Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Bacterial and diazotrophic diversities of endophytes in Dendrobium catenatum determined through barcoded pyrosequencing

Bacterial and diazotrophic diversities of endophytes in Dendrobium catenatum determined through barcoded pyrosequencing

  • Ou Li, 
  • Rong Xiao, 
  • Lihua Sun, 
  • Chenglin Guan, 
  • Dedong Kong, 
  • Xiufang Hu
PLOS
x

Abstract

As an epiphyte orchid, Dendrobium catenatum relies on microorganisms for requisite nutrients. Metagenome pyrosequencing based on 16S rRNA and nifH genes was used to characterize the bacterial and diazotrophic communities associated with D. catenatum collected from 5 districts in China. Based on Meta-16S rRNA sequencing, 22 bacterial phyla and 699 genera were identified, distributed as 125 genera from 8 phyla and 319 genera from 10 phyla shared by all the planting bases and all the tissues, respectively. The predominant Proteobacteria varied from 71.81% (GZ) to 96.08% (YN), and Delftia (10.39–38.42%), Burkholderia (2.71–15.98%), Escherichia/Shigella (4.90–25.12%), Pseudomonas (2.68–30.72%) and Sphingomonas (1.83–2.05%) dominated in four planting bases. Pseudomonas (17.94–22.06%), Escherichia/Shigella (6.59–11.59%), Delftia (9.65–22.14%) and Burkholderia (3.12–11.05%) dominated in all the tissues. According to Meta-nifH sequencing, 4 phyla and 45 genera were identified, while 17 genera and 24 genera from 4 phyla were shared by all the planting bases and all the tissues, respectively. Burkholderia and Bradyrhizobium were the most popular in the planting bases, followed by Methylovirgula and Mesorhizobium. Mesorhizobium was the most popular in different tissues, followed by Beijerinckia, Xanthobacter, and Burkholderia. Among the genera, 39 were completely overlapped with the results based on the 16S rRNA gene. In conclusion, abundant bacteria and diazotrophs were identified in common in different tissues of D. catenatum from five planting bases, which might play a great role in the supply of nutrients such as nitrogen. The exact abundance of phylum and genus on the different tissues from different planting bases need deeper sequencing with more samples.

Introduction

Dendrobium catenatum Lindl. is one of the most popular epiphytic medicinal orchids in South Asia and Southeast Asia [1]. Stems of D. catenatum contain rich polysaccharide and dendrobine contents, which makes them valuable for relieving stomach upsets, promoting body fluid production, nourishing ‘‘yin” and antipyresis [2], relieving throat inflammation and fatigue, reducing peripheral vascular obstruction, and enhancing immunity. Therefore, D. catenatum has been listed in the 2010 edition of the Chinese Pharmacopoeia, under the name ‘Tiepishihu’ as the sole origin of the herb [3]. The many desirable functions have greatly increased the demand for Tiepishihu, thereby resulting in severe depletion of the wild resources of this plant. Therefore, in order to meet the increasing demand, artificial cultivation has become the main source of D. catenatum.

It is well known that fungal endophytes have profound effects on plant ecology, fitness, evolution, and even the diversity and structure of plant communities [4]. As epiphytic or lithophytic plants with minute seeds [5], D. catenatum plants vitally require fungi as their main source of essential nutrients for growth and secondary metabolism [6]. Knowledge of the diversity and functions of endophytic fungi in dendrobium has been accumulating. However, the so-called mycorrhiza helper bacteria (MHB) or “satellite” bacteria were found to be commonly occurring in ectomycorrhiza and in arbuscular mycorriza associations [7] since Bowen and Theodorou first reported the promotion of Rhizopogon luteolus on Pinus radiate roots [8]. The popular MHB are closely associated with mycorrhizal fungi (MF) with taxonomically diverse bacterial groups [9], and the majority of bacteria have a stimulating effect on the mycelial growth and mycorrhiza formation [10]. This indicates that not only single species, but entire microbial communities, may have evolved to live in close association with mycorrhizal fungi and plants.

Plant associated bacteria are commonly recognized to have a great and often favorable impact on plant growth and development, due to nitrogen fixation, production of plant growth regulators, improvement of water uptake and mineral nutrition, and biosynthesis of fungicidal and/or bactericidal substances, thus reducing the number of phytopathogens [11]. For epiphyte or lithophyte D. catenatum, nitrogen-fixation is one of the most important functions provided by the unique ability of diazotrophic bacteria. Therefore, it is worth learning about the diversity of diazotrophic bacteria in the plants of D. catenatum.

However, not much is known about the composition and functional activity of orchid-associated bacteria, let alone diazotrophic bacteria in particular. Some studies on microbial diversity, localization and functional activity of orchid-associated bacteria in several greenhouse and wild-grown terrestrial and epiphytic orchids revealed an abundance of heterotrophic and phototrophic bacteria on the roots of some cultivated tropical orchids of Calanthe, Acampe and Dendrobium genera. As observed through scanning electron microscopy, microbial clusters can occupy the surface and inner root tissues, particularly the velamen, and be submerged within the intercellular matrix. The multilayered structure of the velamen, which may protect the associated bacteria from various biotic and abiotic factors, represents a suitable econiche for colonization by cyanobacteria. The functional role of the isolated strains was proven by high nitrogen-fixing activity of the orchid-associated cyanobacteria, and indole-3-acetic acid (IAA) production by heterotrophic bacteria. The intimate relations of cyanobacteria with their hosts create a unique symbiotic consortium, which guarantees an ecological stability and nutrient supply for both partners [12]. The MHB, with cyanobacteria being a well-studied example, have become the focus of recent interest. Thus, the diversity and roles of other orchid diazotrophic endophytes remain largely unexplored.

High-throughput sequencing is a promising method for investigating microbial community structure and diversity, as it provides enough sequencing depth to cover complex microbial communities [13]. Thus far, it has been applied to analyze microbial communities in both environment [14] and host [15]. However, few studies have been conducted using this method to investigate D. catenatum. Both phylogenetic (e.g., 16S rRNA, gyrB, recA, and ribosomal intergenic regions) and functional marker genes (e.g., amoA, nirS, nirK, nifH, dsrAB, and other biogeochemically important genes) are very useful for studying phylogenetic relationships among different organisms, for analyzing microbial community structure, and/or for monitoring the physiological status and functional activities of microbial populations in natural environments [16].

The purpose of this study is to assess the diversity of bacterial and diazotrophic endophytes in D. catenatum based on metagenome targeting of 16S rRNA and nifH genes. Thess results of this study may also be beneficial for the study of the functions and interactions of bacteria with their hosts. To the best of our knowledge, this study is the first application of PCR-based Illumina Miseq pyrosequencing for the characterization and comparison of multiple D. catenatum samples.

Materials and methods

D. catenatum plant samples

D. catenatum plants were randomly collected from planting bases in Zhejiang (ZJ), Fujian (FJ), Yunnan (YN), Guizhou (GZ) and Guangxi (GX) provinces in China (shown in Table 1). All the planting bases belong to private land, and the owner, one of the co-auther, gave permission to conduct the study on these sites. From each site, five sub-samples were collected and mixed to provide a unique representative sample for endophytic diversity analysis. All the plants (2 years old) exhibited a healthy appearance and were used for total DNA extraction.

thumbnail
Table 1. Location of five geographically distributed planting bases of D. catenatum.

https://doi.org/10.1371/journal.pone.0184717.t001

DNA extraction

The D. catenatum plants were washed under running tap water to remove adhering soil particles and the majority of microbial epiphytes. Root, stem and leaf tissues were separated and sterilized to eliminate remaining microorganisms using a series of treatments consisting of sterile distilled water, 70% ethanol, and sodium hypochlorite solution (2.5% available Cl-). Total genomic DNA was extracted from about 0.2 g fresh plant tissues using the modified cetyltrimethyl ammonium bromide (CTAB) method [17], with the incubating time at 65°C prolonged to 90 min. The quality of the extracted DNA was analyzed using the Nanodrop method. The DNA was stored at −20°C.

Amplicon library construction

The DNA concentration in each sample was measured using a NanoDrop ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA) to verify that samples had concentrations in excess of 50 ng/ul, which is the value required for metagenomic analyses. The fragments of 16S rRNA and nifH genes were amplified using the metagenomic DNA and Super-Therm Taq DNA polymerase (JMR, UK). Primer pair fM1/rC5 was used for amplification of the 16S rRNA gene [18], and PolF/PolR and PolF/ZehrR pairs were used for amplification of the nifH gene [19, 20]. The 5’-fused primer includes an inserted 7 nucleotide ‘barcode’; the barcode is permuted for each sample and allows the identification of individual samples in a mixture in a single pyrosequencing run [21].

Based on the manufacturer’s instructions, the PCR reaction mixture (50 μL) contained 1 μL Taq polymerase (2.5 unit), 3 μL primer set, 1BF-2BR (10 pmol), 5 μL of reaction buffer, 15 mM MgCl2, 5 μL of 2 mM dNTP, 5 μL template DNA, and 28 μL sterilized water. Thirty cycles, with denaturation at 94°C for 60 s, annealing at 54°C for 40 s, and extension at 72°C for 45 s, were followed by a final incubation at 72°C for 8 min. After PCR analysis on agarose gel (1%), the specific bands were purified using a PCR purification kit (Sangon Biotech, Shanghai, China). The barcode-tagged fragments were quantified using a Qubit® dsDNA HS Assay kit with a Qubit2.0 fluorometer (Life Technologies, Grand Island, NY, USA) and pooled in approximately equal concentrations to ensure equal representation of each sample. These samples were sequenced using 2×300 paired-end (PE) Illumina Miseq at the Shanghai Sangon Biotech Co. Ltd (Shanghai, China).

Illumina sequencing and analysis

PCR products with unique indices from each library were taken in equal nanogram quantities and subjected to 200-nucleotide paired-end multiplex sequencing using an Illumina GAIIx sequencer. Image analysis and base calling were performed using Illumina Analysis pipeline (Version 2.2).

After sequencing, FLASH v1.2.7 was used to separate the PE reads from each sample according to their barcode sequence, then overlapped to assemble the tag sequences. With that, the assembled sequence was completed with removing the sequences whose lengths did not meet the minimum quality filter criteria, and removing the barcode sequences and the joint sequences, thus allowing the acquisition of high quality and credibility clean data. The sequences selected above were defined as ‘raw reads’ for each sample. All raw sequence data had been deposited in the NCBI Sequence Read Archive database and the accession number is SRR3035220.

The raw reads were de-noised using pre.cluster’ (http://www.Mothur.org/wiki/Pre.cluster) to remove sequences that were likely due to sequencing errors [22]. Unique sequences were then aligned against SILVA [23] and chimeric sequences were removed using chimera.uchime [24]. After sequence quality control and filtration, the effective sequences of each sample were clustered using QIIME 1.8.0 to build operational taxonomic units (OTUs) with a cutoff value of 97% sequence identity. In order to analyze species composition more accurately, the RDP Classifier of the Ribosomal Database Project (RDP) [25] was used for all OTU representative sequences species classification analysis at a confidence threshold of 80%. Taxonomic assignment from phylum level to genus level was assigned based on the hits, with abundance graphs plotted based on the number of hits. Heatmaps were plotted using MG-RAST. The Venn diagram was constructed with the Venn diagram plotter jquery.venny, a tool developed by genotoul bioinfor (http://bioinfo.genotoul.fr/). Diversity indices were calculated using SPADE software [26]. The rarefaction curve and Alpha rarefaction calculations were performed using Mothur [27], including Chao1, ACE values, Shannon indices and coverage. For beta diversity analysis, NMDS plot based on RDP Classifier taxa was conducted using un-weighted UniFrac distance implemented in QIIME [28].

Meta-nifH sequence analysis was performed using the software package QIIME [29] version 1.8.0, according to the Qiime tutorial (http://qiime.org/) with some modified methods. Chimeric sequences were removed using usearch61 [30] with denovo models. Sequences were clustered into Operational Taxonomic Units (OTUs) at the thresholds of 97%, 95%, 92%, and 90% sequence similarities using uclust. Each OTU was then assigned to a taxonomic group using BLAST [31] (e-value: 1e-10) with the NCBI NT database (Excluding uncultured/environmental sample sequences). Rarefaction and rank abundance curves were calculated from OTU tables generated at 92% similarity using alpha diversity and rank abundance scripts within the QIIME pipeline. β-diversity analysis generated a NMDS plot using un-weighted UniFrac distance implemented in QIIME. Analysis of Variance (ANOVA) was carried out to test the difference between samples using STAMP.

Results

Based on the metagenome sequencing and data analysis, 141,196 16S rRNA gene sequences and 633,624 nifH sequences were obtained from 15 D. catenatum samples, and the number of reads per sample ranged from 7,582 to 12,052 and 23,292 to 61,285 (S1 Table), respectively. GZL had the highest bacterial ratio and ZJS had the lowest bacterial ratio, while GZL had the highest nifH ratio and ZJL had the lowest nifH ratio.

The relationships among the microbial communities of the different samples were investigated using unweighted heat map. In the resulting heat map, different distances were observed between the microbial communities of individual samples (Fig 1). In the case of the bacterial community (Fig 1b), stem and leave samples tended to cluster closer to each other, which was consistent with the results of the NMDS plot (S2 Fig). For the diazotrophic community (Fig 1b), stem and leaf samples appeared to cluster together, while the root samples exhibited another cluster. Therefore, some similarity existed between the bacterial and diazotrophic communities of the samples.

thumbnail
Fig 1. The unweighted heatmap based on 16S rRNA (a) and nifH (b) communities.

https://doi.org/10.1371/journal.pone.0184717.g001

Taxonomic diversity of bacterial endophytes

Rarefaction curves were generated to facilitate comparison of the sequencing effort among the samples (S1a Fig). None of the curves approached a plateau, suggesting that further sequencing would have resulted in more OTUs in each sample. As shown in Table 2, GZ and FJ samples displayed relatively higher species richness than the samples from the other three locations with regard to the number of OTUs, Shannon Wiener, Chao 1, and ACE indices. In total, the Shannon Wiener indices of the roots were higher than those of the stems and leaves with the exception of GX and GZ samples. For the planting bases, GZ showed the highest Shannon Wiener index, followed by FJ and GX, while YN had the lowest index. According to the heat map (Fig 1), the leaf samples tended to cluster together with stems, and roots samples tended to cluster in another distinct group.

Across all the fifteen samples, a total of 22 bacterial phyla were identified, but only four had an average abundance of greater than 1% (S3 Fig), which accounted for 98.35% of the total effective bacterial sequences. The four dominant phyla were Proteobacteria (84.40%), Bacteroidetes (6.13%), Firmicutes (5.59%), and Actinobacteria (2.23%). The results showed that Proteobacteria was the predominant phylum, which was dominated by Gammaproteobacteria (42.53%). On the genus level, there were 699 genera identified from D. catenatum, 12 of which had over 1% reads. The predominance at the phylum and class levels was driven by the high abundance of five dominant genera with over 4.5% reads in average, such as: Pseudomonas (20.04%), Delftia (16.59%), Escherichia/Shigella (9.46%), Burkholderia (6.62%) and Methylobacterium (4.52%).

Diversity between planting bases.

To investigate the influence of environment on the bacterial communities, the diversity between planting bases was compared. As shown in the Venn diagram (Fig 2), the numbers of phyla were different between bases, with ZJ and FJ having the maximum of 16 phyla, followed by GZ (15), GX (12), and YN (11). Among these phyla, 8 phyla existed in all the planting bases, which occupied 98.94% of the total reads, and 14 phyla were shared by partial bases or unique bases (Fig 2). Interestingly, 4 of the 8 phyla in common dominated in all the bases, and their abundances varied with the planting base. The abundance of Proteobacteria varied from 71.81% (GZ) to 96.08% (YN), and that of Actinobacteria from 0.36% (YN) to 4.81% (GZ). The relative abundances of Bacteroidetes and Firmicutes were higher in the GZ samples and lower in the other samples.

thumbnail
Fig 2. The phylum levels (A) and genus levels (B) of species composition of five planting bases samples.

https://doi.org/10.1371/journal.pone.0184717.g002

On the genus level, some differences existed between the samples of the different planting bases. Firstly, the numbers of genera were different. Sample ZJ had the maximum of 465 genera, followed by GZ (462), FJ (433) and GX (412), and YN had the least number of genera with 179, which was similar to the order of the latitudes of the sampling districts (Table 1). Secondly, the types of genera of each base were different. Except for 125 genera shared by all the planting bases, there were 575 genera shared by partial bases, e.g. 203 genera for 1 base, 129 by 2 bases, 104 and 138 by three and four planting bases respectively. The 203 special genera were unevenly distributed in the different bases, with 61 genera in ZJ, 59 in GZ, 42 in GX, 35 in FJ and 6 in YN, respectively. Thirdly, the abundances of genera in common were unevenly distributed in the different planting bases. Taking the 25 dominant genera with more than 1% reads in at least one base as an example, Delftia (10.39–38.42%), Burkholderia (2.71–15.98%), Escherichia/Shigella (4.90–25.12%), Pseudomonas (2.68–30.72%) and Sphingomonas (1.83–2.05%) dominated in four of the planting bases except for YN. For the YN sample, the dominant genera only included Pseudomonas (57.45%) and Delftia (2.72%). In addition, each base had its own dominant genera. There were 7 genera that dominated in the YN sample, e.g. Aeromonas 4.82%, and only 1–3 genera in each of the other four bases, with the prominence of Pantoea (9.29%) and Buttiauxella (6.58%) in GX, and Methylobacterium (17.02%) in GZ (Table 3). Overall, the YN sample exhibited unique bacterial composition, while the other four sample locations shared great similarities as well as some differences in their bacterial compositions.

Diversity between tissue types.

Interestingly, more phyla were identified in leaves (20) than in stems (16) or in roots (13). Among these phyla, 10 phyla were shared by all the tissues, while 7 were shared by two tissues and 5 existed in unique tissues. Of the 10 phyla in common, 4 were dominant, such as: Proteobacteria, Bacteroidetes, Firmicutes, and Actinobacteria, which constituted over 98% of the total reads (S4 Fig). Proteobacteria was the most dominant, with 82–88% of the reads in all the tissues; however, the second-most dominant genera was Bacteroidetes in the root (4.83%) and leaf (7.12%) samples, and Firmicutes was the third, with 7.61% reads in the stem. Lentisphaerae, OD1, Deinococcus-Thermus and Fibrobacteres existed only in the root and stem samples, Spirochaetes was present only in the roots and stems, and Chlamydiae and Armatimonadetes were found only in root and leaf samples.

On the genus level, the dominant genera exhibited much greater variability with tissue type. There were 495–506 genera in each tissue, 319 of which existed in all the tissues, while there were 45–61 in every two tissue types, and 70–78 in unique tissues. The 319 genera in common, which constituted 45.64% of the total reads, comprised the remainder of all the genera with over 0.5% reads. There were 10, 10 and 11 genera with over 1% reads in the root, stem, and leaf tissues, respectively. The dominant genera Pseudomonas (17.94–22.06%), Escherichia/Shigella (6.69–11.59%), Delftia (9.65–22.14%) and Burkholderia (3.12–11.05%) also dominated in all the tissues, with some variation between tissues. The other dominant genera usually dominated in some unique tissues, e.g. Methylobacterium (11.22%) dominated in the leaf, Pantoea (6.05%) in the stem, while Duganella (3.65%) and Buttiauxella (3.60%) were dominant in the root tissues (Table 3). These dominant genera contained most of the bacteria in each tissue, with 65.41%, 67.56% and 70.25% of the reads in the root, stem and leaf samples, respectively. The proportion of abundant and rare species ranged from 48.60% to 60.86% and 39.14% to 51.40% respectively in the D. catenatum samples (S5 Fig).

Functional diversity of diazotrophic endophytes

To investigate the diversity of diazotrophs in D. catenatum, metagenome analysis based on the nifH gene was carried out. The rarefaction curves (S1B Fig) all approach a plateau, indicating that further sequencing would not result in more OTUs in each sample. As shown in S2 Table, a total of 4 diazotroph phyla were identified, and Proteobacteria was the absolute dominant phylum, with Alphaproteobacteria (61.47%) and Betaproteobacteria (22.47%) as the leading classes. On the genus level, 45 genera were altogether identified, with 13 genera with over 1% reads constituting 91.75% of the total, while the remaining 32 genera constituted only 1.98% of the total. Among these dominant genera, most of them belonged to Alphaproteobacteria, such as: Bradyrhizobium (14.85%), Methylovirgula (10.68%), Xanthobacter (8.91%), Sphingomonas (7.13%), Azospirillum (5.53%), Mesorhizobium (5.45%), Methylocystis (4.56%), and Beijerinckia (3.74%), which constituted 60.85% of the total reads. In addition, 3 genera, namely, Burkholderia (13.19%), Azohydromonas (6.12%), and Herbaspirillum (2.83%), were divided into Betaproteobacteria, which constituted 22.14% of the total reads. The other two genera, Micrococcus (6.46%) and Enterobacter (2.30%), belonged to Actinobacteria and Gammaproteobacteria, respectively (Fig 3). According to these results, the potential for nitrogen fixation within the bacterial community is substantial.

thumbnail
Fig 3. The genus level (meta-nifH) of species composition of samples from five planting bases.

https://doi.org/10.1371/journal.pone.0184717.g003

Diversity indices between planting bases.

To determine the effect of environment on the nitrogen-fixers, the diversity between planting bases was compared. As shown in S2 Table, the predominant phylum Proteobacteria accounted for a higher percentage (98.99%) in the samples of ZJ, GX and GZ, and a lower percentage in both FJ (66.14%) and YN (69.84%). For the latter two samples, 33.84% reads belonged to unidentified phyla and 30.00% reads belonged to Actinobacteria, respectively. Interestingly, Firmicutes was observed only in the ZJ, GX, and YN samples with low ratios.

On the genus level, some differences existed between the samples from the different planting bases. Firstly, the numbers of genera were different. Sample YN had the maximum of 40 genera, followed by FJ (33), ZJ and GZ (25), and GX had the least number, with 22 genera. Secondly, the types of genera contained by each base were different. Except for 17 genera shared by all the planting bases, there were 28 genera shared by partial bases, although 26 of those belonged to rare genera with less than 1% of the reads. There were 11 genera for 1 base, 7 for 2 bases, and 5 for both 3 and 4 planting bases, respectively. The 11 special genera were unevenly distributed in different bases, with 6 genera in YN, 3 in FJ, and 1 in both ZJ and GX, respectively. Thirdly, the ratios of the 17 common genera were unevenly distributed in different planting bases (Fig 3). The dominant Burkholderia and Bradyrhizobium were the most popular genera in four planting bases, followed by Methylovirgula and Mesorhizobium. Four genera (Beijerinckia, Xanthobacter, Sphingomonas and Azohydromonas) dominated in two planting bases, and the other four (Micrococcus, Azospirillum, Methylocystis, and Herbaspirillum) dominated in unique bases.

Diversity indices between tissue types.

Comparing the results between different tissues, some differences were observed on the phylum level. In the root samples, the dominant phyla were: Proteobacteria (95.18%), Actinobacteria (4.68%), and Firmicutes (0.09%). The stem samples had the same phyla with different ratios, with much less Proteobacteria (69.46%) and more Actinobacteria (13.72%). However, Proteobacteria absolutely dominated in the leaves (99.39%), with few Actinobacteria (0.04%).

On the genus level, some similarity existed between the samples. Firstly, the numbers of genera were similar between the samples, with 42 for root, 25 for stem and 38 for leaf samples. Secondly, the types of genera were similar among the different tissues. Altogether, there were 24 genera shared by all the tissues, with the only exception of 12 genera shared by root-stem and root-leaf samples (Fig 4), respectively. In the case of the roots, there were nine genera with over 1% of the reads, including six dominant genera, which were Methylovirgula (26.85%), Bradyrhizobium (28.46%), Burkholderia (16.77%), Mesorhizobium (7.07%), Xanthobacter (5.19%), and Micrococcus (4.68%). There were eight genera with over 1% of the reads for the stems, including six dominant genera, which were Burkholderia (19.41%), Azohydromonas (16.40%), Azospirillum (14.74%), Micrococcus (13.73%), Herbaspirillum (6.69%), and Beijerinckia (4.81%). For the leaves, there were ten genera with over 1% reads. The dominant genera were Xanthobacter (22.21%), Sphingomonas (21.75%), Bradyrhizobium (15.35%), Methylocystis (11.70%), Methylovirgula (6.50%), Beijerinckia (5.99%), and Mesorhizobium (5.35%) (Fig 4). It is evident that, with the exception of Mesorhizobium which dominated in all the tissues, and Beijerinckia, Xanthobacter, and Burkholderia dominating in two types of tissues, the other 9 genera dominated in unique tissues. Thus, the results indicate that there exists high similarity regarding genera type, with some differences in terms of the genera abundances.

thumbnail
Fig 4. Venn diagram (A) and bar chart (B) for the genus levels of species composition of three tissue samples.

https://doi.org/10.1371/journal.pone.0184717.g004

Comparison between nifH and 16S rDNA results

To compare the results between 16S rDNA and nifH methods, a Venn diagram was constructed to show their relationship. As shown in Fig 5, 39 out of 45 genera identified from meta-nifH were also contained in the results based on meta-16S rRNA. However, six genera, including Azohydromonas (6.12%), Azorhizobium (0.12%), Roseiflexus (0.0015%), Rhodovulum (0.0003%), Halorhodospira (0.0008%), and Pelobacter (0.0002%), were unexpectedly excluded from the meta-16S rDNA results. When comparing the abundances of given genera, only 2 genera (Burkholderia and Sphingomonas) belong to the dominant bacteria with over 1% ratio in the results obtained using both methods (S3 Table). Some dominant genera from meta-nifH, e.g. Methylovirgula, Bradyrhizobium, Methylocystis and Mesorhizobium, displayed low ratios in the meta-16S rDNA results. Three dominant genera, including Azohydromonas, Xanthobacter and Micrococcus, were not detected through meta-rDNA. Overall, these observations indicate good consistency in nitrogenase gene diversity and microbial community structure from the cross-system comparison.

thumbnail
Fig 5. Venn diagram of genus level of species diversity comparison between 16S rDNA and nifH methods.

https://doi.org/10.1371/journal.pone.0184717.g005

Discussion

It is well known that transplants of different species are difficult to culture without bacteria, hinting at the important ecological roles that bacteria play in plant growth [32]. As epiphytes or lithophytes, orchids principally rely on symbiotic microorganisms for their requisite nutrients. Recently, attention has been paid to the roles of certain bacteria as “satellite” bacteria or mycorrhiza helper bacteria for the promotion of both ectomycorrhiza and the host plant [33]. The roles of orchid-associated bacteria reveal a novel field of symbiosis between microorganisms and orchids.

In this study, the bacterial and diazotrophic diversities in D. catenatum were successfully characterized based on 16S rRNA and nifH genes. Altogether 699 bacterial genera within 22 phyla and 45 diazotrophic genera belonging to 4 phyla were identified from D. catenatum. The results showed that bacterial and diazotrophic diversities of endophytic bacteria varied with both the tissue types and habitats of D. catenatum, which needs further evidence.

Diversity of bacteria in D. catenatum

Based on the meta-16S rRNA, 22 bacterial phyla were in total identified; 12–17 of these which were distributed in each of the planting bases, and 13–20 in each tissue type. On the genus level, there were in total 699 genera identified from D. catenatum; 125 of these genera were shared by all the planting bases, and 319 existed in all the tissue types. The occurrence of endophytic bacteria is the result of a combination of several chance factors, determined by the chances of developing roots coming into contact with effective levels of bacteria that can become endophytic, and deterministic factors, determined by the presence of dedicated genetic systems that enable bacterial-plant crosstalk and an active endophytic colonization process [34].

Genetics are a vital factor shaping an endophytic community [35]. In the case of D. catenatum, within the predominant Proteobacteria (71.81–96.08%), Delftia (10.39–38.42%), Pseudomonas (2.68–30.72%), Burkholderia (2.71–15.98%), Escherichia/Shigella (4.90–25.12%) and Sphingomonas (1.83–2.05%) dominated in four of the planting bases with the exception of YN, which was dominated by Pseudomonas (57.45%), and Delftia (2.72%). In fact, Pseudomonas, Burkholderia and Sphingomonas are widely distributed in other Dendrobium plants, Orchidaceae and other habitat environments. For example, Pseudomonas was reported to exist in dendrobium plants, such as Dendrobium moschatum and Dendrobium crumenatum [36, 37], and many orchid plants, including epipactis species [38], Dactylorhiza maculate [39], Paphiopedilum appletonianum [40], Acampe papillosa [41] and Phalaenopsis spp. [42]. In summary, Pseudomonas is ubiquitous in plants, playing beneficial roles through participation in the carbon and nitrogen cycles and phosphate solubilization [43, 44]. Similar distributions and roles can be found for Burkholderia and Sphingomonas, as mentioned in the following section.

Delftia and Escherichia/Shigella have considerable abundances only in D. catenatum. Usually, the two genera belong to opportunistic pathogens widespread in many habitats, e.g. river [45], soil [46], human gut [47] and plant (Dendrobium aurantiacum) [48, 49]. In recent years, Shigella was identified as a PGPR [50], and Delftia has been found to promote plant growth through suppressing fungal phytopathogens [51], and transforming or degrading multiple organic and inorganic toxins [52, 53]. Regardless of the fact that they are abundant bacteria in D. catenatum but less so in other plants, it is necessary to test the safety and functions of Delftia and Escherichia/Shigella in medicinal plants, which are usually freshly eaten.

The environment is another important factor influencing the bacterial communities. In this study, some genera showed plantinging base-specificity (3.53% Afipia in ZJ, 3.49% Pectobacterium in YN and 2.42% Streptococcus in GZ), some displayed tissue-type specificity (3.23% Salinarimonas in the leaves, 2.02% Enterobacter in the stem), while others showed specificity or dominance in certain tissues from some locations (for example, 17.02% Methylobacterium in GZL, 9.29% Pantoea in GXS, 6.58% Buttiauxella in GXR, 6.88% Duganella in YNR and 4.82% Aeromonas in YNS). Members of Methylobacterium are ubiquitous in leaves, such as those of Platanus orientalis, maize, cotton, sunflower, soybean, clover, winter wheat and rice [5457]. Until now, although no community member of the Methylobacterium population was found to colonize only one plant species exclusively [58, 59], the most abundant genera in the leaves of the present study were Methylobacterium, consisting of the many species such as M. radiotolerans, M. mesophilicum, and M. fujisawaense [60, 61]. Pantoea not only was detected in the stem of D. catenatum, but also in the stem of sweet potato and willow [62]. While Buttiauxella was rarely isolated from roots in previous reports [63], it is possible that the Buttiauxella exists specifically in D. catenatum roots. The base or tissue-specific bacteria displayed the effect of environment on microbial communities. The determination of which resources and functions are important environmental factors requires further study.

In addition to the apparent conspicuous regional characteristics of endophytic bacteria, we also found an interesting phenomenon whereby compared to abundant species, the percentage of rare species determined is also relatively and unusually large, even significantly higher than the abundant species in the YN and GZ samples (S4 Fig). It has long been a belief that abundant species that are represented by perhaps only a few species in the bacterial population play an important role in biotic and abiotic environments. However, the present report provides a testament that rare species also play an important role in biogeochemical cycles and other functional roles in the environment [64]. In addition, the rare species which are unique to a particular sample may play important roles other than those previously predicted. In view of the special efficacy of D. catenatum, further research regarding the importance of such rare endogenous species is needed to fully understand their role in D. catenatum.

Diversity of diazotrophs in D. catenatum

The abundant bacteria may play a vital role in the growth of the plants. As epiphytic or lithophytic Dendrobium, nitrogen fixation might be one of the most important functions of these bacteria in D. catenatum. Expectedly, 4 phyla and 45 genera of diazotrophs were identified, 17 of which were shared by all the planting bases and 24 shared by all the tissue types. Among the planting bases, Burkholderia and Bradyrhizobium were the most popular, followed by Mesorhizobium and Methylovirgula, and Mesorhizobium was the most popular in the different tissues, followed by Beijerinckia, Xanthobacter, and Burkholderia. These species might be a good alternative source of nitrogen nutrition, especially for the naturally-grown plants.

Burkholderia and Mesorhizobium were both encountered in high abundance in all planting bases and tissues, suggesting that they might play a leading role in the microecosystem of D. catenatum. As additional evidence, we obtained many Burkholderia strains from D. catenatum, which exhibited nitrogen-fixation and thus could promote the growth of D. catenatum seedlings (data not shown). In fact, Burkholderia strains are widely distributed in many plants, including maize [65], rice [66], citrus [67], yellow lupine [68], grape [69], banana and pineapple [70], and display nitrogen-fixing activity in plants [71, 72] which grow well in nutrient-poor sites. In the case of Mesorhizobium, it has been successfully used worldwide to permit an effective establishment of the nitrogen-fixing symbiosis with leguminous crop plants [73], and Caragana species in China [74]. In addition, Mesorhizobium has shown multiple functions for improving plant growth, such as producing IAA [75], and solubilizing phosphate [76]. Bradyrhizobium and Beijerinckia have also shown similar characteristics on non-legume plants [77].

Other dominant diazotrophs, such as Sphingomonas, Methylovirgula, and Xanthobacter, are widely distributed in other plants. Sphingomonas constituted the dominant endophytic group in D. catenatum, and is also present in other orchids. Nitrogen-fixing abilities have been confirmed in Sphingomonas bacteria associated with rice [78] and Oryza sativa [79], and the acetylene was transformed into ethylene [80]. We isolated an endophytic Sphingomonas SH1 from D. catenatum, which promoted the growth of D. catenatum through production of phytohormones and nitrogen fixation [81]. Methylovirgula and Xanthobacter are found in water, sediment [82], rice rhizosphere, marigold plants and tree leaves [83]. Interestingly, Methylovirgula and Xanthobacter belong to both diazotrophic and methylotrophic bacteria [84] in different plants [85]. Methylovirgula were specialized to utilize methanol as their sole carbon source [86], and the abundance of Methylovirgula correlated significantly and negatively with the C/N ratio [87]. Xanthobacter flavus grows autotrophically by using the Calvin cycle for the fixation of CO2 [88], and Xanthobacter autotrophicus has the ability to grow on H2/CO2, ketones, alcohols, sugars, carboxylic acids, and aliphatic alkenes [89]. Methylotrophy was reported to be associated with lower plants [90], and it became apparent that extra- and/or intra-cellular symbiotic or mutualistic associations may exist between plants and some methylotrophic strains to make them well suited for survival in stressful environments. Methylovirgula and Xanthobacter reveal a rich distribution of methylotrophic and diazogtrophic bacteria in D. catenatum, the roles of which require further investigation.

Previously, cyanobacteria were found to be the first to show a high activity of nitrogen fixation on orchids. Tsavkelova first studied the localization of phototrophic microorganisms on the roots of Dendrobium moschatum, Acampe papillosa, and Phalaenopsis amabilis [91]. Nostoc, Anabaena, and Calothrix were distributed on the surface of the A. papillosa aerial roots, whereas Nostoc, Oscillatoria, and representatives of the LPP-group (Lyngbia, Phormidium, and Plectonema, incapable of nitrogen fixation) were found on the substrate roots [92]. The functional characteristics and primary role of the orchid cyanobacteria are those of nitrogen fixation and nutrient supply to both the host plant and the “satellite” microorganisms, including mycorrhizae-forming fungi.

Therefore, the diazotrophs found in this study, as well as cyanobacteria, might be vital sources of nutrient nitrogen for the growth of D. catenatum, providing the possibility for them to live as epiphytics or lithophytics under wild conditions.

Difference between bacteria and diazotrophs in D. catenatum

The results of the current study indicate that the two methods were successful for performing genetic and nitrogen-fixing functional diversity analysis. Comparing the results of the two methods, all phyla and 39 genera were completely overlapped with those based on the 16S rRNA gene, which demonstrated the consistence of the taxonomies based on both the 16S rRNA and nifH genes. It further demonstrates the successful use of the nifH gene as both a taxonomic and functional marker and target gene in the study of microbial communities.

However, 6 diazotroph genera (Azohydromonas 6.12%, Azorhizobium 0.12%, Roseiflexus 0.0015%, Rhodovulum 0.0003%, Halorhodospira 0.0008%, and Pelobacter 0.0002%) were not covered in the results based on the 16S rRNA gene. There are three possible reasons for this omission. The first reason is the selectivity of the primers. As mentioned before [18], primer pairs targeting the 16S rRNA gene are useful for amplifying bacteria with the exceptions of cyanobacteria and archae. Thus, this explains why no cyanobacteria were detected here; however, it does not explain the additional genus based on the nifH gene. The second reason might be a failure of the taxonomy of the nifH gene due to horizontal transfer [93], although most of the time it can be used successfully to characterize bacteria. It is reported that some strains displayed two different genera on the basis of the rRNA gene and nifH gene [94, 95]. The nifH gene from horizontal transfer might confuse or enlarge the diversity of bacteria. Some of the above genera might be attributed to horizontal transfer, which needs further evidence to substantiate. Thirdly, the low ratio of diazotrophs, such as those with less than 0.001% ratio, might influence the competitive amplification of the corresponding 16S rRNA gene. This hypothesis is supported by previous studies that indicated that RNA-targeted probing methods cannot detect target species when numbers are <0.01% of all bacteria present [64, 96]. Therefore, the nifH gene is a good alternative target gene for investigating the majority of diazotrophs and their community in plants.

In conclusion, the 16S rRNA and nifH genes were successfully used to investigate the diversity of bacteria and diazotrophs in D. catenatum. Abundant bacteria and diazotrophs were identified in different tissues of D. catenatum from five different planting bases, many of which were found to be common to the various tissues and planting bases. These bacteria might play a significant role in the provision of nitrogen and other nutrients for the growth of D. catenatum, which needs further experimental evidence.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors. All the samples obtained from the planting bases permitted by the owner, one of the co-auther. No endangered or protected species were involved in this study.

Supporting information

S2 Table. The proportion of phylum level from different bases (nifH).

https://doi.org/10.1371/journal.pone.0184717.s002

(XLSX)

S3 Table. Genus level of species diversity comparison between 16S rDNA and nifH.

https://doi.org/10.1371/journal.pone.0184717.s003

(XLSX)

S1 Fig. Shannon curves for 16S rDNA (A) and nifH (B) genes.

Samples were collected at 5 sites (ZJ, Zhejiang; FJ, Fujian; GX, Guangxi; YN, Yunnan; GZ, Guizhou).

https://doi.org/10.1371/journal.pone.0184717.s004

(TIF)

S2 Fig. NMDS plot of the D. catenatum samples.

Most of the samples do not cluster together indicating that the bacterial diversity among the samples is varied and distinct.

https://doi.org/10.1371/journal.pone.0184717.s005

(TIF)

S3 Fig. Relative abundance of bacterial phyla across the samples.

The percentage of sequences is plotted on the Y-axis. Proteobacteria is the predominant phylum in all the samples.

https://doi.org/10.1371/journal.pone.0184717.s006

(TIF)

S4 Fig. The phylum (A) and genus (B) level of bacterial community composition of the three tissues.

https://doi.org/10.1371/journal.pone.0184717.s007

(TIF)

S5 Fig. Percentage of abundant and rare species across the samples.

In all the five sites when compared to abundant species, the percentage of rare species is significantly high indicating the importance of rare species in a community.

https://doi.org/10.1371/journal.pone.0184717.s008

(TIF)

Acknowledgments

The research is supported by National Natural Science Foundation of China (81673537), Zhejiang Natural Science Foundation of China (LY16C030002 & LY15C010005), Key research and development projects of Zhejiang Province (2015C02030), and Zhejiang Province Public Agricultural Project (2014C32117). We are grateful to Dr. Greg Duns for his valuable comments on the manuscript.

References

  1. 1. Jones W, Kuehnle A, Arumuganathan K. Nuclear DNA Content of 26 Orchid (Orchidaceae) Genera with Emphasis on Dendrobium. Ann Bot. 1998;82(2):189–94.
  2. 2. Ma YN, Hou ZN, Liang ZS, Liu L. Quality evaluation of the rare medicinal plant Dendrobium officinale based on volatile constituents, methanol extracts and polysaccharides. Curr Pharm Anal. 2017;13:
  3. 3. China Pharmacopoeia Committee. Pharmacopoeia of the People's Republic of China. China Chemical Industry Press: Beijing, 2005.
  4. 4. Brundrett MC. Understanding the roles of multifunctional mycorrhizal and endophytic fungi. In: Schulz BJE, Boyle CJC, Sieber TN, editors. Microbial Root Endophytes. Berlin: Springer-Verlag; 2006. pp. 281–293.
  5. 5. Smith SE, Read DJ. Mycorrhizal Symbiosis. 2nd ed. London: Academic Press; 1997.
  6. 6. Hou XQ, Guo SX. Interaction between a dark septate endophytic isolate from Dendrobium sp. and roots of D. nobile seedlings. J Integr Plant Biol. 2009;51(4):374–81. pmid:21452589
  7. 7. Meyer JR, Linderman RG, Meyer JR, Linderman RG. Response of subterranean clover to dual inoculation with vesicular-arbuscular mycorrhizal fungi and a plant growth-promoting bacterium, Pseudomonas putida. Soil Biol Biochem. 1986;18(2):185–90.
  8. 8. Bowen GD, Theodorou C. Interactions between bacteria and ectomycorrhizal fungi. Soil Biol Biochem. 1979;11(2):119–26.
  9. 9. Xavier LJC, Germida JJ. Bacteria associated with Glomus clarum spores influence mycorrhizal activity. Soil Biol Biochem. 2003;35(3):471–8.
  10. 10. Frey-Klett P, Garbaye J. Mycorrhiza helper bacteria: A promising model for the genomic analysis of fungal-bacterial interactions. New Phytol. 2005;168(1):4–8. pmid:16159316
  11. 11. Mehnaz S, Lazarovits G. Inoculation effects of Pseudomonas putida, Gluconacetobacter azotocaptans, and Azospirillum lipoferum on corn plant growth under greenhouse conditions. Microb Ecol. 2006;51(3):326–35. pmid:16598630
  12. 12. Rai AN, Soderback E, Bergman B. Cyanobacterium-plant symbioses. New Phytol. 2000;147(3):449–81.
  13. 13. Shendure J, Ji H. Next-generation DNA sequencing. Nat Biotechnol. 2008;26(10):1135–45. pmid:18846087
  14. 14. Qian PY, Wang Y, Lee OO, Lau SC, Yang J, Lafi FF, et al. Vertical stratification of microbial communities in the Red Sea revealed by 16S rDNA pyrosequencing. ISME J. 2011;5(3):507–18. pmid:20668490
  15. 15. Claesson MJ, O'Sullivan O, Wang Q, Nikkilä J, Marchesi JR, Smidt H, et al. Comparative analysis of pyrosequencing and a phylogenetic microarray for exploring microbial community structures in the human distal intestine. PLoS One. 2009;4(8):e6669. pmid:19693277
  16. 16. Zhou J, Kang S, Schadt CW, Garten CT. Spatial scaling of functional gene diversity across various microbial taxa. Proc Natl Acad Sci U S A. 2008;105(22):7768–73. pmid:18509054
  17. 17. Doyle J. Isolation of plant DNA from fresh tissue. Focus. 1990;12:13–5.
  18. 18. Yu J, Zhou XF, Yang SJ, Liu WH, Hu XF. Design and application of specific 16S rDNA-targeted primers for assessing endophytic diversity in Dendrobium officinale using nested PCR-DGGE. Appl Microbiol Biotechnol. 2013;97(22):9825–36. pmid:24127138
  19. 19. Xiao CH, Tang H, Pu LJ, Sun DM, Ma JZ, Yu M, et al. Diversity of nitrogenase (nifH) genes pool in soybean field soil after continuous and rotational cropping. J Basic Microbiol. 2010;50(4):373–9. pmid:20473958
  20. 20. Poly F, Monrozier LJ, Bally R. Improvement in the RFLP procedure for studying the diversity of nifH genes in communities of nitrogen fixers in soil. Res Microbiol. 2001;152(1):95–103. pmid:11281330
  21. 21. Sogin ML, Morrison HG, Huber JA, Mark WD, Huse SM, Neal PR, et al. Microbial diversity in the deep sea and the underexplored "rare biosphere". Proc Natl Acad Sci U S A. 2006;103(32):12115–20. pmid:16880384
  22. 22. Huse SM, Welch DM, Morrison HG, Sogin ML. Ironing out the wrinkles in the rare biosphere through improved OTU clustering. Environ Microbiol. 2010;12(7):1889–98. pmid:20236171
  23. 23. Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J, et al. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res. 2007;35(21):7188–96. pmid:17947321
  24. 24. Bodenhausen N, Horton MW, Bergelson J. Bacterial communities associated with the leaves and the roots of Arabidopsis thaliana. PLoS One. 2013;8(2):e56329. pmid:23457551
  25. 25. Wang Q, Garrity GM, Tiedje JM, Cole JR. Naïve Bayesian Classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol. 2007;73(16):5261–7. pmid:17586664
  26. 26. Hong SH, Bunge J, Jeon SO, Epstein SS. Predicting microbial species richness. Proc Natl Acad of Sci U S A. 2006;103(1):117–22.
  27. 27. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, et al. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol. 2009;75(23):7537–41. pmid:19801464
  28. 28. Lozupone C, Knight R. UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol. 2006;71(12):8228–35.
  29. 29. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7(5):335–6. pmid:20383131
  30. 30. Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26(19):2460. pmid:20709691
  31. 31. Mcdonald D, Clemente JC, Kuczynski J, Rideout JR, Stombaugh J, Wendel D, et al. The biological observation matrix (BIOM) format or: how I learned to stop worrying and love the ome-ome. GigaScience. 2012;1(1):1–6.
  32. 32. Leifert C, Waites WM, Nicholas JR. Bacterial contaminants of micropropagated plant cultures. J Appl Microbiol. 1989;67(4):353–61.
  33. 33. Tsavkelova E. Bacteria Associated With Orchid Roots. In: Maheshwari DK editor. Bacteria in Agrobiology: Plant Growth Responses. Berlin: Spinger; 2011. pp. 221–259.
  34. 34. Reinholdhurek B, Hurek T. Life in grasses: diazotrophic endophytes. Trends Microbiol. 1998;6(4):139–44. pmid:9587190
  35. 35. Bálint M, Tiffin P, Hallström B, O'Hara RB, Olson MS, Fankhauser JD, et al. Host genotype shapes the foliar fungal microbiome of balsam poplar (Populus balsamifera). PLoS One. 2013;8(1):e53987. pmid:23326555
  36. 36. Ja TDS, Tsavkelova EA, Zeng S, Ng TB, Parthibhan S, Dobránszki J, et al. Symbiotic in vitro seed propagation of Dendrobium: fungal and bacterial partners and their influence on plant growth and development. Planta. 2015;242(1):1. pmid:25940846
  37. 37. Sandrasagaran UM, Subramaniam S, Murugaiyah V. New perspective of Dendrobium crumenatum orchid for antimicrobial activity against selected pathogenic bacteria. Pak J Bot. 2014;46(2):719–24.
  38. 38. Jacquemyn H, Lenaerts M, Tyteca D, Lievens B. Microbial diversity in the floral nectar of seven Epipactis (Orchidaceae) species. Microbiologyopen. 2013;2(4):644. pmid:23836678
  39. 39. Shekhovtsova NV, Marakaev OA, Pervushina KA, Osipov GA. The underground organ microbial complexes of moorland spotted orchid Dactylorhiza maculata (L.) Soó (Orchidaceae). Adv Biosci Biotechnol. 2013;04(7):35–42.
  40. 40. Tsavkelova EA, Cherdyntseva TA, Botina SG, Netrusov AI. Bacteria associated with orchid roots and microbial production of auxin. Microbiol Res. 2007;162(1):69–76. pmid:17140781
  41. 41. Tsavkelova EA, Cherdyntseva TA, Netrusov AI. Auxin production by bacteria associated with orchid roots. Microbiology. 2005;74(1):46–53.
  42. 42. Stovold GE, Bradley J, Fahy PC. Acidovorax avenae subsp. cattleyae (Pseudomonas cattleyae) causing leafspot and death of Phalaenopsis orchids in New South Wales. Australas Plant Pathol. 2001;30(1):73–4.
  43. 43. Vyas P, Gulati A. Organic acid production in vitro and plant growth promotion in maize under controlled environment by phosphate-solubilizing fluorescent Pseudomonas. BMC Microbiol. 2009;9(1):174.
  44. 44. Mulet M, Bennasar A, Lalucat J, Garcíavaldés E. An rpoD-based PCR procedure for the identification of Pseudomonas species and for their detection in environmental samples. Mol Cell Probes. 2009;23(3–4):140–7. pmid:19268522
  45. 45. Kenzaka T, Yamaguchi N, Prapagdee B, Mikami E, Nasu M. Bacterial community composition and activity in urban rivers in Thailand and Malaysia. J Health Sci. 2001;47(4):2333–6.
  46. 46. Graff A, Stubner S. Isolation and molecular characterization of thiosulfate-oxidizing bacteria from an Italian rice field soil. Syst Appl Microbiol. 2003;26(3):445–52. pmid:14529188
  47. 47. Robles AV, Guarner F. Linking the gut microbiota to human health. Br J Nutr. 2013;109 Suppl 2:S21–6.
  48. 48. Berg G, Erlacher A, Grube M. The edible plant microbiome: importance and health issues. In Lugtenberg B, editor. Principles of plant-microbe interaction. Cham: Springer. 2015. pp. 419–26.
  49. 49. Vijayabaskar P, Shiyamala V. Antibacterial activities of brown marine algae (Sargassum wightii and Turbinaria ornata) from the gulf of mannar biosphere reserve. Adv Biol Res. 2011;5(2):99–102.
  50. 50. Lu DG, Hou S, Qin SJ, Liu LZ, Ma HY, Liu GC. Preliminary screening of plant growth promoting rhizobacteria of Cerasus sachalinensis and their promotion effects on seedling. Guizhou Agric Sci. 2011;7:169–73.
  51. 51. Prasannakumar SP, Gowtham HG, Hariprasad P, Shivaprasad K, Niranjana SR. Delftia tsuruhatensis WGR-UOM-BT1, a novel rhizobacterium with PGPR properties from Rauwolfia serpentina (L.) Benth. ex Kurz also suppresses fungal phytopathogens by producing a new antibiotic-AMTM. Lett Appl Microbiol. 2015;61(5):460–8. pmid:26258398
  52. 52. Morel MA, Ubalde MC, Braña V, Castrosowinski S. Delftia sp. JD2: a potential Cr(VI)-reducing agent with plant growth-promoting activity. Arch Microbiol. 2011;193(1):63–8. pmid:20857088
  53. 53. Banerjee MR, Yesmin L. Sulfur-oxidizing plant growth promoting rhizobacteria for enhanced canola performance. Unite states patent US 7491535. 2009 Feb 17.
  54. 54. Kang YS, Kim J, Shin HD, Nam YD, Bae JW, Jeon CO, et al. Methylobacterium platani sp. nov., isolated from a leaf of the tree Platanus orientalis. Int J Syst Evol Microbiol. 2008;57:2849–53.
  55. 55. Balachandar D, Raja P, Sundaram S. Genetic and metabolic diversity of pink-pigmented facultative methylotrophs in phyllosphere of tropical plants. Braz J Microbiol. 2008;39:68–73. pmid:24031182
  56. 56. Madhaiyan M, Kim BY, Poonguzhali S, Kwon SW, Song MH, Ryu JH, et al. Methylobacterium oryzae sp. nov., an aerobic, pink-pigmented, facultatively methylotrophic, 1-aminocyclopropane-1-carboxylate deaminase-producing bacterium isolated from rice. 2007;57:326–31.
  57. 57. Knief C, Ramette A, Frances L, Alonso-Blanco C, Vorholt JA. Site and plant species are important determinants of the Methylobacterium community composition in the plant phyllosphere. ISME J. 2010;4(6):719. pmid:20164863
  58. 58. Knief C, Frances L, Cantet F, Vorholt JA. Cultivation-independent characterization of Methylobacterium populations in the plant phyllosphere by automated ribosomal intergenic spacer analysis. Appl Environ Microbiol. 2008;74(7):2218–28. pmid:18263752
  59. 59. Omer ZS, Tombolini R, Gerhardson B. Plant colonization by pink-pigmented facultative methylotrophic bacteria (PPFMs). FEMS Microbiol Ecol. 2004;47(3):319–26. pmid:19712320
  60. 60. Delmotte N, Knief C, Chaffron S, Innerebner G, Roschitzki B, Schlapbach R, et al. Community proteogenomics reveals insights into the physiology of phyllosphere bacteria. Proc Natl Acad Sci U S A. 2009;106(38):16428–33. pmid:19805315
  61. 61. Iguchi H, Yurimoto H, Sakai Y. Interactions of methylotrophs with plants and other heterotrophic bacteria. Microorganisms. 2015;3(2):137–51. pmid:27682083
  62. 62. A CA Jr, Adachi K. Isolation of endophytic diazotroph Pantoea agglomerans and nondiazotroph Enterobacter asburiae from sweet potato stem in Japan. Lett Appl Microbiol. 2004;38(1):19–23. pmid:14687210
  63. 63. Meyer SED, Beuf KD, Vekeman B, Willems A. A large diversity of non-rhizobial endophytes found in legume root nodules in Flanders (Belgium). Soil Biol Biochem. 2015;83:1–11.
  64. 64. Fuhrman JA. Microbial community structure and its functional implications. Nature. 2009;459:193–9. pmid:19444205
  65. 65. Mcinroy JA, Kloepper JW. Survey of indigenous bacterial endophytes from cotton and sweet corn. Plant Soil. 1995;173(2):337–42.
  66. 66. Engelhard M, Hurek T, Reinholdhurek B. Preferential occurrence of diazotrophic endophytes, Azoarcus spp., in wild rice species and land races of Oryza sativa in comparison with modern races. Environ Microbiol. 2000;2(2):131–41. pmid:11220300
  67. 67. Barac T, Taghavi S, Borremans B, Provoost A, Oeyen L, Colpaert JV, et al. Engineered endophytic bacteria improve phytoremediation of water-soluble, volatile, organic pollutants. Nature Biotechnol. 2004;22(5):583–8.
  68. 68. Araújo WL, M W Jr, Aguilar-Vildoso CI, Barroso PA, Saridakis HO, Azevedo JL. Variability and interactions between endophytic bacteria and fungi isolated from leaf tissues of citrus rootstocks. Can J Microbiol. 2001;47(3):229–36. pmid:11315114
  69. 69. Compant S, Duffy B, Nowak J, Clément C, Barka EA. Use of plant growth-promoting bacteria for biocontrol of plant diseases: principles, mechanisms of action, and future prospects. Appl Environ Microbiol. 2005;71(9):4951–9. pmid:16151072
  70. 70. Weber OB, Baldani VLD, Teixeira KRS, Kirchhof G, Baldani JI, Dobereiner J. Isolation and characterization of diazotrophic bacteria from banana and pineapple plants. Plant Soil. 1999;210(1):103–13.
  71. 71. Dalton DA, Kramer S, Azios N, Fusaro S, Cahill E, Kennedy C. Endophytic nitrogen fixation in dune grasses (Ammophila arenaria and Elymus mollis) from Oregon. FEMS Microbiol Ecol. 2004;49(3):469–79. pmid:19712295
  72. 72. Baldani JI, Baldani VL. History on the biological nitrogen fixation research in graminaceous plants: special emphasis on the Brazilian experience. An Acad Bras Cienc. 2005;77(3):549–79. pmid:16127558
  73. 73. Bottomley PJ, Maggard SP. Determination of viability within serotypes of a soil population of Rhizobium leguminosarum bv. trifolii. Appl Environ Microbiol 1990;56(2):533–40. pmid:2407187
  74. 74. Li M, Li Y, Chen WF, Sui XH, Li YJ, Li Y, et al. Genetic diversity, community structure and distribution of rhizobia in the root nodules of Caragana spp. from arid and semi-arid alkaline deserts, in the north of China. Syst Appl Microbiol. 2012;35(4):239–45. pmid:22463809
  75. 75. Ahmad F, Ahmad I, Khan MS. Screening of free-living rhizospheric bacteria for their multiple plant growth promoting activities. Microbiol Res. 2008;163(2):173–81. pmid:16735107
  76. 76. Peix A, Rivas-Boyero AA, Mateos PF, Rodriguez-Barrueco C, MartíNez-Molina E, Velazquez E. Growth promotion of chickpea and barley by a phosphate solubilizing strain of Mesorhizobium mediterraneum under growth chamber conditions. Soil Biol Biochem. 2001;33(1):103–10.
  77. 77. Jarabolorenzo A, Pérezgaldona R, Donatecorrea J, Rivas R, Velázquez E, Hernández M, et al. Genetic diversity of bradyrhizobial populations from diverse geographic origins that nodulate Lupinus spp. and Ornithopus spp. Syst Appl Microbiol. 2003;26(4):611–23. pmid:14666990
  78. 78. Videira SS, Araujo JLSD, Rodrigues LDS, Baldani VLD, Baldani JI. Occurrence and diversity of nitrogen-fixing Sphingomonas bacteria associated with rice plants grown in Brazil. FEMS Microbiol Lett. 2009;293(1):11–9. pmid:19222576
  79. 79. Xie CH, Yokota A. Sphingomonas azotifigens sp. nov., a nitrogen-fixing bacterium isolated from the roots of Oryza sativa. Int J Syst Evol Microbiol. 2006;56(4):889–93.
  80. 80. Hsueh PR, Teng LJ, Yang PC, Chen YC, Pan HJ, Ho SW, et al. Nosocomial infections caused by Sphingomonas paucimobilis: clinical features and microbiological characteristics. Clin Infect Dis. 1998;26(3):676–81. pmid:9524843
  81. 81. Yang S, Zhang X, Cao Z, Zhao K, Wang S, Chen M, et al. Growth-promoting Sphingomonas paucimobilis ZJSH1 associated with Dendrobium catenatum through phytohormone production and nitrogen fixation. Microb Biotechnol. 2014;7(6):611–20. pmid:25142808
  82. 82. Lidstrom-Oconnor ME, Fulton GL, Wopat AE. "Methylobacterium ethanolicum": a syntrophic association of two methylotrophic bacteria. Microbiology. 1983;129(10):3139–48.
  83. 83. Wiegel J. The Genus Xanthobacter. In: Dworkin M, Falkow S, Rosenberg E, Schleifer KH, Stackebrandt E, editors. The Prokaryotes. New York: Springer; 2006. pp. 290–314.
  84. 84. Small FJ, Ensign SA. Carbon dioxide fixation in the metabolism of propylene and propylene oxide by Xanthobacter strain Py2. J Bacteriol. 1995;177(21):6170–5. pmid:7592382
  85. 85. Sahoo K, Dhal NK. Potential microbial diversity in mangrove ecosystems: a review. Indian J Geo-Mar Sci. 2009;38(2):249–56.
  86. 86. Mosin O, Ignatov I. Evolution, metabolism and biotechnological usage of methylotrophic microorganisms. Eur J Mol Biotechnol. 2014;5(3):131–48.
  87. 87. Hoppe B, Krüger D, Kahl T, Arnstadt T, Buscot F, Bauhus J, et al. Erratum: A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies. Sci Rep. 2016;6:10498. pmid:27162103
  88. 88. VDB Er, Baker SC, Raggers RJ, Terpstra P, Woudstra EC, Dijkhuizen L, et al. Primary structure and phylogeny of the Calvin cycle enzymes transketolase and fructosebisphosphate aldolase of Xanthobacter flavus. J Bacteriol. 1996;178(3):888–93. pmid:8550527
  89. 89. Larsen RA, Wilson MM, Guss AM, Metcalf WW. Genetic analysis of pigment biosynthesis in Xanthobacter autotrophicus Py2 using a new, highly efficient transposon mutagenesis system that is functional in a wide variety of bacteria. Arch Microbiol. 2002;178(3):193–201. pmid:12189420
  90. 90. Corpe WA. Methanol-utilizing bacteria associated with green plants. Dev Ind Microbiol. 1982;23:483–494.
  91. 91. Tsavkelova EA, Lobakova ES, Kolomeitseva GL, Cherdyntseva TA, Netrusov AI. Localization of associative cyanobacteria on the roots of epiphytic orchids. Microbiology. 2003;72(1):86–91.
  92. 92. Tsavkelova EA, Lobakova ES, Kolomeĭtseva GL, Cherdyntseva TA, Netrusov AI. Associative cyanobacteria isolated from the roots of epiphytic orchids. Microbiology. 2003;72(1):92–7.
  93. 93. Bolhuis H, Severin I, Confurius-Guns V, Wollenzien UI, Stal LJ. Horizontal transfer of the nitrogen fixation gene cluster in the cyanobacterium Microcoleus chthonoplastes. ISME J. 2009;4(1):121–30. pmid:19741736
  94. 94. Nouioui I, Ghodhbane-Gtari F, Beauchemin NJ, Tisa LS, Gtari M. Phylogeny of members of the Frankia genus based on gyrB, nifH and glnII sequences. Antonie van Leeuwenhoek. 2011;100(4):579–87. pmid:21713368
  95. 95. Zehr JP, Jenkins BD, Short SM, Steward GF. Nitrogenase gene diversity and microbial community structure: a cross-system comparison. Environ Microbiol. 2003;5(7):539–54. pmid:12823187
  96. 96. Koike S, Kobayashi Y. Development and use of competitive PCR assays for the rumen cellulolytic bacteria: Fibrobacter succinogenes, Ruminococcus albus and Ruminococcus flavefaciens. FEMS Microbiol Lett. 2001;204(2):361–6. pmid:11731149