Enrichment of soil rare bacteria in root by an invasive plant Ageratina adenophora

Science of the Total Environment 683 (2019) 202–209

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Enrichment of soil rare bacteria in root by an invasive plant Ageratina adenophora Lin Chen a,b,c, Kai Fang a,b,c, Jie Zhou a,b, Zhi-Ping Yang a,b, Xing-Fan Dong a,b, Guang-Hui Dai d, Han-Bo Zhang a,b,c,⁎ a

State Key Laboratory for Conservation and Utilization of Bio-Resources in Yunnan, Yunnan University, Kunming 650091, China School of Life Sciences, Yunnan University, Kunming 650091, China School of Ecology and Environmental Science, Yunnan University, Kunming 650091, China d Institute of Ecology and Geobotany, School of Ecology and Environment Science, Yunnan University, Kunming 650091, China b c

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Ageratina adenophora selectively accumulate bacteria from soil rare bacteria. • Bacteroidetes dominant in rhizobacteria but Proteobacteria dominant in endophytes. • The accumulated bacteria are different for plant growth at the strain level.

a r t i c l e

i n f o

Article history: Received 24 January 2019 Received in revised form 15 May 2019 Accepted 15 May 2019 Available online 18 May 2019 Editor: Charlotte Poschenrieder Keywords: Ageratina adenophora Enrichment Rare bacteria Rhizosphere Root endosphere

a b s t r a c t The assembly of the root-associated microbiome provides mutual benefits for the host plant and bacteria in soils. It is interesting how invasive plants interact with the local soil microbial community and establish the soil bacterial community in the endosphere of these plants in the short term. In this study, we compared the bacterial community in the rhizosphere with that in the root endosphere of an invasive plant, Ageratina adenophora, using high-throughput sequencing. The results showed that the roots of A. adenophora selectively accumulated the genera Clostridium and Enterobacter, which are rarely distributed in the rhizosphere. This selective accumulation caused a switch in the bacterial composition at the phylum level from Bacteroidetes predominant in the rhizosphere to Proteobacteria dominant in the root endosphere of A. adenophora. Our data indicated the potential existence of a highly conserved signal recognition in which hosts, either invasive or native, enrich the endosphere bacteria, such as Clostridium, Enterobacter, etc., from the rhizosphere. Moreover, the accumulated bacteria were physiologically and genetically different at the strain level and displayed distinct roles in growth between invasive and native plants. The assembly of the bacterial community in the roots may be an advantageous strategy for A. adenophora in competition with native plants. © 2019 Elsevier B.V. All rights reserved.

1. Introduction ⁎ Corresponding author at: School of Life Sciences, Yunnan University, Kunming 650091, China. E-mail address: [email protected] (H.-B. Zhang).

https://doi.org/10.1016/j.scitotenv.2019.05.220 0048-9697/© 2019 Elsevier B.V. All rights reserved.

Plants grow in soil, where they are exposed to a high abundance and diversity of microbes (Tringe et al., 2005). Both plants and microbes are

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209

adapted to exploit their close association for mutual benefit (Zhang et al., 2009). In this symbiotic system, microbes acquire the photosynthetic products of the host from the root exudates and provide the products required by the host, e.g., phytohormones and nutrients (Zhang et al., 2017). Therefore, the assembly of the root-associated microbiome is essential for plant-microbe interactions through soil resource partitioning (Fitzpatrick et al., 2018). Historically, it has been a mystery how plants assemble endosphere bacteria from the rhizosphere soil. Recently, a three-step enrichment model was developed (Bulgarelli et al., 2012; Lundberg et al., 2012; Edwards et al., 2015). In brief, the microbial inoculum pool in bulk soils is enriched near the roots along with the general gradients of carbon concentration, phytochemicals, etc. Then, a more specialized community is further accumulated on the rhizoplane, and finally, certain microorganisms enter the root surface and establish as the endosphere microbiome (Reinhold-Hurek et al., 2015). This model suggested that, due to selective accumulation, there should be a great difference between two microbial compartments: microbes surrounding the roots and those within the roots. This fact has been verified by studies from several plant species, including Arabidopsis, Populus, maize, and rice (Bulgarelli et al., 2012; Lundberg et al., 2012; Peiffer et al., 2013; Schlaeppi et al., 2013; Shakya et al., 2013; Reinhold-Hurek et al., 2015). For example, studies on the root microbial communities across 30 angiosperm species revealed the large differences in diversity and composition between the endosphere and the rhizosphere (Fitzpatrick et al., 2018). Bulgarelli et al. (2012) reported that Arabidopsis thaliana was preferentially colonized by Proteobacteria, Bacteroidetes and Actinobacteria in the endosphere, whereas Planctomycetes, Acidobacteria, and Proteobacteria were dominant in the rhizosphere. As new arrivals in their introduced habitats, invasive plants unavoidably encounter and interact with the local microbial community (Hawkes et al., 2005), which has been considered to play an important role in the invasion process of these plants (Dawson and Schrama, 2016). Callaway et al. (2004) reported that soil microbes from the home range of the invasive plant Centaurea maculosa showed stronger inhibitory effects on its growth than the microbes from soils the weed has invaded in North America. Chen et al. (2016) reported that the invasive plant Alnus trabeculosa increased the soil bacterial diversity in the invaded range. Regarding root endosphere bacteria, most reports have focused on invasive legumes. For example, the invasive leguminous plant Cytisus scoparius in Australia was observed to form nodules with Bradyrhizobium, Rhizobium, and Mesorhizobium (Lafay and Burdon, 2006); a legume from North America, Robinia pseudoacacia, was shown to invade China's barren regions by forming nodules with local bacteria (Wei et al., 2009). Few reports have explored the endosphere microbial community composition and its potential function for invasive nonlegume plants (Dai et al., 2016; Bickford et al., 2018). To date, there is no report comparing the difference between two microbial compartments: microbes surrounding the roots and those within the roots of an invasive plant. This knowledge gap hinders our understanding of whether invasive plants can selectively enrich root bacteria from surrounding soils. Ageratina adenophora (family Asteraceae, syn. Eupatorium adenophorum Spreng.), commonly known as Crofton weed, originates from Mexico and invaded China from Burma beginning in the 1940s. A. adenophora has rapidly spread in southwestern China and is expected to spread into the vast areas of potential habitat in southern and southcentral China based on ecological niche modeling without control measures (Wang and Wang, 2006). Previous studies showed that A. adenophora could change the soil microbial community to benefit its invasion (Niu et al., 2007; Xu et al., 2012; Sun et al., 2013). For example, A. adenophora invasion resulted in a reduction of actinomycetes but an increase of aerobic and anaerobic bacteria (Niu et al., 2007). Xu et al. (2012) reported that A. adenophora invasion significantly increased the number and diversity of nitrogen-fixing bacteria (NFB) but did not change the core bacterial community, including 16 operational

203

taxonomic units (OTUs) belonged to Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria across a range of geographical sites. Nonetheless, it is unclear whether A. adenophora can selectively accumulate bacteria in roots from local soils. In this study, highthroughput sequencing technology was used to explore whether the specific rhizobacteria were enriched as endosphere bacteria in the roots of A. adenophora. The potential ecological effects, specifically the plant growth-promoting effect, of the enriched bacteria were also discussed. 2. Materials and methods 2.1. Field site and sampling approach The sampling was performed in July in 2016 at three invaded sites in Yunnan Province, China, including Simao County (Lat 22°44′57″N, Lon 100°48′47″E, with an elevation of 1310 m), Lancang County (Lat 22°37′37″N, Lon 99°47′49″E, with an elevation of 1680 m), and Cangyuan County (Lat 23°20′34″N, Lon 99°20′35″E, with an elevation of 1150 m). At each sampling site, a 20 m × 20 m quadrat was selected. Six individuals, each at least 5 m apart, were randomly selected for the collection of the rhizosphere soil. Plants were dug out and lightly shaken; then, the soil remaining attached to the root surface was carefully collected with a brush and treated as the rhizosphere soil. Both the roots and rhizosphere soils of plants were packed into polyethylene bags and were placed on ice packs in a cooler after collection and immediately transported to the laboratory, where they were stored at 4 °C until processing. For each individual, 10 g of rhizosphere soils was collected and mixed. Only three randomly selected individuals were sequenced to represent the rhizosphere soils in a given sampling site. However, 10 g of the middle part of the fine roots, approximately 5 cm from the main root of each individual, was removed and collected. All root samples were mixed and sent to be sequenced. Roots were sonicated for 10 min in an ultrasonic bath cleaner (Branson, CT, USA) (Richter-Heitmann et al., 2016). The tissues were surface disinfected for 5 min in 70% CH3CH2OH followed by 5 min in 0.5% NaClO and washed with sterile water. The washed water from the final washing was inoculated on tryptic soy agar (TSA) culture to verify the successful surface sterilization. 2.2. Soil characteristics and climate data of sampling sites Measurements of soil characteristics were conducted according to the methods described by Lu (2000). Soil pH was determined in a soil suspension prepared with a soil-water ratio of 1:2.5 (w/v) with a pH meter model 610 A (Fisher Scientific Inc., Salt Lake City, UT). Soil organic matter (SOM) was determined by oxidation with potassium dichromate. Soil total nitrogen (TN) was measured through dry combustion analysis with a Vario MAX-CN Elemental Analyzer (Elementar, Germany). The available nitrogen (AN) was measured using the alkalihydrolysis and diffusion method. Total soil phosphorus (TP) and potassium (TK) were extracted and then determined via the perchloric acid digestion method and spectrophotometry protocols, respectively. The total available phosphorus (AP) was determined according to the methods described by Blakemore (Blakemore et al., 1972). The available potassium (AK) was extracted for 30 min with 1 M NH4OAc (soil-solution ratio of 1:10) and then analyzed with atomic absorption spectrophotometry (Lanyon and Heald, 1982). Precipitation and temperature data were calculated with Worldclim database version 2.0 using the R v3.4.3 package “raster”. 2.3. Bacterial culture For the isolation of bacteria, 200 mg of surface sterilized roots was grinded in 1 mL sterile water and plated on three culture media, including yeast mannitol agar (YMA), TSA, and 523 (Kado and Heskett, 1970),

204

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209

to incubate for 24–36 h at 25 °C; the single strain was transformed to new culture and cultured for another 24–36 h and used for DNA extraction. 2.4. DNA extraction and sequencing For purified isolates, DNA was extracted using a DNA Isolation Kit (Bio Teke, Beijing, China). The bacterial 16 S ribosomal RNA gene was PCR amplified in triplicate using 27 F and 1492 R (Wilmotte et al., 1993), and the nucleotide sequences of 50 isolates have been deposited in GenBank under accession numbers MK779071-MK779120. For highthroughput sequencing, the total DNA of 200 mg fine roots was extracted using the CTAB method for each root sample (Murray and Thompson, 1980), and 0.5 g soil was extracted with a commercial DNA extraction kit for each rhizosphere soil (PowerSoil® DNA Isolation Kit, MoBio). DNA quality was monitored by 0.8% agarose gel electrophoresis, and extracted DNA was diluted to a concentration of 1 ng/μL and stored at −20 °C until further processing. The diluted DNA was used as a template for PCR amplification of bacterial 16 S rRNA genes with the barcoded primers and HiFi Hot Start Ready Mix (KAPA). For bacterial diversity analysis, V3-V4 variable regions of 16 S rRNA genes were amplified with universal primers 343 F and 798 R (Nossa et al., 2010). Amplicon quality was visualized using gel electrophoresis, purified with AMPure XP beads (Agencourt), and amplified for another round of PCR. After purification with the AMPure XP beads again, the final amplicon was quantified using a Qubit dsDNA assay kit. Equal amounts of purified amplicon were pooled for subsequent sequencing at Shanghai OE Biotech. Co., Ltd. (Shanghai, China) using the MiSeq platform. 2.5. Bioinformatic analysis of high-throughput sequencing Raw sequencing data were in FASTQ format. Paired-end reads were then preprocessed using Trimmomatic software v0.36 (Bolger et al., 2014) to detect and remove ambiguous bases (N). We also removed low-quality sequences with average quality score below 20 using a sliding window trimming approach. After trimming, paired-end reads were assembled using FLASH software v1.2.11 (Reyon et al., 2012). Parameters of assembly were 10 bp of minimal overlapping, 200 bp of maximum overlapping and 20% of maximum mismatch rate. Sequences were performed for further denoising as follows: reads with ambiguous, homologous sequences or below 200 bp were abandoned. Reads with 75% of bases above Q20 were retained. Then, reads with chimeras were detected and removed. These two steps were achieved using QIIME software v1.8.0 (Caporaso et al., 2010). Clean reads were subjected to primer sequence removal and clustered to generate OTUs using UPARSE software with a 97% similarity cutoff (Edgar, 2013). The representative read of each OTU was selected using the QIIME package. All representative reads were annotated and blasted against the Silva database (Version 123) using RDP classifier (confidence threshold was 70%) (Wang et al., 2007). 2.6. Data analysis of high-throughput sequencing All reads obtained from 27 samples were divided into 3515 OTUs. We first deleted chloroplast, mitochondria, and the OTUs below 5 reads and then subsampled the remaining 22,565 reads (minimized read sample) using MOTHUR v1.35.1 “sub. sample”. All data were analyzed according to data description based on subsampled data. The rarefaction curve for each sample was calculated using MOTHUR v1.35.1 “rarefaction.single” and plotted using GraphPad Prism v7 (GraphPad Software, Inc., CA, USA). Alpha diversity was calculated using the R “vegan” package and plotted using GraphPad Prism v7. The MannWhitney U test was used to test the differences in alpha diversity between groups. A heatmap of top 30 genera was generated with the R “pheatmap” package. Nonmetric multidimensional scaling (NMDS) based on Morisita-Horn dissimilarity and Jaccard dissimilarity was

performed using the R v3.4.3 “vegan” package to investigate the differences among root bacterial communities in the three invaded sites. The differences between bacteria in the roots and rhizosphere soil bacteria were also determined using NMDS. Permutational multivariate analysis of variance (PERMANOVA) was used to test the statistically significant differences among sites or sources based on bacterial OTU richness or phylogenetic diversity of bacteria (0, 1 matrix) and was performed using the R package “vegan”. Venn diagrams were generated by the R “VennDiagram” package. Other graphs were plotted by GraphPad Prism v7. 2.7. Growth-promoting activity of bacterial OTUs The cultured strains grouped into the same OTU according to the 97% cutoff in high-throughput sequencing were selected for plant growth experiments, including OTU6 (strain number: X170; GenBank accession number: MK779102) and OTU30 (strain number: X154; GenBank accession number: MK779095) (Enterobacteriaceae). Two native plants (Hypoestes triflora and Reinwardtia indica) as well as A. adenophora were collected in Xishan Forest Park (Lat 24°58′34″N, Lon 102°37′5″E, with an elevation of 2080 m). Healthy, same-size seedlings were selected for cultivation. Roots of plants were surface disinfected for 3 min in 70% CH3CH2OH followed by 3 min in 0.5% NaClO and rinsed in sterile water and placed in a sterilized plastic cup (9 cm (diameter) × 15 cm (high)) with 200 g of sterile soil. The bacteria used for the inoculation were incubated in tryptic soy broth (TSB) for 24 h and diluted to 0.5–0.6 OD at 600 nm using a Bio-Rad SmartSpec Plus spectrophotometer (Bio-Rad, Hercules, CA, USA). After 7 days of planting, 2 mL bacterial suspensions were added to the roots of the plants and the control was added to the sterilized bacterial suspension. Each combination of plant and bacterial OTU included over 3 replicates. Then, the plants were grown 60 days (25 °C, humidity 70%, daylight 12 h and night 12 h, illumination intensity 80–100 μmol·m−2·s−1) until harvest to measure the biomass of roots and shoots. During this period, sterilized water was added to maintain the moisture as needed. The biomass of roots and shoots with individual was plotted by GraphPad Prism v7 and the differences between treatment and control were tested with the MannWhitney U test using the R package “stats”. 3. Results 3.1. Profile of the bacteria under high-throughput sequencing In total, 18 roots and 9 rhizosphere soils were sequenced, and 893,743 valid tags were generated by sequencing with a range from 22,583 to 38,430 for each sample and a range of valid tags from 77.70% to 93.70% (Table S1, also see dataset raw data). Subsequently, we first deleted chloroplast, mitochondria, and the OTUs under 5 reads and then subsampled to the minimized reads sample, i.e., 22,565 reads. The rarefaction curve reached saturation at ~12,000 sequences, and the species number in rhizosphere soils was higher than that in roots (Fig. 1a). Similarly, the alpha diversity of rhizosphere soils was significantly higher than that of roots (P b 0.00001, MannWhitney U test). However, there was no significant difference among sites in either roots or rhizosphere soils (P N 0.05, Mann-Whitney U test, Fig. 1b). The reads belonged to 27 phyla, 41 classes (dataset subsampled data). Firmicutes (40.91%), Proteobacteria (39.51%), Bacteroidetes (14.94%), and Actinobacteria (2.35%) were the dominant phyla. Clostridia (39.14%), Gammaproteobacteria (19.54%), Bacteroidia (12.91%), Betaproteobacteria (11.07%), Alphaproteobacteria (7.47%), Actinobacteria (2.16%), Bacilli (1.29%), and Flavobacteriia (1.00%) were the dominant classes. Dominant genera, including Clostridium (28.38%), Bacteroides (8.92%), Enterobacter (7.75%), Dickeya (4.16%), Anaerosporobacter (1.89%), Rhizobium (1.58%), Ruminococcus (1.47%),

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209

(a) 1200

400

12000 18000 Sequences

0 .9

4

0 .6

2

0 .3

0

0 .0

24000

SM

6000

6

S LC (n= 3 CYS(n ) = S M S(n 3 ) R = LC (n 3) = CYR(n 6) R( =6) n S M =6 ) S( LC n= 3 CYS(n ) = SM S ( n 3 ) R = LC (n 3) = CYR(n 6) R( =6) n= 6)

0 0

1 .2

*** ***

Simpson index

Species

800

(b) 8 Shannon index

SMS LCS CYS SMR LCR CYR

205

Fig. 1. Rarefaction curves (a) and alpha diversity of bacteria (b). SMS: rhizosphere soil sample of SM sites; LCS: rhizosphere soil sample of LC sites; CYS: rhizosphere soil sample of CY sites; SMR: root samples of SM sites; LCR: root samples of LC sites; CYR: root samples of CY sites. The red column represents Shannon index on the left Y axis, and the blue column represents the Simpson index on the right Y axis. The n in parentheses indicates the duplication number of sampling sites. Error bars represent one SD. ***, P ≤ 0.001.

and Pseudomonas (1.24%), accounted for 55.39% of the total reads (dataset subsampled data). 3.2. Enrichment of rare soil bacteria in roots by the invasive plant Ageratina adenophora We inferred that A. adenophora might enrich specific bacteria from rhizosphere soils and cause a significant transformation of endosphere bacteria. We first indicated that the bacterial community in the roots was different only in components (Jaccard dissimilarity, P = 0.001, Table S2, Fig. S1a) rather than in abundance (Morisita-Horn dissimilarity, P = 0.062, Table S2, Fig. S1b), whereas those in rhizosphere soils were significantly different in both component and abundance from different sampling sites (Table S3). The differences in bacteria community of rhizosphere soil might be caused by the different soil characteristics and climate conditions of the sampling sites (Table 1). Moreover, we verified that the bacterial communities were completely divided into

root and rhizosphere soil groups (Fig. 2, P = 0.001 for both Jaccard and Morisita-Horn dissimilarity), explained primarily by isolation sources (roots vs. rhizosphere soils) rather than by geographic sites, although there were differences in soil characteristics and climate among sampling sites (Table 1, Table S4). The results indicated that the variation in endosphere bacteria was smaller than that in rhizosphere soil bacteria across the three invaded sites, and there were significant differences between bacteria in the roots and rhizobacteria. We further explored the community differences between the roots and rhizosphere soils at different phylogenetic levels. At the phylum level, Firmicutes was common both in roots (45.68%) and rhizosphere (29.78%). Proteobacteria was significantly higher in the roots (46.59%) than in the rhizosphere soils (24.76%) (Mann-Whitney U test, P = 0.009), but Bacteroidetes was dominant in rhizosphere soils, with a percentage of 39.06% (P = 0.000) (Fig. 3a). At the class level, Clostridia were present in both rhizosphere soils and roots, with a percentage of 27.47% and 44.98%, respectively (P N 0.05). Similarly, Beta- and Alpha-

Table 1 Soil characteristics and climate data of sampling sites.

SM LC CY

APrec (mm)

Prec (July, mm)

ATemp (°C)

Temp (July, °C)

pH

SOM (g/kg)

TN (g/kg)

TP (g/kg)

TK (g/kg)

AN (mg/kg)

AP (mg/kg)

AK (mg/kg)

1500 1530 1414

323 322 301

19.0 17.6 18.2

22.6 20.9 21.9

6.3 5.9 7.5

23.82 59.74 16.13

1.49 2.84 1.21

0.52 0.45 0.81

20.11 4.61 10.47

82.32 145.64 44.63

22.99 0.88 3.1

134.76 71 65.5

APrec: annual precipitation, Prec: precipitation (July), ATemp: annual temperature, Temp: temperature (July), SOM: soil organic matter, TN: total nitrogen, TP: total soil phosphorus, TK: total soil potassium, AN: available nitrogen, AP: total available phosphorus, AK: available potassium.

(a)

(b) 1

NMDS 2

0. 0

Sources

−0.5 −1.0

0 Sites

0 NMDS 1

1

Sources

−1 −2

Stress=0.1050 −1

NMDS 2

Sites

0. 5

Stress=0.0906 −2 −1 0 NMDS 1

1

CYR CYS LCR LCS SMR SMS

2

Fig. 2. NMDS analysis for comparing bacterial occurrence on different invaded ranges and sources. (a) Morisita-Horn dissimilarity; (b) Jaccard dissimilarity. Red: root samples; blue: rhizosphere soil samples. Source, bacterial communities from roots vs. rhizosphere soils; Sites, bacterial communities from different sampling sites (CY, LC, and SM); the line length indicates the ratio of the total variation in bacterial community in NMDS analysis.

206

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209

(c)

(a) Relative abundance %

1 00

Acidobacteria Cyanobacteria Gemmatimonadetes Fusobacteria Actinobacteria Bacteroidetes Proteobacteria Firmicutes

50

Azospirillum Magnetospirillum Rhizobacter Acidovorax Caulobacter Flavobacterium Sphingobium Sphingomonas Novosphingobium Bacteroides Ruminiclostridium Escherichia Lachnospiraceae Alistipes Ruminococcus Prevotella Lachnoclostridium Faecalibacterium Blautia Clostridium Enterobacter Rhizobium Pseudomonas Dickeya Tolumonas Anaerosporobacter Aeromonas Roseateles Paenibacillus Uliginosibacterium

En do

(b)

Rh iz o

0

Relative abundance %

1 00 Epsilonproteobacteria Gemmatimonadetes Sphingobacteriia Deltaproteobacteria Flavobacteriia Fusobacteriia Bacilli Actinobacteria Alphaproteobacteria Betaproteobacteria Gammaproteobacteria Bacteroidia Clostridia

50

En d

o

Rh iz o

10 8 6 4 2 0

LCS.2 LCS.1 LCS.3 SMS.3 SMS.2 SMS.1 CYS.3 CYS.1 CYS.2 SMR.3 SMR.2 SMR.1 LCR.6 LCR.5 LCR.3 LCR.1 LCR.2 CYR.3 CYR.2 CYR.1 CYR.6 CYR.4 CYR.5 SMR.5 SMR.4 SMR.6 LCR.4

0

12

Fig. 3. Relative abundances of bacterial distribution in the rhizosphere soils and roots at different phylogenetic levels. (a) Relative abundances of bacteria at the phylum level. (b) Relative abundances of bacteria at the class level. (c) Heatmap of relative abundances of the top 30 fungal genera from the different sites. The relative abundances are expressed as the richness of bacteria transformed by log2(x + 1) from a given isolation source. Red genera represent the significantly more abundant genera in rhizosphere soils than in roots (P b 0.05, Mann-Whitney U test). Blue genera represent the significantly more abundant genera in roots than in rhizosphere soils (P b 0.05, Mann-Whitney U test). Black genera represent no distinct difference between the two sources (P N 0.05, Mann-Whitney U test).

(b)

(c)

OTU548 OTU78 OTU154 OTU33 OTU32 OTU15 OTU59 OTU14 OTU127 OTU27 OTU2 OTU106 OTU39 OTU1631 OTU2875 OTU9 OTU30 OTU50 OTU80 OTU457 OTU289 OTU41 OTU38 OTU84 OTU3 OTU204 OTU13 OTU11 OTU6 OTU1

OTU869 OTU616 OTU309 OTU776 OTU600 OTU2906 OTU1233 OTU491 OTU520 OTU462 OTU585 OTU630 OTU580 OTU591 OTU618 OTU628 OTU322 OTU593 OTU91 OTU2953 OTU647 OTU236 OTU289 OTU594 OTU579 OTU577 OTU18 OTU586 OTU11 OTU573

(a) S R

1592

1292

327

R = 78.68% S = 3.12%

0

1

2

3

Relative abundance %

4

5

6

S R

R = 4.47% S = 45.21%

0

1

2

3

4

Relative abundance %

Fig. 4. Asymmetrical distribution of OTUs in the rhizosphere soils and roots. (a) Venn diagram of OTU number of roots or rhizosphere soils. (b) Cumulative histogram of top 30 bacteria in the roots distributed in the rhizosphere and root endosphere. (c) Cumulative histogram of top 30 rhizobacteria distributed in the rhizosphere and root endosphere. The relative abundances are expressed as the richness of bacteria transformed by log2(x + 1) from a given isolation source. Red OTU is the abundance of OTUs in both roots and rhizosphere soils. S: Bacteria in rhizosphere soils; R: bacteria in root endosphere.

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209

0 .4

0 .5

OTU6 ns

0 .4

0 .2 0 .1

Biomass / g

Biomass / g

0 .3

207

OTU30 ns

0 .3 0 .2 0 .1 0 .0

Sh oo to Sh f A oo A Sh t of oo R I t Ro of H ot T o Ro f AA ot Ro of R ot I of H T

Sh oo to Sh f A oo A Sh t of oo R I t Ro of H ot T o Ro f AA ot Ro of R ot I of H T

0 .0

Fig. 5. Plant growth promotion of Enterobacteriaceae isolated from A. adenophora root. Red: treatment; blue: control. The middle of the line segment represents the mean, and the two sides represent the positive and negative standard deviations. AA, Ageratina adenophora; HT, Hypoestes triflora; RI, Reinwardtia indica. ns: no significance between treatment and control in each group (P N 0.05, Mann-Whitney U test).

proteobacteria were the common groups in both environments. However, Bacteroidia was dominant in rhizosphere soils, with a relative abundance of rhizosphere soils/roots = 35.56%/1.58% (P = 0.000), while Gammaproteobacteria was dominant in roots, with a relative abundance of rhizosphere soils/roots = 3.62%/27.51% (P = 0.000) (Fig. 3b). At the genus level, the heatmap showed that bacterial communities were divided into root and rhizosphere soil groups, and genera were divided into three groups (Fig. 3c). The abundances of Bacteroides, Ruminococcus, Prevotella, Faecalibacterium, Lachnoclostridium, and Lachnospiraceae were significantly higher in rhizosphere soils than in roots (P b 0.05, Mann-Whitney U test). In contrast, the abundances of Clostridium, Enterobacter, Dickeya, Anaerosporobacter, Rhizobium, Pseudomonas, Tolumonas, Roseateles were dominant in roots (P b 0.05, Mann-Whitney U test). Some genera, e.g., Azospirillum, Flavobacterium, Sphingobium, Acidovorax, Magnetospirillum, and Novosphingobium, were evenly present both in the roots and rhizosphere soils. Overall, 1292 OTUs were present in both rhizosphere soils and roots, but 1592 OTUs were present only in rhizosphere soils and 327 OTUs only in roots (Fig. 4a). More than 79% of OTUs in roots also occurred in rhizosphere soils, indicating the high overlap between the rhizosphere and root endosphere. We compared the distribution of the top 30 OTUs of bacteria in the roots and rhizobacteria (Fig. 4b, c). Only two OTUs, OTU11 (Burkholderiales) and OTU289 (Oxalobacteraceae), were abundant both in roots and rhizosphere soils. Among the remaining 56 OTUs, 3 OTUs only occurred in rhizosphere soils and 8 OTUs only occurred in roots. The top 30 OTUs in roots, accounting for 78.68% of bacteria in the roots, were from a small number of bacteria in rhizosphere soils (3.12%). The dominant OTUs in roots, OTU1 (36.12%, Clostridium) and OTU6 (10.53%, Enterobacter), occurred in rhizosphere soils, with abundances of 0.06% and 0.27%, respectively. In contrast, the top 30 OTUs of rhizosphere soils, accounting for 45.22% of bacteria in rhizosphere soils, had low overlap with bacteria in the roots (Fig. 4c). Therefore, the bacteria in the roots most likely tend to be selectively recruited from rare rhizobacteria. Among the cooccurring OTUs, 13 OTUs abundant as root endophytes included OTU1 (Clostridium, 36.12%), OTU6 (Enterobacter, 10.53%), OTU13 (Dickeya, 3.56%), OTU11 (Burkholderiales, 3.41%), OTU204 (Dickeya, 2.68%), OTU3 (Anaerosporobacter, 2.61%), OTU84 (Clostridium, 2.47%), OTU38 (Enterobacteriaceae, 1.67%), OTU457 (Enterobacteriaceae, 1.37%), OTU41 (Clostridium, 1.23%), OTU80 (Rhizobium, 1.18%), OTU50 (Roseateles, 1.17%), and OTU30 (Enterobacteriaceae, 1.00%). Most of the OTUs belonged to the genera Clostridium (39.83%) and Dickeya (6.24%) and family Enterobacteriaceae (14.57%), accounting for over 60% of reads.

3.3. Plant growth-promoting experiment We selected two representative strains from OTU6 and OTU30 (Enterobacteriaceae) for the plant growth-promoting experiment. OTU6 marginally promoted the growth of A. adenophora but inhibited native plants H. triflora and R. indica, whereas the opposite effect was observed for OTU30 (Fig. 5).

4. Discussion Currently, knowledge is very limited about the root bacteria associated with invasive plants, although these bacteria have been considered to play an important role in the invasion process (Hong et al., 2015). More importantly, there is no previous report on whether invasive plants can selectively enrich root bacteria from surrounding soils. In this study, we analyzed the bacterial community in the root endosphere and rhizosphere of A. adenophora in the invaded ranges. We found that A. adenophora selectively accumulated the bacteria in the roots from rare rhizosphere bacteria, causing a significant switch of bacterial composition at the phylum level, from Bacteroidetes predominant in the rhizosphere to Proteobacteria predominant in the root endosphere (Fig. 3), mainly due to the enrichment of genera Clostridium and Enterobacter (Gammaproteobacteria) in the root endosphere, represented by a high abundance of OTU1 (Clostridium) and OTU6 (Enterobacter). The bacterial communities in the rhizosphere soils varied more significantly than those in the root endosphere across the three sites (Fig. S1, Tables S2, S3). Again, the sources (root endosphere vs. rhizosphere soils) played the most important role in the variation of bacterial community (P = 0.001) rather than the geographic sites (P N 0.05, Table S4). These data indicated A. adenophora accumulated the conserved endosphere bacteria in the invaded ranges. Previously, the local microclimate and soil have been considered to be able to homogenize the root endosphere community (Hoffman and Arnold, 2008; Pan et al., 2008; Long et al., 2009). In Arabidopsis thaliana, both the rhizosphere and endosphere communities are strongly influenced by distinct geochemical characteristics in bulk soils (Lundberg et al., 2012). In this study, however, we found great variation in rhizosphere soil, whereas endosphere bacterial communities were relatively homogeneous across the invaded sites (Fig. 3, Tables S3, S4), suggesting that the host A. adenophora played the key role in establishing the conserved bacterial community in the root endosphere. Clostridium and Enterobacter are common endobacteria in native plants and crops. For example, Johnston-Monje and Raizada (2011)

208

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209

reported that these two genera were conserved in a variety of Zea genotypes. Enterobacter-like bacteria were dominant in roots of rice cultivated in low-pH soil (Hardoim et al., 2012). Enterobacter was also found as the bacteria in the roots in two medicinal plants, Hypericum perforatum and Ziziphora capitata (Egamberdieva et al., 2017). In a study of 30 angiosperm species, Enterobacteriaceae and Clostridiaceae were verified as common bacterial families in the root endosphere (Fitzpatrick et al., 2018). Again, Enterobacteriaceae was common in the root endosphere of A. thaliana (Lundberg et al., 2012). These reports suggested that there was a similar molecular signal recognition for accumulation of these specific groups of bacteria in the root endosphere by a variety of plants. Traditionally, molecular signal recognition match is considered to have developed via a long-term coevolution, e.g., the enrichment of the conserved endosphere bacteria occurred in legumes and their nitrogen-fixing bacteria, in which the latter can be attracted to infect the legume root endosphere by the root exudate flavonoids acting as the inducers of genes responsible for the nodulation process of rhizobia (Wang et al., 2011; Liu et al., 2017). There is also evidence that the composition of root exudates, e.g., phytohormones, plays a central role in the recruitment of bacteria from the bulk soil into the rhizosphere and, finally, into the root endosphere of plant hosts (Bais et al., 2006; Long et al., 2009; Micallef et al., 2009). Interestingly, A. adenophora is native to Mexico and has been present in China for only 70–80 years (Wang and Wang, 2006). Our data verified that, like most native plants, A. adenophora recruited the common endobacteria Clostridium and Enterobacter from the rhizosphere across a large geographic range (Fig. 3). These data indicated that, on one hand, there may exist a highly conserved signal recognition, when hosts, either invasive or native, enriched certain endosphere bacteria, such as Clostridium, Enterobacter, etc. On the other hand, the adaptation may be rapidly achieved in a short term to develop the molecular signal match between invasive plant hosts and these groups of local bacteria. It is thus an interesting question for future studies exploring the molecular mechanisms involved in the recruitment of root bacteria by invasive hosts. The endosphere community has been considered to have high potential for use in plant growth promotion, the improvement of plant stress resistance, biocontrol of pathogens, and bioremediation (Taghavi et al., 2010; Sessitsch et al., 2012). The direct plant growthpromoting activity of Enterobacteriaceae sp. was reported in several species of poplar (Taghavi et al., 2010). In the present case, the dominant genus Clostridium was not successfully cultured due to its obligate anaerobic status. Fortunately, two strains of Enterobacteriaceae from OTU6 and OTU30, respectively, were obtained and used for the plant growth-promoting experiment. The strain from OTU6 produced marginally positive feedback to A. adenophora growth but negative feedback to native plants; however, the opposite response occurred for the strain from OTU30 (Fig. 5). The data indicated that different strains played distinct roles in the direct plant growth-promoting activity of the invasive host, although the altered soil microbes as a whole were considered to play an important role in the invasion process of A. adenophora by facilitating the invasive host but inhibiting the native plants (Niu et al., 2007). Nonetheless, our growth experiment only used the plants from one invaded site, Xishan Forest Park in Kunming. It would be interesting to further explore whether there exist different effects of bacteria-plant combinations, including the same geographical source (bacteria and plants from one site) as well as different sources (bacteria and plants from different sites). Moreover, it is important to clarify the effects of multiple endosphere bacteria on the growth of A. adenophora, including the combination of strains from phylogenetically close or distant groups. In addition, it should be verified whether these endosphere bacteria play a role beyond in plant growth-promoting effects, e.g., in the improvement of the stress resistance of A. adenophora. In contrast, Dickeya is well known as a pathogen with a broad host range (Ma et al., 2007). Members of the genus Dickeya can affect a wide range of plant hosts worldwide, particularly banana,

Chrysanthemum spp., Dianthus spp., maize, potato, and tomato, and can survive in soil for the maximum period of 12 months (Toth et al., 2011; Ansermet et al., 2016). Therefore, A. adenophora enriched Dickeya and might benefit through an indirect effect described by the accumulation of local pathogens hypothesis (Eppinga et al., 2010). For example, the accumulated pathogens might spill back onto co-occurring native species, exacerbating the effects of invasions (Flory and Clay, 2013). Mangla et al. found that Chromolaena odorata accumulated high concentrations of soil pathogens to create a negative feedback on native plant species (Mangla et al., 2008). 5. Conclusion Our data verified that A. adenophora selectively accumulated bacteria, primarily Clostridium and Enterobacter, from the rhizosphere soils in the invaded ranges, suggesting that the recognition mechanism between plant host and certain bacteria in the roots may be conserved and not require a long-term interaction between invasive plants and local soil bacteria. Nonetheless, these accumulated bacteria may be physiologically and genetically distinct at the strain level because phylogenetically similar strains displayed different roles in the growth of the invasive host itself and of native plants. The molecular mechanisms involved require further exploration, representing an important future line of inquiry in the field invasive biology. Authors' contributions L C experimental design, data analysis and manuscript writing; L C, K F, J Z, Z-P, Y, X-F, D, G-H, D, H-B, Z soil sampling collection, seedling collection, experimental performance; H-B, Z design of the project, data analysis and manuscript writing. All authors have read and approve the submission of this manuscript. Declaration of Competing Interest No conflict of interest. Acknowledgements This work was funded by the National Natural Science Foundation of China (grant Nos. 31770585, 31360153). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2019.05.220. References Ansermet, M., Schaerer, S., Kellenberger, I., Tallant, M., Dupuis, B., 2016. Influence of seedborne and soil-carried inocula of Dickeya spp. on potato plant transpiration and symptom expression. Eur. J. Plant Pathol. 145, 459–467. Bais, H.P., Weir, T.L., Perry, L.G., Gilroy, S., Vivanco, J.M., 2006. The role of root exudates in rhizosphere interactions with plants and other organisms. Annu. Rev. Plant Biol. 57, 233–266. Bickford, W.A., Goldberg, D.E., Kowalski, K.P., Zak, D.R., 2018. Root endophytes and invasiveness: no difference between native and non-native Phragmites in the Great Lakes Region. Ecosphere 9, e02526. Blakemore, L.C., Searle, P.L., Daly, B.K., 1972. Methods for chemical analysis of soils. New Zealand Soil Bureau Report 10 A. Government Printer, Wellington. Bolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. Bulgarelli, D., Rott, M., Schlaeppi, K., van Themaat, E.V.L., Ahmadinejad, N., Assenza, F., et al., 2012. Revealing structure and assembly cues for Arabidopsis root-inhabiting bacterial microbiota. Nature 488, 91–95. Callaway, R.M., Thelen, G.C., Rodriguez, A., Holben, W.E., 2004. Soil biota and exotic plant invasion. Nature 427, 731–733. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., et al., 2010. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336.

L. Chen et al. / Science of the Total Environment 683 (2019) 202–209 Chen, X., Zhang, X., Xe, Zhu, Zhang, H., Liang, X., Lei, Y., et al., 2016. Exotic plant Alnus trabeculosa alters the composition and diversity of native rhizosphere bacterial communities of Phragmites australis. Pedosphere 26, 108–119. Dai, Z.C., Fu, W., Wan, L.Y., Cai, H.H., Wang, N., Qi, S.S., et al., 2016. Different growth promoting effects of endophytic bacteria on invasive and native clonal plants. Front. Plant Sci. 7, 706. Dawson, W., Schrama, M., 2016. Identifying the role of soil microbes in plant invasions. J. Ecol. 104, 1211–1218. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–998. Edwards, J., Johnson, C.S., Santos-Medellín, C., Lurie, E., Natarajkumar, P., Bhatnagar, S., et al., 2015. Structure, variation, and assembly of the root-associated microbiomes of rice. Proc. Natl. Acad. Sci. U. S. A. 112, E911–E920. Egamberdieva, D., Wirth, S., Behrendt, U., Ahmad, P., Berg, G., 2017. Antimicrobial activity of medicinal plants correlates with the proportion of antagonistic endophytes. Front. Microbiol. 8, 199. Eppinga, M.B., Rietkerk, M., Dekker, S.C., de Ruiter, P.C., der Putten, W.H.V., 2010. Accumulation of local pathogens: a new hypothesis to explain exotic plant invasions. Oikos 114, 168–176. Fitzpatrick, C.R., Copeland, J., Wang, P.W., Guttman, D.S., Kotanen, P.M., Johnson, M.T.J., 2018. Assembly and ecological function of the root microbiome across angiosperm plant species. Proc. Natl. Acad. Sci. U. S. A. 115, E1157–E1165. Flory, S.L., Clay, K., 2013. Pathogen accumulation and long-term dynamics of plant invasions. J. Ecol. 101, 607–613. Hardoim, P.R., Hardoim, C.C.P., Overbeek, L.S.V., Elsas, J.D.V., 2012. Dynamics of seedborne rice endophytes on early plant growth stages. PLoS One 7, e30438. Hawkes, C.V., Wren, I., Herman, D.J., Firestone, M.K., 2005. Plant invasion alters nitrogen cycling by modifying the soil nitrifying community. Ecol. Lett. 8, 976–985. Hoffman, M.T., Arnold, A.E., 2008. Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees. Mycol. Res. 112, 331–344. Hong, Y., Liao, D., Hu, A., Wang, H., Chen, J., Khan, S., et al., 2015. Diversity of endophytic and rhizoplane bacterial communities associated with exotic Spartina alterniflora and native mangrove using Illumina amplicon sequencing. Can. J. Microbiol. 61, 723–733. Johnston-Monje, D.M., Raizada, M., 2011. Conservation and diversity of seed associated endophytes in Zea across boundaries of evolution, ethnography and ecology. PLoS One 6, e20396. Kado, C.I., Heskett, M.G., 1970. Selective media for isolation of Agrobacterium, Corynebacterium, Erwinia, Pseudomonas, and Xanthomonas. Phytopathology 60, 969–976. Lafay, B., Burdon, J., 2006. Molecular diversity of rhizobia nodulating the invasive legume Cytisus scoparius in Australia. J. Appl. Microbiol. 100, 1228–1238. Lanyon, L.E., Heald, W.R., 1982. Methods of soil analysis. Agronomy monograph Part 2. Agronomy Society of America and Soil Science Society of America, Madison. Liu, Y.C., Qin, X.M., Xiao, J.X., Tang, L., Wei, C.Z., Wei, J.J., et al., 2017. Intercropping influences component and content change of flavonoids in root exudates and nodulation of Faba bean. J. Plant Interact. 12, 187–192. Long, H.H., Sonntag, D.G., Schmidt, D.D., Baldwin, I.T., 2009. The structure of the culturable root bacterial endophyte community of Nicotiana attenuata is organized by soil composition and host plant ethylene production and perception. New Phytol. 185, 554–567. Lu, R., 2000. Analysis methods on soil agro-chemistry. China Agricultural Science and Technology Press, Beijing. Lundberg, D.S., Lebeis, S.L., Paredes, S.H., Yourstone, S., Gehring, J., Malfatti, S., et al., 2012. Defining the core Arabidopsis thaliana root microbiome. Nature 488, 86–90. Ma, B., Hibbing, M.E., Kim, H.S., Reedy, R.M., Yedidia, I., Breuer, J., et al., 2007. Host range and molecular phylogenies of the soft rot enterobacterial genera Pectobacterium and Dickeya. Phytopathology 97, 1150–1163. Mangla, S., Inderjit, Callaway, R.M., 2008. Exotic invasive plant accumulates native soil pathogens which inhibit native plants. J. Ecol. 96, 58–67. Micallef, S., Shiaris, M.P., Colon-Carmona, A., 2009. Influence of Arabidopsis thaliana accessions on rhizobacterial communities and natural variation in root exudates. J. Exp. Bot. 60, 1729–1742. Murray, M.G., Thompson, W.F., 1980. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res. 8, 4321–4325.

209

Niu, H.B., Liu, W.X., Wan, F.H., Liu, B., 2007. An invasive aster (Ageratina adenophora) invades and dominates forest understories in China: altered soil microbial communities facilitate the invader and inhibit natives. Plant Soil 294, 73–85. Nossa, C., Oberdorf, E.W., Yang, L., Aas, J., Paster, B., DeSantis, T., et al., 2010. Design of 16S rRNA gene primers for 454 pyrosequencing of the human foregut microbiome. World J. Gastroenterol. 16, 4135–4144. Pan, J.J., Baumgarten, A.M., May, G., 2008. Effects of host plant environment and Ustilago maydis infection on the fungal endophyte community of maize (Zea mays). New Phytol. 178, 147–156. Peiffer, J.A., Spor, A., Koren, O., Jin, Z., Tringe, S.G., Dangl, J.L., et al., 2013. Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proc. Natl. Acad. Sci. U. S. A. 110, 6548–6553. Reinhold-Hurek, B., Bü, W., Burbano, C.S., Sabale, M., Hurek, T., 2015. Roots shaping their microbiome: global hotspots for microbial activity. Annu. Rev. Phytopathol. 53, 403–424. Reyon, D., Tsai, S.Q., Khayter, C., Foden, J.A., Sander, J.D., Joung, J.K., 2012. FLASH assembly of TALENs for high-throughput genome editing. Nat. Biotechnol. 30, 460–465. Richter-Heitmann, T., Eickhorst, T., Knauth, S., Friedrich, M.W., Schmidt, H., 2016. Evaluation of strategies to separate root-associated microbial communities: a crucial choice in rhizobiome research. Front. Microbiol. 7, 773. Schlaeppi, K., Dombrowski, N., Oter, R.G., Themaat, EVLv, Schulze-Lefert, P., 2013. Quantitative divergence of the bacterial root microbiota in Arabidopsis thaliana relatives. Proc. Natl. Acad. Sci. U. S. A. 111, 585–592. Sessitsch, A., Hardoim, P.R., Döring, J., Weilharter, A., Krause, A., Woyke, T., et al., 2012. Functional characteristics of an endophyte community colonizing rice roots as revealed by metagenomic analysis. Mol. Plant-Microbe Interact. MPMI 25, 28–36. Shakya, M., Gottel, N., Castro, H., Yang, Z.M.K., Gunter, L., Labbé, J., et al., 2013. A multifactor analysis of fungal and bacterial community structure in the root microbiome of mature Populus deltoides trees. PLoS One 8, e76382. Sun, X., Gao, C., Guo, L., 2013. Changes in soil microbial community and enzyme activity along an exotic plant Eupatorium adenophorum invasion in a Chinese secondary forest. Chin. Sci. Bull. 58, 4101–4108. Taghavi, S., Dvd, Lelie, Hoffman, A., Zhang, Y.B., Walla, M.D., Vangronsveld, J., et al., 2010. Genome sequence of the plant growth promoting endophytic bacterium Enterobacter sp. 638. PLoS Genet. 6, e1000943. Toth, I.K., Jv, Beckhoven, Saddler, G.S., Lojkowska, E., Hélias, V., Pirhonen, M., et al., 2011. Dickeya species: an emerging problem for potato production in Europe. Plant Pathol. 60, 385–399. Tringe, S.G., Cv, Mering, Kobayashi, A., Salamov, A., Chen, K., Chang, H.W., et al., 2005. Comparative metagenomics of microbial communities. Science 308, 554–557. Wang, R., Wang, Y.Z., 2006. Invasion dynamics and potential spread of the invasive alien plant species Ageratina adenophora (Asteraceae) in China. Divers. Distrib. 12, 397–408. Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., 2007. Naïve Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl. Environ. Microbiol. 73, 5261–5267. Wang, D., Yang, S., Tang, F., Zhu, H., 2011. Symbiosis specificity in the legume–rhizobial mutualism. Cell. Microbiol. 14, 334–342. Wei, G., Chen, W., Zhu, W., Chen, C., Young, J.P.W., Bontemps, C., 2009. Invasive Robinia pseudoacacia in China is nodulated by Mesorhizobium and Sinorhizobium species that share similar nodulation genes with native American symbionts. FEMS Microbiol. Ecol. 68, 320–328. Wilmotte, A., Van der Auwera, G., De Wachter, R., 1993. Structure of the 16 S ribosomal RNA of the thermophilic cyanobacterium Chlorogloeopsis HTF (‘Mastigocladus laminosus HTF’) strain PCC7518, and phylogenetic analysis. FEBS Lett. 317, 96–100. Xu, C.W., Yang, M.Z., Chen, Y.J., Chen, L.M., Zhang, D.Z., Mei, L., et al., 2012. Changes in nonsymbiotic nitrogen-fixing bacteria inhabiting rhizosphere soils of an invasive plant Ageratina adenophora. Appl. Soil Ecol. 54, 32–38. Zhang, H., Sun, Y., Xie, X., Kim, M.S., Dowd, S.E., Paré, P.W., 2009. A soil bacterium regulates plant acquisition of iron via deficiency-inducible mechanisms. Plant J. 58, 568–577. Zhang, Y., Jin, X., Riera, N., Jin, T., Li, J., Wang, N., 2017. Huanglongbing impairs the rhizosphere-to-rhizoplane enrichment process of the citrus root-associated microbiome. Microbiome 5, 97.