other microorganisms, as
19
well as numerous microbial metabolites, is used as the fermentation starter of Hong
20
Qu glutinous rice wine, a traditional alcoholic beverage. Two widely-used types of
21
Hong Qu, namely Wuyi Qu (WYQ) and Gutian Qu (GTQ), were thoroughly
22
compared for their fermentation properties, volatile profiles, and microbiota structures
23
in this study. Significantly higher color value, glucoamylase and α-amylase activities
24
were discovered in WYQ. And substantial variation in volatile components and
25
microbial communities were also observed between them. It was identified that
26
bacterial genus Burkholderia dominated GTQ (71.62%) and Bacillus dominated
27
WYQ (44.73%), while Monascus purpureus was the most abundant fungal species in
28
both types of starters (76.99%). In addition, 213 bacterial genera and 150 fungal
29
species with low-abundance were also detected. Since the Linear Discriminant
30
Analysis Effect Size algorithm, 14 genus-level bacterial taxa and 10 species-level
31
fungal taxa could be utilized to distinguish these two types of starters. Moreover, the
32
potential correlation of the volatile components and microbiota within WYQ and
33
GTQ were further analyzed, by utilizing Partial Least Squares Discriminant Analysis.
34
Ultimately, this study provides detailed insight into the volatile profiles and microbial
35
communities presented in Hong Qu.
AC C
EP
TE D
M AN U
SC
RI PT
18
36 37
Keywords: Hong Qu glutinous rice wine starter; bacterial community; fungal
38
community; high throughput sequencing; Monascus purpureus; volatile components
39 2
ACCEPTED MANUSCRIPT 40
1. Introduction Chinese yellow rice wine, which may date back as early as the seventh
42
millennium before Christ (B.C.), is a traditional fermented alcoholic beverage that is
43
brewed from glutinous rice or wheat (McGovern et al., 2004). Hong Qu glutinous rice
44
wine, primarily produced in Fujian province, China, is one of the most ancient in this
45
particular category. The distinct characteristics of Hong Qu glutinous rice wine is the
46
utilization of Hong Qu as the fermentation starter, which produces a brilliant
47
bright-red color, fine sweet flavor, and also offers a healthcare functionality to the
48
wine (Park et al., 2016). Through an empirical solid fermentation process, Hong Qu is
49
inoculated with diverse bacteria and fungi, along with numerous pigments, enzymes
50
and other metabolites, facilitating the formation of alcohol and its unique flavor.
51
Similar to other traditional fermented foods, Hong Qu is generally prepared under
52
non-sterile fermentation conditions. Thus, a wide variety of microbes are presented in
53
Hong Qu, and the quality of starters will tend to vary from region to region, resulting
54
in the uncontrollability of the fermentation process of Hong Qu glutinous rice wine.
55
Thus, to fully understand microbiota structures, as well as the fermentation properties
56
and volatile profiles, and identify the key microorganism responsible for the aromatic
57
forming or other fermentation activities, will help to establish next generation starters,
58
which have a limited number of pure strains, and so as to improve the controllability.
AC C
EP
TE D
M AN U
SC
RI PT
41
59
Despite the variation of microbes within Hong Qu, two distinguishing types of
60
Hong Qu can still be recognized, which are Wuyi Qu (WYQ) and Gutian Qu (GTQ).
61
In general, WYQ presents a somewhat darker color and has a higher saccharifying 3
ACCEPTED MANUSCRIPT and liquefying power than that of GTQ (based on experience). Moreover, a dramatic
63
difference in flavor can easily be identified between these two types of Hong Qu, as
64
well as the rice wines that are fermented from them. Diverse microbiota within the
65
starters may be the primary factor or cause relating to these differences
RI PT
62
Bacteria and fungi, particularly filamentous fungi and yeasts, play an essential
67
role in the quality of Chinese rice wine fermentation starters, as their involvement in
68
the saccharification, liquification, alcoholic fermentation and flavor formation during
69
the wine brewing process. Therefore, numerous studies have been conducted to
70
adequately characterize the bacterial and fungal diversity of Chinese rice wine
71
fermentation starters (Guan et al., 2012; Lu et al., 2008). Due to the different
72
processing methods, Hong Qu have distinct aromatic and microbial profiles with other
73
rice wine starters. Previously our group has investigated the bacterial and fungal
74
community structures in WYQ. Briefly, based on culture-dependent or PCR-RFLP
75
and PCR-DGGE or MALDI-TOF mass spectrometry fingerprinting methods, 16
76
filamentous fungi species, including Monascus purpureus, Aspergillus flavus,
77
Aspergillus niger, Rhizopus oryzae and et al (Lv, Huang, et al., 2012), and 2 yeast
78
species, including Saccharomycopsis fibuligera and Saccharomyces cerevisiae (Lv,
79
Huang, Zhang, et al., 2013), and 16 bacterial species belonging to 6 bacterial genera
80
(Bacillus,
81
Lactococcus)(Lv et al., 2016) were identified from WYQ. Nevertheless, in terms of
82
GTQ, the research is limited, although this variety is a more extensively used one in
83
folk. The recent expansion of the next-generation sequencing technology and data
AC C
EP
TE D
M AN U
SC
66
Staphylococcus,
Leuconostoc,
4
Pediococcus,
Lactobacillus
and
ACCEPTED MANUSCRIPT 84
mining strategies could provide a more clarifying and precise picture of the microbial
85
community of fermentation starters and, thus, enable researchers to obtain a more
86
global insight into the functions and interactions of various microorganisms. In the present study, through solid phase microextraction (SPME)-GC/MS
88
analysis and high throughput sequencing of bacterial 16S rRNA genes and fungal ITS
89
rRNA genes upon Illumina HiSeq platform, the volatile profiles and microbial
90
community structures of WYQ and GTQ were investigated and compared. In addition,
91
the association between volatile components and microorganisms within these two
92
types of starters were also calculated through Partial Least Squares Discriminant
93
Analysis (PLS-DA) modelling, thus to preliminarily explore the roles of the microbes
94
as they relate to the aroma formation of Hong Qu. This study may provide a better
95
understanding of the microbial community in Hong Qu and their contribution to the
96
fermentation process, which may be helpful for the development of new starters with
97
pure strains and the improvement of the controllability of Hong Qu glutinous rice
98
wine brewing.
99
2. Materials and methods
101
SC
M AN U
TE D
EP
AC C
100
RI PT
87
2.1 Sample collection
Five Wuyi Hong Qu (WYQ) were collected from local markets of Jian’ou,
102
Fuzhou, Fuqing, Gutian, and Yongchun cities of Fujian Province, China; moreover,
103
five Gutian Hong Qu (GTQ) were also collected from local markets of Gutian city of
104
Fujian Province, China. All the collected samples are exhibited in Figure 1. After
105
collection, the samples were then ground into fine powder and appropriately stored at 5
ACCEPTED MANUSCRIPT 106
-20 °C until analysis.
107
2.2 Color value analysis In general, three colors (red, orange and yellow), including both water-soluble
109
and alcohol-soluble ones, are produced by Monascus stains, which with a maximum
110
absorption at 510 nm, 465 nm, and 410 nm, respectively. Normally, A410, A465, and
111
A510 are used to estimate the content of this three pigments (Yang et al., 2005; Zhang
112
et al., 2017). In this study, distilled water or 70% (v/v) ethanol were used for pigment
113
extraction: 0.2 g ground Hong Qu powder was suspended in 10 mL distilled water or
114
70% (v/v) ethanol at 60 °C for 60 min; then the extracted solution was centrifuged at
115
8000 rpm for 15min. Subsequently, appropriate dilution of supernate was measured
116
for absorbance at 410 nm, 465nm and 510 nm with a spectrophotometer (Hitachi
117
U-1900, Tokyo, Japan). The color value was defined as the summation of absorbance
118
units at 410 nm, 465nm and 510 nm× dilution factor per gram of dry samples (U/g).
119
2.3 Fermentative power analysis
TE D
M AN U
SC
RI PT
108
Moreover the fermentative power, including α-amylase, glucoamylase and
121
protease, of the ten Hong Qu samples were also determined as previously described
122
(Lv, Huang, et al., 2012). Briefly, the iodine colorimetry method was used to
123
determine the α-amylase activities of the samples (De Moraes et al., 1995). One unit
124
(U) of α-amylase activity was defined as the quantity of enzymes required to
125
hydrolyze 1 mg of starch in 10 min at 40 °C. The dinitrosalicylic (DNS) acid method
126
was then utilized to determine the glucoamylase activities of the samples (Bernfeld,
127
1955). One unit (U) of glucoamylase activity was defined as the amount of enzymes
AC C
EP
120
6
ACCEPTED MANUSCRIPT releasing 1 mg of glucose from the substrate in 60 minutes under strict assaying
129
conditions. Next, the protease activity was determined, according to the method
130
described by Farley and Ikasari (Farley and Ikasari, 1992). One unit (U) of protease
131
activity was defined as the amount of enzyme that liberated 1 µg tyrosine per min
132
under assaying conditions. All the experiments were conducted in triplicate.
133
2.4 Volatile profiles analyses
RI PT
128
Volatile profiles of the ten starters were analyzed by utilizing SPME–GC/MS
135
methods(Luo et al., 2008). Briefly, 1.00 g of each ground Hong Qu powder, 2.00 g of
136
NaCl and 5 mL distilled water were transferred to a 30 mL vial. This vial was then
137
tightly capped with a silicon septum and a 50/30 µm divinylbenzene/carboxen/poly
138
(dimethylsiloxane) (DVB/CAR/PDMS) coated fibre (Supelco, Inc., Bellefonte, PA,
139
USA) was inserted into the headspace of the vial for the volatile compound extraction
140
at 50 °C water bath for 40 min.
TE D
M AN U
SC
134
GC/MS analyses were performed on an Agilent 7890B gas chromatograph
142
(Agilent, Palo Alto, CA, USA) coupled with an Agilent 5973C mass spectrometer
143
(Agilent, Palo Alto, CA, USA). Separation of compounds was performed on an
144
HP-5MS column (30.0 m×0.25 mm×0.25µm, Agilent, USA). Helium was used as a
145
carrier gas at a constant flow rate of 1 mL/min. Oven temperature was maintained at
146
40 °C for 5 min, programmed at 5 °C/min to 250 °C and held for 5 min. The interface
147
temperature was set at 280 °C. The mass spectrometer was operated in electron
148
impact mode with the electron energy set at 70 eV and a scan range of 30–550 m/z.
149
The temperature of MS source and quadrupole was set at 230 and 150 °C, respectively.
AC C
EP
141
7
ACCEPTED MANUSCRIPT Each compound was identified using the National Institute of Standards and
151
Technology (NIST) library (11 L). The relative percentages of the detected peaks
152
were obtained by peak-area normalization. Afterward, based on the volatile profile
153
data set, Principal Component Analysis (PCA) was used to visualize the differences
154
among the starter samples, by using R software (Version 3.2.5) with the vegan, ade4
155
and ggplot2 packages.
156
2.5 Bacterial and fungal DNA extraction and Illumina sequencing
SC
RI PT
150
Bacterial and fungal DNA was extracted from 0.2 g of each ground Hong Qu
158
powder samples by utilizing a rapid DNA extraction kit (BioTeke Corporation,
159
Beijing, China), following the instruction of the manufacturer. The extracted bacterial
160
and fungal DNA were then checked by agarose gel electrophoresis.
M AN U
157
Bacterial primers 341-F (5’-CCT AYG GGR BGC ASC AG-3’) and 806-R (5’-
162
GGA CTA CNN GGG TAT CTA AT-3’) with specific barcode were used to amplify
163
the V3–V4 region of bacterial 16S rRNA genes. The fungal primers ITS5-1737-F
164
(5’-GGA AGT AAA AGT CGT AAC AAG G-3’) and ITS2-2043-R (5’-GCT GCG
165
TTC TTC ATC GAT GC-3’) with specific barcode were used to amplify the ITS1
166
regions of the fungal ITS rRNA genes. The sequencing libraries of bacterial 16S
167
rRNA genes and fungal ITS rRNA genes were generated for high throughput
168
sequencing utilizing the TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina,
169
San Diego,
170
platform (Illumina, San Diego, USA) by Novogene Bioinformatics Technology Co.,
171
Ltd. (Beijing, China).
AC C
EP
TE D
161
USA). Then both libraries were sequenced on an Illumina HiSeq2500
8
ACCEPTED MANUSCRIPT 172
2.6 Bioinformatic analysis Raw sequencing reads obtained from the Illumina platform were then merged
174
using FLASH software (Version 1.2.7)(Magoč and Salzberg, 2011), and filtered with
175
the QIIME software (Version 1.7)(Caporaso et al., 2010). All quality filtered
176
sequencing reads were then clustered into operational taxonomic units (OTUs) with a
177
threshold of 97% sequence similarity, by utilizing UPARSE software (Version
178
7.0)(Edgar, 2013). The representative sequence for each bacterial OTU was then
179
annotated by utilizing the GreenGene Database based on RDP classifier (Version 2.2)
180
algorithm (DeSantis et al., 2006). For each fungal representative sequence, the Unite
181
Database was used based on Blast algorithm which was calculated by QIIME
182
software, so as to annotate taxonomic information (Kõljalg et al., 2013). The OTUs
183
abundance information was normalized, by utilizing a standard sequence number
184
corresponding to the sample with the least sequences. Subsequent the differences of
185
samples in OTU-level were evaluated through the PCA, by using R software.
186
Moreover, the linear discriminant analysis (LDA) effect size (LEfSe) algorithm was
187
performed to identify the representative bacterial and fungal taxa of each type of
188
starters, by utilizing the Huttenhower Lab Galaxy Server (Goecks et al., 2010). The
189
relative abundances of the representative taxa were further visualized with heatmap
190
and hierarchical clustering (with complete linkage), by utilizing R software with a
191
pheatmap package. Subsequently, the possible relation between the major volatile
192
components with relative content over 1.0 % and the representative bacterial and
193
fungal taxa was explored, by utilizing a PLS-DA modelling, and plotted in a
AC C
EP
TE D
M AN U
SC
RI PT
173
9
ACCEPTED MANUSCRIPT 194
concatenation panel, by using R software with the mixOmics package. Other data are expressed as mean ± SD. Furthermore, the statistical significance
196
between WYQ and GTQ were analyzed with “Independent-Samples T-Test” using
197
SPSS software (Version 19.0.0), while the significance threshold was established at
198
0.05.
199
3. Results
200
3.1 Color value and fermentative power of WYQ and GTQ
SC
RI PT
195
As shown in Figure 1, the appearance of the ten starters were similar; however, in
202
general, GTQ appears slightly redder, while WYQ appears slightly darker in color.
203
The pigment extracted from these two types of Hong Qu exhibited remarkable
204
differences (Figure 2). For example, GTQ had a significantly higher content of water
205
soluble and ethanol soluble pigments (P = 0.013 and 0.037, respectively). However, it
206
was also noticed that WYQ5 had higher color values of water extraction and ethanol
207
extraction than those of other WYQ samples, and even higher than those of some
208
GTQ samples. The enzymatic properties of the ten collected starter samples were also
209
investigated in the present study. Significantly higher glucoamylase and α-amylase
210
activities were discovered in WYQ than those in GTQ, whereas, in terms of protease,
211
there were no significant differences found between them (Table 1). It is also worth
212
noting that, the color value and enzymatic activities were diverse among samples,
213
even within the same type.
214
3.2 Volatile components in WYQ and GTQ
215
AC C
EP
TE D
M AN U
201
The volatile components presented in WYQ and GTQ were analyzed with 10
ACCEPTED MANUSCRIPT SPME-GC/MS approach. As summarized in Table 2, a total of 66 volatile compounds
217
were identified from the ten starter samples, including 10 alcohols, 6 acids, 13 esters,
218
14 aldehydes and ketones, 7 aromatic compounds, 5 lactones, 4 alkanes and
219
cycloalkanes and 7 nitrogen-containing compounds. Based on the normalized peak
220
area percentages of individual components, PCA score plot was performed to
221
visualize the differences between these two types of starters (Figure 3A). It was
222
shown that the volatile profiles of WYQ and GTQ exhibited little similarity to each
223
other. In addition, the 5 samples of WYQ seemed more close to each other; while
224
samples belonging to GTQ distributed more widely, which indicated that GTQ present
225
higher variations in volatile profiles than WYQ. Of these volatile components, 13
226
components, including Ethyl butyrate, Ethyl isobutyrate, Butanoic acid, Ethyl Acetate,
227
Ethanol, Octanoic acid, Ethyl caproate, Hexanoic acid, Hexyl alcohol, Ethyl caprylate,
228
Hexanal, 2-Heptanone and 2-Nonanone, were defined as the major volatile
229
components with relative contents over 1.00 %. They jointly constituted from 86.40%
230
to 94.31% of the total peak area, and thus covered the major information contained in
231
the volatile profiles.
232
3.3 Overall bacterial and fungal structures comparison of WYQ and GTQ
SC
M AN U
TE D
EP
AC C
233
RI PT
216
The bacterial and fungal components in the traditional fermentation starters WYQ
234
and GTQ were investigated by Illumina HiSeq sequencing analyses. As for the
235
bacterial community, a total of 321,142 quality filtered sequencing reads
236
corresponding to the V3–V4 region of bacterial 16S rRNA genes, about 426 bp in
237
length, were obtained. All sequences were clustered into 6650 OTUs with a 97% 11
ACCEPTED MANUSCRIPT similarity level, ranging from 430 (GTQ4) to 868 (WYQ3). Following the high
239
throughput sequencing of the fungal ITS rRNA genes amplicons, 757,328 quality
240
filtered sequencing reads, with an average length of about 224 bp, were generated
241
from the ten starter samples and assigned to 895 OTUs with a 97% similarity level.
242
Good's coverage estimation values were within the range of 98.9-100%, indicating
243
adequate sequence coverage to reliably describe the full bacterial and fungal
244
communities presented in all the samples (Supplementary Table S1).
SC
RI PT
238
The bacterial and fungal communities within the two types of starters were
246
compared by using PCA analysis, based on the total OTUs’ relative abundance
247
information (Figure 3B and C). Both PCA score plots depicted 2 divided groups,
248
which revealed relatively low similarities in bacterial and fungal communities
249
between GTQ and WYQ. Moreover, it was also revealed that the bacterial
250
communities within WYQ were dramatically distinct from one another, while the
251
GTQ patterns were quite well clustered, which indicated more similarity of bacterial
252
profile within GTQ samples.
253
3.4 Bacterial and fungal profiles of Hong Qu
TE D
EP
AC C
254
M AN U
245
The bacterial OTU sequences were annotated into different taxa from the
255
GreenGene Database. However, due to the high sequence homology of the V3–V4
256
region of the bacterial 16S rRNA genes, the obtained bacterial reads cannot be
257
accurately annotated beyond the genus level. Thus, the taxonomic information of
258
bacteria at the genus level were recorded, and a total of 215 bacterial genera were
259
identified across the ten starter samples. The 10 most abundant genera were: 12
ACCEPTED MANUSCRIPT Burkholderia,
Bacillus,
Erwinia,
Klebsiella,
Ochrobactrum,
Staphylococcus,
261
Shewanella, Agrobacterium, Acinetobacter and Lactococcus, which, taken as a
262
composite, comprised from 69.69% (WYQ4) to 96.50% (GTQ4) of all sequences.
263
The relative abundance of the 10 most predominant bacterial genera are depicted in
264
Table 3. The phylotypes of Bacillus and Burkholderia represented the most
265
predominant bacteria presented in WYQ (Circa 45% and 8% of relative abundance,
266
respectively), while Burkholderia were the most abundant bacteria in GTQ,
267
accounting for approximately 72 % of the total sequences. When comparing the
268
bacterial genera taxonomic differences between WYQ and GTQ, GTQ exhibited a
269
significantly higher relative abundance of Burkholderia (P = 0.000), whereas WYQ
270
had a significantly higher Bacillus proportion (P = 0.000).
M AN U
SC
RI PT
260
The ITS1 regions of fungal ITS rRNA genes facilitated the species level
272
taxonomic annotation of fungi from the Unite Database. Following the sequencing of
273
these fungal genes, 151 different species were identified from the Unite Database, of
274
which the 10 most abundant fungal species were: Monascus purpureus,
275
Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Eurotiomycetes sp.,
276
Fusarium pseudensiforme, Rhizopus oryzae, Rhizopus microsporus, Aspergillus sp.
277
and Agaricomycetes sp., which, when combined together, constituted from 51.85%
278
(WYQ3) to 99.54% (GTQ2) of all sequences, as shown in Table 3. Of these,
279
Monascus purpureus represented the largest fraction in both WYQ and GTQ, which
280
accounted for 55.77 ± 16.29% and 98.21 ± 0.82% of the total fungal sequences,
281
respectively. Generally, the fungal profile in WYQ was different from that in GTQ.
AC C
EP
TE D
271
13
ACCEPTED MANUSCRIPT Specifically, GTQ had a significantly higher relative abundance of M. purpureus (P =
283
0.000), whereas WYQ had a significantly higher relative abundance of
284
Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Eurotiomycetes sp.,
285
Rhizopus oryzae and Aspergillus sp. (P < 0.05).
286
3.5 Representative bacteria and fungi in Hong Qu
RI PT
282
To identify the specific bacterial and fungal taxa within each type of sample, the
288
LEfSe analysis with 3.0 as the threshold on the LDA score for discriminative features
289
was applied (Figure 4). In the comparison of the bacterial communities in WYQ and
290
GTQ, 14 genus-level bacterial taxa were verified as being differentially abundant in
291
these two particular types of starters (Figure 4A). Of these, 13 taxa were enriched in
292
WYQ, namely, Bacillus, Klebsiella, Ochrobactrum, Staphylococcus, Acinetobacter,
293
Agrobacterium, Shewanella, Lactobacillus, Microbacterium, Brachybacterium,
294
Comamonas, Pseudomonas and Sphingomonas. Whereas, genus Bacillus had the
295
highest LDA score of 5.44, followed by Klebsiella with an LDA of 4.55,
296
Ochrobactrum with an LDA of 4.40 and Staphylococcus with an LDA of 4.10.
297
Whereas only genus Burkholderia was enriched in GTQ, with an LDA of 5.52.
M AN U
TE D
EP
AC C
298
SC
287
For fungal LDA analysis, 9 species-level fungal taxa were found to be enriched in
299
WYQ, including Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Rhizopus
300
oryzae, Fusarium pseudensiforme, Eurotiomycetes sp., Rhizopus microsporus,
301
Aspergillus sp. and Meyerozyma guilliermondii, as depicted in Figure 4C. Of these,
302
species Saccharomyces sp. and Aspergillus flavus had an LDA score higher than 4.0
303
(4.63 and 4.46, respectively). The fungal group enriched in GTQ was only Monascus 14
ACCEPTED MANUSCRIPT 304
purpureus, with an LDA score of 5.02. Based on the relative abundance of these 14 bacterial taxa and 10 fungal taxa,
306
heatmap plots and hierarchical clustering were then employed, so as to provide a
307
visual and overall comparison for differentiating the two types of starters, as reflected
308
in Figures 4B and D. Unsurprisingly, the clustering depicted two clear separated
309
groups in both bacterial and fungal communities, which indicated that these taxa were
310
the most representative microbial groups in WYQ or GTQ.
311
3.6 Correlation analysis between major volatile components and representative
312
microbiota in Hong Qu
M AN U
SC
RI PT
305
The relationship of the 13 major volatile components, 14 representative bacterial
314
taxa and 10 representative fungal taxa of the both types of starters were explored by
315
using PLS-DA algorithm and visualized by using R software. Using a correlation
316
cut-off of 0.9, the association among the three data sets were depicted in Figure 5. It
317
was shown that 2-Nonanone was positively correlated with Saccharomyces sp.;
318
Ethanol was positively correlated with Saccharomyces sp. and Rhizopus microsporus;
319
2-Heptanone was negatively correlated with Agrobacterium, Brachybacterium,
320
Meyerozyma guilliermondii and Fusarium pseudensiforme; Hexyl alcohol was
321
positively correlated with Aspergillus flavus; Ethyl caproate was positively correlated
322
with Agrobacterium; Octanoic acid was negatively correlated with Bacillus and
323
Rhizopus oryzae.
AC C
EP
TE D
313
324
Based on the concatenation panel, correlation between representative bacteria and
325
fungi was complicated. A total of 22 pair-wise associations between the two data sets 15
ACCEPTED MANUSCRIPT were identified, of which Monascus purpureus was negatively correlated with
327
Bacillus, Agrobacterium and Sphingomonas; Burkholderia was negatively correlated
328
with Aspergillus niger and Rhizopus oryzae; as for Bacillus, in addition to the
329
negative correlation with Monascus purpureus, it was also positively correlated with
330
Rhizopus oryzae.
331
4. Discussion
RI PT
326
The major manufacturing process of Hong Qu includes the preparation of
333
polished round grain rice; inoculation of previously prepared ferments; cultivation of
334
microorganisms; and finally dehydration for preservation. Different ferments are used
335
for WYQ and GTQ, and passed down batch to batch. Hence, the unique cultures
336
likely established in these two types of starters, potentially lead to the distinctive
337
flavor of the starters and final traditional brewed wines. However, at this time, the
338
properties of these two types of starters have not yet been fully characterized and
339
compared. Thus, the color value, fermentative power, volatile profiles and, more
340
importantly, the bacterial and fungal community structures of WYQ and GTQ were
341
investigated and compared in the present study.
M AN U
TE D
EP
AC C
342
SC
332
In general, WYQ has darker appearance (Figure 1), but lower content of
343
Monascus pigments than GTQ (Figure 2). The color presented in GTQ may mainly
344
derived from Monascus sp., while, in terms of WYQ, beyond Monascus sp., other
345
microbial pigments may also contribute to the color of WYQ. In addition, higher
346
glucoamylase and α-amylase activities were discovered in WYQ than those in GTQ.
347
And remarkable different volatile profiles were also observed in these two types of 16
ACCEPTED MANUSCRIPT 348
starters. The varied fermentation powers and volatile components presented in the
349
starters were mainly derived from intricate microorganisms harbored in the starters. Following the high throughput sequencing analysis, the bacterial and fungal
351
profiles of these two types of starters were further investigated. It was established that
352
the major bacterial genus presented in WYQ and GTQ was Bacillus and Burkholderia,
353
respectively, each of which were also identified as the most representative bacteria of
354
these two types of starters by using LEfSe algorithm. Bacillus, a gram-positive genus,
355
can survive under low-moisture and high-temperature conditions (Ma et al., 2014);
356
thus, it is also frequently detected in various traditional fermented foods, such as
357
Daqu (Wang et al., 2008), nuruk (Song et al., 2013), fermented soya bean (Kiers et al.,
358
2000) etc. Several members within the Bacillus genus may play an important role in
359
the brewing process of some fermented foods, as they can secrete various hydrolytic
360
enzymes, including amylases and proteases (Simonen and Palva, 1993); moreover,
361
they can produce a broad range of volatile compounds, which include pyrazines,
362
aldehydes, ketones and alcohols (Azokpota et al., 2010). Bacterial genus Burkholderia,
363
an environmental ubiquitous gram-negative bacteria is frequently found in open
364
fermentation starters, such as Daqu (Li et al., 2015). Some species within the
365
Burkholderia genus can synthesize short chain alkyl esters (Dutta and Dasu, 2011),
366
which may contribute to the flavor of the final products.
AC C
EP
TE D
M AN U
SC
RI PT
350
367
Previously, our group investigated the bacterial communities of WYQ, based on
368
the PCR-DGGE approach (Lv, Weng, et al., 2012). Several species within the genus
369
of Bacillus, including B. ginsengihumi, B. megaterium or B. aryabhattai, B. subtilis, B. 17
ACCEPTED MANUSCRIPT methylotrophicus and B. amyloliquefaciens were identified in most WYQ samples.
371
DGGE is considered to be a low-resolution and low-sensitivity technique, and in this
372
study Illumina HiSeq sequencing analysis was performed, targeting the V3–V4 region
373
of the bacterial 16S rRNA genes, so as to further explore the bacterial communities in
374
WYQ and GTQ, where numerous other specific details were revealed. Apart from
375
Bacillus and Burkholderia, the genera Erwinia, Klebsiella, Ochrobactrum and some
376
other low-abundant taxa also presented in Hong Qu, although some of which were
377
related to spoilage or pathogenic bacteria, such as Klebsiella, Staphylococcus,
378
Shewanella, etc. The existence of these bacteria may raise concerns about the safety
379
of the final products, and they may also be responsible for the acidification and
380
spoilage of wine. However, interestingly, in our another study, during the traditional
381
brewing process of Hong Qu glutinous rice wine with WYQ as the starter, some
382
bacterial species (such as Bacillus sp., P. acidilactici, L. brevis and P. pentosaceus)
383
decreased as the fermentation progressed (Lv, Huang, Chen, et al., 2013). Thus, some
384
factors may influence the growth of these spoilage or pathogenic bacteria. Such as, it
385
was revealed in this study that M. purpureus has a strong negative correlation with
386
Bacillus, Agrobacterium and Sphingomonas. In addition, A. niger and R. oryzae were
387
also found to be negatively correlated to Burkholderia.
SC
M AN U
TE D
EP
AC C
388
RI PT
370
Another noteworthy feature of the bacterial profile in Hong Qu is that several
389
genera within the order Lactobacillales, which primarily include Lactococcus,
390
Lactobacillus, Pediococcus, Streptococcus, Weissella, Enterococcus and Leuconostoc
391
were detected across all samples. The bacteria order Lactobacillales, produces lactic 18
ACCEPTED MANUSCRIPT acid as the major metabolic end product of carbohydrate fermentation, and is usually
393
referred to as lactic acid bacteria (LAB). LAB plays a critical role in various
394
fermented foods such as cheese, yoghurt, pickles, wine and rice wine, by enhancing
395
the microbial safety and, thus, offering pleasant sensory and nutritional benefits
396
(Leroy and De Vuyst, 2004). Although bacteria within this division consist of a small
397
portion of the total microorganisms, and together account for a mere 0.7% of all
398
bacterial sequence, they experience rapid growth during the traditional brewing
399
process of Hong Qu glutinous rice wine (Lv, Huang, Chen, et al., 2013). After
400
inoculation, LAB can rapidly acidify the starchy substrates through the production of
401
lactic acid and, thus, suppress the growth of various acid-intolerant pathogens. From
402
this perspective, the presence of LAB may be one of the inhibiting factors for
403
pathogens. Moreover, the production of acetic acid, ethanol, aroma compounds,
404
esterases, lipases and alcohol acetyl transferases by LAB is highly significant for
405
enhancing the flavor forming attributes of rice wine.
TE D
M AN U
SC
RI PT
392
However, the most distinguishing feature of the Hong Qu from other starters is
407
that it contains large populations of Monascus sp. strains, which are a unique
408
red-pigment producing filamentous fungi. The red appearance and relative high color
409
values of all Hong Qu implied the presence of Monascus. In GTQ, the proportion of
410
M. purpureus greatly exceeded those of WYQ, which was reflected in the higher color
411
values (Figure 2). Similarly, the higher color values of WYQ5 than other WYQ
412
samples was attributed to its relatively higher M. purpureus content. Moreover, as
413
confirmed by the LEfSe analysis, M. purpureus was distinctly and dramatically
AC C
EP
406
19
ACCEPTED MANUSCRIPT enriched in GTQ. M. purpureus has a long history of being used for making red rice
415
wine, red soybean cheese and Anka (red rice) in Southern China, Japan and Southeast
416
Asia (Lin et al., 2008). As the secondary metabolites, Monascus pigments have been
417
well studied and reviewed in relation to their structures, biosynthetic pathway,
418
fermentation processes, physicochemical properties detection methods, functions, and
419
molecular biological activity (Feng et al., 2012; Lin et al., 2008). Due to the wide
420
range of biological activities of Monascus pigments, Hong Qu glutinous rice wine is
421
believed to possess a variety of nutritional benefits, such as anti-inflammatory activity,
422
cancer cell cytotoxic activity, anti-hypertensive activity, cholesterol-lowering activity
423
and various other beneficial attributes (Park et al., 2016). Moreover, Monascus
424
pigments possess antimicrobial activities (Feng et al., 2012), establishing it as being
425
an active pathogen inhibitor in the Hong Qu, as reflected in this study by negatively
426
correlating with Bacillus, Agrobacterium and Sphingomonas. It is also worth to notice
427
that, in addition to pigments, Monascus can also produce mycotoxin citrinin, which
428
may raise concern of the safety of Hong Qu and Hong Qu glutinous rice wine.
EP
TE D
M AN U
SC
RI PT
414
In addition to M. purpureus, some other fungal taxa, which include
430
Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Eurotiomycetes sp.,
431
Rhizopus oryzae, Aspergillus sp., etc., also accounted for a considerable proportion of
432
fungi in Hong Qu; particularly prevalent in WYQ. Many members of the fungal genus
433
Saccharomyces are considered very important in the fermentation of foods, especially
434
S. cerevisiae, which is widely used in making wine, bread, and beer, and is an
435
effective ethanol producer; often referred to as “yeast” (Kurtzman, 1994). During the
AC C
429
20
ACCEPTED MANUSCRIPT brewing process of rice wine, the population of Saccharomyces increase dramatically,
437
along with the increase of ethanol, which will result in the decrease of alcohol
438
intolerance among microorganisms. Therefore, this may be another reason for the
439
reduction of some pathogens during wine brewing. As the predominant filamentous
440
fungi in Daqu, the genus Aspergillus has the ability to secrete large quantities of a
441
broad range of different enzymes into its environment, which may contribute to the
442
saccharification of the starch in wine mash (Oda et al., 2006; Zheng et al., 2011). R.
443
oryzae, with a strong amylase production ability, is commonly found in various types
444
of amylolytic fermentation starters (Choi et al., 2012; Dung et al., 2007). Through
445
synthesis of volatile compounds, R. oryzae also makes a great contribution to the
446
flavor formation of wine. Taken as a whole, WYQ had significantly higher
447
proportions of filamentous fungi and yeast that have strong enzyme synthesis abilities,
448
such as Saccharomyces and Aspergillus, making WYQ a more powerful fermentation
449
starter. This is also consistent with the results of the comparison of enzymatic
450
properties of the ten collected starter samples. As a matter of fact, WYQ has long
451
been regarded as a “stronger” starter by local brewers. In addition, Aspergillus may
452
also contribute to the darker color of WYQ.
SC
M AN U
TE D
EP
AC C
453
RI PT
436
The contributions of all these bacteria and fungi were also preliminarily explored
454
by digging the correlation between volatile components and microbial communities. A
455
concatenation panel was established to visualize their intricate correlation. However,
456
most volatile components are secondary metabolites, it may be not sufficient by only
457
using PLS-DA modelling to interpret their relationships. More importantly, the 21
ACCEPTED MANUSCRIPT correlations present here were obtained by mathematical model, more well designed
459
studies are required to validate these correlations. Nevertheless, this concatenation
460
panel indicated a potential roles of microorganism in aromatic forming, as well as a
461
complicated symbiotic relationship between various bacteria and fungi, which
462
provided clues for the further studies.
463
5. Conclusions
RI PT
458
In the current study, the fermentative properties, volatile profiles and bacterial
465
and fungal communities of the two major types of traditional fermentation starters of
466
Hong Qu glutinous rice wine were compared. Significantly higher color value,
467
glucoamylase and α-amylase activities were discovered in WYQ. And substantial
468
variation in volatile components and microbial communities were also observed in
469
these two types of starters. In addition, 14 genus-level bacterial taxa and 10
470
species-level fungal taxa were identified as the representative microorganism in WYQ
471
and GTQ by LEfSe analysis; and their interrelationship and potential roles in aroma
472
forming were also preliminarily explored by PLS-DA modeling. This study provides a
473
more comprehensive and thorough insight into the properties of Hong Qu, and may be
474
helpful in developing next generation starters for Hong Qu glutinous rice wine, which
475
with limited number of pure strains.
476
Acknowledgment
AC C
EP
TE D
M AN U
SC
464
477
This study was supported by the National Natural Science Foundation of China
478
(No. 31371820 and 31501494) and the Natural Science Foundation of Fujian Province
479
(No. 2016J01707). 22
ACCEPTED MANUSCRIPT 480
Reference
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
Azokpota, P., Hounhouigan, J.D., Annan, N.T., Odjo, T., Nago, M.C., Jakobsen, M., 2010. Volatile compounds profile and sensory evaluation of Beninese condiments produced by inocula of Bacillus subtilis. Journal of the Science of Food and Agriculture 90, 438-444. Bernfeld, P., 1955. Amylases, alpha and beta. Methods in enzymology I, 149-158. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pena,
RI PT
A.G., Goodrich, J.K., Gordon, J.I., 2010. QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335-336.
Choi, J.-S., Jung, S.-T., Choi, J.-H., Choi, H.-S., Baek, S.-Y., Yeo, S.-H., 2012. Quality characteristics of wheat nuruks by storage conditions of liquid starters using Rhizopus oryzae N174. Microbiology and Biotechnology Letters 40, 319-324.
De Moraes, L., Astolfi-Filho, S., Oliver, S., 1995. Development of yeast strains for the efficient
SC
utilisation of starch: evaluation of constructs that express α-amylase and glucoamylase separately or as bifunctional fusion proteins. Applied microbiology and biotechnology 43, 1067-1076. DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Huber, T., Dalevi, D., Hu, P.,
M AN U
Andersen, G.L., 2006. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and environmental microbiology 72, 5069-5072. Dung, N., Rombouts, F., Nout, M., 2007. Characteristics of some traditional Vietnamese starch-based rice wine fermentation starters (men). LWT-Food Science and Technology 40, 130-135. Dutta, K., Dasu, V.V., 2011. Synthesis of short chain alkyl esters using cutinase from Burkholderia cepacia NRRL B2320. Journal of Molecular Catalysis B: Enzymatic 72, 150-156. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature
TE D
methods 10, 996-998.
Farley, P.C., Ikasari, L., 1992. Regulation of the secretion of Rhizopus oligosporus extracellular carboxyl proteinase. Microbiology 138, 2539-2544.
Feng, Y., Shao, Y., Chen, F., 2012. Monascus pigments. Applied microbiology and biotechnology 96, 1421-1440.
EP
Goecks, J., Nekrutenko, A., Taylor, J., 2010. Galaxy: a comprehensive approach for supporting accessible, reproducible, and transparent computational research in the life sciences. Genome biology 11, 1.
AC C
Guan, Z.B., Zhang, Z.H., Cao, Y., Chen, L.L., Xie, G.F., Lu, J., 2012. Analysis and comparison of bacterial communities in two types of ‘wheat Qu’, the starter culture of Shaoxing rice wine, using nested PCR‐DGGE. Journal of the Institute of Brewing 118, 127-132. Kiers, J.L., Rombouts, F., Nout, M., 2000. In vitro digestibility of Bacillus fermented soya bean. International journal of food microbiology 60, 163-169. Kõljalg, U., Nilsson, R.H., Abarenkov, K., Tedersoo, L., Taylor, A.F., Bahram, M., Bates, S.T., Bruns, T.D., Bengtsson‐Palme, J., Callaghan, T.M., 2013. Towards a unified paradigm for sequence‐based identification of fungi. Molecular ecology 22, 5271-5277. Kurtzman, C.P., 1994. Molecular taxonomy of the yeasts. Yeast 10, 1727-1740. Leroy, F., De Vuyst, L., 2004. Lactic acid bacteria as functional starter cultures for the food fermentation industry. Trends in Food Science & Technology 15, 67-78. Li, P., Liang, H., Lin, W.-T., Feng, F., Luo, L., 2015. Microbiota dynamics associated with environmental conditions and potential roles of cellulolytic communities in traditional Chinese cereal starter 23
ACCEPTED MANUSCRIPT solid-state fermentation. Applied and environmental microbiology 81, 5144-5156. Lin, Y.-L., Wang, T.-H., Lee, M.-H., Su, N.-W., 2008. Biologically active components and nutraceuticals in the Monascus-fermented rice: a review. Applied microbiology and biotechnology 77, 965-973. Lu, J., Cao, Y., Fang, H., Li, W.-j., Xie, G.-f., Zou, H.-j., HU, Z.-m., 2008. Fungal community of wheat Qu of Shaoxing rice wine. Journal of Food Science and Biotechnology 2, e23. Luo, T., Fan, W., Xu, Y., 2008. Characterization of Volatile and Semi‐Volatile Compounds in Chinese Spectrometry. Journal of the Institute of Brewing 114, 172–179.
RI PT
Rice Wines by Headspace Solid Phase Microextraction Followed by Gas Chromatography‐Mass Lv, X.-C., Huang, R.-L., Chen, F., Zhang, W., Rao, P.-F., Ni, L., 2013. Bacterial community dynamics during the traditional brewing of Wuyi Hong Qu glutinous rice wine as determined by culture-independent methods. Food Control 34, 300-306.
Lv, X.-C., Huang, X.-L., Zhang, W., Rao, P.-F., Ni, L., 2013. Yeast diversity of traditional alcohol culture-independent methods. Food Control 34, 183-190.
SC
fermentation starters for Hong Qu glutinous rice wine brewing, revealed by culture-dependent and Lv, X.-C., Huang, Z.-Q., Zhang, W., Rao, P.-F., Ni, L., 2012. Identification and characterization of filamentous fungi isolated from fermentation starters for Hong Qu glutinous rice wine brewing. The
M AN U
Journal of general and applied microbiology 58, 33-42.
Lv, X.-C., Jia, R.-B., Chen, J.-H., Zhou, W.-B., Li, Y., Xu, B.-X., Liang, Y.-T., Liu, B., Chen, S.-J., Tian, Y.-T., 2016. Development of Reverse Transcription Quantitative Real-Time PCR (RT-qPCR) Assays for Monitoring Saccharomycopsis fibuligera. Food Analytical Methods, 1-11.
Lv, X.-C., Weng, X., Zhang, W., Rao, P.-F., Ni, L., 2012. Microbial diversity of traditional fermentation starters for Hong Qu glutinous rice wine as determined by PCR-mediated DGGE. Food Control 28, 426-434.
TE D
Ma, K., Maeda, T., You, H., Shirai, Y., 2014. Open fermentative production of l-lactic acid with high optical purity by thermophilic Bacillus coagulans using excess sludge as nutrient. Bioresource technology 151, 28-35.
Magoč, T., Salzberg, S.L., 2011. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27, 2957-2963.
EP
McGovern, P.E., Zhang, J., Tang, J., Zhang, Z., Hall, G.R., Moreau, R.A., Nuñez, A., Butrym, E.D., Richards, M.P., Wang, C.-s., 2004. Fermented beverages of pre-and proto-historic China. Proceedings of the National Academy of Sciences of the United States of America 101, 17593-17598. Oda, K., Kakizono, D., Yamada, O., Iefuji, H., Akita, O., Iwashita, K., 2006. Proteomic analysis of
AC C
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
extracellular proteins from Aspergillus oryzae grown under submerged and solid-state culture conditions. Applied and environmental microbiology 72, 3448-3457. Park, K.H., Liu, Z., Park, C.-S., Ni, L., 2016. Microbiota associated with the starter cultures and brewing process of traditional Hong Qu glutinous rice wine. Food Science and Biotechnology 25, 649-658. Simonen, M., Palva, I., 1993. Protein secretion in Bacillus species. Microbiological Reviews 57, 109-137. Song, S.H., Lee, C., Lee, S., Park, J.M., Lee, H.-J., Bai, D.-H., Yoon, S.-S., Choi, J.B., Park, Y.-S., 2013. Analysis of microflora profile in Korean traditional nuruk. J. Microbiol. Biotechnol 23, 40-46. Wang, C.-l., Shi, D.-j., Gong, G.-l., 2008. Microorganisms in Daqu: a starter culture of Chinese Maotai-flavor liquor. World Journal of Microbiology and Biotechnology 24, 2183-2190. Yang, S., Zhang, H., Li, Y., Qian, J., Wang, W., 2005. The ultrasonic effect on biological characteristics of Monascus sp. Enzyme and microbial technology 37, 139-144. 24
ACCEPTED MANUSCRIPT 567 568 569 570
Zhang, X., Liu, W., Chen, X., Cai, J., Wang, C., He, W., 2017. Effects and Mechanism of Blue Light on Monascus in Liquid Fermentation. Molecules 22, 385. Zheng, X.W., Tabrizi, M.R., Nout, M., Han, B.Z., 2011. Daqu—a traditional Chinese liquor fermentation starter. Journal of the Institute of Brewing 117, 82-90.
AC C
EP
TE D
M AN U
SC
RI PT
571
25
ACCEPTED MANUSCRIPT Figure captions Figure 1 Ten Hong Qu samples (5 WYQ and 5 GTQ) obtained from different geographical
RI PT
areas of Fujian province, China.
Figure 2 Color values of the water and ethanol extractions of the ten starter samples. The color value was recorded as the summation of absorbance units at 410 nm, 465nm and 510
SC
nm× dilution factor per gram dry samples (U/g). Each value represented the mean ± SD of
M AN U
three determinations. P values indicated the statistical difference between data sets.
Figure 3 PCA plots of volatile components and microbial communities of the ten Hong Qu samples. (A) PCA plots based on the normalized peak area percentages of volatile
TE D
components; (B) PCA plots based on the relative abundance of bacterial 16S gene OTUs; (C) PCA plots based on the relative abundance of fungal ITS gene OTUs.
EP
Figure 4 Representative bacteria and fungi of WYQ and GTQ. (A) LEfSe comparison of
AC C
bacterial communities at genus-level between WYQ and GTQ; (B) Heatmap comparison and hierarchical clustering dendrogram based on the relative abundance of 14 representative bacteria genera across all samples; (C) LEfSe comparison of fungal communities at species-level between WYQ and GTQ; (D) Heatmap comparison and hierarchical clustering dendrogram based on the relative abundance of 10 representative fungal species across all samples.
26
ACCEPTED MANUSCRIPT Figure 5 Correlation analysis between major volatile components and representative microbiota of WYQ and GTQ by PLS-DA modeling. Red lines in the circle represent positive correlation, while blue lines represent negative correlation between volatile
RI PT
components and microbiota. Red blocks on the circle represent the 13 major volatile components, blue blocks represent the 14 genus-level representative bacterial taxa and green blocks represent the 10 species-level representative fungal taxa. Blue lines and orange lines
SC
outside the circle represent the relative content of volatile components or microbiota
AC C
EP
TE D
M AN U
presented in GTQ and WYQ, respectively.
27
ACCEPTED MANUSCRIPT Table 1 Glucoamylase, α-amylase and protease activities (U/g) of the ten Hong Qu samples * α-Amylase activity
Protease activity
WYQ1 WYQ2 WYQ3 WYQ4 WYQ5
103.35±4.89 141.13±1.15 158.94±1.57 121.59±4.84 96.62±1.70
862.05±20.36 1043.42±22.48 1269.60±26.65 904.73±29.87 603.33±9.09
1.75±0.22 3.35±1.24 4.32±0.17 3.98±0.93 1.41±0.54
GTQ1 GTQ2 GTQ3 GTQ4 GTQ5
66,01±3.20 62.10±1.21 58.41±4.64 62.53±3.21 81.20±1.15
471.57±13.23 486.82±22.92 271.84±18.03 188.63±10.83 142.86±19.27
1.85±0.22 2.14±0.22 1.02±0.15 0.92±0.30 1.80±0.30
0.001
0.057
0.001 §
EP
TE D
M AN U
Values are presented as means ±SD (n= 3); P value of WYQ vs GTQ.
AC C
*
SC
P value§
RI PT
Glucoamylase activity
28
ACCEPTED MANUSCRIPT Table 2. Identification and relative contents of volatile compounds in the ten Hong Qu samples. No.
WYQ, relative content (%) b
Compounds a
GTQ, relative content (%)b
WYQ1
WYQ2
WYQ3
WYQ4
WYQ5
GTQ1
GTQ2
GTQ3
GTQ4
GTQ5
Alcohols Ethanol
3.10
4.61
3.94
2.56
3.89
3.70
3.21
2.72
0.75
4.01
2
Isopentyl alcohol
0.16
0.56
0.15
0.43
0.13
0.13
0.10
nd
nd
0.18
3
Hexyl alcohol
1.10
2.21
3.51
0.88
0.89
0.43
0.38
0.52
0.39
0.66
4
Gentanol
0.21
0.39
0.52
0.13
0.35
0.13
nd
0.17
0.29
0.12
5
3-Octenol
nd
0.29
0.48
0.12
nd
nd
nd
nd
nd
nd
6
Benzyl alcohol
0.13
0.20
0.30
0.09
nd
nd
nd
nd
nd
nd
7
3,5-Octadien-2-ol
nd
nd
nd
nd
nd
nd
nd
nd
0.10
nd
8
Linalool
0.39
0.44
0.42
0.19
0.39
0.33
0.31
0.26
0.35
0.25
9
Phenylethyl Alcohol
0.49
0.75
0.60
0.46
0.46
0.52
0.32
0.99
2.06
0.76
Cyclohexanol, 1-methyl-
nd
nd
nd
nd
nd
11
Acetic acid
nd
nd
0.60
0.43
0.65
12
Propanoic acid, 2-methyl-
1.77
nd
13
Butanoic acid
18.21
9.50
14
Hexanoic acid
2.27
1.68
15
Octanoic acid
1.12
2.82
0.23
nd
1.00
0.06
nd
nd
nd
nd
0.54
0.14
0.43
0.18
nd
0.12
nd
0.04
1.85
nd
0.96
0.13
8.46
10.57
8.18
9.38
20.69
4.66
10.06
18.79
0.93
1.46
1.29
0.80
2.65
1.16
nd
2.89
0.95
1.38
1.78
1.46
1.76
2.08
2.00
1.35
nd
26.26
10.58
nd
0.06
nd
nd
nd
1.00
1.02
0.86
0.95
1.57
0.93
0.99
0.67
0.93
17.67
17.75
18.80
14.19
16.54
18.97
12.74
20.68
16.22
17.89
43.83
46.34
49.87
35.85
42.24
47.25
38.39
51.77
46.35
43.69
0.09
nd
nd
nd
0.06
0.06
0.05
nd
nd
nd
0.92
0.87
1.02
0.50
1.19
2.00
1.15
1.36
nd
0.92
nd
nd
nd
nd
0.08
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
0.05
nd
nd
nd
0.75
1.09
0.80
0.44
0.67
0.55
0.46
0.50
0.78
0.80
0.58
0.95
0.91
0.47
0.69
0.61
0.57
0.70
0.66
0.50
4-Hexyl-2,5-dioxofuran-3-ac 16
0.15
M AN U
Acids
SC
10
RI PT
1
etic acid Esters Ethyl Acetate
18
Ethyl isobutyrate
19
Ethyl butyrate
20
Propyl butyrate
21
Ethyl caproate
22
Isopentyl isobutyrate
23
Isoamyl butyrate
24
Nonanal
25
Ethyl benzoate
26
Ethyl caprylate
0.81
0.85
0.90
0.41
1.72
nd
2.25
0.68
0.77
nd
27
Ethyl nonanoate
nd
nd
nd
nd
nd
0.06
0.06
nd
nd
nd
28
Ethyl caprate
0.09
0.15
0.12
0.06
0.18
0.19
0.30
0.13
0.17
0.23
29
Ethyl Oleate
nd
0.10
nd
nd
nd
0.07
nd
0.12
0.15
nd
AC C
EP
TE D
17
Aldehydes and Ketones 30
Pentanal
nd
nd
nd
nd
nd
0.42
nd
nd
0.47
nd
31
3-Penten-2-one, (E)-
nd
nd
nd
nd
nd
1.28
1.43
1.33
1.45
nd
32
Hexanal
nd
nd
nd
nd
nd
2.62
nd
1.66
4.74
nd
33
Furfural
nd
nd
nd
nd
nd
nd
nd
nd
0.96
nd
34
2-Heptanone
0.73
0.89
0.75
nd
1.11
1.11
1.38
1.60
1.30
0.69
35
Heptanal
0.53
0.66
0.37
0.23
0.56
0.81
0.64
0.79
1.13
0.54
36
Benzaldehyde
0.32
0.41
0.23
0.22
0.35
0.56
0.34
1.08
0.84
0.35
29
ACCEPTED MANUSCRIPT Octanal
nd
0.49
0.38
nd
0.62
0.54
nd
nd
nd
nd
38
3-Octen-2-one
nd
nd
nd
nd
nd
0.10
nd
nd
nd
nd
39
Benzeneacetaldehyde
0.12
nd
nd
0.20
nd
nd
nd
nd
0.16
nd
40
2-Tridecenal, (E)-
nd
nd
nd
nd
nd
nd
nd
nd
0.11
nd
41
2-Nonanone
1.02
1.12
0.99
0.35
2.55
0.95
2.81
0.98
1.75
0.91
42
Decanal
0.13
0.31
0.20
0.10
0.12
0.16
0.08
0.16
0.16
0.32
43
2-Undecanone
nd
0.09
0.10
nd
0.10
nd
0.10
nd
nd
nd
44
Ethylbenzene
0.45
0.42
0.39
0.13
0.45
0.48
0.44
nd
0.15
0.43
45
o-Xylene
0.23
nd
nd
nd
0.23
nd
nd
nd
nd
nd
46
1,3-Xylene
nd
0.18
0.20
0.10
nd
nd
0.21
0.17
nd
nd
47
p-Xylene
nd
nd
nd
nd
nd
0.24
nd
nd
nd
nd
48
Phenol
nd
nd
nd
nd
nd
0.11
nd
0.14
0.15
nd
49
2,4-Di-tert-butylphenol
0.07
0.14
nd
0.09
nd
50
Butylated Hydroxytoluene
nd
0.07
0.06
nd
SC
Aromatic compounds
RI PT
37
0.16
0.38
0.63
Ethyl 2-(5-methyl-5-vinyltetrahydr 51 ofuran-2-yl)propan-2-yl carbonate Dehydromevalonic lactone
nd
0.95
53
trans-3-Methyl-4-octanolide
nd
nd
54
gamma-Octanoic lactone
nd
0.11
55
gamma-Nonanolactone
nd
0.18
Alkanes and cycloalkanes 56
cis-2-Methyl-3-pentene
57
Styrene 3-Ethyl-2-methyl-1,3-hexadi
58 59
Cyclooctane
0.28
0.06
0.08
0.12
nd
0.12
nd
0.27
nd
0.23
0.06
0.30
0.17
0.16
0.91
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
0.18
nd
nd
0.06
nd
0.07
0.05
nd
0.10
nd
0.15
nd
nd
nd
nd
nd
nd
nd
0.57
nd
0.38
nd
0.78
nd
nd
nd
nd
nd
0.33
0.36
0.37
0.16
nd
0.36
0.30
0.32
0.20
0.25
nd
nd
nd
nd
nd
nd
nd
nd
0.10
nd
nd
nd
nd
nd
nd
nd
0.14
0.58
EP
en
0.13
0.10
TE D
52
nd
M AN U
Lactones
nd
nd
nd
Nitrogen-containing compounds 60
Oxime-, methoxy-phenyl-_
61
2,6-Dimethyl pyrazine
62
nd
nd
0.05
0.08
nd
nd
0.08
0.82
0.09
0.31
nd
0.19
nd
0.40
0.20
0.51
0.09
0.26
Ethylpyrazine
nd
nd
nd
nd
nd
0.18
nd
nd
nd
nd
63
Pyrrole-2-carboxaldehyde
nd
0.11
0.12
nd
nd
nd
nd
nd
0.26
nd
64
2-ethenyl-6-methylpyrazine
nd
nd
nd
nd
nd
nd
nd
nd
0.27
nd
65
1,3,5-Triazine-2,4,6-triamine
nd
nd
nd
nd
nd
0.71
2.83
0.68
0.61
0.72
nd
nd
nd
nd
nd
0.11
nd
0.15
nd
nd
AC C
nd
0.20
Pyrazine, 66 3-ethyl-2,5-dimethyla b
Identified by comparison with mass spectra; Relative content, percent normalized peak areas.
30
ACCEPTED MANUSCRIPT Table 3 Relative abundance of the ten most abundant bacteria and fungi in the ten Hong Qu samples*.
WYQ, relative abundance (%)
P§
2
3
4
5
1
2
3
4
5
10.18
3.03
5.08
11.33
10.82
34.80
85.21
69.91
95.01
73.16
0.000
Bacillus
40.39
57.62
51.47
32.49
41.66
0.33
0.65
0.41
0.16
0.54
0.000
Erwinia
14.04
1.28
1.93
4.42
5.81
50.35
1.90
21.20
0.61
14.77
0.224
Klebsiella
8.47
1.79
1.39
14.52
9.52
8.43
1.06
0.79
0.29
1.12
0.139
Ochrobactrum
0.50
2.72
6.27
0.40
0.28
0.05
0.13
0.05
0.01
0.05
0.125
Staphylococcus
0.92
4.37
1.77
2.37
0.10
0.05
0.15
0.12
0.03
0.07
0.036
Shewanella
0.29
2.38
1.58
0.31
0.46
0.08
0.63
0.15
0.05
0.14
0.103
Agrobacterium
0.71
1.45
1.44
0.33
0.74
0.06
0.09
0.10
0.03
0.09
0.005
Acinetobacter
0.09
0.37
0.16
3.11
0.41
0.07
0.09
0.09
0.02
0.07
0.221
Lactococcus
0.28
0.28
0.45
0.41
0.28
Others
24.12
24.70
28.47
30.31
29.92
1
2
3
4
5
Monascus purpureus
68.48
51.49
30.80
Saccharomyces sp
5.17
5.67
8.28
Aspergillus flavus
1.00
6.17
7.45
Aspergillus niger
1.25
1.35
1.11
Eurotiomycetes sp
0.41
0.92
1.07
Fusarium pseudensiforme
0.25
1.47
1.45
Rhizopus oryzae
0.13
0.73
1.02
Rhizopus microsporus
0.07
Aspergillus sp
0.13
Agaricomycetes sp
23.12
SC 0.49
0.37
0.30
0.29
0.988
5.52
9.61
6.81
3.50
9.69
―
1
2
3
4
5
56.26
71.83
99.09
97.97
98.86
97.03
98.12
0.000
5.52
3.18
0.07
0.11
0.12
0.13
0.20
0.000
1.73
2.30
0.03
0.03
0.04
0.13
0.09
0.022
2.17
0.01
0
0
0
0
0
0.009
0.38
0.44
0.17
0.19
0.13
0.20
0.26
0.015
0
0.05
0
0
0
0.02
0
0.094
0.97
0.31
0
0
0
0
0
0.007
0.12
0.40
0.80
0.12
0.03
0.10
0.06
0.07
0.10
0.130
0.28
0.27
0.18
0.11
0.02
0.02
0.03
0.04
0.22
0.040
0
0
0
0
0
1.12
0
0
0
0.347
31.80
48.15
31.99
21.63
0.59
0.46
0.76
2.38
1.01
―
EP
Others
0
0.26
M AN U
Fungi (species level)
RI PT
1
Burkholderia
TE D
Bacteria ( genus level)
GTQ, relative abundance (%)
AC C
* The colors range from blue to red indicates relative abundances range from low to high in the same row; § P value of WYQ vs GTQ
31
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1
32
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 2
33
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 3
34
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 4
35
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 5
36
ACCEPTED MANUSCRIPT Highlights A total of 66 volatiles, 215 bacteria and 151 fungi were detected from WYQ and GTQ.
RI PT
The predominant bacterial genera in Hong Qu are Burkholderia and Bacillus. Monascus purpureus is the most abundant fungal species in Hong Qu.
14 bacterial taxa and 10 fungal taxa can be utilized to distinguish WYQ and
SC
GTQ.
M AN U
The correlation of volatile components and microbiota were preliminarily
AC C
EP
TE D
explored.
S0740-0020(17)30093-X
DOI:
10.1016/j.fm.2017.07.019
Reference:
YFMIC 2834
To appear in:
Food Microbiology
Received Date: 31 January 2017 Revised Date:
17 July 2017
Accepted Date: 25 July 2017
Please cite this article as: Liu, Z., Wang, Z., Lv, X., Zhu, X., Chen, L., Ni, L., Comparison study of the volatile profiles and microbial communities of Wuyi Qu and Gutian Qu, two major types of traditional fermentation starters of Hong Qu glutinous rice wine, Food Microbiology (2017), doi: 10.1016/ j.fm.2017.07.019. 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 1
Comparison study of the volatile profiles and microbial communities of Wuyi Qu
2
and Gutian Qu, two major types of traditional fermentation starters of Hong Qu
3
glutinous rice wine
RI PT
4
Zhibin Liu1, Zhiyao Wang1, Xucong Lv2, Xiaoping Zhu1, Liling Chen1, Li Ni1*
5 6 1
Institute of Food Science & Technology, Fuzhou University, Fuzhou 350108, P.R. China
8
2
National Engineering Research Center of JUNCAO Technology, Fujian Agriculture and
M AN U
9
SC
7
Forestry University, Fuzhou, Fujian 350002, China
10
Corresponding author: Dr. Li Ni, Institute of Food Science & Technology, Fuzhou
12
University, No. 2 Xueyuan Road, Fuzhou 350108, P.R. China
13
Tel/ Fax: +86-591-22866378
14
Email: [email protected]
EP
16
AC C
15
TE D
11
1
ACCEPTED MANUSCRIPT 17
Abstract Hong Qu, which mainly contains Monascus sp. and other microorganisms, as
19
well as numerous microbial metabolites, is used as the fermentation starter of Hong
20
Qu glutinous rice wine, a traditional alcoholic beverage. Two widely-used types of
21
Hong Qu, namely Wuyi Qu (WYQ) and Gutian Qu (GTQ), were thoroughly
22
compared for their fermentation properties, volatile profiles, and microbiota structures
23
in this study. Significantly higher color value, glucoamylase and α-amylase activities
24
were discovered in WYQ. And substantial variation in volatile components and
25
microbial communities were also observed between them. It was identified that
26
bacterial genus Burkholderia dominated GTQ (71.62%) and Bacillus dominated
27
WYQ (44.73%), while Monascus purpureus was the most abundant fungal species in
28
both types of starters (76.99%). In addition, 213 bacterial genera and 150 fungal
29
species with low-abundance were also detected. Since the Linear Discriminant
30
Analysis Effect Size algorithm, 14 genus-level bacterial taxa and 10 species-level
31
fungal taxa could be utilized to distinguish these two types of starters. Moreover, the
32
potential correlation of the volatile components and microbiota within WYQ and
33
GTQ were further analyzed, by utilizing Partial Least Squares Discriminant Analysis.
34
Ultimately, this study provides detailed insight into the volatile profiles and microbial
35
communities presented in Hong Qu.
AC C
EP
TE D
M AN U
SC
RI PT
18
36 37
Keywords: Hong Qu glutinous rice wine starter; bacterial community; fungal
38
community; high throughput sequencing; Monascus purpureus; volatile components
39 2
ACCEPTED MANUSCRIPT 40
1. Introduction Chinese yellow rice wine, which may date back as early as the seventh
42
millennium before Christ (B.C.), is a traditional fermented alcoholic beverage that is
43
brewed from glutinous rice or wheat (McGovern et al., 2004). Hong Qu glutinous rice
44
wine, primarily produced in Fujian province, China, is one of the most ancient in this
45
particular category. The distinct characteristics of Hong Qu glutinous rice wine is the
46
utilization of Hong Qu as the fermentation starter, which produces a brilliant
47
bright-red color, fine sweet flavor, and also offers a healthcare functionality to the
48
wine (Park et al., 2016). Through an empirical solid fermentation process, Hong Qu is
49
inoculated with diverse bacteria and fungi, along with numerous pigments, enzymes
50
and other metabolites, facilitating the formation of alcohol and its unique flavor.
51
Similar to other traditional fermented foods, Hong Qu is generally prepared under
52
non-sterile fermentation conditions. Thus, a wide variety of microbes are presented in
53
Hong Qu, and the quality of starters will tend to vary from region to region, resulting
54
in the uncontrollability of the fermentation process of Hong Qu glutinous rice wine.
55
Thus, to fully understand microbiota structures, as well as the fermentation properties
56
and volatile profiles, and identify the key microorganism responsible for the aromatic
57
forming or other fermentation activities, will help to establish next generation starters,
58
which have a limited number of pure strains, and so as to improve the controllability.
AC C
EP
TE D
M AN U
SC
RI PT
41
59
Despite the variation of microbes within Hong Qu, two distinguishing types of
60
Hong Qu can still be recognized, which are Wuyi Qu (WYQ) and Gutian Qu (GTQ).
61
In general, WYQ presents a somewhat darker color and has a higher saccharifying 3
ACCEPTED MANUSCRIPT and liquefying power than that of GTQ (based on experience). Moreover, a dramatic
63
difference in flavor can easily be identified between these two types of Hong Qu, as
64
well as the rice wines that are fermented from them. Diverse microbiota within the
65
starters may be the primary factor or cause relating to these differences
RI PT
62
Bacteria and fungi, particularly filamentous fungi and yeasts, play an essential
67
role in the quality of Chinese rice wine fermentation starters, as their involvement in
68
the saccharification, liquification, alcoholic fermentation and flavor formation during
69
the wine brewing process. Therefore, numerous studies have been conducted to
70
adequately characterize the bacterial and fungal diversity of Chinese rice wine
71
fermentation starters (Guan et al., 2012; Lu et al., 2008). Due to the different
72
processing methods, Hong Qu have distinct aromatic and microbial profiles with other
73
rice wine starters. Previously our group has investigated the bacterial and fungal
74
community structures in WYQ. Briefly, based on culture-dependent or PCR-RFLP
75
and PCR-DGGE or MALDI-TOF mass spectrometry fingerprinting methods, 16
76
filamentous fungi species, including Monascus purpureus, Aspergillus flavus,
77
Aspergillus niger, Rhizopus oryzae and et al (Lv, Huang, et al., 2012), and 2 yeast
78
species, including Saccharomycopsis fibuligera and Saccharomyces cerevisiae (Lv,
79
Huang, Zhang, et al., 2013), and 16 bacterial species belonging to 6 bacterial genera
80
(Bacillus,
81
Lactococcus)(Lv et al., 2016) were identified from WYQ. Nevertheless, in terms of
82
GTQ, the research is limited, although this variety is a more extensively used one in
83
folk. The recent expansion of the next-generation sequencing technology and data
AC C
EP
TE D
M AN U
SC
66
Staphylococcus,
Leuconostoc,
4
Pediococcus,
Lactobacillus
and
ACCEPTED MANUSCRIPT 84
mining strategies could provide a more clarifying and precise picture of the microbial
85
community of fermentation starters and, thus, enable researchers to obtain a more
86
global insight into the functions and interactions of various microorganisms. In the present study, through solid phase microextraction (SPME)-GC/MS
88
analysis and high throughput sequencing of bacterial 16S rRNA genes and fungal ITS
89
rRNA genes upon Illumina HiSeq platform, the volatile profiles and microbial
90
community structures of WYQ and GTQ were investigated and compared. In addition,
91
the association between volatile components and microorganisms within these two
92
types of starters were also calculated through Partial Least Squares Discriminant
93
Analysis (PLS-DA) modelling, thus to preliminarily explore the roles of the microbes
94
as they relate to the aroma formation of Hong Qu. This study may provide a better
95
understanding of the microbial community in Hong Qu and their contribution to the
96
fermentation process, which may be helpful for the development of new starters with
97
pure strains and the improvement of the controllability of Hong Qu glutinous rice
98
wine brewing.
99
2. Materials and methods
101
SC
M AN U
TE D
EP
AC C
100
RI PT
87
2.1 Sample collection
Five Wuyi Hong Qu (WYQ) were collected from local markets of Jian’ou,
102
Fuzhou, Fuqing, Gutian, and Yongchun cities of Fujian Province, China; moreover,
103
five Gutian Hong Qu (GTQ) were also collected from local markets of Gutian city of
104
Fujian Province, China. All the collected samples are exhibited in Figure 1. After
105
collection, the samples were then ground into fine powder and appropriately stored at 5
ACCEPTED MANUSCRIPT 106
-20 °C until analysis.
107
2.2 Color value analysis In general, three colors (red, orange and yellow), including both water-soluble
109
and alcohol-soluble ones, are produced by Monascus stains, which with a maximum
110
absorption at 510 nm, 465 nm, and 410 nm, respectively. Normally, A410, A465, and
111
A510 are used to estimate the content of this three pigments (Yang et al., 2005; Zhang
112
et al., 2017). In this study, distilled water or 70% (v/v) ethanol were used for pigment
113
extraction: 0.2 g ground Hong Qu powder was suspended in 10 mL distilled water or
114
70% (v/v) ethanol at 60 °C for 60 min; then the extracted solution was centrifuged at
115
8000 rpm for 15min. Subsequently, appropriate dilution of supernate was measured
116
for absorbance at 410 nm, 465nm and 510 nm with a spectrophotometer (Hitachi
117
U-1900, Tokyo, Japan). The color value was defined as the summation of absorbance
118
units at 410 nm, 465nm and 510 nm× dilution factor per gram of dry samples (U/g).
119
2.3 Fermentative power analysis
TE D
M AN U
SC
RI PT
108
Moreover the fermentative power, including α-amylase, glucoamylase and
121
protease, of the ten Hong Qu samples were also determined as previously described
122
(Lv, Huang, et al., 2012). Briefly, the iodine colorimetry method was used to
123
determine the α-amylase activities of the samples (De Moraes et al., 1995). One unit
124
(U) of α-amylase activity was defined as the quantity of enzymes required to
125
hydrolyze 1 mg of starch in 10 min at 40 °C. The dinitrosalicylic (DNS) acid method
126
was then utilized to determine the glucoamylase activities of the samples (Bernfeld,
127
1955). One unit (U) of glucoamylase activity was defined as the amount of enzymes
AC C
EP
120
6
ACCEPTED MANUSCRIPT releasing 1 mg of glucose from the substrate in 60 minutes under strict assaying
129
conditions. Next, the protease activity was determined, according to the method
130
described by Farley and Ikasari (Farley and Ikasari, 1992). One unit (U) of protease
131
activity was defined as the amount of enzyme that liberated 1 µg tyrosine per min
132
under assaying conditions. All the experiments were conducted in triplicate.
133
2.4 Volatile profiles analyses
RI PT
128
Volatile profiles of the ten starters were analyzed by utilizing SPME–GC/MS
135
methods(Luo et al., 2008). Briefly, 1.00 g of each ground Hong Qu powder, 2.00 g of
136
NaCl and 5 mL distilled water were transferred to a 30 mL vial. This vial was then
137
tightly capped with a silicon septum and a 50/30 µm divinylbenzene/carboxen/poly
138
(dimethylsiloxane) (DVB/CAR/PDMS) coated fibre (Supelco, Inc., Bellefonte, PA,
139
USA) was inserted into the headspace of the vial for the volatile compound extraction
140
at 50 °C water bath for 40 min.
TE D
M AN U
SC
134
GC/MS analyses were performed on an Agilent 7890B gas chromatograph
142
(Agilent, Palo Alto, CA, USA) coupled with an Agilent 5973C mass spectrometer
143
(Agilent, Palo Alto, CA, USA). Separation of compounds was performed on an
144
HP-5MS column (30.0 m×0.25 mm×0.25µm, Agilent, USA). Helium was used as a
145
carrier gas at a constant flow rate of 1 mL/min. Oven temperature was maintained at
146
40 °C for 5 min, programmed at 5 °C/min to 250 °C and held for 5 min. The interface
147
temperature was set at 280 °C. The mass spectrometer was operated in electron
148
impact mode with the electron energy set at 70 eV and a scan range of 30–550 m/z.
149
The temperature of MS source and quadrupole was set at 230 and 150 °C, respectively.
AC C
EP
141
7
ACCEPTED MANUSCRIPT Each compound was identified using the National Institute of Standards and
151
Technology (NIST) library (11 L). The relative percentages of the detected peaks
152
were obtained by peak-area normalization. Afterward, based on the volatile profile
153
data set, Principal Component Analysis (PCA) was used to visualize the differences
154
among the starter samples, by using R software (Version 3.2.5) with the vegan, ade4
155
and ggplot2 packages.
156
2.5 Bacterial and fungal DNA extraction and Illumina sequencing
SC
RI PT
150
Bacterial and fungal DNA was extracted from 0.2 g of each ground Hong Qu
158
powder samples by utilizing a rapid DNA extraction kit (BioTeke Corporation,
159
Beijing, China), following the instruction of the manufacturer. The extracted bacterial
160
and fungal DNA were then checked by agarose gel electrophoresis.
M AN U
157
Bacterial primers 341-F (5’-CCT AYG GGR BGC ASC AG-3’) and 806-R (5’-
162
GGA CTA CNN GGG TAT CTA AT-3’) with specific barcode were used to amplify
163
the V3–V4 region of bacterial 16S rRNA genes. The fungal primers ITS5-1737-F
164
(5’-GGA AGT AAA AGT CGT AAC AAG G-3’) and ITS2-2043-R (5’-GCT GCG
165
TTC TTC ATC GAT GC-3’) with specific barcode were used to amplify the ITS1
166
regions of the fungal ITS rRNA genes. The sequencing libraries of bacterial 16S
167
rRNA genes and fungal ITS rRNA genes were generated for high throughput
168
sequencing utilizing the TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina,
169
San Diego,
170
platform (Illumina, San Diego, USA) by Novogene Bioinformatics Technology Co.,
171
Ltd. (Beijing, China).
AC C
EP
TE D
161
USA). Then both libraries were sequenced on an Illumina HiSeq2500
8
ACCEPTED MANUSCRIPT 172
2.6 Bioinformatic analysis Raw sequencing reads obtained from the Illumina platform were then merged
174
using FLASH software (Version 1.2.7)(Magoč and Salzberg, 2011), and filtered with
175
the QIIME software (Version 1.7)(Caporaso et al., 2010). All quality filtered
176
sequencing reads were then clustered into operational taxonomic units (OTUs) with a
177
threshold of 97% sequence similarity, by utilizing UPARSE software (Version
178
7.0)(Edgar, 2013). The representative sequence for each bacterial OTU was then
179
annotated by utilizing the GreenGene Database based on RDP classifier (Version 2.2)
180
algorithm (DeSantis et al., 2006). For each fungal representative sequence, the Unite
181
Database was used based on Blast algorithm which was calculated by QIIME
182
software, so as to annotate taxonomic information (Kõljalg et al., 2013). The OTUs
183
abundance information was normalized, by utilizing a standard sequence number
184
corresponding to the sample with the least sequences. Subsequent the differences of
185
samples in OTU-level were evaluated through the PCA, by using R software.
186
Moreover, the linear discriminant analysis (LDA) effect size (LEfSe) algorithm was
187
performed to identify the representative bacterial and fungal taxa of each type of
188
starters, by utilizing the Huttenhower Lab Galaxy Server (Goecks et al., 2010). The
189
relative abundances of the representative taxa were further visualized with heatmap
190
and hierarchical clustering (with complete linkage), by utilizing R software with a
191
pheatmap package. Subsequently, the possible relation between the major volatile
192
components with relative content over 1.0 % and the representative bacterial and
193
fungal taxa was explored, by utilizing a PLS-DA modelling, and plotted in a
AC C
EP
TE D
M AN U
SC
RI PT
173
9
ACCEPTED MANUSCRIPT 194
concatenation panel, by using R software with the mixOmics package. Other data are expressed as mean ± SD. Furthermore, the statistical significance
196
between WYQ and GTQ were analyzed with “Independent-Samples T-Test” using
197
SPSS software (Version 19.0.0), while the significance threshold was established at
198
0.05.
199
3. Results
200
3.1 Color value and fermentative power of WYQ and GTQ
SC
RI PT
195
As shown in Figure 1, the appearance of the ten starters were similar; however, in
202
general, GTQ appears slightly redder, while WYQ appears slightly darker in color.
203
The pigment extracted from these two types of Hong Qu exhibited remarkable
204
differences (Figure 2). For example, GTQ had a significantly higher content of water
205
soluble and ethanol soluble pigments (P = 0.013 and 0.037, respectively). However, it
206
was also noticed that WYQ5 had higher color values of water extraction and ethanol
207
extraction than those of other WYQ samples, and even higher than those of some
208
GTQ samples. The enzymatic properties of the ten collected starter samples were also
209
investigated in the present study. Significantly higher glucoamylase and α-amylase
210
activities were discovered in WYQ than those in GTQ, whereas, in terms of protease,
211
there were no significant differences found between them (Table 1). It is also worth
212
noting that, the color value and enzymatic activities were diverse among samples,
213
even within the same type.
214
3.2 Volatile components in WYQ and GTQ
215
AC C
EP
TE D
M AN U
201
The volatile components presented in WYQ and GTQ were analyzed with 10
ACCEPTED MANUSCRIPT SPME-GC/MS approach. As summarized in Table 2, a total of 66 volatile compounds
217
were identified from the ten starter samples, including 10 alcohols, 6 acids, 13 esters,
218
14 aldehydes and ketones, 7 aromatic compounds, 5 lactones, 4 alkanes and
219
cycloalkanes and 7 nitrogen-containing compounds. Based on the normalized peak
220
area percentages of individual components, PCA score plot was performed to
221
visualize the differences between these two types of starters (Figure 3A). It was
222
shown that the volatile profiles of WYQ and GTQ exhibited little similarity to each
223
other. In addition, the 5 samples of WYQ seemed more close to each other; while
224
samples belonging to GTQ distributed more widely, which indicated that GTQ present
225
higher variations in volatile profiles than WYQ. Of these volatile components, 13
226
components, including Ethyl butyrate, Ethyl isobutyrate, Butanoic acid, Ethyl Acetate,
227
Ethanol, Octanoic acid, Ethyl caproate, Hexanoic acid, Hexyl alcohol, Ethyl caprylate,
228
Hexanal, 2-Heptanone and 2-Nonanone, were defined as the major volatile
229
components with relative contents over 1.00 %. They jointly constituted from 86.40%
230
to 94.31% of the total peak area, and thus covered the major information contained in
231
the volatile profiles.
232
3.3 Overall bacterial and fungal structures comparison of WYQ and GTQ
SC
M AN U
TE D
EP
AC C
233
RI PT
216
The bacterial and fungal components in the traditional fermentation starters WYQ
234
and GTQ were investigated by Illumina HiSeq sequencing analyses. As for the
235
bacterial community, a total of 321,142 quality filtered sequencing reads
236
corresponding to the V3–V4 region of bacterial 16S rRNA genes, about 426 bp in
237
length, were obtained. All sequences were clustered into 6650 OTUs with a 97% 11
ACCEPTED MANUSCRIPT similarity level, ranging from 430 (GTQ4) to 868 (WYQ3). Following the high
239
throughput sequencing of the fungal ITS rRNA genes amplicons, 757,328 quality
240
filtered sequencing reads, with an average length of about 224 bp, were generated
241
from the ten starter samples and assigned to 895 OTUs with a 97% similarity level.
242
Good's coverage estimation values were within the range of 98.9-100%, indicating
243
adequate sequence coverage to reliably describe the full bacterial and fungal
244
communities presented in all the samples (Supplementary Table S1).
SC
RI PT
238
The bacterial and fungal communities within the two types of starters were
246
compared by using PCA analysis, based on the total OTUs’ relative abundance
247
information (Figure 3B and C). Both PCA score plots depicted 2 divided groups,
248
which revealed relatively low similarities in bacterial and fungal communities
249
between GTQ and WYQ. Moreover, it was also revealed that the bacterial
250
communities within WYQ were dramatically distinct from one another, while the
251
GTQ patterns were quite well clustered, which indicated more similarity of bacterial
252
profile within GTQ samples.
253
3.4 Bacterial and fungal profiles of Hong Qu
TE D
EP
AC C
254
M AN U
245
The bacterial OTU sequences were annotated into different taxa from the
255
GreenGene Database. However, due to the high sequence homology of the V3–V4
256
region of the bacterial 16S rRNA genes, the obtained bacterial reads cannot be
257
accurately annotated beyond the genus level. Thus, the taxonomic information of
258
bacteria at the genus level were recorded, and a total of 215 bacterial genera were
259
identified across the ten starter samples. The 10 most abundant genera were: 12
ACCEPTED MANUSCRIPT Burkholderia,
Bacillus,
Erwinia,
Klebsiella,
Ochrobactrum,
Staphylococcus,
261
Shewanella, Agrobacterium, Acinetobacter and Lactococcus, which, taken as a
262
composite, comprised from 69.69% (WYQ4) to 96.50% (GTQ4) of all sequences.
263
The relative abundance of the 10 most predominant bacterial genera are depicted in
264
Table 3. The phylotypes of Bacillus and Burkholderia represented the most
265
predominant bacteria presented in WYQ (Circa 45% and 8% of relative abundance,
266
respectively), while Burkholderia were the most abundant bacteria in GTQ,
267
accounting for approximately 72 % of the total sequences. When comparing the
268
bacterial genera taxonomic differences between WYQ and GTQ, GTQ exhibited a
269
significantly higher relative abundance of Burkholderia (P = 0.000), whereas WYQ
270
had a significantly higher Bacillus proportion (P = 0.000).
M AN U
SC
RI PT
260
The ITS1 regions of fungal ITS rRNA genes facilitated the species level
272
taxonomic annotation of fungi from the Unite Database. Following the sequencing of
273
these fungal genes, 151 different species were identified from the Unite Database, of
274
which the 10 most abundant fungal species were: Monascus purpureus,
275
Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Eurotiomycetes sp.,
276
Fusarium pseudensiforme, Rhizopus oryzae, Rhizopus microsporus, Aspergillus sp.
277
and Agaricomycetes sp., which, when combined together, constituted from 51.85%
278
(WYQ3) to 99.54% (GTQ2) of all sequences, as shown in Table 3. Of these,
279
Monascus purpureus represented the largest fraction in both WYQ and GTQ, which
280
accounted for 55.77 ± 16.29% and 98.21 ± 0.82% of the total fungal sequences,
281
respectively. Generally, the fungal profile in WYQ was different from that in GTQ.
AC C
EP
TE D
271
13
ACCEPTED MANUSCRIPT Specifically, GTQ had a significantly higher relative abundance of M. purpureus (P =
283
0.000), whereas WYQ had a significantly higher relative abundance of
284
Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Eurotiomycetes sp.,
285
Rhizopus oryzae and Aspergillus sp. (P < 0.05).
286
3.5 Representative bacteria and fungi in Hong Qu
RI PT
282
To identify the specific bacterial and fungal taxa within each type of sample, the
288
LEfSe analysis with 3.0 as the threshold on the LDA score for discriminative features
289
was applied (Figure 4). In the comparison of the bacterial communities in WYQ and
290
GTQ, 14 genus-level bacterial taxa were verified as being differentially abundant in
291
these two particular types of starters (Figure 4A). Of these, 13 taxa were enriched in
292
WYQ, namely, Bacillus, Klebsiella, Ochrobactrum, Staphylococcus, Acinetobacter,
293
Agrobacterium, Shewanella, Lactobacillus, Microbacterium, Brachybacterium,
294
Comamonas, Pseudomonas and Sphingomonas. Whereas, genus Bacillus had the
295
highest LDA score of 5.44, followed by Klebsiella with an LDA of 4.55,
296
Ochrobactrum with an LDA of 4.40 and Staphylococcus with an LDA of 4.10.
297
Whereas only genus Burkholderia was enriched in GTQ, with an LDA of 5.52.
M AN U
TE D
EP
AC C
298
SC
287
For fungal LDA analysis, 9 species-level fungal taxa were found to be enriched in
299
WYQ, including Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Rhizopus
300
oryzae, Fusarium pseudensiforme, Eurotiomycetes sp., Rhizopus microsporus,
301
Aspergillus sp. and Meyerozyma guilliermondii, as depicted in Figure 4C. Of these,
302
species Saccharomyces sp. and Aspergillus flavus had an LDA score higher than 4.0
303
(4.63 and 4.46, respectively). The fungal group enriched in GTQ was only Monascus 14
ACCEPTED MANUSCRIPT 304
purpureus, with an LDA score of 5.02. Based on the relative abundance of these 14 bacterial taxa and 10 fungal taxa,
306
heatmap plots and hierarchical clustering were then employed, so as to provide a
307
visual and overall comparison for differentiating the two types of starters, as reflected
308
in Figures 4B and D. Unsurprisingly, the clustering depicted two clear separated
309
groups in both bacterial and fungal communities, which indicated that these taxa were
310
the most representative microbial groups in WYQ or GTQ.
311
3.6 Correlation analysis between major volatile components and representative
312
microbiota in Hong Qu
M AN U
SC
RI PT
305
The relationship of the 13 major volatile components, 14 representative bacterial
314
taxa and 10 representative fungal taxa of the both types of starters were explored by
315
using PLS-DA algorithm and visualized by using R software. Using a correlation
316
cut-off of 0.9, the association among the three data sets were depicted in Figure 5. It
317
was shown that 2-Nonanone was positively correlated with Saccharomyces sp.;
318
Ethanol was positively correlated with Saccharomyces sp. and Rhizopus microsporus;
319
2-Heptanone was negatively correlated with Agrobacterium, Brachybacterium,
320
Meyerozyma guilliermondii and Fusarium pseudensiforme; Hexyl alcohol was
321
positively correlated with Aspergillus flavus; Ethyl caproate was positively correlated
322
with Agrobacterium; Octanoic acid was negatively correlated with Bacillus and
323
Rhizopus oryzae.
AC C
EP
TE D
313
324
Based on the concatenation panel, correlation between representative bacteria and
325
fungi was complicated. A total of 22 pair-wise associations between the two data sets 15
ACCEPTED MANUSCRIPT were identified, of which Monascus purpureus was negatively correlated with
327
Bacillus, Agrobacterium and Sphingomonas; Burkholderia was negatively correlated
328
with Aspergillus niger and Rhizopus oryzae; as for Bacillus, in addition to the
329
negative correlation with Monascus purpureus, it was also positively correlated with
330
Rhizopus oryzae.
331
4. Discussion
RI PT
326
The major manufacturing process of Hong Qu includes the preparation of
333
polished round grain rice; inoculation of previously prepared ferments; cultivation of
334
microorganisms; and finally dehydration for preservation. Different ferments are used
335
for WYQ and GTQ, and passed down batch to batch. Hence, the unique cultures
336
likely established in these two types of starters, potentially lead to the distinctive
337
flavor of the starters and final traditional brewed wines. However, at this time, the
338
properties of these two types of starters have not yet been fully characterized and
339
compared. Thus, the color value, fermentative power, volatile profiles and, more
340
importantly, the bacterial and fungal community structures of WYQ and GTQ were
341
investigated and compared in the present study.
M AN U
TE D
EP
AC C
342
SC
332
In general, WYQ has darker appearance (Figure 1), but lower content of
343
Monascus pigments than GTQ (Figure 2). The color presented in GTQ may mainly
344
derived from Monascus sp., while, in terms of WYQ, beyond Monascus sp., other
345
microbial pigments may also contribute to the color of WYQ. In addition, higher
346
glucoamylase and α-amylase activities were discovered in WYQ than those in GTQ.
347
And remarkable different volatile profiles were also observed in these two types of 16
ACCEPTED MANUSCRIPT 348
starters. The varied fermentation powers and volatile components presented in the
349
starters were mainly derived from intricate microorganisms harbored in the starters. Following the high throughput sequencing analysis, the bacterial and fungal
351
profiles of these two types of starters were further investigated. It was established that
352
the major bacterial genus presented in WYQ and GTQ was Bacillus and Burkholderia,
353
respectively, each of which were also identified as the most representative bacteria of
354
these two types of starters by using LEfSe algorithm. Bacillus, a gram-positive genus,
355
can survive under low-moisture and high-temperature conditions (Ma et al., 2014);
356
thus, it is also frequently detected in various traditional fermented foods, such as
357
Daqu (Wang et al., 2008), nuruk (Song et al., 2013), fermented soya bean (Kiers et al.,
358
2000) etc. Several members within the Bacillus genus may play an important role in
359
the brewing process of some fermented foods, as they can secrete various hydrolytic
360
enzymes, including amylases and proteases (Simonen and Palva, 1993); moreover,
361
they can produce a broad range of volatile compounds, which include pyrazines,
362
aldehydes, ketones and alcohols (Azokpota et al., 2010). Bacterial genus Burkholderia,
363
an environmental ubiquitous gram-negative bacteria is frequently found in open
364
fermentation starters, such as Daqu (Li et al., 2015). Some species within the
365
Burkholderia genus can synthesize short chain alkyl esters (Dutta and Dasu, 2011),
366
which may contribute to the flavor of the final products.
AC C
EP
TE D
M AN U
SC
RI PT
350
367
Previously, our group investigated the bacterial communities of WYQ, based on
368
the PCR-DGGE approach (Lv, Weng, et al., 2012). Several species within the genus
369
of Bacillus, including B. ginsengihumi, B. megaterium or B. aryabhattai, B. subtilis, B. 17
ACCEPTED MANUSCRIPT methylotrophicus and B. amyloliquefaciens were identified in most WYQ samples.
371
DGGE is considered to be a low-resolution and low-sensitivity technique, and in this
372
study Illumina HiSeq sequencing analysis was performed, targeting the V3–V4 region
373
of the bacterial 16S rRNA genes, so as to further explore the bacterial communities in
374
WYQ and GTQ, where numerous other specific details were revealed. Apart from
375
Bacillus and Burkholderia, the genera Erwinia, Klebsiella, Ochrobactrum and some
376
other low-abundant taxa also presented in Hong Qu, although some of which were
377
related to spoilage or pathogenic bacteria, such as Klebsiella, Staphylococcus,
378
Shewanella, etc. The existence of these bacteria may raise concerns about the safety
379
of the final products, and they may also be responsible for the acidification and
380
spoilage of wine. However, interestingly, in our another study, during the traditional
381
brewing process of Hong Qu glutinous rice wine with WYQ as the starter, some
382
bacterial species (such as Bacillus sp., P. acidilactici, L. brevis and P. pentosaceus)
383
decreased as the fermentation progressed (Lv, Huang, Chen, et al., 2013). Thus, some
384
factors may influence the growth of these spoilage or pathogenic bacteria. Such as, it
385
was revealed in this study that M. purpureus has a strong negative correlation with
386
Bacillus, Agrobacterium and Sphingomonas. In addition, A. niger and R. oryzae were
387
also found to be negatively correlated to Burkholderia.
SC
M AN U
TE D
EP
AC C
388
RI PT
370
Another noteworthy feature of the bacterial profile in Hong Qu is that several
389
genera within the order Lactobacillales, which primarily include Lactococcus,
390
Lactobacillus, Pediococcus, Streptococcus, Weissella, Enterococcus and Leuconostoc
391
were detected across all samples. The bacteria order Lactobacillales, produces lactic 18
ACCEPTED MANUSCRIPT acid as the major metabolic end product of carbohydrate fermentation, and is usually
393
referred to as lactic acid bacteria (LAB). LAB plays a critical role in various
394
fermented foods such as cheese, yoghurt, pickles, wine and rice wine, by enhancing
395
the microbial safety and, thus, offering pleasant sensory and nutritional benefits
396
(Leroy and De Vuyst, 2004). Although bacteria within this division consist of a small
397
portion of the total microorganisms, and together account for a mere 0.7% of all
398
bacterial sequence, they experience rapid growth during the traditional brewing
399
process of Hong Qu glutinous rice wine (Lv, Huang, Chen, et al., 2013). After
400
inoculation, LAB can rapidly acidify the starchy substrates through the production of
401
lactic acid and, thus, suppress the growth of various acid-intolerant pathogens. From
402
this perspective, the presence of LAB may be one of the inhibiting factors for
403
pathogens. Moreover, the production of acetic acid, ethanol, aroma compounds,
404
esterases, lipases and alcohol acetyl transferases by LAB is highly significant for
405
enhancing the flavor forming attributes of rice wine.
TE D
M AN U
SC
RI PT
392
However, the most distinguishing feature of the Hong Qu from other starters is
407
that it contains large populations of Monascus sp. strains, which are a unique
408
red-pigment producing filamentous fungi. The red appearance and relative high color
409
values of all Hong Qu implied the presence of Monascus. In GTQ, the proportion of
410
M. purpureus greatly exceeded those of WYQ, which was reflected in the higher color
411
values (Figure 2). Similarly, the higher color values of WYQ5 than other WYQ
412
samples was attributed to its relatively higher M. purpureus content. Moreover, as
413
confirmed by the LEfSe analysis, M. purpureus was distinctly and dramatically
AC C
EP
406
19
ACCEPTED MANUSCRIPT enriched in GTQ. M. purpureus has a long history of being used for making red rice
415
wine, red soybean cheese and Anka (red rice) in Southern China, Japan and Southeast
416
Asia (Lin et al., 2008). As the secondary metabolites, Monascus pigments have been
417
well studied and reviewed in relation to their structures, biosynthetic pathway,
418
fermentation processes, physicochemical properties detection methods, functions, and
419
molecular biological activity (Feng et al., 2012; Lin et al., 2008). Due to the wide
420
range of biological activities of Monascus pigments, Hong Qu glutinous rice wine is
421
believed to possess a variety of nutritional benefits, such as anti-inflammatory activity,
422
cancer cell cytotoxic activity, anti-hypertensive activity, cholesterol-lowering activity
423
and various other beneficial attributes (Park et al., 2016). Moreover, Monascus
424
pigments possess antimicrobial activities (Feng et al., 2012), establishing it as being
425
an active pathogen inhibitor in the Hong Qu, as reflected in this study by negatively
426
correlating with Bacillus, Agrobacterium and Sphingomonas. It is also worth to notice
427
that, in addition to pigments, Monascus can also produce mycotoxin citrinin, which
428
may raise concern of the safety of Hong Qu and Hong Qu glutinous rice wine.
EP
TE D
M AN U
SC
RI PT
414
In addition to M. purpureus, some other fungal taxa, which include
430
Saccharomyces sp., Aspergillus flavus, Aspergillus niger, Eurotiomycetes sp.,
431
Rhizopus oryzae, Aspergillus sp., etc., also accounted for a considerable proportion of
432
fungi in Hong Qu; particularly prevalent in WYQ. Many members of the fungal genus
433
Saccharomyces are considered very important in the fermentation of foods, especially
434
S. cerevisiae, which is widely used in making wine, bread, and beer, and is an
435
effective ethanol producer; often referred to as “yeast” (Kurtzman, 1994). During the
AC C
429
20
ACCEPTED MANUSCRIPT brewing process of rice wine, the population of Saccharomyces increase dramatically,
437
along with the increase of ethanol, which will result in the decrease of alcohol
438
intolerance among microorganisms. Therefore, this may be another reason for the
439
reduction of some pathogens during wine brewing. As the predominant filamentous
440
fungi in Daqu, the genus Aspergillus has the ability to secrete large quantities of a
441
broad range of different enzymes into its environment, which may contribute to the
442
saccharification of the starch in wine mash (Oda et al., 2006; Zheng et al., 2011). R.
443
oryzae, with a strong amylase production ability, is commonly found in various types
444
of amylolytic fermentation starters (Choi et al., 2012; Dung et al., 2007). Through
445
synthesis of volatile compounds, R. oryzae also makes a great contribution to the
446
flavor formation of wine. Taken as a whole, WYQ had significantly higher
447
proportions of filamentous fungi and yeast that have strong enzyme synthesis abilities,
448
such as Saccharomyces and Aspergillus, making WYQ a more powerful fermentation
449
starter. This is also consistent with the results of the comparison of enzymatic
450
properties of the ten collected starter samples. As a matter of fact, WYQ has long
451
been regarded as a “stronger” starter by local brewers. In addition, Aspergillus may
452
also contribute to the darker color of WYQ.
SC
M AN U
TE D
EP
AC C
453
RI PT
436
The contributions of all these bacteria and fungi were also preliminarily explored
454
by digging the correlation between volatile components and microbial communities. A
455
concatenation panel was established to visualize their intricate correlation. However,
456
most volatile components are secondary metabolites, it may be not sufficient by only
457
using PLS-DA modelling to interpret their relationships. More importantly, the 21
ACCEPTED MANUSCRIPT correlations present here were obtained by mathematical model, more well designed
459
studies are required to validate these correlations. Nevertheless, this concatenation
460
panel indicated a potential roles of microorganism in aromatic forming, as well as a
461
complicated symbiotic relationship between various bacteria and fungi, which
462
provided clues for the further studies.
463
5. Conclusions
RI PT
458
In the current study, the fermentative properties, volatile profiles and bacterial
465
and fungal communities of the two major types of traditional fermentation starters of
466
Hong Qu glutinous rice wine were compared. Significantly higher color value,
467
glucoamylase and α-amylase activities were discovered in WYQ. And substantial
468
variation in volatile components and microbial communities were also observed in
469
these two types of starters. In addition, 14 genus-level bacterial taxa and 10
470
species-level fungal taxa were identified as the representative microorganism in WYQ
471
and GTQ by LEfSe analysis; and their interrelationship and potential roles in aroma
472
forming were also preliminarily explored by PLS-DA modeling. This study provides a
473
more comprehensive and thorough insight into the properties of Hong Qu, and may be
474
helpful in developing next generation starters for Hong Qu glutinous rice wine, which
475
with limited number of pure strains.
476
Acknowledgment
AC C
EP
TE D
M AN U
SC
464
477
This study was supported by the National Natural Science Foundation of China
478
(No. 31371820 and 31501494) and the Natural Science Foundation of Fujian Province
479
(No. 2016J01707). 22
ACCEPTED MANUSCRIPT 480
Reference
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
Azokpota, P., Hounhouigan, J.D., Annan, N.T., Odjo, T., Nago, M.C., Jakobsen, M., 2010. Volatile compounds profile and sensory evaluation of Beninese condiments produced by inocula of Bacillus subtilis. Journal of the Science of Food and Agriculture 90, 438-444. Bernfeld, P., 1955. Amylases, alpha and beta. Methods in enzymology I, 149-158. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pena,
RI PT
A.G., Goodrich, J.K., Gordon, J.I., 2010. QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335-336.
Choi, J.-S., Jung, S.-T., Choi, J.-H., Choi, H.-S., Baek, S.-Y., Yeo, S.-H., 2012. Quality characteristics of wheat nuruks by storage conditions of liquid starters using Rhizopus oryzae N174. Microbiology and Biotechnology Letters 40, 319-324.
De Moraes, L., Astolfi-Filho, S., Oliver, S., 1995. Development of yeast strains for the efficient
SC
utilisation of starch: evaluation of constructs that express α-amylase and glucoamylase separately or as bifunctional fusion proteins. Applied microbiology and biotechnology 43, 1067-1076. DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Huber, T., Dalevi, D., Hu, P.,
M AN U
Andersen, G.L., 2006. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and environmental microbiology 72, 5069-5072. Dung, N., Rombouts, F., Nout, M., 2007. Characteristics of some traditional Vietnamese starch-based rice wine fermentation starters (men). LWT-Food Science and Technology 40, 130-135. Dutta, K., Dasu, V.V., 2011. Synthesis of short chain alkyl esters using cutinase from Burkholderia cepacia NRRL B2320. Journal of Molecular Catalysis B: Enzymatic 72, 150-156. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature
TE D
methods 10, 996-998.
Farley, P.C., Ikasari, L., 1992. Regulation of the secretion of Rhizopus oligosporus extracellular carboxyl proteinase. Microbiology 138, 2539-2544.
Feng, Y., Shao, Y., Chen, F., 2012. Monascus pigments. Applied microbiology and biotechnology 96, 1421-1440.
EP
Goecks, J., Nekrutenko, A., Taylor, J., 2010. Galaxy: a comprehensive approach for supporting accessible, reproducible, and transparent computational research in the life sciences. Genome biology 11, 1.
AC C
Guan, Z.B., Zhang, Z.H., Cao, Y., Chen, L.L., Xie, G.F., Lu, J., 2012. Analysis and comparison of bacterial communities in two types of ‘wheat Qu’, the starter culture of Shaoxing rice wine, using nested PCR‐DGGE. Journal of the Institute of Brewing 118, 127-132. Kiers, J.L., Rombouts, F., Nout, M., 2000. In vitro digestibility of Bacillus fermented soya bean. International journal of food microbiology 60, 163-169. Kõljalg, U., Nilsson, R.H., Abarenkov, K., Tedersoo, L., Taylor, A.F., Bahram, M., Bates, S.T., Bruns, T.D., Bengtsson‐Palme, J., Callaghan, T.M., 2013. Towards a unified paradigm for sequence‐based identification of fungi. Molecular ecology 22, 5271-5277. Kurtzman, C.P., 1994. Molecular taxonomy of the yeasts. Yeast 10, 1727-1740. Leroy, F., De Vuyst, L., 2004. Lactic acid bacteria as functional starter cultures for the food fermentation industry. Trends in Food Science & Technology 15, 67-78. Li, P., Liang, H., Lin, W.-T., Feng, F., Luo, L., 2015. Microbiota dynamics associated with environmental conditions and potential roles of cellulolytic communities in traditional Chinese cereal starter 23
ACCEPTED MANUSCRIPT solid-state fermentation. Applied and environmental microbiology 81, 5144-5156. Lin, Y.-L., Wang, T.-H., Lee, M.-H., Su, N.-W., 2008. Biologically active components and nutraceuticals in the Monascus-fermented rice: a review. Applied microbiology and biotechnology 77, 965-973. Lu, J., Cao, Y., Fang, H., Li, W.-j., Xie, G.-f., Zou, H.-j., HU, Z.-m., 2008. Fungal community of wheat Qu of Shaoxing rice wine. Journal of Food Science and Biotechnology 2, e23. Luo, T., Fan, W., Xu, Y., 2008. Characterization of Volatile and Semi‐Volatile Compounds in Chinese Spectrometry. Journal of the Institute of Brewing 114, 172–179.
RI PT
Rice Wines by Headspace Solid Phase Microextraction Followed by Gas Chromatography‐Mass Lv, X.-C., Huang, R.-L., Chen, F., Zhang, W., Rao, P.-F., Ni, L., 2013. Bacterial community dynamics during the traditional brewing of Wuyi Hong Qu glutinous rice wine as determined by culture-independent methods. Food Control 34, 300-306.
Lv, X.-C., Huang, X.-L., Zhang, W., Rao, P.-F., Ni, L., 2013. Yeast diversity of traditional alcohol culture-independent methods. Food Control 34, 183-190.
SC
fermentation starters for Hong Qu glutinous rice wine brewing, revealed by culture-dependent and Lv, X.-C., Huang, Z.-Q., Zhang, W., Rao, P.-F., Ni, L., 2012. Identification and characterization of filamentous fungi isolated from fermentation starters for Hong Qu glutinous rice wine brewing. The
M AN U
Journal of general and applied microbiology 58, 33-42.
Lv, X.-C., Jia, R.-B., Chen, J.-H., Zhou, W.-B., Li, Y., Xu, B.-X., Liang, Y.-T., Liu, B., Chen, S.-J., Tian, Y.-T., 2016. Development of Reverse Transcription Quantitative Real-Time PCR (RT-qPCR) Assays for Monitoring Saccharomycopsis fibuligera. Food Analytical Methods, 1-11.
Lv, X.-C., Weng, X., Zhang, W., Rao, P.-F., Ni, L., 2012. Microbial diversity of traditional fermentation starters for Hong Qu glutinous rice wine as determined by PCR-mediated DGGE. Food Control 28, 426-434.
TE D
Ma, K., Maeda, T., You, H., Shirai, Y., 2014. Open fermentative production of l-lactic acid with high optical purity by thermophilic Bacillus coagulans using excess sludge as nutrient. Bioresource technology 151, 28-35.
Magoč, T., Salzberg, S.L., 2011. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27, 2957-2963.
EP
McGovern, P.E., Zhang, J., Tang, J., Zhang, Z., Hall, G.R., Moreau, R.A., Nuñez, A., Butrym, E.D., Richards, M.P., Wang, C.-s., 2004. Fermented beverages of pre-and proto-historic China. Proceedings of the National Academy of Sciences of the United States of America 101, 17593-17598. Oda, K., Kakizono, D., Yamada, O., Iefuji, H., Akita, O., Iwashita, K., 2006. Proteomic analysis of
AC C
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
extracellular proteins from Aspergillus oryzae grown under submerged and solid-state culture conditions. Applied and environmental microbiology 72, 3448-3457. Park, K.H., Liu, Z., Park, C.-S., Ni, L., 2016. Microbiota associated with the starter cultures and brewing process of traditional Hong Qu glutinous rice wine. Food Science and Biotechnology 25, 649-658. Simonen, M., Palva, I., 1993. Protein secretion in Bacillus species. Microbiological Reviews 57, 109-137. Song, S.H., Lee, C., Lee, S., Park, J.M., Lee, H.-J., Bai, D.-H., Yoon, S.-S., Choi, J.B., Park, Y.-S., 2013. Analysis of microflora profile in Korean traditional nuruk. J. Microbiol. Biotechnol 23, 40-46. Wang, C.-l., Shi, D.-j., Gong, G.-l., 2008. Microorganisms in Daqu: a starter culture of Chinese Maotai-flavor liquor. World Journal of Microbiology and Biotechnology 24, 2183-2190. Yang, S., Zhang, H., Li, Y., Qian, J., Wang, W., 2005. The ultrasonic effect on biological characteristics of Monascus sp. Enzyme and microbial technology 37, 139-144. 24
ACCEPTED MANUSCRIPT 567 568 569 570
Zhang, X., Liu, W., Chen, X., Cai, J., Wang, C., He, W., 2017. Effects and Mechanism of Blue Light on Monascus in Liquid Fermentation. Molecules 22, 385. Zheng, X.W., Tabrizi, M.R., Nout, M., Han, B.Z., 2011. Daqu—a traditional Chinese liquor fermentation starter. Journal of the Institute of Brewing 117, 82-90.
AC C
EP
TE D
M AN U
SC
RI PT
571
25
ACCEPTED MANUSCRIPT Figure captions Figure 1 Ten Hong Qu samples (5 WYQ and 5 GTQ) obtained from different geographical
RI PT
areas of Fujian province, China.
Figure 2 Color values of the water and ethanol extractions of the ten starter samples. The color value was recorded as the summation of absorbance units at 410 nm, 465nm and 510
SC
nm× dilution factor per gram dry samples (U/g). Each value represented the mean ± SD of
M AN U
three determinations. P values indicated the statistical difference between data sets.
Figure 3 PCA plots of volatile components and microbial communities of the ten Hong Qu samples. (A) PCA plots based on the normalized peak area percentages of volatile
TE D
components; (B) PCA plots based on the relative abundance of bacterial 16S gene OTUs; (C) PCA plots based on the relative abundance of fungal ITS gene OTUs.
EP
Figure 4 Representative bacteria and fungi of WYQ and GTQ. (A) LEfSe comparison of
AC C
bacterial communities at genus-level between WYQ and GTQ; (B) Heatmap comparison and hierarchical clustering dendrogram based on the relative abundance of 14 representative bacteria genera across all samples; (C) LEfSe comparison of fungal communities at species-level between WYQ and GTQ; (D) Heatmap comparison and hierarchical clustering dendrogram based on the relative abundance of 10 representative fungal species across all samples.
26
ACCEPTED MANUSCRIPT Figure 5 Correlation analysis between major volatile components and representative microbiota of WYQ and GTQ by PLS-DA modeling. Red lines in the circle represent positive correlation, while blue lines represent negative correlation between volatile
RI PT
components and microbiota. Red blocks on the circle represent the 13 major volatile components, blue blocks represent the 14 genus-level representative bacterial taxa and green blocks represent the 10 species-level representative fungal taxa. Blue lines and orange lines
SC
outside the circle represent the relative content of volatile components or microbiota
AC C
EP
TE D
M AN U
presented in GTQ and WYQ, respectively.
27
ACCEPTED MANUSCRIPT Table 1 Glucoamylase, α-amylase and protease activities (U/g) of the ten Hong Qu samples * α-Amylase activity
Protease activity
WYQ1 WYQ2 WYQ3 WYQ4 WYQ5
103.35±4.89 141.13±1.15 158.94±1.57 121.59±4.84 96.62±1.70
862.05±20.36 1043.42±22.48 1269.60±26.65 904.73±29.87 603.33±9.09
1.75±0.22 3.35±1.24 4.32±0.17 3.98±0.93 1.41±0.54
GTQ1 GTQ2 GTQ3 GTQ4 GTQ5
66,01±3.20 62.10±1.21 58.41±4.64 62.53±3.21 81.20±1.15
471.57±13.23 486.82±22.92 271.84±18.03 188.63±10.83 142.86±19.27
1.85±0.22 2.14±0.22 1.02±0.15 0.92±0.30 1.80±0.30
0.001
0.057
0.001 §
EP
TE D
M AN U
Values are presented as means ±SD (n= 3); P value of WYQ vs GTQ.
AC C
*
SC
P value§
RI PT
Glucoamylase activity
28
ACCEPTED MANUSCRIPT Table 2. Identification and relative contents of volatile compounds in the ten Hong Qu samples. No.
WYQ, relative content (%) b
Compounds a
GTQ, relative content (%)b
WYQ1
WYQ2
WYQ3
WYQ4
WYQ5
GTQ1
GTQ2
GTQ3
GTQ4
GTQ5
Alcohols Ethanol
3.10
4.61
3.94
2.56
3.89
3.70
3.21
2.72
0.75
4.01
2
Isopentyl alcohol
0.16
0.56
0.15
0.43
0.13
0.13
0.10
nd
nd
0.18
3
Hexyl alcohol
1.10
2.21
3.51
0.88
0.89
0.43
0.38
0.52
0.39
0.66
4
Gentanol
0.21
0.39
0.52
0.13
0.35
0.13
nd
0.17
0.29
0.12
5
3-Octenol
nd
0.29
0.48
0.12
nd
nd
nd
nd
nd
nd
6
Benzyl alcohol
0.13
0.20
0.30
0.09
nd
nd
nd
nd
nd
nd
7
3,5-Octadien-2-ol
nd
nd
nd
nd
nd
nd
nd
nd
0.10
nd
8
Linalool
0.39
0.44
0.42
0.19
0.39
0.33
0.31
0.26
0.35
0.25
9
Phenylethyl Alcohol
0.49
0.75
0.60
0.46
0.46
0.52
0.32
0.99
2.06
0.76
Cyclohexanol, 1-methyl-
nd
nd
nd
nd
nd
11
Acetic acid
nd
nd
0.60
0.43
0.65
12
Propanoic acid, 2-methyl-
1.77
nd
13
Butanoic acid
18.21
9.50
14
Hexanoic acid
2.27
1.68
15
Octanoic acid
1.12
2.82
0.23
nd
1.00
0.06
nd
nd
nd
nd
0.54
0.14
0.43
0.18
nd
0.12
nd
0.04
1.85
nd
0.96
0.13
8.46
10.57
8.18
9.38
20.69
4.66
10.06
18.79
0.93
1.46
1.29
0.80
2.65
1.16
nd
2.89
0.95
1.38
1.78
1.46
1.76
2.08
2.00
1.35
nd
26.26
10.58
nd
0.06
nd
nd
nd
1.00
1.02
0.86
0.95
1.57
0.93
0.99
0.67
0.93
17.67
17.75
18.80
14.19
16.54
18.97
12.74
20.68
16.22
17.89
43.83
46.34
49.87
35.85
42.24
47.25
38.39
51.77
46.35
43.69
0.09
nd
nd
nd
0.06
0.06
0.05
nd
nd
nd
0.92
0.87
1.02
0.50
1.19
2.00
1.15
1.36
nd
0.92
nd
nd
nd
nd
0.08
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
0.05
nd
nd
nd
0.75
1.09
0.80
0.44
0.67
0.55
0.46
0.50
0.78
0.80
0.58
0.95
0.91
0.47
0.69
0.61
0.57
0.70
0.66
0.50
4-Hexyl-2,5-dioxofuran-3-ac 16
0.15
M AN U
Acids
SC
10
RI PT
1
etic acid Esters Ethyl Acetate
18
Ethyl isobutyrate
19
Ethyl butyrate
20
Propyl butyrate
21
Ethyl caproate
22
Isopentyl isobutyrate
23
Isoamyl butyrate
24
Nonanal
25
Ethyl benzoate
26
Ethyl caprylate
0.81
0.85
0.90
0.41
1.72
nd
2.25
0.68
0.77
nd
27
Ethyl nonanoate
nd
nd
nd
nd
nd
0.06
0.06
nd
nd
nd
28
Ethyl caprate
0.09
0.15
0.12
0.06
0.18
0.19
0.30
0.13
0.17
0.23
29
Ethyl Oleate
nd
0.10
nd
nd
nd
0.07
nd
0.12
0.15
nd
AC C
EP
TE D
17
Aldehydes and Ketones 30
Pentanal
nd
nd
nd
nd
nd
0.42
nd
nd
0.47
nd
31
3-Penten-2-one, (E)-
nd
nd
nd
nd
nd
1.28
1.43
1.33
1.45
nd
32
Hexanal
nd
nd
nd
nd
nd
2.62
nd
1.66
4.74
nd
33
Furfural
nd
nd
nd
nd
nd
nd
nd
nd
0.96
nd
34
2-Heptanone
0.73
0.89
0.75
nd
1.11
1.11
1.38
1.60
1.30
0.69
35
Heptanal
0.53
0.66
0.37
0.23
0.56
0.81
0.64
0.79
1.13
0.54
36
Benzaldehyde
0.32
0.41
0.23
0.22
0.35
0.56
0.34
1.08
0.84
0.35
29
ACCEPTED MANUSCRIPT Octanal
nd
0.49
0.38
nd
0.62
0.54
nd
nd
nd
nd
38
3-Octen-2-one
nd
nd
nd
nd
nd
0.10
nd
nd
nd
nd
39
Benzeneacetaldehyde
0.12
nd
nd
0.20
nd
nd
nd
nd
0.16
nd
40
2-Tridecenal, (E)-
nd
nd
nd
nd
nd
nd
nd
nd
0.11
nd
41
2-Nonanone
1.02
1.12
0.99
0.35
2.55
0.95
2.81
0.98
1.75
0.91
42
Decanal
0.13
0.31
0.20
0.10
0.12
0.16
0.08
0.16
0.16
0.32
43
2-Undecanone
nd
0.09
0.10
nd
0.10
nd
0.10
nd
nd
nd
44
Ethylbenzene
0.45
0.42
0.39
0.13
0.45
0.48
0.44
nd
0.15
0.43
45
o-Xylene
0.23
nd
nd
nd
0.23
nd
nd
nd
nd
nd
46
1,3-Xylene
nd
0.18
0.20
0.10
nd
nd
0.21
0.17
nd
nd
47
p-Xylene
nd
nd
nd
nd
nd
0.24
nd
nd
nd
nd
48
Phenol
nd
nd
nd
nd
nd
0.11
nd
0.14
0.15
nd
49
2,4-Di-tert-butylphenol
0.07
0.14
nd
0.09
nd
50
Butylated Hydroxytoluene
nd
0.07
0.06
nd
SC
Aromatic compounds
RI PT
37
0.16
0.38
0.63
Ethyl 2-(5-methyl-5-vinyltetrahydr 51 ofuran-2-yl)propan-2-yl carbonate Dehydromevalonic lactone
nd
0.95
53
trans-3-Methyl-4-octanolide
nd
nd
54
gamma-Octanoic lactone
nd
0.11
55
gamma-Nonanolactone
nd
0.18
Alkanes and cycloalkanes 56
cis-2-Methyl-3-pentene
57
Styrene 3-Ethyl-2-methyl-1,3-hexadi
58 59
Cyclooctane
0.28
0.06
0.08
0.12
nd
0.12
nd
0.27
nd
0.23
0.06
0.30
0.17
0.16
0.91
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
0.18
nd
nd
0.06
nd
0.07
0.05
nd
0.10
nd
0.15
nd
nd
nd
nd
nd
nd
nd
0.57
nd
0.38
nd
0.78
nd
nd
nd
nd
nd
0.33
0.36
0.37
0.16
nd
0.36
0.30
0.32
0.20
0.25
nd
nd
nd
nd
nd
nd
nd
nd
0.10
nd
nd
nd
nd
nd
nd
nd
0.14
0.58
EP
en
0.13
0.10
TE D
52
nd
M AN U
Lactones
nd
nd
nd
Nitrogen-containing compounds 60
Oxime-, methoxy-phenyl-_
61
2,6-Dimethyl pyrazine
62
nd
nd
0.05
0.08
nd
nd
0.08
0.82
0.09
0.31
nd
0.19
nd
0.40
0.20
0.51
0.09
0.26
Ethylpyrazine
nd
nd
nd
nd
nd
0.18
nd
nd
nd
nd
63
Pyrrole-2-carboxaldehyde
nd
0.11
0.12
nd
nd
nd
nd
nd
0.26
nd
64
2-ethenyl-6-methylpyrazine
nd
nd
nd
nd
nd
nd
nd
nd
0.27
nd
65
1,3,5-Triazine-2,4,6-triamine
nd
nd
nd
nd
nd
0.71
2.83
0.68
0.61
0.72
nd
nd
nd
nd
nd
0.11
nd
0.15
nd
nd
AC C
nd
0.20
Pyrazine, 66 3-ethyl-2,5-dimethyla b
Identified by comparison with mass spectra; Relative content, percent normalized peak areas.
30
ACCEPTED MANUSCRIPT Table 3 Relative abundance of the ten most abundant bacteria and fungi in the ten Hong Qu samples*.
WYQ, relative abundance (%)
P§
2
3
4
5
1
2
3
4
5
10.18
3.03
5.08
11.33
10.82
34.80
85.21
69.91
95.01
73.16
0.000
Bacillus
40.39
57.62
51.47
32.49
41.66
0.33
0.65
0.41
0.16
0.54
0.000
Erwinia
14.04
1.28
1.93
4.42
5.81
50.35
1.90
21.20
0.61
14.77
0.224
Klebsiella
8.47
1.79
1.39
14.52
9.52
8.43
1.06
0.79
0.29
1.12
0.139
Ochrobactrum
0.50
2.72
6.27
0.40
0.28
0.05
0.13
0.05
0.01
0.05
0.125
Staphylococcus
0.92
4.37
1.77
2.37
0.10
0.05
0.15
0.12
0.03
0.07
0.036
Shewanella
0.29
2.38
1.58
0.31
0.46
0.08
0.63
0.15
0.05
0.14
0.103
Agrobacterium
0.71
1.45
1.44
0.33
0.74
0.06
0.09
0.10
0.03
0.09
0.005
Acinetobacter
0.09
0.37
0.16
3.11
0.41
0.07
0.09
0.09
0.02
0.07
0.221
Lactococcus
0.28
0.28
0.45
0.41
0.28
Others
24.12
24.70
28.47
30.31
29.92
1
2
3
4
5
Monascus purpureus
68.48
51.49
30.80
Saccharomyces sp
5.17
5.67
8.28
Aspergillus flavus
1.00
6.17
7.45
Aspergillus niger
1.25
1.35
1.11
Eurotiomycetes sp
0.41
0.92
1.07
Fusarium pseudensiforme
0.25
1.47
1.45
Rhizopus oryzae
0.13
0.73
1.02
Rhizopus microsporus
0.07
Aspergillus sp
0.13
Agaricomycetes sp
23.12
SC 0.49
0.37
0.30
0.29
0.988
5.52
9.61
6.81
3.50
9.69
―
1
2
3
4
5
56.26
71.83
99.09
97.97
98.86
97.03
98.12
0.000
5.52
3.18
0.07
0.11
0.12
0.13
0.20
0.000
1.73
2.30
0.03
0.03
0.04
0.13
0.09
0.022
2.17
0.01
0
0
0
0
0
0.009
0.38
0.44
0.17
0.19
0.13
0.20
0.26
0.015
0
0.05
0
0
0
0.02
0
0.094
0.97
0.31
0
0
0
0
0
0.007
0.12
0.40
0.80
0.12
0.03
0.10
0.06
0.07
0.10
0.130
0.28
0.27
0.18
0.11
0.02
0.02
0.03
0.04
0.22
0.040
0
0
0
0
0
1.12
0
0
0
0.347
31.80
48.15
31.99
21.63
0.59
0.46
0.76
2.38
1.01
―
EP
Others
0
0.26
M AN U
Fungi (species level)
RI PT
1
Burkholderia
TE D
Bacteria ( genus level)
GTQ, relative abundance (%)
AC C
* The colors range from blue to red indicates relative abundances range from low to high in the same row; § P value of WYQ vs GTQ
31
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1
32
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 2
33
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 3
34
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 4
35
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 5
36
ACCEPTED MANUSCRIPT Highlights A total of 66 volatiles, 215 bacteria and 151 fungi were detected from WYQ and GTQ.
RI PT
The predominant bacterial genera in Hong Qu are Burkholderia and Bacillus. Monascus purpureus is the most abundant fungal species in Hong Qu.
14 bacterial taxa and 10 fungal taxa can be utilized to distinguish WYQ and
SC
GTQ.
M AN U
The correlation of volatile components and microbiota were preliminarily
AC C
EP
TE D
explored.