Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
Composition and predictive functional analysis of bacterial
communities in the surface seawater of the Changjiang
Estuary
Dong-Mei Wu 1 , Jian-Xin Wang 1 , Xiao-Hui Liu 1 , Ying-Ping Fan 1 , Ran Jiang 1 , Ming-Hua Liu 1 , Shuai-Bing
Wang 1 , Xue-Zhu Liu Corresp. 1
1 Marine Science and Technology College, Zhejiang Ocean University, Zhoushan, China
Corresponding Author: Xue-Zhu Liu
Email address: [email protected]
The objective of this study was to characterize the structure and function of microbial
communities in surface seawater from the Changjiang Estuary and adjacent areas, China.
Sample water was collected at 12 sites and environmental parameters were measured.
Community structure was analyzed using high-throughput sequencing of 16S rDNA genes.
Predictive metagenomic approach was used to predict the function of bacterial
communities. Result showed that sample site A0102 had the highest bacterial abundance
and diversity. The heatmap indicated that different samples could be clustered into six
groups. Phylogenetic analysis showed that Proteobacteria was the predominant phylum in
all samples, followed by Bacteroidetes and Actinobacteria. Alphaproteobacteria and
Gammaproteobacteria were the dominant classes. The analysis of predictive metagenomic
showed carbon fixation pathways in prokaryotes, nitrogen metabolism, carbon fixation in
photosynthetic organisms, photosynthesis and polycyclic aromatic hydrocarbon
degradation were enriched in all samples. Redundancy analysis (RDA) identified that
dissolved oxygen (DO) and PO43– concentration had positive correlations with the bacterial
communities while chemical oxygen demand (COD), dissolved oxygen (DO) and PO43–
concentration were significantly associated with microbial functional diversity. This study
adds to our knowledge of functional and taxonomic composition of microbial communities.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
1 Composition and predictive functional analysis of bacterial
2 communities in the surface seawater of the Changjiang
3 Estuary4
5
6
7
8
9 Dong-Mei Wu #, Jian-Xin Wang #, Xiao-Hui Liu, Ying-Ping Fan, Ran Jiang, Ming-Hua Liu,
10 Shuai-Bing Wang, Xue-Zhu Liu*
11 Marine Microorganism Ecological & Application Lab, Zhejiang Ocean University, 1 South
12 Haida Road, Zhejiang 316022, China.
13 #These authors contributed equally to this study and share the first authorship.
14 *Correspondence author. Tel.: +86 13575617505; fax: +86 580-8180982; E-mail:
16
17 KEY WORDS: Bacterial community structure; 16S rDNA; Environmental factors; PICRUSt
18 ABSTRACT
19 The objective of this study was to characterize the structure and function of microbial
20 communities in surface seawater from the Changjiang Estuary and adjacent areas, China. Sample
21 water was collected at 12 sites and environmental parameters were measured. Community
22 structure was analyzed using high-throughput sequencing of 16S rDNA genes. Predictive
23 metagenomic approach was used to predict the function of bacterial communities. Result showed
24 that sample site A0102 had the highest bacterial abundance and diversity. The heatmap indicated
25 that different samples could be clustered into six groups. Phylogenetic analysis showed that
26 Proteobacteria was the predominant phylum in all samples, followed by Bacteroidetes and
27 Actinobacteria. Alphaproteobacteria and Gammaproteobacteria were the dominant classes. The
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
28 analysis of predictive metagenomic showed carbon fixation pathways in prokaryotes, nitrogen
29 metabolism, carbon fixation in photosynthetic organisms, photosynthesis and polycyclic
30 aromatic hydrocarbon degradation were enriched in all samples. Redundancy analysis (RDA)
31 identified that dissolved oxygen (DO) and PO43– concentration had positive correlations with the
32 bacterial communities while chemical oxygen demand (COD), dissolved oxygen (DO) and PO43–
33 concentration were significantly associated with microbial functional diversity. This study adds
34 to our knowledge of functional and taxonomic composition of microbial communities.
35 INTRODUCTION
36 Bacterioplankton communities are an important microorganisms in the marine
37 ecosystems, which greatly affect material cycling, energy flow and the ocean food web.
38 Microorganisms in aquatic ecosystems are very sensitive to changes in environmental conditions
39 and thus bacterial community composition can act as an environmental indicator(Paerl et al.
40 2003). Determining physicochemical factors and bacterial community structures can increase our
41 understanding of microbial ecology. To unravel the functional potential of bacteria is beneficial
42 for understanding their roles in biogeochemical cycling.
43 The Changjiang Estuary, also called the Yangtze River. It located offshore from the
44 mouth of the Changjiang River.(Chen et al. 1999). Because of the mixture of Changjiang Dulited
45 Water with the Taiwan Warm Current (TWC), this region is extremely complicated and
46 dynamic(Jiao et al. 2007; Zhang et al. 1999). Many studies on bacterial diversity in the
47 Changjiang Estuary have focused on ammonia-oxidizing bacteria (AOB), For example,
48 molecular biological techniques were used to analyze the community structure and diversity of
49 AOB in Changjiang Estuary sediments and adjacent waters in the East China Sea (Chen et al.
50 2014). Liu reported a relationship between bacterial abundance and concentrations of phosphate
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
51 in the Changjiang River (Liu et al. 2009). Sala, Vieira and colleagues studied the spatial
52 distribution of bacterial communities (Sala et al. 2008; Vieira et al. 2008). However, there is
53 relatively little information on the diversity and abundance of the whole bacterial population in
54 surface seawater of the Changjiang Estuary and adjacent areas. A growing number of studies
55 have focused on the functional potential of bacteria in marine sediments (Graves et al. 2016;
56 Kirchman et al. 2015). Information on the function of microbial communities in the surface
57 seawater are poorly understood.
58 In this study, we used high-throughput sequencing technology targeted to 16S rDNA
59 genes to analyze bacterial diversity and used predicted metagenomic analysis to compare
60 microbial functions. Meanwhile, we estimated the bacterial biomass based on 4,6-diamidino-2-
61 phenylindole (DAPI) fluorescence direct counts, and used clustering analyses to assess the
62 correlations between bacterial community structure, functions and environmental factors. Our
63 research is of benefit for understanding the bacterial abundance, diversity , functions and
64 distribution in the Changjiang Estuary and adjacent waters.
65
66 MATERIALS AND METHODS
67 Sampling areas and sampling
68 In July 2015, samples were collected using an SBE 32 sampler (Sea-Bird Electronics,
69 Washington, USA) at 2 m depth in surface seawater from 12 sites in the Changjiang Estuary and
70 adjacent areas (Figure 1). Table 1 lists the latitude and longitude of the sampling sites. Samples
71 for DAPI fluorescence examination were collected in 10-mL sterile cryopreservation tubes,
72 preserved with buffered glutaraldehyde (final concentration 1%), and stored in the dark at
73 ambient temperature for 15 min. Subsequently the samples were stored in airtight plastic bottles
74 at −20°C for the duration of the cruise, and at −80°C after returning to the laboratory. Samples
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
75 for DNA isolation were collected using a vacuum pump suction filter with a vacuum of 20 kPa
76 (Porter & Feig 1980). One liter samples were filtered first through 3-μm then 0.22-μm pore size
77 polycarbonate nucleopore membranes (Merck Millipore Ltd., USA). The 0.22-μm filters were
78 preserved in 5-ml sterile cryopreservation tubes at −20°C during the cruise and at −80°C after
79 returning to the laboratory.
80 Environmental parameters
81 Samples were pretreated according to specifications for marine monitoring (National
82 Standards of People’s Republic of China, GB 17378.5, 2007) before chemical parameter analysis
83 (Heijs et al. 2008). PO43– was measured using a QuAAtro continuous flow analyzer (SEAL
84 Analytical, Hamburg, Germany). Dissolved oxygen (DO), NO2− and NH4
+ levels were measured
85 using a spectrophotometer (752 UV/visible spectrophotometer, Shanghai-Hengping, China).
86 Chemical oxygen demand (COD) was measured using the alkaline potassium permanganate.
87 DAPI fluorescence direct counts
88 Seawater samples were stained with DAPI by adding 1 mL of pretreated sample to 1 mL
89 of 20 μg/mL DAPI. Staining was performed in a darkened room over 30 min with occasional
90 swirling of the reaction tube. The mixed solution was then passed through pre-wetted black
91 polycarbonate nucleopore membrane filters (pore size 0.2 μm) using a 5-mL syringe fitted with a
92 needle. The membrane was observed under a fluorescence microscope using a flat-field 100× oil
93 immersion lens, and a minimum of 30 cells per filter were counted in a minimum of 20 fields of
94 view. Bacterial density in the original sample was calculated by using the formula (cell/mL) = (N
95 × At)/(Ag × Vf), where N is the number of cells counted, At is the effective area of the filter (in
96 mm2 or μm2), Ag is the area of the counting grid (in mm2 or μm2), and Vf is the volume of
97 diluted sample filtered (in mL)(Jr & Pratt 1994).
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
98 DNA extraction and PCR amplification
99 Bacterial 16S rRNA genes were analyzed to determine the bacterial community structure
100 and diversity. Genomic DNA was extracted using the FastPrep®-24 rapid nucleic acid extraction
101 kit (MP Biomedicals, USA) following the manufacturer’s instructions. Phylogenetically
102 diagnostic sequences were amplified using the bacterial 16S rRNA universal primers 515F (5ʹ-
103 GTGCCAGCMGCCGCGG-3ʹ) and 907R (5ʹ-CCGTCAATTCMTTTRAGTTT-3ʹ). Amplified
104 DNA was verified by electrophoresis of PCR mixtures in 1.2 % agarose in 1X TAE buffer and
105 purified using a protein nucleic acid detector (Bio-RAD,USAA).
106 Sequencing and phylogenetic analysis
107 Samples were sent for sequencing on a Miseq platform. Sequencing data were cleaned
108 using the software package Qiime and then clustered to operational taxonomic units (OTUs) with
109 a complete linkage algorithm at a 97% sequence identity level. Abundance-based coverage
110 estimators, observed_otus , the Chao1, Shannon, and Simpson parameters were estimated for
111 alpha diversity analysis. A rarefaction curve was also analyzed using QIIME software
112 (http://qiime.org/scripts). Taxa with proportions <0.01% were grouped as “others”. With the
113 VEGAN package in the integrated suite of software facilities R (Oksanen et al. 2009),
114 redundancy analysis (RDA)was used to examine the correlations between community variations
115 ,community functions and environmental parameters. The heatmap.2 program within the gplots
116 package was used to paint the heat map. PICRUSt, a bioinformatics tool designed to address the
117 functional potential in different sites using 16S ribosomal DNA sequences(Langille et al. 2013).
118 For this analysis, the closed-reference OTU picking protocol was performed using
119 QIIME1.9.0(Caporaso et al. 2010). Sequences are aligned with the Greengenes database (vers.
120 13.5)(Desantis et al. 2006). The OTU table was created after rarefying samples to 29652
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
121 sequences and the gene copies were normalized, then using the PICRUSt for further analysis.
122 This normalization helped us avoid overestimation of some groups of
123 microorganisms(Urbanová et al., 2014). For example, without normalization the estimate of the
124 relative abundance of Proteobacteria can be up to twice as high. It performed functional analysis
125 for COGs and KEGG ortholots. Here, we used the KEGG ortholots (KOs). The relative
126 abundance of functional categories was generated using the OTU table of assigned taxa and their
127 relative distribution in different samples(Urbanová et al., 2014). In the KEGG database,
128 functions were grouped into three level subgroups based on different KEGG functional gene
129 ontology affiliation (i.e. metabolism, cellularprocesses, environmental processing).
130 Accession numbers
131 All the sequences in this study have been submitted to the NCBI-SRA public database
132 (http://www.ncbi.nlm.nih.gov/sra/SRP104573) under the ID: SRP104573 (all the twelve samples
133 of the Changjiang Estuary).
134 RESULTS
135 Environmental parameters and bacterial counts
136 The environmental parameters measured are given in Table 1. Samples from site A0102
137 had the highest COD while A0502 had the lowest; B0202 had the highest NH4+ concentration
138 while A0302, A0402 and B0402 were the lowest; DO was highest at site C0402 and lowest at
139 C0102; NO2− concentration was relatively lower at all sites. DAPI fluorescence direct counts of 1
140 or 2 mL of seawater are shown in Table 1; site A0102 had the highest total bacterial count and
141 C0102 the lowest.
142 Bacterial community structure
143 A total of 445025 16S rRNA gene sequences were obtained from the 12 sample sites.
144 The sequences with insufficient quality or sequences that could not be adequately assigned were
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
145 not included, such as chimera sequences. 442894 sequences were retained for further analysis.
146 To compare the diversities and richness of the bacterial communities, Chao1 and ACE estimates
147 and the Shannon-Weaver diversity index were calculated (Table 2). The richness, estimated by
148 number of OTUs, Chao1 and ACE indices, showed that the highest bacterial richness was at site
149 A0102 with the lowest at A0502. Similarly, the Shannon and Simpson diversity indices indicated
150 that site A0102 had the highest bacterial diversity with lowest diversity at A0202. Rarefaction
151 curves show the diversity and richness of each sample (Figure 2). A0102 had the highest
152 bacterial diversity and richness while A0502 was the lowest.
153 QIIME software was used to identify sequences obtained or to find sequences similar to
154 those obtained here. Overall, there were 36 phyla in the samples. The composition and structure
155 of the bacterial communities in the different samples were compared at the phylum level (Figure
156 3) : Proteobacteria (67.8%), Acidobacteria (0.8%), Actinobacteria (12.2%), Bacteroidetes
157 (11.7%), Chlorobi (0.1%), Chloroflexi (0.2%), Cyanobacteria (2.5%), Firmicutes (0.1%),
158 Gemmatimonadetes (0.3%), Nitrospirae (0.2%), OP3 (0.1%), PAUC34f (0.1%), Planctomycetes
159 (0.1%), SAR406 (2%), SBR1093 (0.1%), Tenericutes (0.1%), Verrucomicrobia (0.8%), ZB3
160 (0.2%), and unassigned (0.3%). “Others” were present in very low abundance. In current study,
161 Proteobacteria were the major component in each library, followed by Actinobacteria and
162 Bacteroidetes. Alphaproteobacteria (38.7%) and Gammaproteobacteria (22.8%) were the
163 predominant classes observed and were in each samples. Firmicutes, SBR1093, Tenericutes, and
164 ZB3 were also detected in all samples (with very low abundance).
165 Similarity of bacterial communities in the seawater samples
166 The heatmap in Figure 4 shows the abundance of the bacteria in the 12 samples. The
167 different samples could be clustered into six groups: A0202; B0302/C0102; A0402/C0402;
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
168 A0502/B0402; A0102; A0302/B0102/B0202/C0302. The communities of B0302 and C0102
169 were highly similar, which had Proteobacteria and Actinobacteria as the dominant phyla. A0402
170 and C0402 were similar and had Proteobacteria, Actinobacteria and Cyanobacteria as the
171 dominant phyla while B0402 and A0502 had the Proteobacteria, Bacteroidetes and SAR406 as
172 the dominant phyla. The A0302, B0102, B0202 and C0302 were highly similar, which had the
173 Proteobacteria, Actinobacteria and Bacteroidetes as the dominant phyla. However, the
174 communities of samples A0102 and A0202 were distant from the other samples. A0102 had the
175 Acidobacteria, Proteobacteria and Actinobacteria as the dominant phyla while the A0202 had the
176 Proteobacteria and Actinobacteria as the dominant phyla. The abundance of bacteria was
177 obtained directly from the heatmap. Proteobacteria were dominant, followed by Actinobacteria
178 and Bacteroidetes, while the least abundant were Firmicutes, Chlorobi, PAUC34f, OP3 and
179 SBR1093.
180 Contribution of environmental factors to bacterial community structure and functional
181 genens
182 Our study indicated that environmental factors might be important determinants of
183 structure and function of microbial communities in the surface water samples. To explore how
184 environmental parameters influenced the bacterial community composition and functional
185 diversity, RDA was performed. Result indicated that PO43− and DO showed positive
186 relationships with taxonomic composition (Figure 5). The first two axes explained 39.4% of the
187 taxonomic information. The envfit showed that PO43− had significant influence on the bacterial
188 communities (P < 0.05) (table 3). Meanwhile, the PO43−, DO and COD were significantly
189 associated with microbial functional diversity. The first two axes explained 72% of the function
190 information. Envfit showed that PO43− and COD had significant influences on the functional
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
191 genes (P < 0.05) (Table 4). PO43- was identified as a major environmental factor structuring the
192 microbial community and contributing to the function of microbial communities. The mantel test
193 (Table 5) indicated that functional genes showed higher relationships to environmental factors
194 than taxonomic composition.
195 Metagenome analysis
196 Based on the 16S rRNA gene copy number of detected phylotype, we predicted the
197 functional profiles of bacterial communities among the 12 water samples. The relative
198 abundance of functional profiles were similar in most of samples. We observed that amino acid
199 metabolism, carbohydrate metabolism, membrane transport and energy metabolism were
200 pronounced enriched in all samples (Figure 6). We also analyzed functional profiles that were
201 involved in the bacterial community adaptation to environment and nutritional conditions
202 (Figure 7). At the individual pathway level, we found that oxidative phosphorylation, carbon
203 fixation pathways in prokaryotes, photosynthesis, nitrogen metabolism, carbon fixation in
204 photosynthetic organisms, polycyclic aromatic hydrocarbon degradation, sulfur metabolism and
205 methane metabolism were enriched in all samples. However, there were also pronounced
206 differences of the functional categories among the samples. Photosynthesis and Polycyclic
207 aromatic hydrocarbon degradation were markedly enriched in A0102. Oxidative phosphorylation,
208 Carbon fixation in photosynthetic organisms, Carbon fixation pathways in prokaryotes and
209 Nitrogen metabolism were pronounced enriched in C0102. These two sites were close to the
210 estuary.
211 DISCUSSION
212 Bacterial abundance and its correlation with environmental factors
213 The bacterial biomass was counted by DAPI fluorescence direct enumeration. Sample
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
214 A0102 had the highest total bacterial counts while A0402 had the lowest. Bacterioplankton are
215 typically present at 108–10 9 cells/L in seawater (Zeng et al. 2014). The average bacterial
216 biomass was 1.76×105 cells/mL, which conformed with previous reports: Zhao (2003) found
217 bacterial abundance ranging from 3.05×105 cells/mL to 1.36×106 cells/mL in the East China Sea
218 (Zhao et al. 2003). However, our result was higher than the values reported by Lin (1998) from
219 the Pacific Ocean and Prydz Bay in Antarctica (Lin & Zeng 1998). In recent years the
220 Changjiang Estuary has become extensively polluted, which provides rich nutrients for bacterial
221 growth. This may lead to the higher bacterial abundance in the Changjiang Estuary. We also
222 observed that the bacterial abundance was higher in the estuary compared to the open sea. For
223 example, sampling sites A0102, A0202 and A0302 (Figure 1) had higher bacterial biomass than
224 A0402 and B0402. Thereby, we deduced that the estuary had higher nutrient levels than the
225 open sea. This may result from the effect of coastal flow. The Subei coastal current and Yellow
226 Sea Coastal Water could bring ample nutrients to the estuary. Yang et al. reported that under the
227 effect of coastal flow on the coastal side, the content of oxygen and nutrients near the
228 Changjiang Estuary was obviously higher than that in the open sea (Yang et al. 2008).
229 Additionally, when the northward TWC, southward Subei coastal current, and Yellow Sea
230 coastal current mix with the freshwater of the Changjiang River in the estuary, there could be
231 ample nutrients to contribute to bacterial growth. Nutrients are supplied to the Changjiang
232 Estuary by the Changjiang River, the TWC respectively, at different times, by different means
233 and with different intensities (Yang et al. 2008). The Changjiang Estuary is a complex ecosystem,
234 the required material sources for microbial growth is varied, including the terrigenous runoff,
235 vertical mixing and resuspension(Shiah & Ducklow 1995). Thereby, the abundance of bacteria
236 and it’s distribution may be influenced by many environmental factors.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
237
238 Diversity of bacteria
239 We used 16S rDNA gene sequences, which were analyzed according to a taxon cutoff set
240 at 97% similarity, to assess the biodiversity of bacterial communities in this environment.
241 Proteobacteria, Actinobacteria and Bacteroidetes were the dominant phyla. Alphaproteobacteria
242 and Gammaproteobacteria were the most abundant subphyla (classes), which was consistent with
243 previous study: Feng et al. reported that Alphaproteobacteria (23.4%) and Gammaproteobacteria
244 (31.7%) were the most abundant phylum in the Changjiang Estuary, followed by Firmicutes
245 (6.4%), Bacteroidetes (4.6%) and Actinobacteria (4.1%) (Feng et al. 2009). Skeiguchi (2002)
246 also found that Alphaproteobacteria and Gammaproteobacteria were the predominant taxa in the
247 Changjiang Estuary (Sekiguchi et al. 2002). Proteobacteria was the dominant phylum in all sites
248 around Xiamen Island (Shan et al. 2015) (including 1569 OTUs, accounting for 49.62%–76.84%
249 of the total reads at different sites, among which Alphaproteobacteria and Gammaproteobacteria
250 were the predominant classes in seawater).
251 In our study, Alphaproteobacteria was the dominant class and distributed in all samples
252 with high abundance, which concurred with previous studies showing that Alphaproteobacteria
253 were the dominant group in open and coastal ocean environments (Bernhard et al. 2005; FO et al.
254 1999). Previous study found that Alphaproteobacteria related to the process of Sulfide(Li et al.
255 2008). Betaproteobacteria had their highest abundance at site A0102. Previous studies have
256 found that Betaproteobacteria had positive correlations with low salinity (Liu et al. 2009; Mosier
257 & Francis 2008). In our study, this site was close to the offshore. The east China sea coastal
258 current, originated in the Yangtze estuary and hangzhou bay area, the dilute water from the
259 Yangtze river and the qiantang river, all of them flow to the south, which led to the low salinity
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
260 in the south of fujian and zhejiang coast. This may led to the high richness of Betaproteobacteria
261 in the A0102. Gammaproteobacteria were distributed in all sites with relatively high abundance.
262 Gammaproteobacteria played important role in the anoxic nitrification-manganese reduction
263 processes(Freitag & Prosser 2003). Hence ,the high abundance of Gammaproteobacteria may be
264 closely related to the carbon-nitrogen cycle, which was conformed with our study that the carbon
265 fixation pathways in prokaryotes and nitrogen metabolism were significantly enriched in all
266 samples. Deltaproteobacteria was also distributed in all samples but with very low abundance.
267 Deltaproteobacteria can degrade naphthalene, alkylbenzenes and benzene (Musat et al. 2009).
268 Actinobacteria and Bacteroidetes were also dominant taxa in all sites. Actinobacteria play
269 important role in the degradation of organic pollutants(He-Yang et al. 2012), which was
270 consistent with our study that Actinobacteria had their highest abundance at site C0302 and the
271 nitrodoluene degradation, chloroalkane and chloroalkene degradation and naphthalene
272 degradation were significantly enriched in C0302. Bacteroidetes was able to degradebiological
273 macromolecules, such as, Chitin, AGAR, DNA and cellulose (Reichenbach 2006). In our study,
274 xenobiotic biodegradation pathways were enriched in all samples including Chitin and AGAR.
275 Cyanobacteria live in relatively pristine coastal environments and play an important role in
276 primary production (Sun et al. 2003). Acidobacteria were mainly distributed at sampling sites
277 A0102, A0202, B0102 and B0202. Barns et al. reported that Acidobacteria have the ability to
278 withstand metal-contaminated, acidic environments and may also be widely distributed in
279 radionuclide-contaminated environments (Barns et al. 2007). In addition, such bacteria are
280 commonly abundant in the terrestrial environment, such as the soil (Wang et al.,2010). In our
281 study, these four sites were close to the estuary and affected by terrigenous, which led to the
282 higher abundance of Acidobacteria.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
283 The relationship between community structure , functional diversity and the
284 environmental factors
285 Bacterial communities are always closely related to their environments and factors such
286 as salinity, temperature and nutrients can be used as an indicator of marine ecosystem status
287 (Andersson et al. 2010; Herlemann et al. 2011). In our study, the results of RDA showed that DO
288 and PO43– were the most important environmental factors that influenced the distribution of the
289 bacterial community structure in surface seawater of the Changjiang Estuary and adjacent areas.
290 Bacterial communities in the water and sediment of the Tama River were influenced by organic
291 matter in the water (Sugita et al. 1983). Joshua et al. reported that phosphate concentration
292 predicted the diversity of bacteria (Ladau et al. 2013). The bacterial community structure
293 responds quickly to changes in DO (Yan et al. 2008). Consistent with these findings, PO43– and
294 DO had the most significant correlations with the bacterial communities in our study. In our
295 study, we found that NH4+ had no correlation with the bacterial community structure. Recent
296 study has revealed that the bacterial ability of using NO3- was higher than we thought
297 (Middelburg & Nieuwenhuize 2000). The high abundance of NO3- could meet the needs of the N
298 source for the bacteria, which weakens the bacteria dependence on nitrogen. We also observed
299 that PO43-, DO and COD were markedly associated with microbial functional diversity.
300 Interestingly, from the Mantel test, we found that the functional genes was significantly
301 correlated with environmental conditions than the taxonomic genera, indicating that functional
302 gene patterns may be more sensitive to environmental conditions than taxonomic composition.
303 Similarly, functional genes appeared to be more appropriate than ‘species’ information in
304 addressing questions regarding bacterial community assembly(Burke et al. 2011).
305 Bacterioplankton are essential players in the release of phosphorus and nitrogen fixation in
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
306 aquatic ecosystems(Wang & Chen 2008; Yang et al. 2013). We observed that PO43- was the most
307 important factors affecting the bacterioplankton community function and composition. The
308 Changjiang Estuary and its adjacent areas are a complex area system which mixes with the TWC.
309 The TWC originated from the Taiwan Strait, extended along the coast of Fujian and Zhejiang
310 Provinces to the north and met with Changjiang Estuary. Therefore we speculated that the TWC
311 may bring abundant phosphate to the area. Similarly, the Changjiang River could also transport
312 phosphate to the Changjiang Estuary. Yang et al. found that the sources of phosphate in the
313 Changjiang estuary included the Changjiang River, the TWC, the cyclone-type eddy and the
314 32°N upwelling (Yang et al. 2008). These may result in the high phosphate concentration in the
315 Changjiang Estuary, variation of which correlated significantly with the bacterial communities
316 and functions.
317 Predictive Functional Analysis
318 PICRUSt provides a prediction of microbiome function based on marker genes, but not
319 an actual measurement of such function. The NSTI scores was used to evaluate the predictive
320 accuracy of PICRUSt. The PICRUSt predictions of genomes tend to be less accurate in poorly
321 environment, for there are relatively few reference genome sequences available(Cleary et al.
322 2015). In our study, the NSTI was relatively lower. Langille et al. reported that the accuracy of
323 PICRUSt decreased with increasing NSTI scores. We found that oxidative phosphorylation,
324 carbon fixation pathways in prokaryotes, photosynthesis, nitrogen metabolism, carbon fixation in
325 photosynthetic organisms, polycyclic aromatic hydrocarbon degradation, sulfur metabolism and
326 methane metabolism were enriched in all samples. The polycyclic aromatic hydrocarbon
327 degradation was significantly enriched in A0102 and C0102. However, it was lower in B0402
328 and A0502. Previous study reported that petroleum content decreased from the alongshore to the
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
329 open sea(Lei et al. 2014), which was consistent with our study. Site A0102 and C0102 were
330 close to the shore while B0402 and A0502 were in the open sea. The nitrogen metabolism was
331 markedly enriched in all samples. The relationship between taxonomic composition and nitrogen
332 sources was reported in some other river system(ClaraRuiz-González et al. 2013). This
333 correlation may be attributed to the Proteobacteria, which includes many members involved in
334 nitrogen cycling(Breitbart et al. 2009; Yang et al. 2013). However, further studies are needed to
335 determine how the detected bacterial members stimulate nutrient cycling processes in this
336 ecosystem(Yan et al. 2015). The photosynthesis was markedly enriched in A0102. Cyanobacteria
337 is widely distributed in the ocean and can do photosynthesis(Sun et al. 2003). In our study, we
338 observed that Cyanobacteria had the highest abundance in A0102, which may lead to the
339 enrichment of photosynthesis. The phosphonate and phosphinate metabolism was also enriched
340 in all samples. With the development of science and technology, eutrophication phenomenon
341 was increasingly serious in the Changjiang Estuary. It may lead to the enrichment of the
342 phosphonate and phosphinate metabolism.
343 CONCLUSIONS
344 The bacterial biomass was high in surface seawater of the Changjiang Estuary and surrounding
345 areas and differed among sites. We detected 36 phyla, including Proteobacteria, Acidobacteria,
346 Actinobacteria, Bacteroidetes and unclassified, as well as others with low abundance.
347 Proteobacteria was the dominant phylum, followed by Actinobacteria and Bacteroidetes.
348 Diversity analysis indicated that sampling site A0102 had the highest diversity of bacteria. RDA
349 result showed that PO43– was the main environmental factors that influenced the distribution of
350 bacterial communities and functions. Analysis of predictive metagenomic showed that Oxidative
351 phosphorylation, Carbon fixation pathways in prokaryotes, Photosynthesis, Nitrogen
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
352 metabolism, Carbon fixation in photosynthetic organisms, Polycyclic aromatic hydrocarbon
353 degradation, Sulfur metabolism, Methane metabolism were enriched in all samples. These
354 findings expand our current understanding on bacterial structure and function in the Changjiang
355 Estuary and it’s adjacent areas.
356
357 Acknowledgements
358 This work was supported by the National Natural Science Foundation of China [grant numbers
359 31270160, J1310037] and the Natural Science Foundation of Zhejiang Province, China [grant
360 number LY12C03003].
361 References
362
363 Andersson AF, Riemann L and Bertilsson S. 2010. Pyrosequencing reveals contrasting seasonal dynamics of taxa
364 within Baltic Sea bacterioplankton communities. Isme Journal 4: 171-181.
365 Barns SM, Cain EC, Sommerville L and Kuske CR. 2007. Acidobacteria phylum sequences in uranium-contaminated
366 subsurface sediments greatly expand the known diversity within the phylum. Applied & Environmental
367 Microbiology 73: 3113-3116.
368 Bernhard AE, Colbert D, Mcmanus J and Field KG. 2005. Microbial community dynamics based on 16S rRNA gene
369 profiles in a Pacific Northwest estuary and its tributaries. FEMS microbiology ecology 52: 115–128.
370 Breitbart M, Hoare A, Nitti A, Siefert J, Haynes M, Dinsdale E, Edwards R, Souza V, Rohwer F and Hollander D. 2009.
371 Metagenomic and stable isotopic analyses of modern freshwater microbialites in Cuatro Cienegas, Mexico.
372 Environmental microbiology 11: 16-34.
373 Burke C, Steinberg P, Rusch D, Kjelleberg S and Thomas T. 2011. Bacterial community assembly based on functional
374 genes rather than species. Proceedings of the National Academy of Sciences 108: 14288-14293.
375 Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK and
376 Gordon JI. 2010. QIIME allows analysis of high-throughput community sequencing data. Nature Methods 7: 335-
377 336.
378 Chen J, Li D, Chen B, Hu F, Zhu H and Liu C. 1999. The processes of dynamic sedimentation in the Changjiang
379 Estuary. Journal of Sea Research 41: 129-140.
380 Chen Y, Zhen Y, He H, Lu X, Mi T and Yu Z. 2014. Diversity, Abundance, and Spatial Distribution of Ammonia-
381 Oxidizing β-Proteobacteria in Sediments from Changjiang Estuary and Its Adjacent Area in East China Sea. Microbial
382 ecology 67: 788-803.
383 ClaraRuiz-González, LorenzoProia, IsabelFerrera, Gasol JM and SergiSabater. 2013. Effects of large river dam
384 regulation on bacterioplankton community structure. FEMS microbiology ecology 84: 316–331.
385 Cleary DFR, Voogd NJD, Polónia ARM, Freitas R and Gomes NCM. 2015. Composition and Predictive Functional
386 Analysis of Bacterial Communities in Seawater, Sediment and Sponges in the Spermonde Archipelago, Indonesia.
387 Microbial ecology 70: 889-903.
388 Desantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, Huber T, Dalevi D, Hu P and Andersen GL. 2006.
389 Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied &
390 Environmental Microbiology 72: 5069-5072.
391 Feng BW, Li XR, Wang JH, Hu ZY, Meng H, Xiang LY and Quan ZX. 2009. Bacterial diversity of water and sediment in
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
392 the Changjiang estuary and coastal area of the East China Sea. FEMS microbiology ecology 70: 80-92.
393 FO G, BM F and R A. 1999. Bacterioplankton compositions of lakes and oceans: a first comparison based on
394 fluorescence in situ hybridization. Applied and environmental microbiology 65: 3721-3726.
395 Freitag TE and Prosser JI. 2003. Community Structure of Ammonia-Oxidizing Bacteria within Anoxic Marine
396 Sediments. Applied & Environmental Microbiology 69: 1359-1371.
397 Graves CJ, Makrides E, Schmidt V, Giblin A, Cardon Z and Rand D. 2016. Functional responses of salt marsh
398 microbial communities to long-term nutrient enrichment. Applied and environmental microbiology 82: 2862-2871.
399 He-Yang LI, Chen MX and Jiao NZ. 2012. The community composition of bacterioplankton at typical station in the
400 Changjiang Estuary. Acta Oceanologica Sinica 34: 183-188.
401 Heijs SK, Laverman AM, Forney LJ, Hardoim PR and van Elsas JD. 2008. Comparison of deep-sea sediment microbial
402 communities in the Eastern Mediterranean. FEMS microbiology ecology 64: 362-377.
403 Herlemann DP, Labrenz M, Jürgens K, Bertilsson S, Waniek JJ and Andersson AF. 2011. Transitions in bacterial
404 communities along the 2000 km salinity gradient of the Baltic Sea. Isme Journal 5: 1571-1579.
405 Jiao N, Zhang Y, Zeng Y, Gardner WD, Mishonov AV, Richardson MJ, Hong N, Pan D, Yan XH and Jo YH. 2007.
406 Ecological anomalies in the East China Sea: impacts of the Three Gorges Dam? Water Research 41: 1287-1293.
407 Jr KR and Pratt JR. 1994. Use of fluorochromes for direct enumeration of total bacteria in environmental samples:
408 past and present. Microbiological Reviews 58: 603-615.
409 Kirchman DL, Hanson TE, Cottrell MT and Hamdan LJ. 2015. Metagenomic analysis of organic matter degradation
410 in methane‐rich Arctic Ocean sediments. Limnology & Oceanography 59: 548-559.
411 Ladau J, Sharpton TJ, Finucane MM, Jospin G, Kembel SW, O'Dwyer J, Koeppel AF, Green JL and Pollard KS. 2013.
412 Global marine bacterial diversity peaks at high latitudes in winter. Isme Journal 7: 1669-1677.
413 Langille MGI, Zaneveld J, Caporaso JG, Mcdonald D, Dan K, Reyes JA, Clemente JC, Burkepile DE, Thurber RLV and
414 Knight R. 2013. Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences.
415 Nature biotechnology 31: 814-821.
416 Lei LI, Jiang M, Shen X and Wang Y. 2014. Spatial distribution and pollution analysis of the petroleum hydrocarbons
417 in water from the Changjiang Estuary and its adjacent areas. Environmental Chemistry 33:1366-1372.
418 Li T, Wang P and Wang P. 2008. Bacterial and archaeal diversity in surface sediment from the south slope of the
419 South China Sea. Acta Microbiologica Sinica 48: 323-329.
420 Lin Y and Zeng H. 1998. Ddistribution characteristics of bacterial number in the Pacific Ocean, the Pryts Bay of
421 Antarctica and its adjacent waters. Acta Oceanologica Sinica 20:81-86.
422 Liu CC, Xiang LY, Zhang HF and Wang JH. 2009. Distribution and relationship between heterotrophic bacteria and
423 environmental factors in Changjiang Estuary. Marine Environmental Science 28: 1-0001-04
424 Middelburg JJ and Nieuwenhuize J. 2000. Nitrogen uptake by heterotrophic bacteria and phytoplankton in the
425 nitrate-rich Thames estuary. Marine Ecology Progress 203: 13-21.
426 Mosier AC and Francis CA. 2008. Relative abundance and diversity of ammonia-oxidizing archaea and bacteria in
427 the San Francisco Bay estuary. Environmental microbiology 10: 3002-3016.
428 Musat F, Galushko A, Jacob J, Widdel F, Kube M, Reinhardt R, Wilkes H, Schink B and Rabus R. 2009. Anaerobic
429 degradation of naphthalene and 2-methylnaphthalene by strains of marine sulfate-reducing bacteria.
430 Environmental microbiology 11: 209-219.
431 Oksanen J, Blanchet G, Kindt R, Legendre P, O’Hara B, Simpson GL, Solymos P, Stevens M and Wagner H. 2009.
432 Vegan: Community Ecology Package, 1.17-The R Project for Statistical Computing. Website xiii–xx.
433 Paerl HDJ, Moisander PH, Noble RT, Piehler MF, Tf PJS, Twomey L and Valdes LM. 2003. Microbial indicators of
434 aquatic ecosystem change: current applications to eutrophication studies. FEMS microbiology ecology 46: 233–246.
435 Porter KG and Feig YS. 1980. The use of DAPI for identifying and counting aquatic microflora1. Limnology &
436 Oceanography 25: 943-948.
437 Reichenbach H. 2006. The Order Cytophagales. Springer New York.
438 Sala MM, Terrado R, Lovejoy C, Unrein F and Pedrós-Alió C. 2008. Metabolic diversity of heterotrophic
439 bacterioplankton over winter and spring in the coastal Arctic Ocean. Environmental Microbiology 10: 942-949.
440 Sekiguchi H, Koshikawa H, Hiroki M, Murakami S, Xu K, Watanabe M, Nakahara T, Zhu M and Uchiyama H. 2002.
441 Bacterial distribution and phylogenetic diversity in the Changjiang estuary before the construction of the Three
442 Gorges Dam. Microbial ecology 43: 82-91.
443 Shan D, Wei G, Li M, Wang W, Li X, Gao Z and Shao Z. 2015. Distribution and diversity of bacterioplankton
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
444 communities in subtropical seawater around Xiamen Island, China. Microbiological Research 175: 16-23.
445 Shiah FK and Ducklow HW. 1995. Regulation of bacterial abundance and production by substrate supply and
446 bacterivory: A mesocosm study. Microbial ecology 30: 239-255.
447 Sugita H, Oshima K, Tamura M and Deguchi Y. 1983. Bacterial flora in the gastrointestine of freshwater fishes in the
448 river. Nihon-suisan-gakkai-shi 49: 1387-1395.
449 Sun S, Xiao T and Yue H. 2003. Distribution character of Synechococcus spp.in the east china sea and the yellow sea
450 in autumn and spring. Oceanologia Et Limnologia Sinica 34:161–168.
451 Urbanová Z and Bárta J. 2014. Microbial community composition and in silico predicted metabolic potential reflect
452 biogeochemical gradients between distinct peatland types. Fems Microbiology Ecology 90:633-646.
453 Vieira RP, Gonzalez AM, Cardoso AM, Oliveira DN, Albano RM, Clementino MM, Martins OB and Paranhos R. 2008.
454 Relationships between bacterial diversity and environmental variables in a tropical marine environment, Rio de
455 Janeiro. Environmental Microbiology 10: 189–199.
456 Wang CX. 2010. Distribution and Diversity of Acidobacteria in Tropical Rain Forest Soil of Xishuangbanna.
457 Microbiology China 37:24–29.
458
459 Wang Y and Chen F. 2008. Decomposition and phosphorus release from four different size fractions of Microcystis
460 spp.taken from Lake Taihu,China. Journal of Environmental Sciences 20: 891-896.
461 Yan Q, Yuhe YU, Feng W, Zhigang YU and Chen H. 2008. Plankton community composition in the Three Gorges
462 Reservoir Region revealed by PCR-DGGE and its relationships with environmental factors. Journal of Environmental
463 Sciences 20: 732-738.
464 Yan Q, Bi Y, Deng Y, He Z, Wu L, Van Nostrand JD, Shi Z, Li J, Wang X and Hu Z. 2015. Impacts of the Three Gorges
465 Dam on microbial structure and potential function. Scientific reports 5: 8605.
466 Yang, Ping, Zhimei, Qing, Wang, Huihui, Hong and Shen. 2013. Decrease of NH4^+-N by bacterioplankton
467 accelerated the removal of cyanobacterial blooms in aerated aquatic ecosystem. Journal of Environmental Sciences
468 25: 2223-2228.
469 Yang D, Wang F, Miao Z, Chen Y and Tong Y. 2008. The main phosphorous sources in the Changjiang estuary.
470 Chinese Journal of Oceanology & Limnology 26: 486-497.
471 Zeng YX, Yu Y, Qiao ZY, Jin HY and Li HR. 2014. Diversity of bacterioplankton in coastal seawaters of Fildes
472 Peninsula, King George Island, Antarctica. Archives of Microbiology 196: 137-147.
473 Zhang J, Zhang ZF, Liu SM, Wu Y, Xiong H and Chen HT. 1999. Human impacts on the large world rivers: Would the
474 Changjiang (Yangtze River) be an illustration? Global Biogeochemical Cycles 13: 1099-1106.
475 Zhao SJ, Tian X and Yue HD. 2003. THE DISTRIBUTION OF HETEROTROPHIC BACTERIA AND SOME RELATED
476 FACTORS IN THE EAST CHINA AND YELLOW SEAS DURING FALL. Oceanologia Et Limnologia Sinica 34:305-311.
477
478
479 Table 1. Environmental parameters, bacteria direct counts, and location of the 12 sample sites
NH4+ PO4
3- COD DO NO2- Total bacterial count
Sample (mg/L) (mg/L) (mg/L) (ml/L) (mg/L) (cells/L) Longitude Latitude
A0102 0.07 0.05 2.86 5.48 0 3.40×105 122°12′42″ 30°58′48″A0202 0.11 0.05 3.63 5.66 0.01 2.62×105 122°20′42″ 30°56′48″A0302 0.03 0.02 0.8 5.8 0.01 1.76×105 122°41′18″ 30°59′54″A0402 0.03 0.01 1.37 6.06 0.01 1.52×104 122°49′30″ 30°59′30″A0502 0.04 0 0.36 6.46 0 2.12×105 122°59′54″ 30°59′42″B0102 0.12 0.05 1.56 4.7 0.01 1.32×105 122°07′00″ 30°45′24″B0202 0.52 0.04 2.02 5.35 0.01 2.80×105 122°20′30″ 30°46′54″B0302 0.11 0.03 0.92 5.95 0.01 2.16×105 122°41′48″ 30°44′06″
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
B0402 0.03 0 0.92 5.95 0 9.94×104 123°03′48″ 30°42′42″C0102 0.18 0.05 2.06 2.54 0 9.04×104 122°06′54″ 30°25′54″C0302 0.04 0.03 1.14 5.14 0 1.65×105 122°42′18″ 30°25′48″C0402 0.06 0.01 0.55 6.27 0.01 1.24×105 122°53′00″ 30°28′18″
480 COD: Chemical Oxygen Demand DO: Dissolved Oxygen
481 Table 2. Bacterial diversity indices
Sample Diversity indices
No. of
sequences OTUs
Bias-corrected
Chao1
Abundance-base
Coverage
Estimator
Shannon
index(H)
Simpson
index(D)
A0102 39176 1096 1199 1174 7.4 0.02
A0202 30107 482 549 516 4.92 0.16
A0302 40387 607 733 757 5.75 0.05
A0402 39799 473 602 592 5.63 0.05
A0502 31670 408 551 532 5.63 0.04
B0102 43482 749 851 850 6.42 0.03
B0202 36889 794 990 982 6.49 0.03
B0302 41777 707 1001 969 5.78 0.05
B0402 30993 490 615 599 5.82 0.05
C0102 33672 681 780 783 6.32 0.03
C0302 32430 638 827 791 5.9 0.04
C0402 44643 532 674 689 5.61 0.05
482
483
484
485
486 Table 3. Redundancy analysis of environmental factors and bacterial community structure in the Changjian estuary
487 and adjacent areas.
488
RDA1 RDA2 r2 Pr(>r)
NH4+ -78.33×10-2 62.12×10-2 32.99×10-2 0.171
PO43- -99.99×10-2 1.27×10-2 49.06×10-2 0.049*
COD -95.65×10-2 29.18×10-2 26.40×10-2 0.24
DO 79.55×10-2 60.60×10-2 47.27×10-2 0.054
NO2- 5.32×10-2 99.86×10-2 38.92×10-2 0.106
489 *P<0.05. Number of permutations: 999.
490
491
492 Table 4. Redundancy analysis of environmental factors and functional genes in the Changjiang estuary and adjacent
493 areas.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
494
RDA1 RDA2 r2 Pr(>r)
NH4+ 73.76×10-2 -67.52×10-2 11.06×10-2 0.467
PO43- -5.82×10-2 -99.83×10-2 49.40×10-2 0.037*
COD -74.15×10-2 -67.09×10-2 52.16×10-2 0.036*
DO -53.45×10-2 84.52×10-2 53.17×10-2 0.082
NO2- -57.48×10-2 -81.83×10-2 15.20×10-2 0.485
495 *P<0.05. Number of permutations: 999.
496
497
498
499 Table 5. Summary results from Mantel tests performed at the functional gene level or phylogenetic genus level.
R2 P
Functional genes vs environmental factors R2=0.5298 0.0095
Phlylogenetic genera vs environmental factors R2=0.3314 0.0675
500
501
502
503
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
504
505 Figure 1. Distribution of the 12 sampling sites in the Changjiang Estuary and adjacent areas.
506
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
507
508 Figure 2. Rarefaction curves for the 12 samples
509
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
510
511 Figure 3. Community structure of the 12 water samples at the phylum level
512
513
514
515
516
517
518
519
520
521
522
523
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
524
525 Figure 4. Heat map of the 12 water samples based on the abundance similarity of the bacteria. Columns represent the
526 different samples; rows represent the bacteria. Red represents the highest bacterial abundance, lightblue the lowest.
527 The “*” represent the relative abundance of samples > 0.1 while the “+” represent the relative abundance > 0.0
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
528
529 Figure 5. Redundancy analysis (RDA) shows the relationships between environmental variables and the bacteria
530 communities (a); relationships between environmental variables and the functional genes (b).
531
532
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
533
534 Figure 6. Heatmap showing the differences among the 12 investigated communities based on the KEGG orthology
535 groups. Columns represent the different samples; rows represent the bacteria. Red represents the highest bacterial
536 abundance, lightblue the lowest. The “*” represent the gene counts > 600000 while the “+” represent the gene
537 counts >300000
538
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
539
540
541 Figure 7. The relative abundance of some imputed functional profiles in the 12 water samples.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017
542
543
544
545
546
547
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3079v1 | CC BY 4.0 Open Access | rec: 10 Jul 2017, publ: 10 Jul 2017