This study examined the bacterial and archaeal diversity from a worldwide range of wetlands soils and sediments using a meta-analysis approach. All available 16S rRNA gene sequences recovered from wetlands in public databases were retrieved. In November 2012, a total of 12677 bacterial and 1747 archaeal sequences were collected in GenBank. All the bacterial sequences were assigned into 6383 operational taxonomic units (OTUs 0.03), representing 31 known bacterial phyla, predominant with Proteobacteria (2791 OTUs), Bacteroidetes (868 OTUs), Acidobacteria (731 OTUs), Firmicutes (540 OTUs), and Actinobacteria (418 OTUs). The genus
Wetlands, which were estimated to be 45% of the total value of global natural ecosystems [
Initial sequence sets were obtained from the GenBank (
Sequences were aligned with Kalign [
The accession numbers for all sequences analyzed in this study were available from the corresponding author. The sequences were currently maintained in an in-house ARB database of 16S rRNA gene sequences for wetlands. A copy of this database and the sequence alignment were also available by request from the corresponding author.
This study was conducted as a meta-analysis ground on publicly available 16S rRNA gene sequences recovered from wetland soils worldwide. The sequences dataset collected from Genbank and RDP database was analyzed no matter their previously assigned taxonomic information or other analyses.
To address the long-term question of understanding microorganisms from wetland soil habitats, this study first aimed at characterizing prokaryotic communities inhabiting wetland soils. The prokaryotic microorganisms from wetland soil habitats drive the biogeochemical cycles of elements and may be a source of novel halophilic enzymes. Thus, we studied the diversity of prokaryotic microorganisms from wetland soils with meta-analysis approach.
Totally 14318 sequences longer than 250 bp were retrieved from GenBank and RDP databases. The sequences were mostly about 800 bp long, followed by approximately 600 bp (Figure
Diversity statistics for Archaea, Bacteria, and “Major” phylum groups. Coverage = #OTUs/rarefaction estimate; OTU and abundance were calculated using a 0.03 dissimilarity cut-off.
Group | Total sequences | Unclassified to phylum | Number of OTUs | ACE | Chao1 | Rarefaction estimation | Current coverage (%) |
---|---|---|---|---|---|---|---|
Bacteria | 12583 | none | 6383 | 30581 | 17176 | 15768 | 41 |
Pro | 5763 | 2791 | 12472 | 7245 | 6811 | 40 | |
Act | 783 | 418 | 2280 | 1088 | 1033 | 46 | |
Aci | 1345 | 731 | 2972 | 1693 | 1602 | 28 | |
Fir | 973 | 540 | 3595 | 1856 | 1915 | 54 | |
Bact | 2244 | 868 | 2700 | 1887 | 1601 | 59 | |
Archaea | 1735 | none | 521 | 1131 | 884 | 883 | 83 |
Eur | 925 | 418 | 681 | 505 | 504 | 62 | |
Cre | 810 | 197 | 442 | 311 | 320 | 41 |
Distribution of the length of retrieved 16S rRNA sequences.
Treemap of observed prokaryotic taxons shown in their hierarchical order. Treemap showing taxonomic ranking of all taxa for all retrieved sequences. The size of each box is proportional to the number of sequences assigned to that taxon with respect to the entire dataset. The placement of boxes is arbitrary with respect to boxes within the same taxonomic rank and does not correspond to any form of phylogeny or relatedness.
Of the archaeal sequences analyzed, all of them were classified within two phyla: Euryarchaeota and Crenarchaeota, representing 925 and 810 sequences, respectively.
The Proteobacteria was the largest and most diverse phylum in the present dataset. It comprised a total of 5637 sequences, approximately 44.8% of the bacterial sequences, assigned to 466 known genera. There are 2791 OTUs generated, with a Simpson diversity index of 0.0020. All six classes within the Proteobacteria were represented, but the Delta-, Gamma-, Beta-, and Alphaproteobacteria together represented over 99% of the proteobacterial sequences (Figure
Treemap of observed Proteobacteria taxons shown in their hierarchical order.
Classes in Proteobacteria showed various tendencies in different wetlands. The wide distribution of Gammaproteobacteria and Deltaproteobacteria in marine sediment has been documented, and most of them were involved in sulfur reduction under anaerobic conditions [
Deltaproteobacteria was the largest class in the phylum, with 1627 sequences (28.9% of the proteobacteria).
For the class Gammaproteobacteria, 1456 sequences were identified. It was the second largest class in Proteobacteria. Approximately 12.6% of gammaproteobacterial sequences (184 sequences) were assigned to the genus
The 1420 betaproteobacterial sequences were identified in Proteobacteria. The genus
The fourth largest proteobacterial class was Alphaproteobacteria, with 1090 sequences (over 19.3%). The dominating genus
Bacteroidetes was the second abundant phylum in the present dataset, including 2244 sequences (nearly 17.8% of all bacterial sequences), which were assigned to 109 known genera, with 868 OTUs and a Simpson diversity index of 0.0007 (Figure
Treemap of observed Bacteroidetes taxons shown in their hierarchical order.
The most frequently observed genus in Flavobacteria was
Acidobacteria was the third largest phylum in our dataset, including 1345 sequences assigned to 29 genera. Acidobacteria is a new phylum, whose members are physiologically diverse and ubiquitous in soils, but are underrepresented in culture at present. There were 731 OTUs identified, with a Simpson diversity index of 0.0031 (Figure
Treemap of observed Acidobacteria taxons shown in their hierarchical order.
The fourth largest phylum was the Firmicutes, assigned into 973 sequences and 540 OTUs with a Simpson diversity index of 0.0041 (Figure
Treemap of observed Firmicutes taxons shown in their hierarchical order.
In total, about 45% of the Firmicutes sequences were classified to the class Clostridia, and nearly 36% were classified into the class Bacilli. The Clostridia (sulfite-reducing bacteria) is an anaerobic and highly polyphyletic bacterium, while Bacilli can be obligate aerobes or facultative anaerobes. There was a long record of evidence to suggest that both of them were the abundant taxa in sewage sludge [
The class Negativicutes represented 178 sequences. The genus
As the fifth abundant phylum, Actinobacteria represented 783 sequences, clustered into 418 OTUs, with a Simpson diversity index of 0.0054. All of Acidobacteria sequences were classified to the class Actinobacteria and over 66% of them belonged to order Actinomycetales (Figure
Treemap of observed Actinobacteria taxons shown in their hierarchical order.
In addition to the five phyla described above, 26 minor phyla with 1601 sequences were also observed based on the dataset. Of these minor phyla, only the phyla Chloroflexi (2.96%), Planctomycetes (2.77%), Cyanobacteria (2.28%), and Verrucomicrobia (1.28%) represented more than 1% of all the bacterial sequences and accounted for over 73% of all minor phyla sequences (Figure
Some known genera were represented in these “minor phyla.” The most abundant of the minor phyla, Chloroflexi, comprised 372 sequences. Members of the Chloroflexi are generally found in intertidal sediment and moderately acidic wetland [
Euryarchaeota comprised 925 sequences, approximately 53.3% archaeal sequences. They were clustered into 418 OTUs with a Simpson diversity index of 0.0054 (Figure
Treemap of observed Euryarchaeota taxons shown in their hierarchical order.
Methanomicrobia contributes a large proportion of methane emission in wetlands, no matter in cold area or in subtropical places [
The largest genus in class Thermoplasmata was
Crenarchaeota owned less abundant sequences than Euryarchaeota in the dataset, with 810 sequences. Crenarchaeota diversity was lower, with only 197 OTUs generated and a Simpson diversity index of 0.0443. It suggested that Crenarchaeota was more related to aerobic metabolisms in the water and surface sediment [
All of the Crenarchaeota sequences were assigned to the class Thermoprotei (Figure
Treemap of observed Crenarchaeota taxons shown in their hierarchical order.
For all of the bacterial groups, the ACE value of richness was the greatest, while the majority of corresponding rarefaction estimates were the lowest (Table
The present results showed that the coverage of microbial diversity in wetlands was remaining rather low. Rarefaction analysis of Bacteria showed that only sampling at the phylum (0.20 phylogenetic distance) level has begun to reach a horizontal plateau. The other sampling at the taxonomic ranks was still projecting upward (Figure
Estimates of current taxonomic coverage for Archaea and Bacteria.
Distance | Number of Current OTUs | Rarefaction estimation | Coveragea (%) |
---|---|---|---|
Archaea | |||
0.03 | 521 | 883 | 59 |
0.05 | 364 | 587 | 62 |
0.10 | 190 | 278 | 68 |
0.20 | 82 | 91 | 90 |
Bacteria | |||
0.03 | 6383 | 15768 | 40 |
0.05 | 5042 | 9854 | 51 |
0.10 | 2937 | 4617 | 63 |
0.20 | 954 | 1118 | 85 |
Rarefaction curve for the Archaea (a) and Bacteria (b) with different dissimilarity cut-off.
Sufficient coverage and depth were provided to explore an individual sample or compare multiple samples through multiplexing, with the developing of second generation sequencing technologies. Moreover, new sequences dataset could be added to the composite datasets analyzed in this study to increase our knowledge on the diversity of this ecosystem. The knowledge on the diversity may shine light on the understanding of the microbiomes of wetlands and define the significance of individual microbia. It is also suited for continuous following of the succession variation of the diversity of wetlands. However, the beta diversity was hardly determined because most of studies could not contain large sequence datasets and detailed information with same methodologies and sequence submission criteria. A “core group” was defined after analyzing seven municipal sludge digesters [
Now that analysis of 16S rRNA gene sequences can provide insight into the functional diversity of wetlands, the metabolic functions of organisms are getting more concerned. For a good comprehension of the metabolic capacities of these organisms, metagenomic studies techniques such as SIP and MAR-FISH should be used more frequently. Cultivation-based studies are also needed to define the functions of uncharacterized species of bacteria and archaea in wetlands.
The present dataset generated from GenBank and RDP databases was largely dominated by Proteobacteria. Approximately 40% of sequences and OTUs belonged to Proteobacteria. Our results showed that (1) nearly 56% of the archaeal and 45% of the bacterial species-level diversity in wetlands have been witnessed; (2) sequences from the bacterial phyla Proteobacteria, Bacteroidetes, Chloroflexi, Firmicutes, Actinobacteria, and archaeal class were well represented by the available sequences and the corresponding microorganisms were probably important participants in the wetland environments; (3) the global diversity contains numerous groups for which there was no close cultured representative, especially the majority of sequences assigned to the phyla Chloroflexi and Bacteroidetes. Therefore future studies should utilize multiple approaches to characterize the microbial diversity and its function in wetlands.
The authors declare that there is no conflict of interests regarding the publication of this paper.
Xiaofei Lv, Junbao Yu, and Yuqin Fu contributed equally to this work.
The authors would like to thank the Project of National Science & Technology Pillar Program in “12th Five Year” period (2011BAC02B01), National Natural Science Foundation for Distinguished Young Scholar of Shandong Province (no. JQ201114), the National Science Foundation of China (41301333), and the CAS/SAFEA International Partnership Program for Creative Research Teams “Representative environmental processes and resources effects in coastal zone.” They also would like to thank the Yellow River Delta Wetland Ecological Experimental Station, CAS, for providing experimental and residential place for this study.