Accepted Manuscript Title: Characterization of rhizosphere and endophytic bacterial communities from leaves, stems and roots of medicinal Stellera chamaejasme L Author: Hui Jin Xiao-Yan Yang Zhi-Qiang Yan Quan Liu Xiu-Zhuang Li Ji-Xiang Chen Deng-Hong Zhang Li-Ming Zeng Bo Qin PII: DOI: Reference:
S0723-2020(14)00074-5 http://dx.doi.org/doi:10.1016/j.syapm.2014.05.001 SYAPM 25620
To appear in: Received date: Revised date: Accepted date:
26-1-2014 30-4-2014 2-5-2014
Please cite this article as: H. Jin, X.-Y. Yang, Z.-Q. Yan, Q. Liu, X.Z. Li, J.-X. Chen, D.-H. Zhang, L.-M. Zeng, B. Qin, Characterization of rhizosphere and endophytic bacterial communities from leaves, stems and roots of medicinal Stellera chamaejasme L, Systematic and Applied Microbiology (2014), http://dx.doi.org/10.1016/j.syapm.2014.05.001 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.
1
Characterization of rhizosphere and endophytic bacterial
2
communities from leaves, stems and roots of medicinal Stellera
3
chamaejasme L.
4
Hui Jina, Xiao-Yan Yanga, Zhi-Qiang Yana, Quan Liua, Xiu-Zhuang Lia, Ji-Xiang
5
Chenb, Deng-Hong Zhangc, Li-Ming Zenga, Bo Qina, d, *
Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of
cr
a
Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China c
School of Petrochemical Engineering, Lanzhou University of Technology, Lanzhou 730050, China
us
b
Sino-U.S. Centers for Grazingland Ecosystem Sustainability, Key Laboratory of Grassland Ecosystem Education
Ministry, College of Prataculture, Gansu Agricultural University, Lanzhou 730070, China
State Key Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou 730000, China
an
d
* Corresponding author at: Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, E-mail address: [email protected] (B. Qin).
M
18 Tianshui Middle Road, Lanzhou 730000, China. Tel.: +86 931 4968372; fax: +86 931 8277088.
d
Running title: Rhizosphere and endophytic bacterial communities of S. chamaejasme
CA, correspondence analysis.
te
Abbreviations: NA, nutrient agar[CJR1]; OTUs, operational taxonomic units; PCA, principal component analysis;
Ac ce p
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
ip t
1
Page 1 of 28
2
ABSTRACT
23
A diverse array of bacteria that inhabit the rhizosphere and different plant organs play
24
a crucial role in plant health and growth. Therefore, a general understanding of these
25
bacterial communities and their diversity is necessary. Using the 16S rRNA gene
26
clone library technique, the bacterial community structure and diversity of the
27
rhizosphere and endophytic bacteria in Stellera chamaejasme compartments were
28
compared and clarified for the first time. Grouping of the sequences obtained showed
29
that members of the Proteobacteria (43.2%), Firmicutes (36.5%) and Actinobacteria
30
(14.1%) were dominant in both samples. Other groups that were consistently found,
31
albeit at lower abundance, were Bacteroidetes (2.1%), Chloroflexi (1.9%), and
32
Cyanobacteria (1.7%). The habitats (rhizosphere vs endophytes) and organs (leaf,
33
stem and root) structured the community, since the Wilcoxon signed rank test
34
indicated that more varied bacteria inhabited the rhizosphere compared to the organs
35
of the plant. In addition, correspondence analysis also showed that differences were
36
apparent in the bacterial communities associated with these distinct habitats.
37
Moreover, principal component analysis revealed that the profiles obtained from the
38
rhizosphere and roots were similar, whereas leaf and stem samples clustered together
39
on the opposite side of the plot from the rhizosphere and roots. Taken together, these
41 42 43 44 45
cr
us
an
M
d
te
Ac ce p
40
ip t
22
results suggested that, although the communities associated with the rhizosphere and organs shared some bacterial species, the associated communities differed in structure and diversity.
Keywords:
Plant-associated bacteria
46
Bacterial community structure
47
Stellera chamaejasme L.
48
16S rRNA gene
49
Biodiversity
50
Page 2 of 28
3
50
Introduction Plants harbor a diverse array of bacteria that inhabit different plant organs and
52
tissues, including the rhizosphere, leaves, stems, roots, seeds and fruits [38]. The
53
bacteria in the rhizosphere benefit plant growth by increasing the availability of
54
mineral nutrients, production of phytohormones, degradation of phytotoxic
55
compounds and suppression of pathogens [29,41]. Endophytic bacteria are able to
56
contribute to plant growth and development directly or indirectly through biological
57
control, plant growth-promoting effects, endophytic nitrogen-fixing activity, and other
58
actions [4,49]. Consequently, plant-bacterial associations have gained more and more
59
research attention in recent years for their potential biotechnological applications
60
[9,35]. Thus, it is crucial to understand the community structure and diversity of
61
plant-associated bacteria and their relationships in rhizosphere soil and plant organs.
an
us
cr
ip t
51
Most studies on the structure of the rhizobacteria and plant-associated bacterial
63
communities in different species have been performed using both culture-dependent
64
and
65
biodiversity studies of the rhizosphere and endophytic bacterial community are
66
somewhat limited. The major problem is that only a minor fraction of the bacterial
67
populations can be recovered [34]. Culture-independent methods, based on analysis of
68
DNA extracted directly from rhizosphere and plant samples, represents an alternative
69
that overcomes these limitations. Moreover, cultivation-independent methods allow a
71 72 73 74 75 76
[1,13,51].
However,
culture-dependent
te
d
approaches
Ac ce p
70
culture-independent
M
62
comparatively rapid analysis of a large number of samples and provide reliable information on the community composition of the rhizosphere and endophytic bacteria [5,42]. Although most of the notable studies on rhizosphere and endophytic bacteria have been focused on crops, horticultural plant species and wild grasses [5,6,8,16], it is still important to explore other plant species for their associated organisms, particularly those of important toxic plants that have the advantage of being able to produce medicines and pesticides [22].
77
Stellera chamaejasme L. (Thymelaeaceae), a toxic perennial plant, has a wide
78
geographical range from southern Russia to northern China and Mongolia, southwards
79
as far as the dry regions of the western Himalayas, the Tibetan Plateau and
80
southwestern China [58]. It has become an important pharmacological plant resource,
81
and has been used as a raw material for developing various kinds of pesticides and
Page 3 of 28
4
medicines. Most studies have demonstrated that extracts of S. chamaejasme show
83
significant nematicidal, pesticidal, antimicrobial, phytotoxic, anticancer, antihepatitis
84
B virus and immunomodulating activities [10,31,55,57]. Previous ecological studies
85
on this toxic plant have been focused on population distribution and dynamics, its
86
effect on soil nutrient pools and dynamics, a phylogenetic and phylogeographical
87
study, reproductive biology, and endophytic fungi [19,45,58,59]. However, detailed
88
knowledge concerning the bacterial community associated with S. chamaejasme is
89
still limited.
cr
ip t
82
Therefore, this study addressed two fundamental questions: (1) How does the
91
bacterial community structure differ between the rhizosphere and different organs of S.
92
chamaejasme? and (2) Do different habitats, such as rhizosphere and different tissues,
93
have the potential to influence the structure of bacterial communities? In order to
94
address these questions, the bacterial community structure and diversity in the
95
rhizosphere, as well as in leaves, stems and roots of S. chamaejasme were analyzed
96
using the 16S rRNA gene clone library technique. The results provided insights into
97
the relationships between the rhizosphere and endophytic bacteria in different organs
98
of the plant.
99
Materials and methods
101 102 103 104 105 106 107
an
M
d
te
Sample collection and surface sterilization
Ac ce p
100
us
90
Samples were collected from the Cuiying mountain area (35°56′ N, 104°08′ E) in
the Yuzhong campus of Lanzhou University on June 8, 2012. Some basic properties of the sampling site were given in the reference of Jin et al. [19]. The sampling site was not privately-owned or protected in any way. It was easy to identify the plant species since S. chamaejasme was in bloom at the sampling time, and 10 healthy looking plants were collected using sterile spades and gloves. The spades were sterilized between different individual plants. The distance between single plants was at least 50
108
m depending on the distribution within the vegetation cover. All whole plants were
109
immediately put in sterile polystyrene bags and brought to the laboratory on ice.
110
The roots of S. chamaejasme were shaken vigorously to separate them from loose
111
soil, and the remaining soil closely adhering to the roots (up to 2.5 mm around the
112
root) was pooled and considered as rhizosphere soil [24]. One gram of rhizosphere
113
soil was weighed from each plant and then pooled as a composite rhizosphere sample
Page 4 of 28
5
in sterile laboratory conditions. The composite leaf, stem and root samples were
115
processed in a similar way. Healthy, symptomless leaves, stems and roots were
116
collected from ten individual S. chamaejasme plants. One gram of leaves, stems and
117
roots was weighed from each plant and pooled to make one sample, respectively.
118
Samples were numbered Rs for rhizosphere, L for leaf, St for stem, and R for root.
ip t
114
The leaf, stem, and root samples were rinsed under running tap water for 10 min,
120
immersed in 70% ethanol with shaking for 3 min, followed by fresh sodium
121
hypochlorite solution (2.5% available Cl-) for 5 min and 70% ethanol for 30 s, and
122
finally washed three times with sterile water [46]. Aliquots of the final rinsing water
123
were spread on nutrient agar (NA)
medium plates and cultured for 3 d at 28
us
[CJR2]solid
cr
119
for detection of bacterial colonies in order to examine the effect of the surface
125
sterilization. The samples that were not contaminated, based on the culture-dependent
126
sterility test, were cut into 0.5-1.0 cm pieces and stored at -80
127
DNA-based analyses.
128
DNA extraction and PCR amplification of the bacterial 16S rRNA gene
for further
M
an
124
Total community DNA was extracted from 0.5 g of rhizosphere soil using the ZR
130
Soil Microbe DNA KitTM (Zymo Research, Orange, CA, USA), following the
131
manufacturer’s instructions. For the endophytic fraction, each sample (0.5 g) was
132
frozen with liquid nitrogen, and quickly ground into a fine powder in a pre-cooled
133
sterilized mortar and pestle. Each sample was added to the bead tubes from the kit
135 136 137 138 139 140
te
Ac ce p
134
d
129
following the manufacturer’s instructions. DNA preparations were visualized after electrophoresis in 0.8% agarose gels stained with ethidium bromide under UV light. DNA was further cleaned up by ethanol precipitation and then resuspended in 30 µL sterile Milli-Q water.
Bacterial 16S rRNA genes were PCR amplified using bacteria-specific primers
799F (5′ -AAC AGG ATT AGA TAC CCT G- 3′) and 1492R (5′ -GGT TAC CTT GTT ACG ACT T -3′) that can separate bacterial products and plant chloroplast
141
products [6,37]. Each 50 µL PCR contained PCR buffer (Promega, Madison, WI), 50
142
ng of DNA template, 2.5 mM MgCl2, 200 µM of each dNTP, 0.5 mg mL-1 bovine
143
serum albumin (BSA), 15 pmol of each primer, and 2.5 units Taq polymerase. Cycling
144
conditions were 94
145
30 sec, 72
146
product was isolated by separating the PCR products on 2% low melt agarose gel and
for 2 min, followed by 25 cycles of 94
for 1 min, and with a final extension of 72
for 30 sec, 55
for
for 10 min. The bacterial
Page 5 of 28
6
excising a band of agarose with a size of approximately 740 bp. DNA was extracted
148
from the gel using the QIAquick gel extraction kit (Qiagen). The concentrations of the
149
purified DNA were estimated by comparison with an EZ Load DNA mass standard
150
(Bio-Rad).
151
Construction of the 16S rRNA gene clone library
The purified PCR-amplified 16S rRNA gene fragments were cloned into the
153
pGEM-T vector (Promega, Madison, WI, USA) according to the manufacturer’s
154
instructions. Escherichia coli JM109 competent cells (Promega) were transformed
155
with the ligation products and spread onto Luria-Bertani (LB) agar plates containing
156
ampicillin (100 µg mL-1) and X-gal/IPTG on the surface for standard blue and white
157
screening. Approximately 120 positive clones per sample were randomly selected and
158
cultured for 24 h at 37
159
ampicillin. Plasmid DNA extraction was performed using the QIAprep Spin Miniprep
160
Kit (Qiagen) according to the manufacturer’s instructions, and screened by the PCR
161
program (described above) using pGEM-T vector-specific primers T7 and SP6. The
162
isolated DNAs with correct inserts were confirmed on 0.8% agarose gels and purified
163
using the DNA Fragment Quick Purification/Recover Kit (Dingguo, Beijing, China)
164
following the protocol provided. Purified PCR products were sequenced with an ABI
165
3730 automated sequencer (Invitrogen, Shanghai, China).
166 167 168 169 170 171 172
us
cr
152
Ac ce p
ip t
147
te
d
M
an
in 1.5 mL of LB broth supplemented with 100 µg mL-1
Sequence analyses and phylogenetic tree construction All the 16S rRNA gene sequences obtained in this study were edited manually using
BioEdit, then aligned with ClustalW and grouped together based on sequence similarity. All assembled sequences were checked for possible chimeric artefacts by the UCHIME algorithm [12]. Potential and identified chimeric sequences were excluded from subsequent analyses. The remaining sequences within each library were grouped into operational taxonomic units (OTUs) based on 3% sequence
173
divergence using the furthest-neighbor algorithm in MOTHUR software [39]. The
174
get.oturep command in Mothur was used to select representative sequences and they
175
were used for all subsequent analyses [39,44]. The taxonomic affiliation of each
176
representative sequence was determined by sequence similarity searches against the
177
EzTaxon-e database (http://eztaxon-e.ezbiocloud.net/) [23]. Sequences with >97%
178
similarity were assigned to the same species. To construct phylogenetic trees, the
Page 6 of 28
7
sequences with 97% or higher sequence similarities were aligned with the
180
corresponding EzTaxon-e database sequences (93 type strains with greater than 97%
181
similarity) from representative members of selected families using the ClustalW
182
program. Neighbor-joining phylogenetic trees were constructed from dissimilar
183
distance and pairwise comparisons with the Jukes-Cantor distance model using
184
MEGA version 5.0 [48]. Bootstrapping with 1,000 replicates was used to assign
185
confidence levels to the nodes in the trees.
186
Statistical analysis
cr
ip t
179
The MOTHUR software was applied in order to define OTU rarefaction curves.
188
Species richness and diversity indices (Chao 1, non-parametric Shannon’s diversity,
189
Simpson diversity index, Shannon’s evenness, and Good’s coverage) were also
190
determined by the MOTHUR program at the 0.03 level [39]. The Wilcoxon signed
191
rank test was used to test for the significance of differences in the composition of
192
bacterial communities from the different habitats [20]. A correspondence analysis
193
(CA) was carried out in order to explore patterns of variation in the composition of the
194
bacterial community across samples [14]. Principal component analysis (PCA) was
195
conducted for evaluating the relationship between bacterial community structures.
196
The matrix used for the PCA was a quantitative abundance matrix of all OTUs
197
detected from each clone library [25]. The statistical significance for PCA was
198
assessed by analysis of variance of PCA scores between different habitats performed
200 201 202 203 204
an
M
d
te
Ac ce p
199
us
187
using SPSS software (version 16.01; SPSS). Nucleotide sequence accession numbers The representative 16S rRNA gene sequences generated by this study were
submitted to GenBank under the accession numbers KF385119-KF385192 (rhizosphere samples), KF385038-KF385088 (leaf samples), KF385193-KF385237 (stem samples) and KF385089-KF385118 (root samples) (Table S1).
205
Results and discussion
206
Analysis of sequencing data
207
The UCHIME algorithm was used to identify sequences with a high probability
208
of being chimeric and, as a result, 12 chimeric sequences were discarded. A total of
Page 7 of 28
8
468 sequences were obtained (115 from the rhizosphere library, 116 from the leaf
210
library, 118 from the stem library and 119 from the root library). Sequences were
211
clustered into OTUs based on 3% sequence divergence, which is widely used to
212
approximate species-level similarity. It was suggested that the plant-associated
213
bacterial community was reasonably well characterized by the sampling effort, since
214
the slope of a rarefaction curve was high in the beginning and then gradually
215
decreased, causing the curve to level off (Fig. 1). Interestingly, the rarefaction curves
216
of the leaf and stem samples were higher than the root samples, while the rhizosphere
217
sample was higher than the leaf and stem samples.
218
Phylogenetic profiles and taxonomic distribution of the 16S rRNA gene clones among
219
the samples
us
cr
ip t
209
There were 200 OTUs for all 468 sequences, which were comprised of 74, 51, 45
221
and 30 OTUs for the rhizosphere, leaf, stem and root samples, respectively. More than
222
80% of the OTUs (164) were composed of a single sequence. However, there were
223
five OTUs which contained 20 or more sequences, with OTUl04, OTUst31, OTUr04,
224
OTUrs20 and OTUr08 containing 56, 47, 34, 26 and 20 sequences, respectively,
225
indicating unevenness of OTU distribution in the bacterial communities (Tables 1 and
226
S1). All the 200 representative sequences clustered into eight groups (phyla or classes)
227
according to the taxonomic classification of the EzTaxon-e database (Table S1). These
228
bacterial
230 231 232 233 234 235
M
d
te
groups
were
Proteobacteria
Ac ce p
229
an
220
(43.2%,
including
17.9%
for
the
gamma-subdivision, 16.7% for the alpha-subdivision, 7.5% for the beta-subdivision, and 1.1% for the delta-subdivision), Firmicutes (36.5%), Actinobacteria (14.1%), Bacteroidetes (2.1%), Chloroflexi (1.9%), Cyanobacteria (1.7%), Acidobacteria (0.2%), and Verrucomicrobia (0.2%). These results generally demonstrated high bacterial diversity across the rhizosphere, leaves, stems and roots of S. chamaejasme. The bacterial taxa observed were similar to the findings from other studies that used culture-independent techniques to describe taxa abundances [5,11,50,52].
236
Among all representative sequences, most of them showed 99-100% sequence
237
similarity, whereas a small number of them showed >97 to 99% sequence similarity.
238
More than 97% sequence similarity is accepted for bacterial identification, and
S0723-2020(14)00074-5 http://dx.doi.org/doi:10.1016/j.syapm.2014.05.001 SYAPM 25620
To appear in: Received date: Revised date: Accepted date:
26-1-2014 30-4-2014 2-5-2014
Please cite this article as: H. Jin, X.-Y. Yang, Z.-Q. Yan, Q. Liu, X.Z. Li, J.-X. Chen, D.-H. Zhang, L.-M. Zeng, B. Qin, Characterization of rhizosphere and endophytic bacterial communities from leaves, stems and roots of medicinal Stellera chamaejasme L, Systematic and Applied Microbiology (2014), http://dx.doi.org/10.1016/j.syapm.2014.05.001 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.
1
Characterization of rhizosphere and endophytic bacterial
2
communities from leaves, stems and roots of medicinal Stellera
3
chamaejasme L.
4
Hui Jina, Xiao-Yan Yanga, Zhi-Qiang Yana, Quan Liua, Xiu-Zhuang Lia, Ji-Xiang
5
Chenb, Deng-Hong Zhangc, Li-Ming Zenga, Bo Qina, d, *
Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of
cr
a
Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China c
School of Petrochemical Engineering, Lanzhou University of Technology, Lanzhou 730050, China
us
b
Sino-U.S. Centers for Grazingland Ecosystem Sustainability, Key Laboratory of Grassland Ecosystem Education
Ministry, College of Prataculture, Gansu Agricultural University, Lanzhou 730070, China
State Key Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou 730000, China
an
d
* Corresponding author at: Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, E-mail address: [email protected] (B. Qin).
M
18 Tianshui Middle Road, Lanzhou 730000, China. Tel.: +86 931 4968372; fax: +86 931 8277088.
d
Running title: Rhizosphere and endophytic bacterial communities of S. chamaejasme
CA, correspondence analysis.
te
Abbreviations: NA, nutrient agar[CJR1]; OTUs, operational taxonomic units; PCA, principal component analysis;
Ac ce p
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
ip t
1
Page 1 of 28
2
ABSTRACT
23
A diverse array of bacteria that inhabit the rhizosphere and different plant organs play
24
a crucial role in plant health and growth. Therefore, a general understanding of these
25
bacterial communities and their diversity is necessary. Using the 16S rRNA gene
26
clone library technique, the bacterial community structure and diversity of the
27
rhizosphere and endophytic bacteria in Stellera chamaejasme compartments were
28
compared and clarified for the first time. Grouping of the sequences obtained showed
29
that members of the Proteobacteria (43.2%), Firmicutes (36.5%) and Actinobacteria
30
(14.1%) were dominant in both samples. Other groups that were consistently found,
31
albeit at lower abundance, were Bacteroidetes (2.1%), Chloroflexi (1.9%), and
32
Cyanobacteria (1.7%). The habitats (rhizosphere vs endophytes) and organs (leaf,
33
stem and root) structured the community, since the Wilcoxon signed rank test
34
indicated that more varied bacteria inhabited the rhizosphere compared to the organs
35
of the plant. In addition, correspondence analysis also showed that differences were
36
apparent in the bacterial communities associated with these distinct habitats.
37
Moreover, principal component analysis revealed that the profiles obtained from the
38
rhizosphere and roots were similar, whereas leaf and stem samples clustered together
39
on the opposite side of the plot from the rhizosphere and roots. Taken together, these
41 42 43 44 45
cr
us
an
M
d
te
Ac ce p
40
ip t
22
results suggested that, although the communities associated with the rhizosphere and organs shared some bacterial species, the associated communities differed in structure and diversity.
Keywords:
Plant-associated bacteria
46
Bacterial community structure
47
Stellera chamaejasme L.
48
16S rRNA gene
49
Biodiversity
50
Page 2 of 28
3
50
Introduction Plants harbor a diverse array of bacteria that inhabit different plant organs and
52
tissues, including the rhizosphere, leaves, stems, roots, seeds and fruits [38]. The
53
bacteria in the rhizosphere benefit plant growth by increasing the availability of
54
mineral nutrients, production of phytohormones, degradation of phytotoxic
55
compounds and suppression of pathogens [29,41]. Endophytic bacteria are able to
56
contribute to plant growth and development directly or indirectly through biological
57
control, plant growth-promoting effects, endophytic nitrogen-fixing activity, and other
58
actions [4,49]. Consequently, plant-bacterial associations have gained more and more
59
research attention in recent years for their potential biotechnological applications
60
[9,35]. Thus, it is crucial to understand the community structure and diversity of
61
plant-associated bacteria and their relationships in rhizosphere soil and plant organs.
an
us
cr
ip t
51
Most studies on the structure of the rhizobacteria and plant-associated bacterial
63
communities in different species have been performed using both culture-dependent
64
and
65
biodiversity studies of the rhizosphere and endophytic bacterial community are
66
somewhat limited. The major problem is that only a minor fraction of the bacterial
67
populations can be recovered [34]. Culture-independent methods, based on analysis of
68
DNA extracted directly from rhizosphere and plant samples, represents an alternative
69
that overcomes these limitations. Moreover, cultivation-independent methods allow a
71 72 73 74 75 76
[1,13,51].
However,
culture-dependent
te
d
approaches
Ac ce p
70
culture-independent
M
62
comparatively rapid analysis of a large number of samples and provide reliable information on the community composition of the rhizosphere and endophytic bacteria [5,42]. Although most of the notable studies on rhizosphere and endophytic bacteria have been focused on crops, horticultural plant species and wild grasses [5,6,8,16], it is still important to explore other plant species for their associated organisms, particularly those of important toxic plants that have the advantage of being able to produce medicines and pesticides [22].
77
Stellera chamaejasme L. (Thymelaeaceae), a toxic perennial plant, has a wide
78
geographical range from southern Russia to northern China and Mongolia, southwards
79
as far as the dry regions of the western Himalayas, the Tibetan Plateau and
80
southwestern China [58]. It has become an important pharmacological plant resource,
81
and has been used as a raw material for developing various kinds of pesticides and
Page 3 of 28
4
medicines. Most studies have demonstrated that extracts of S. chamaejasme show
83
significant nematicidal, pesticidal, antimicrobial, phytotoxic, anticancer, antihepatitis
84
B virus and immunomodulating activities [10,31,55,57]. Previous ecological studies
85
on this toxic plant have been focused on population distribution and dynamics, its
86
effect on soil nutrient pools and dynamics, a phylogenetic and phylogeographical
87
study, reproductive biology, and endophytic fungi [19,45,58,59]. However, detailed
88
knowledge concerning the bacterial community associated with S. chamaejasme is
89
still limited.
cr
ip t
82
Therefore, this study addressed two fundamental questions: (1) How does the
91
bacterial community structure differ between the rhizosphere and different organs of S.
92
chamaejasme? and (2) Do different habitats, such as rhizosphere and different tissues,
93
have the potential to influence the structure of bacterial communities? In order to
94
address these questions, the bacterial community structure and diversity in the
95
rhizosphere, as well as in leaves, stems and roots of S. chamaejasme were analyzed
96
using the 16S rRNA gene clone library technique. The results provided insights into
97
the relationships between the rhizosphere and endophytic bacteria in different organs
98
of the plant.
99
Materials and methods
101 102 103 104 105 106 107
an
M
d
te
Sample collection and surface sterilization
Ac ce p
100
us
90
Samples were collected from the Cuiying mountain area (35°56′ N, 104°08′ E) in
the Yuzhong campus of Lanzhou University on June 8, 2012. Some basic properties of the sampling site were given in the reference of Jin et al. [19]. The sampling site was not privately-owned or protected in any way. It was easy to identify the plant species since S. chamaejasme was in bloom at the sampling time, and 10 healthy looking plants were collected using sterile spades and gloves. The spades were sterilized between different individual plants. The distance between single plants was at least 50
108
m depending on the distribution within the vegetation cover. All whole plants were
109
immediately put in sterile polystyrene bags and brought to the laboratory on ice.
110
The roots of S. chamaejasme were shaken vigorously to separate them from loose
111
soil, and the remaining soil closely adhering to the roots (up to 2.5 mm around the
112
root) was pooled and considered as rhizosphere soil [24]. One gram of rhizosphere
113
soil was weighed from each plant and then pooled as a composite rhizosphere sample
Page 4 of 28
5
in sterile laboratory conditions. The composite leaf, stem and root samples were
115
processed in a similar way. Healthy, symptomless leaves, stems and roots were
116
collected from ten individual S. chamaejasme plants. One gram of leaves, stems and
117
roots was weighed from each plant and pooled to make one sample, respectively.
118
Samples were numbered Rs for rhizosphere, L for leaf, St for stem, and R for root.
ip t
114
The leaf, stem, and root samples were rinsed under running tap water for 10 min,
120
immersed in 70% ethanol with shaking for 3 min, followed by fresh sodium
121
hypochlorite solution (2.5% available Cl-) for 5 min and 70% ethanol for 30 s, and
122
finally washed three times with sterile water [46]. Aliquots of the final rinsing water
123
were spread on nutrient agar (NA)
medium plates and cultured for 3 d at 28
us
[CJR2]solid
cr
119
for detection of bacterial colonies in order to examine the effect of the surface
125
sterilization. The samples that were not contaminated, based on the culture-dependent
126
sterility test, were cut into 0.5-1.0 cm pieces and stored at -80
127
DNA-based analyses.
128
DNA extraction and PCR amplification of the bacterial 16S rRNA gene
for further
M
an
124
Total community DNA was extracted from 0.5 g of rhizosphere soil using the ZR
130
Soil Microbe DNA KitTM (Zymo Research, Orange, CA, USA), following the
131
manufacturer’s instructions. For the endophytic fraction, each sample (0.5 g) was
132
frozen with liquid nitrogen, and quickly ground into a fine powder in a pre-cooled
133
sterilized mortar and pestle. Each sample was added to the bead tubes from the kit
135 136 137 138 139 140
te
Ac ce p
134
d
129
following the manufacturer’s instructions. DNA preparations were visualized after electrophoresis in 0.8% agarose gels stained with ethidium bromide under UV light. DNA was further cleaned up by ethanol precipitation and then resuspended in 30 µL sterile Milli-Q water.
Bacterial 16S rRNA genes were PCR amplified using bacteria-specific primers
799F (5′ -AAC AGG ATT AGA TAC CCT G- 3′) and 1492R (5′ -GGT TAC CTT GTT ACG ACT T -3′) that can separate bacterial products and plant chloroplast
141
products [6,37]. Each 50 µL PCR contained PCR buffer (Promega, Madison, WI), 50
142
ng of DNA template, 2.5 mM MgCl2, 200 µM of each dNTP, 0.5 mg mL-1 bovine
143
serum albumin (BSA), 15 pmol of each primer, and 2.5 units Taq polymerase. Cycling
144
conditions were 94
145
30 sec, 72
146
product was isolated by separating the PCR products on 2% low melt agarose gel and
for 2 min, followed by 25 cycles of 94
for 1 min, and with a final extension of 72
for 30 sec, 55
for
for 10 min. The bacterial
Page 5 of 28
6
excising a band of agarose with a size of approximately 740 bp. DNA was extracted
148
from the gel using the QIAquick gel extraction kit (Qiagen). The concentrations of the
149
purified DNA were estimated by comparison with an EZ Load DNA mass standard
150
(Bio-Rad).
151
Construction of the 16S rRNA gene clone library
The purified PCR-amplified 16S rRNA gene fragments were cloned into the
153
pGEM-T vector (Promega, Madison, WI, USA) according to the manufacturer’s
154
instructions. Escherichia coli JM109 competent cells (Promega) were transformed
155
with the ligation products and spread onto Luria-Bertani (LB) agar plates containing
156
ampicillin (100 µg mL-1) and X-gal/IPTG on the surface for standard blue and white
157
screening. Approximately 120 positive clones per sample were randomly selected and
158
cultured for 24 h at 37
159
ampicillin. Plasmid DNA extraction was performed using the QIAprep Spin Miniprep
160
Kit (Qiagen) according to the manufacturer’s instructions, and screened by the PCR
161
program (described above) using pGEM-T vector-specific primers T7 and SP6. The
162
isolated DNAs with correct inserts were confirmed on 0.8% agarose gels and purified
163
using the DNA Fragment Quick Purification/Recover Kit (Dingguo, Beijing, China)
164
following the protocol provided. Purified PCR products were sequenced with an ABI
165
3730 automated sequencer (Invitrogen, Shanghai, China).
166 167 168 169 170 171 172
us
cr
152
Ac ce p
ip t
147
te
d
M
an
in 1.5 mL of LB broth supplemented with 100 µg mL-1
Sequence analyses and phylogenetic tree construction All the 16S rRNA gene sequences obtained in this study were edited manually using
BioEdit, then aligned with ClustalW and grouped together based on sequence similarity. All assembled sequences were checked for possible chimeric artefacts by the UCHIME algorithm [12]. Potential and identified chimeric sequences were excluded from subsequent analyses. The remaining sequences within each library were grouped into operational taxonomic units (OTUs) based on 3% sequence
173
divergence using the furthest-neighbor algorithm in MOTHUR software [39]. The
174
get.oturep command in Mothur was used to select representative sequences and they
175
were used for all subsequent analyses [39,44]. The taxonomic affiliation of each
176
representative sequence was determined by sequence similarity searches against the
177
EzTaxon-e database (http://eztaxon-e.ezbiocloud.net/) [23]. Sequences with >97%
178
similarity were assigned to the same species. To construct phylogenetic trees, the
Page 6 of 28
7
sequences with 97% or higher sequence similarities were aligned with the
180
corresponding EzTaxon-e database sequences (93 type strains with greater than 97%
181
similarity) from representative members of selected families using the ClustalW
182
program. Neighbor-joining phylogenetic trees were constructed from dissimilar
183
distance and pairwise comparisons with the Jukes-Cantor distance model using
184
MEGA version 5.0 [48]. Bootstrapping with 1,000 replicates was used to assign
185
confidence levels to the nodes in the trees.
186
Statistical analysis
cr
ip t
179
The MOTHUR software was applied in order to define OTU rarefaction curves.
188
Species richness and diversity indices (Chao 1, non-parametric Shannon’s diversity,
189
Simpson diversity index, Shannon’s evenness, and Good’s coverage) were also
190
determined by the MOTHUR program at the 0.03 level [39]. The Wilcoxon signed
191
rank test was used to test for the significance of differences in the composition of
192
bacterial communities from the different habitats [20]. A correspondence analysis
193
(CA) was carried out in order to explore patterns of variation in the composition of the
194
bacterial community across samples [14]. Principal component analysis (PCA) was
195
conducted for evaluating the relationship between bacterial community structures.
196
The matrix used for the PCA was a quantitative abundance matrix of all OTUs
197
detected from each clone library [25]. The statistical significance for PCA was
198
assessed by analysis of variance of PCA scores between different habitats performed
200 201 202 203 204
an
M
d
te
Ac ce p
199
us
187
using SPSS software (version 16.01; SPSS). Nucleotide sequence accession numbers The representative 16S rRNA gene sequences generated by this study were
submitted to GenBank under the accession numbers KF385119-KF385192 (rhizosphere samples), KF385038-KF385088 (leaf samples), KF385193-KF385237 (stem samples) and KF385089-KF385118 (root samples) (Table S1).
205
Results and discussion
206
Analysis of sequencing data
207
The UCHIME algorithm was used to identify sequences with a high probability
208
of being chimeric and, as a result, 12 chimeric sequences were discarded. A total of
Page 7 of 28
8
468 sequences were obtained (115 from the rhizosphere library, 116 from the leaf
210
library, 118 from the stem library and 119 from the root library). Sequences were
211
clustered into OTUs based on 3% sequence divergence, which is widely used to
212
approximate species-level similarity. It was suggested that the plant-associated
213
bacterial community was reasonably well characterized by the sampling effort, since
214
the slope of a rarefaction curve was high in the beginning and then gradually
215
decreased, causing the curve to level off (Fig. 1). Interestingly, the rarefaction curves
216
of the leaf and stem samples were higher than the root samples, while the rhizosphere
217
sample was higher than the leaf and stem samples.
218
Phylogenetic profiles and taxonomic distribution of the 16S rRNA gene clones among
219
the samples
us
cr
ip t
209
There were 200 OTUs for all 468 sequences, which were comprised of 74, 51, 45
221
and 30 OTUs for the rhizosphere, leaf, stem and root samples, respectively. More than
222
80% of the OTUs (164) were composed of a single sequence. However, there were
223
five OTUs which contained 20 or more sequences, with OTUl04, OTUst31, OTUr04,
224
OTUrs20 and OTUr08 containing 56, 47, 34, 26 and 20 sequences, respectively,
225
indicating unevenness of OTU distribution in the bacterial communities (Tables 1 and
226
S1). All the 200 representative sequences clustered into eight groups (phyla or classes)
227
according to the taxonomic classification of the EzTaxon-e database (Table S1). These
228
bacterial
230 231 232 233 234 235
M
d
te
groups
were
Proteobacteria
Ac ce p
229
an
220
(43.2%,
including
17.9%
for
the
gamma-subdivision, 16.7% for the alpha-subdivision, 7.5% for the beta-subdivision, and 1.1% for the delta-subdivision), Firmicutes (36.5%), Actinobacteria (14.1%), Bacteroidetes (2.1%), Chloroflexi (1.9%), Cyanobacteria (1.7%), Acidobacteria (0.2%), and Verrucomicrobia (0.2%). These results generally demonstrated high bacterial diversity across the rhizosphere, leaves, stems and roots of S. chamaejasme. The bacterial taxa observed were similar to the findings from other studies that used culture-independent techniques to describe taxa abundances [5,11,50,52].
236
Among all representative sequences, most of them showed 99-100% sequence
237
similarity, whereas a small number of them showed >97 to 99% sequence similarity.
238
More than 97% sequence similarity is accepted for bacterial identification, and
