Changes in structure and assembly of a species-rich soil natural community with contrasting nutrient availability upon establishment of a plant-beneficial Pseudomonas in the wheat rhizosphere

doi:10.21203/rs.3.rs-2745406/v1

Download PDF

Research Article

Changes in structure and assembly of a species-rich soil natural community with contrasting nutrient availability upon establishment of a plant-beneficial Pseudomonas in the wheat rhizosphere

https://doi.org/10.21203/rs.3.rs-2745406/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 29 Sep, 2023

Read the published version in Microbiome →

You are reading this latest preprint version

Background:

Plant-beneficial bacterial inoculants are of great interest in agriculture as they have the potential to promote plant growth and health. However, the inoculation of the rhizosphere microbiome often results in a suboptimal or transient colonization, which is due to a variety of factors that influence the fate of the inoculant. To better understand the fate of plant-beneficial inoculants in complex rhizosphere microbiomes, composed by hundreds of genotypes and multifactorial selection mechanisms, controlled studies with high-complexity soil microbiomes are needed.

Results:

We analysed early compositional changes in a taxa-rich natural soil bacterial community, both in exponential nutrient-rich or stationary nutrient-limited growth conditions (i.e., growing and stable communities, respectively), upon inoculation by the plant-beneficial bacterium Pseudomonas protegens in a bulk soil or a wheat rhizosphere environment. P. protegens successfully established in all conditions tested, being more abundant in the rhizosphere of the stable community. Nutrient availability was a major factor driving microbiome composition and structure as well as the underlying assembly processes. While access to nutrients resulted in communities being mainly assembled by homogeneous selection, stochastic processes dominated in the nutrient-deprived conditions. We also observed an increased rhizosphere selection effect on nutrient-limited conditions, resulting in higher numbers of amplicon sequence variants (ASVs) whose relative abundance was enriched. The inoculation with P. protegens produced discrete changes, some of which involved other Pseudomonas. Direct competition between Pseudomonas strains partially failed to replicate differences observed in the microbiome and pointed to a more complex interaction network.

Conclusions:

The results obtained in this study show that nutrient availability is a major driving force of microbiome composition, structure, and diversity both in the bulk soil and the wheat rhizosphere and determines the assembly processes governing early microbiome development. The successful establishment of the inoculant was facilitated by the wheat rhizosphere and produced discrete changes among other members of the microbiome. Direct competition between Pseudomonas strains only partially explained microbiome changes and revealed that indirect interactions or spatial distribution in the rhizosphere or soil interface could be crucial for the survival of certain bacteria.

Microbiome

nutrient availability

plant-beneficial inoculant

inoculant proliferation

rhizosphere

wheat

bulk soil

microbial communities

Pseudomonas

bacterial competition

The use of bacterial inoculants in agriculture has emerged as an ecological and sustainable alternative to chemical fertilizers and pesticides. These inoculants have the ability to promote plant growth and health through different mechanisms, including solubilization of nutrients [1], protection against abiotic stresses such as drought or salinity [2, 3], enhancing the plant development by production of phytohormones [4], antagonizing soilborne phytopathogens [5] or killing pest insects [6, 7]. Despite these properties, field application of bacterial inoculants faces many challenges, notably their inability to efficiently compete and establish within the resident root microbiota during the time needed to exert their beneficial effect [8], often requiring multiple inoculations to ensure their function [9].

The assembly of the rhizosphere microbiome starts early during plant development with the release of root exudates containing carbon-rich compounds that create a niche capable of supporting bacterial growth in high numbers [10, 11]. These root exudates produce a rapid change in the bulk soil microbiome (not influenced by roots) towards a more specialised and metabolically active community (i.e., the rhizosphere microbiome) whose members are able to exploit the resources of the new niche [12, 13]. The rhizosphere microbiome is in part determined by the composition of the plant root exudates, which vary across genotypes, plant-developmental stages or in response to environmental changes [14, 15]. In addition, root exudates also contain multiple plant secondary metabolites (e.g., flavones, benzoxazinoids and coumarins) involved in microbiome signalling processes by facilitating or inhibiting the growth of susceptible bacteria, thus impacting microbiome assembly and functioning [16–18]. The interactions between the microbiome members will also determine the rhizosphere microbiome structure and composition [19]. Bacteria can compete with other bacteria for resources or space using contact-dependent or diffusible weaponry [20–22] or display positive, mutualistic relationships, which are often found dominating the rhizosphere [23, 24].

The effective establishment of the rhizosphere microbiome by plant-beneficial bacterial inoculants is dependent on their ability to proliferate under the conditions and competing factors in the rhizosphere. Notably, a poor adaptability of the inoculant to varying environmental conditions, such as soil pH [25], soil humidity or regional climate might result in a suboptimal and/or transient colonization [8, 26, 27]. In addition, the fate of the inoculant also depends on biotic interactions, which are often intricate, expanding to multiple trophic levels and higher orders of interactions [28–30]. These can include complex networks involving competition, cooperation [23] or interaction with other microbes. Also grazing by phagotrophic protists [31] and infection by bacteriophages [32, 33], can prevent an effective colonisation of the rhizosphere microbiome by inoculants. An important factor governing inoculant proliferation is the availability of nutrients in the rhizosphere environment to support the community growth and development [34–36], which will be different from the bulk soil nutrients, often scarce, limiting bacterial growth and soil microbial processes [37]. Compositional changes of such microbiomes in response to the proliferation of an inoculant could provide important information that will improve our understanding of how plant-beneficial bacterial inoculants interact within the rhizosphere microbiome.

Soil and rhizosphere microbiomes are considered the most diverse ecosystems on earth [38], harbouring hundreds of genotypes. This challenges the use of natural microbiomes for elucidating microbiome invasion strategies of plant-beneficial inoculants [26, 27, 39–42]. Replicable and simplified communities are often obtained in standardized systems through top-down approaches to further study how they react to stressors, including the introduction of new community members [43–45]. The extreme of such reductionistic approaches is the study of individual bacterial isolates or of synthetic communities. They can be of great importance for investigating mechanistic effects that however might not always be observed under natural conditions. This challenge might be eased by using species-rich natural communities directly derived from the environment [46, 47].

In the present study, we analysed the potential for the model plant-beneficial bacterium Pseudomonas protegens CHA0^T [5, 22, 48] to establish in the rhizosphere of wheat (Triticum aestivum) as a function of the physiological state of a reproducible, species-rich bulk soil natural bacterial community (NatCom) [46]. Two different community growing states were evaluated based on nutrient availability to study differences of an actively growing community (i.e., by dilution of the NatCom and addition of nutrients) versus a stationary, nutrient-limited stable NatCom, with the aim to simulate two nutritional scenarios that occur in natural soils.

Experimental design

A previously isolated and characterized soil natural bacterial community (NatCom) was grown and prepared in microcosms as detailed in [46]. Briefly, microcosms consisting of soil matrix (0.5-3 mm washed and sterile riverbank silt) and 10% (v/w) of soil extract (SE, autoclaved and filtered aqueous top forest soil solution, see [46] for details) as source of nutrients were inoculated with 10% (w/w) of the soil NatCom [46]. Community growth was tracked weekly by counting of colony forming units (CFUs) on R2A medium (Sigma-Aldrich) for about one month until an apparent community stationary phase was achieved, reaching ~ 10⁸ CFUs g^− 1 of soil matrix. Two community states were chosen: (1) the growing NatCom and (2) the stable NatCom. For the growing NatCom, the community was diluted (10% (w/w)) into new soil matrix and soil extract as described above. This was considered the growing NatCom, as the community had access to new nutrients (i.e., soil extract) to grow. The stable NatCom corresponded to the above mentioned NatCom in an overall community stationary growth phase, which had not been provided with new nutrients for a month. For each community, growing or stable, part was inoculated with Pseudomonas protegens CHA0^T tagged with mCherry (CHA0::attTn7-mcherry; gentamycin resistant (Gm^R) [49]) to obtain 10⁵ CFUs g^− 1 of soil matrix, while the other part remained uninoculated. Cells of P. protegens CHA0^T-mcherry (P. protegens CHA0 hereafter) were prepared from an overnight culture in nutrient yeast broth (NYB, per 1 L: 25 g of nutrient broth no. 2, Oxoid, and 5 g of yeast extract, Oxoid), supplemented with gentamycin sulphate (Gm; 10 µg mL^− 1). Cells were harvested by centrifugation at 5,500 x g for 5 min and washed with mineral medium (MM: for 1 L; 1 g of NH₄Cl; 3.49 g of Na₂HPO₄·2H₂O; 2.77 g of KH₂PO₄, pH adjusted with NaOH 10 M to 6.8, autoclaved at 121 ºC for 20 min) supplemented with 0.1 mM succinate. Cells were diluted to a final optical density at 600 nm (OD₆₀₀) of 0.1 (~ 10⁷ CFUs mL^− 1). One mL of this cell suspension was thoroughly mixed into 100 g of soil matrix.

Twenty g of soil matrix containing the growing NatCom or the stable NatCom, inoculated or not with P. protegens CHA0, were transferred into sterile 3.3 borosilicate glass tubes (diameter Ø25 mm x 200 mm, Fisher Scientific) and, for the rhizosphere samples, a surface-disinfected and pregerminated seed of wheat (Triticum aestivum, cv. Arina) was transferred into the system in sterile conditions. Tubes without wheat were considered as bulk soil. All tubes were closed with sterile Magenta™ 2-way polypropylene caps (Sigma-Aldrich) to allow gas exchange. The tubes were kept in a Percival PGC-7L2 plant growth chamber at 22/18 ºC, 16/8 h light/dark photoperiod (light intensity, 160 µE m^− 2 s^− 1), with a relative chamber humidity of 70% until tubes were sampled. Wheat seeds were surface disinfected in 4% NaClO for 15 min and then washed ten times with 40 mL of sterile distilled water. Disinfected seeds were placed on 1.5% (w/v) agar (Agar bacteriological no. 1, Oxoid) plates and germinated for 2 days in darkness in the growth chamber before transferring them into the tubes. Sixteen tubes per condition and sampling time (see below) were prepared.

Microcosms were sampled 1, 5, 7 and 9 days post inoculation (dpi). Per sampling time, plants were carefully removed from the tubes and roots were gently shaken to remove loosely adhering soil matrix particles. Roots with the adhering soil matrix were then cut off at their point of emergence from the seed. This was considered to be the rhizosphere, which also encompasses the rhizoplane. The rhizospheres of four plants where pooled and weighed. Five g of four individual bulk soil samples were also combined into one and weighed. Four replicates per condition and timepoint were carried out (i.e., a total of 16 individual microcosms per condition and sampling time, pooled by four). Twenty mL of MM without carbon source were added to each pool of samples and vortexed for 15 min. Samples were then centrifuged at 300 x g for 1 min to pellet soil matrix debris. Then, 10 mL of the supernatants were collected and centrifuged at 7,000 x g for 15 min to pellet the bacterial cells, which were frozen at -20 ºC until DNA extraction.

Fresh weight was assessed for each pool of four shoots or roots, which were at first washed to remove adhering soil matrix particles. The individual length of shoots was measured, and average values calculated per pool of four. To assess the dry weight, roots and shoots were then fully dried for 2 weeks at 37 ºC.

Dna Extraction, Library Preparation, Sequencing And Asvs Identification

Total DNA extractions were carried out using the DNeasy PowerSoil Pro kit (Qiagen) following the manufacturer’s instructions, and DNA concentrations were measured using the Qubit™ dsDNA BR assay kit (Invitrogen). DNA samples were stored at -20 ºC. To prepare the small ribosomal subunit (16S rRNA) gene amplicon libraries, 10 ng of extracted DNA per sample were used to amplify the V3-V4 region of the 16S rRNA gene following the Illumina 16S Metagenomic Sequencing Library protocol [50] using the A Nextera XT index kit (v2, Illumina) for indexing. Samples were then quantified and pooled in equal amounts for sequencing. Pooled samples were spiked with 25% PhiX control DNA and paired-end sequenced at the Lausanne Genomic Technologies Facility (Lausanne, Switzerland), using an Illumina MiSeq v3 instrument running for 300 cycles.

The obtained raw sequence reads were quality-filtered and trimmed using fastp v0.23.2 [51] with adaptor autodetection and default parameters. Quality-filtered reads were processed following the DADA2 v1.16.0 [52] pipeline [53]. Briefly, after plotting read-quality scores, sequences were truncated when the average mean quality score dropped below 28 and using a maximum number of expected errors of 2 and 3 for forward and reverse reads, respectively. After merging error-corrected reads into sequences, the distribution of lengths was inspected and those with a length between 401 and 428 nucleotides were kept. The rest of the steps and parameters were kept as in the pipeline specified above. Sequences obtained after chimera removal were considered as amplicon sequence variants (ASVs). The number of surviving sequences per step throughout the pipeline are detailed in the Supplementary Table S1. Taxonomy was assigned using the SILVA non-redundant (nr) database, v138 [54] at 99% sequence identity. ASVs were aligned with MAFFT v7.508 [55] to then construct a maximum-likelihood (ML) phylogeny using IQ-TREE v2.2.0.3 [56] with best-fitting nucleotide substitution automatic model detection and 1,000 ultrafast bootstrap replicates.

Diversity And Composition Analyses

ASV sequences, taxonomy, metadata and phylogeny were imported into phyloseq v1.42.0 R package [57]. ASVs classified as mitochondria or chloroplasts were removed, and a prevalence filtering threshold of 0.5% across samples was applied, resulting in a total of 1,509 ASVs. Relative abundances at the bacterial class taxonomic rank were analysed and represented within the 100 most abundant ASVs, using the plot_bar function within phyloseq R package. Observed ASVs and the Shannon diversity index were calculated and plotted with the plot_richness phyloseq R function. Phylogenetic diversity (Faith’s PD) was calculated using the pd function within the picante v1.8.2 R package [58]. Non-metric multidimensional scaling (NMDS) analyses were performed with the ordinate phyloseq R function by first transforming relative abundances into a Bray-Curtis dissimilarity matrix using the vegdist function within the vegan v 2.6-4 R package [59] and a k = 2. Bray-Curtis dissimilarities were also used for a hierarchical clustering by complete-linkage and heatmap representation using the Heatmap function within the ComplexHeatmap v2.9.2 R package [60].

Null Community Model For Assembly Processes

The β-nearest taxon index (βNTI) and Raup-Crick-based Bray-Curtis (RC_bray) were used to determine the contribution of deterministic and stochastic assembly processes as previously described [61, 62]. Briefly, the βNTI measures the degree of deviation of the β-mean-nearest taxon distance (βMNTD) from the null expectations based on 1,000 random shuffles of ASVs across the phylogenetic tree. Values of |βNTI| > 2 indicate deterministic selection, which can be further partitioned into heterogeneous (βNTI > 2) or homogenous (β < 2) selection. While heterogeneous selection implies that selective pressures drive communities to divergent configurations, in homogeneous selection these selective pressures push communities towards a common composition [63]. The remaining community pairs with |βNTI| < 2 indicates that the community is mainly assembled by stochastic processes. RC_bray can be used to further classify this stochastic fraction. Values of RCbray < -0.95 indicate communities influenced by homogenising dispersal (taxonomically more similar than expected; populations are capable of interactions, allowing members to freely exchange), while RCbray > 0.95 suggest dispersal limitation; populations are unable to mix leading to development via ecological drift. Values of |RCbray| < 0.95 indicate an undominated processes, where no single assembly process is capable of explaining variation [63]. The βNTI, RC_bray and assembly processes based on entire-community null models [61] were calculated using the qpen.cm function from the iCAMP v1.5.12 R package [64].

Ecological Association Inference And Differential Abundance

Association inference between ASVs was evaluated using sparse inverse covariance estimation for ecological association inference (SPIEC-EASI) with the SpiecEasi v1.1.2 R package [65], using the Meinshausen and Bühlmann’s neighborhood selection method [66]. ASVs count table was split by tested condition and grouping timepoints 5, 7 and 9. The top 150 most abundant ASVs were retrieved. A lambda minimum ratio of 1e^− 1, nlambda = 20, and 100 replicates were chosen. Networks were visualized using the ggnetwork v0.5.10 R package. Modularity was calculated via greedy optimization of modularity using the igraph v1.3.0 R package. Keystone taxa were identified using the scaled Kleinberg’s hub centrality score [67] and a threshold of 0.7.

Differential abundance of ASVs across samples was determined using DESeq2 v1.38.0 R package [68], which normalizes ASVs by the median ratio of ASV counts relative to geometric mean per ASV [69]. Significance was assessed using the Wald test and a local estimation of dispersion. ASVs with a |log₁₀ Fold Change (FC)| ≥ 2.5 and an adjusted p-value (adj. p) < 0.01 were considered as significant.

Pseudomonas sp. ASV1168 and ASV1173 isolation and constitutive tagging

Axenic cultures of different bacterial strains from the NatCom were previously obtained and characterized [46]. We screened these isolates to search for the closest relatives of ASV1169 and ASV1173 using blastn [70]. Two Pseudomonas isolates were identified, one isolate with a 16S rRNA gene sequence identity of 98.83 with both ASV1168 and ASV1169 (hereafter Pseudomonas sp. ASV1168) and another isolate with 99.06% sequence identity with ASV1173 (hereafter Pseudomonas sp. ASV1173) over the full ASV length (427 nucleotides). Strain ASV1168 was constitutively tagged with gfp2 (ASV1168::attTn7-gfp2; Gm^R, as previously described [71]), while a spontaneous rifampicin resistant (Rif^R) mutant of strain ASV1173 was obtained by successive plating on nutrient agar (NA; Oxoid, CM0003) supplemented with 200 µg · mL^− 1 of rifampicin.

Competition Assays

Pseudomonas protegens CHA0^T-mcherry Gm^R, Pseudomonas sp. ASV1168-gfp2 Gm^R and ASV1173-Rif^R were grown overnight in NYB supplemented with Gm 10 ng · mL^− 1 or Rif 200 ng · mL^− 1, respectively. Cultures were restarted in fresh media (1:100 (v/v)) until an exponential growth was achieved (OD₆₀₀ = 0.4–0.6). Cells were collected by centrifugation at 5,500 x g for 5 min at 4 ºC, washed twice with MM and finally resuspended in SE. The OD₆₀₀ was adjusted to 0.1 in SE, corresponding to ca. 10^− 7 CFUs · mL^− 1. Serial dilutions were plated onto R2A medium to calculate initial CFU load. Cultures were mixed in pairs (1:1) or the three strains together (1:1:1). Two mL of the resulting cell mixtures were inoculated into 20 g of soil matrix (ca. 10^− 5 CFUs · g^− 1, which also accounts for 10% SE (v/w)) contained in glass tubes as the bulk soil condition, and a pregerminated surface-disinfected wheat seed was added for the rhizosphere condition, as described above. Sixteen replicates per condition were carried out. Microcosms were maintained in the plant growth chamber under the same conditions as described above. After nine days, samples were pooled by four and bacteria were extracted from the rhizosphere, or the bulk soil as described above. CFUs were counted by plating triplicate serial dilutions onto R2A medium. The competition index (CI) was calculated using the formula: CI = (Sx at t9 – Sx at t0) / (Sy t9 - Sy t0), where Sx and Sy represent the CFUs · g^− 1 of soil matrix of the two strains being compared, recovered at the final timepoint (t9 days) or inoculated (t0), as previously described [22]. Plant growth was also evaluated as described above.

Statistical Analyses

Multiple comparisons between the relative abundance of different taxonomic ranks across community states, environments and inoculation pattern were assessed with the Kruskal-Wallis rank sum test within the agricolae v1.4-5 R package [72], using normalized ASV counts by cumulative sum scaling (CSS) as previously described [73]. Post hoc tests were performed with the Fisher’s least significant difference (LSD) criterium and p-values were corrected using the false discovery rate (fdr). Differences between Shannon diversity, pairwise Bray-Curtis dissimilarities and βNTI across community states, environments and inoculation pattern were calculated using the Wilcoxon rank sum test within the stat_compare_means function of ggpubr R package. In addition, the effect of different variables on Bray-Curtis dissimilarities were assessed using PERMANOVA (permutational multivariate analyses of variance), with the adonis2 function within the vegan R package, using 9999 permutations. Spearman correlation between Shannon diversity or pairwise Bray-Curtis dissimilarities and sampling time were calculated using the stat_cor function within the ggpubr R package, and data was fitted to a general additive model (GAM), using a k = 3 with the geom_smooth function with the ggplot2 v3.3.6 R package [74] and a confidence interval of 0.95.

Pseudomonas protegens proliferates best in stable resident communities under accessibility to the rhizosphere niche

We assessed the proliferation performance of the plant-beneficial inoculant Pseudomonas protegens CHA0 in response to plant roots (wheat) when exposed to a natural soil bacterial community (NatCom) in a growing versus a stable state. Simultaneously, we analysed the relative abundance of the different bacterial classes in the growing and stable NatComs. Overall, samples of the growing NatCom were dominated by Gammaproteobacteria (Fig. 1A; Supplementary Table S2), followed by Alphaproteobacteria, Bacteroidia, Actinobacteria and Bacilli. Conversely, the stable NatCom samples were dominated by Bacilli, followed by Alphaproteobacteria, Gammaproteobacteria, Bacteroidia and Planctomycetes. These bacterial classes are commonly found in soils worldwide [75, 76] and were also previously detected in the original soil NatCom used in this work [46]. However, the unusual higher abundance of Gammaproteobacteria in the growing community compared to other soils [75, 76] might indicate specific changes resulting from the initial high availability of nutrients (i.e., growing conditions), causing a dominance of fast-growing bacteria. These differences are better highlighted in a comparison of normalized relative class abundances between samples from the growing against the stable conditions (Fig. 1B, Supplementary Fig. S1). Notably, Acidimicrobiia, Bacilli, Planctomycetes, Polyangia and Thermoleophilia were significantly more abundant in the stable condition, while Actinobacteria and Gammaproteobacteria were more abundant in the growing community state. No differences in the relative abundance of Alphaproteobacteria and Bdellovibrionia were observed for most of the comparisons. Interestingly, bacterial classes whose relative abundances increased in stable, nutrient-limited conditions, showed a reduction in the wheat rhizosphere (Fig. 1B, Supplementary Fig. S1). Although root exudates are rich in organic compounds [34], they also contain signalling molecules that could inhibit the growth of specific taxa, including Bacilli [16]. The inverse effect was observed for Gammaproteobacteria, whose abundances in the wheat rhizosphere increased in stable conditions in comparison to the corresponding bulk soil sample (Fig. 1B). In contrast, under growing conditions, most bacterial classes were not significantly different in their abundance in the wheat rhizosphere in comparison to the bulk soil. An exception were Acidimicrobiia and Actinobacteria, whose abundances increased (Fig. 1B). This might be due to a specific exploitation of root exudates. The differences observed between both community states (i.e., growing and stable) can likely be attributed to the initial access to nutrient niches in the growing condition, which allows the rapid growth of part of the population, resulting in the observed class differences.

Inoculation with P. protegens CHA0 did not significantly affect the relative abundances of the dominant bacterial classes. Notable exceptions were lower abundances of Actinobacteria, Alphaproteobacteria and Planctomycetes classes in the P. protegens-inoculated wheat rhizosphere under stable conditions (Fig. 1B, Supplementary Fig. S1). After removing P. protegens ASV counts from the Gammaproteobacteria calculations to avoid an artificial inflation, an increase in the abundance of this class was observed in the uninoculated wheat rhizosphere of stable conditions (Fig. 1B), suggesting a rhizosphere-specific effect. The fact that the largest changes in response to the plant-beneficial inoculant occur within the wheat rhizosphere under stable, nutrient-limited conditions, is probably the result of the selective effect of secreted root exudates. These contain nutrient-rich compounds that create a specific new niche within the otherwise niche-limited bulk soil, which is accessible to specific root-targeting microbiota [11, 48]. Indeed, the highest abundance of the ASV matching P. protegens CHA0 was detected in the rhizosphere of the stable NatCom (Fig. 1AC, average relative abundance across sampling times of 14.11%), which was significantly higher than in the bulk soil condition of the stable community (average relative abundance of 2.01%) and compared to the growing conditions (average relative abundance of 1.79% in the rhizosphere or 1.63% in bulk soil). The inoculant was therefore able to efficiently reach the rhizosphere microbiome and persisted for at least 9 dpi (Fig. 1C, Supplementary Table S2).

Diversity is impacted by nutrient limitation and wheat roots but not by Pseudomonas protegens proliferation

We next evaluated whether the diversity of the rhizosphere microbiome was influenced by the inoculant or by the nutrient availability. Growing NatComs exhibited significantly higher Shannon alpha diversity than stable ones (Fig. 1D), likely as a result of an initial higher availability in nutrients. Within the growing NatCom, the wheat rhizosphere significantly increased all diversity indexes (Shannon diversity, observed ASVs and Faith’s phylogenetic diversity), while the stable NatCom only showed increased Shannon diversity (Supplementary Fig. S2, Supplementary Table S3). This may point to a rhizosphere enrichment effect, and can be attributed to the secretion of specific nutrients by the wheat roots [77]. Inoculation of P. protegens CHA0 did not result in significant differences in diversity, except in the rhizosphere of the stable condition, which showed a reduced Shannon diversity (Fig. 1D). However, the number of observed ASVs and the phylogenetic diversity remained constant here (Supplementary Fig. S2). This reduction in the Shannon diversity in the wheat rhizosphere of the stable NatCom is due to the population of P. protegens CHA0 (average relative abundance across timepoints of 14.11%, Fig. 1AC), which proliferated in the niche that otherwise other NatCom bacteria would colonise. There was mostly no significant correlation of sampling time with Shannon diversity (Spearman correlation, p-value > 0.05; Fig. 1E), except for a positive correlation in the wheat rhizosphere under growing conditions (R = 0.8, p-value = 0.0019). This may be explained by the contribution of root exudates in addition to the nutrients contained in the soil extract.

Importantly, the inoculation with P. protegens CHA0 also produced a discrete increase in wheat growth compared to non-inoculated conditions (Supplementary Fig. S3). Mean weight values of fresh and dry shoots in the growing or stable conditions were higher in the inoculated samples, as were shoot lengths and root weights. The beneficial effect on plant growth of P. protegens CHA0 is in line with previous research reporting the strain’s growth-enhancing effects on different plant species [32, 78] and is related to its capacity to solubilize nutrients [79] and to synthesize phytohormones [80].

Community succession is influenced by nutrient availability, wheat roots and Pseudomonas protegens proliferation

Community succession was evaluated based on Bray-Curtis dissimilarities. Samples from both community states showed different compositional succession over a period of 9 days (Fig. 2AB) and were clearly different based on hierarchical clustering (Supplementary Fig. S4). In case of the growing NatCom, the presence of the inoculant led to measurable divergence of bulk and rhizosphere trajectories (Fig. 2A). Both in presence or absence of the P. protegens CHA0 inoculant, bulk soil and wheat rhizosphere communities followed different trajectories (PERMANOVA p-value = 0.0369). For the stable NatCom all bulk soil samples clustered together, regardless of the sampling time or the inoculation pattern (Fig. 2B, Supplementary Fig. S4), whereas the corresponding wheat rhizosphere communities diverged, depending on inoculation with P. protegens CHA0. This shows that they are undergoing successional changes, coherent with the assembly of the root microbiome [81], the proliferation of P. protegens CHA0, or a combination of both. The finding that no differences in trajectories in the bulk soil communities of stable conditions were detected in absence or presence of the inoculant (Fig. 2B, Supplementary Table S4), is likely due to the inability of the community to grow given the nutrient-deprived environment. Interreplicate variability measured as the distance of each replicate to the centroid, showed a moderate increase in variability with sampling time in the growing NatCom (Fig. 2C), probably caused by a still-evolving community, while this was only observed in the wheat rhizosphere of the stable condition, which could indicate that part of the stable NatCom population can grow using the root exudates as carbon source [34, 36]. In addition, sampling time positively correlated with Bray-Curtis dissimilarities in growing NatCom samples, except in the inoculated wheat rhizosphere (Fig. 2D) while in the stable NatCom a significant correlation with time was only observed in the rhizosphere of wheat.

These results indicate that nutrient availability must have been a major determinant for the different community trajectories between growing and stable NatComs. The successional changes that followed, responded to the presence of wheat roots or the inoculation with P. protegens CHA0, except in the bulk conditions of the stable NatCom, in which the slowed degree of compositional changes across sampling times and no effect of the inoculant point to changes related to other phenomena such as competition for growth-limiting nutrients [82].

Assembly Processes Are Governed By Nutrient Availability But Do Not Impact Overall Association Networks

Processes driving the assembly of the communities were assessed based on entire-community null models. The community state alone (i.e., growing or stable) influenced the deviation from the null models (βNTI, Fig. 3A), with growing state being the most deviated. While the growing NatCom was dominated by deterministic processes, mainly homogeneous selection representing the 90%, the assembly of the stable NatCom was driven by stochastic processes (63% of undominated processes and 35% of homogenizing dispersal, Supplementary Table S5). Environment (bulk soil or wheat rhizosphere) was a significant factor in model deviation (Fig. 3B), but inoculation was not, except in case of the stable community rhizosphere (Fig. 3C). The attributed assembly processes were similar, but the fraction of undominated processes increased up to 74% in the stable community rhizosphere (Fig. 3B). Inoculation of P. protegens CHA0 reduced the proportion of homogeneous selection in favour of undominated processes in the stable community, both for the bulk soil and the wheat rhizosphere (Fig. 3C, Supplementary Table S5).

The dominance of homogeneous selection within the growing NatCom suggests that the addition of nutrients drives the community succession [63, 83], which was expected as the replicability of the NatCom is based upon this concept [46]. However, in the absence of a dominating force (i.e., nutrients), the stable NatCom drifted apart mainly due to stochastic processes of turnover and absence of net growth [84]. Surprisingly, the proportion of undominated stochastic processes in the wheat rhizosphere of stable conditions was even higher than in the bulk soil and could be related to the observed interreplicate variability (Fig. 3B). This increased stochasticity is possibly linked to the early stages of rhizosphere microbiome formation, where root exudates serve as carbon sources for the bacteria [10], which may not be homogenously distributed in the vicinity of the roots. In addition, signalling molecules may enhance or impair the growth of certain taxa [16, 17], as would the plant immune responses [18]. Competition between the members of the community by multiple mechanisms and different levels of spatial exclusion [85] likely contribute as well to a complex mixture of processes that control the rhizosphere microbiome assembly.

The inference of taxa associations across treatments showed that communities overall were dominated by positive interactions, being about three times more abundant than the negative ones (Fig. 3DE). This finding contrasts with a current assumption that competition would be the dominant type of interaction between bacterial species [86]. However, positive correlations have been found to dominate in the rhizosphere microbiome [23]. Network modularity was almost independent of the community state, environment, or the inoculation with P. protegens CHA0, ranging from 0.7049 in the bulk soil of not inoculated stable NatCom to 0.8071 in the bulk soil of the inoculated stable NatCom (Fig. 3DE). This is in agreement with the modularity scores observed in other studies focusing on early rhizosphere microbiome and bulk soil assemblages [23]. Interesting differences, however, were found among the attributed keystone taxa in the different conditions and treatments. This attribution is based on the hub centrality score and indicated, for example, Singulisphaera as a keystone taxon in all networks from samples inoculated with P. protegens CHA0, regardless of the community state or the environment. In non-inoculated samples, Pirellula had a dominating role except for the wheat rhizosphere of the growing community where other taxa such as Agromyces emerged as keystone. Both Singulisphaera and Pirellula are largely understudied members from the Planctomycetes class, usually ubiquitous in moderately acidophilic or mesophilic terrestrial habitats [87]. This difference may not be a direct effect of inoculation with P. protegens CHA0, but rather resulting from an indirect process affecting the interaction network of taxa. The reason is that the abundances of the two Planctomycetes members were not different in presence or absence of the inoculant (see below) and might suggest that both genera exhibit similar niche exploitation in our microcosm conditions.

The Assembly Of The Natcom-derived Wheat Rhizosphere Microbiome Selects Specific Taxa

Differential abundance analyses showed that the wheat rhizosphere environment produced a significant change in ASVs belonging to the Actinobacteriota, Bacteroidota, Firmicutes, Proteobacteria and Verrucomicrobiota phyla (Supplementary Fig. S5). The number of ASVs that were significantly enriched in the wheat rhizosphere was roughly two times higher in the stable NatCom compared to the growing condition, with 33 ASVs enriched in the wheat rhizosphere compared to 12 ASVs specifically enriched in the bulk soil inoculated with P. protegens CHA0 (and 29 compared to 15 ASVs in the non-inoculated systems, respectively). These changes reflect what one would expect for an assembling selecting specialized rhizosphere microbiome [11, 88]. The enrichment of specific taxa in the wheat rhizosphere also agrees with the increase in diversity previously observed, both in stable and growing conditions (Fig. 1D) and the divergence of rhizosphere communities from their bulk soil counterpart (Fig. 2A). Taxa specifically enriched in the wheat rhizosphere belong to known plant-associated genera, notably Flavobacterium, Paenibacillus, Rhizobium group, Enterobacter and Pseudomonas [89, 90], also found associated to the wheat-rhizosphere [29, 88]. The specific enrichment of ASVs in the stable bulk soil (12 and 5 ASVs in the inoculated and non-inoculated conditions, respectively), might correspond to bacteria negatively affected by the plant (e.g., by allelopathic signalling molecules [16] or by plant immune responses [18]) or to bacteria that can secure limiting nutrients through different scavenging mechanisms, such as through the use of siderophores [82]. No specific bulk enrichment of ASVs was observed for the growing community condition.

Pseudomonas ASVs became differentially enriched depending on community states (Supplementary Fig. S5). In growing conditions, Pseudomonas ASV1173 was enriched in the rhizosphere of both, inoculated and non-inoculated samples, whereas in the stable conditions, Pseudomonas ASV1142 was found enriched, regardless of the inoculation pattern. This suggested an additional component of selection of potential competitors to CHA0 depending on the state of the resident community.

The establishment of Pseudomonas protegens alters the relative abundance of other NatCom Pseudomonas ASVs

We further explored differential changes in the relative abundance of ASVs in the growing or stable NatComs in response to the inoculation with P. protegens CHA0. Overall, the number of changing ASV relative abundances was limited to a few taxa (Fig. 4A, Supplementary Table S6), agreeing with previous studies in which the introduction of plant-beneficial inoculants produce little and transient changes in the overall rhizosphere microbiome (at least at this level of taxa resolution) [27, 91]. Most changes occurred in the wheat rhizosphere (Fig. 4A). Under nutrient-limited conditions, we observed a strain-specific selection by root exudates and/or the presence of the inoculant (e.g. Paenibacillus, Flavobacterium and Pantoea, Fig. S4). In the growing conditions, changes affected Pseudomonas ASVs different from the inoculant, i.e., ASV1168 and ASV1173, suggesting direct or indirect competition with the inoculant. The Pseudomonas genus is a highly diverse bacterial taxa, usually found in soils or associated with plants [92, 93], in which kin competition has been previously reported [22, 94].

A detailed exploration of the top five most abundant Pseudomonas ASVs (Fig. 4BC) revealed that in growing conditions, ASV1142 (assigned to P. putida, Supplementary Table S7) dominates the Pseudomonas fraction of the communities, with up to ca. 30% of the relative abundance of the total microbiome at the first sampling time (Fig. 4B). However, in stable, nutrient-limited conditions, irrespective of P. protegens CHA0 inoculation, ASV1142 became scarce, suggesting a nutrient-based growth limitation rather than competition with the inoculant (Fig. 4B). In contrast, the abundance of ASV1142 significantly increased in the wheat rhizosphere compared to stable bulk soil conditions (Fig. 4B, Supplementary Fig. S5). However, ASV1168, ASV1169 (both assigned to P. turikhanskensis) and ASV1173 (assigned to P. koreensis, Supplementary Table S7) showed a reduced relative abundance when co-inoculated with CHA0 in the wheat rhizosphere of growing conditions (Fig. 4C), which might indicate competition. In fact, ASV1168, ASV1169, ASV1173 and CHA0 (ASV1148) all belong to different subgroups within the P. fluorescens complex of species, largely known for their positive interaction with plants [92, 93], which make them likely dwellers of the wheat rhizosphere environment.

To verify this further, we conducted competition experiments (in absence of resident NatComs) between the three closely related Pseudomonas strains (Fig. 4), i.e., P. protegens CHA0 (ASV1148), and Pseudomonas sp. ASV1168 and ASV1173, both in pairs and in triplets. The results showed that CHA0 and ASV1168 are able to coexist, reaching similar proportions either in bulk soil or in the wheat rhizosphere (Fig. 5A, Supplementary Table S8). In contrast, both strains were able to outcompete Pseudomonas sp. ASV1173 in pairs, to a higher extent in the bulk soil than in the wheat rhizosphere (Fig. 5A). Co- or triple inoculation with P. protegens CHA0 increased the shoot and root biomass of the wheat plants (Fig. 5B), but the response was higher with the CHA0-ASV1173 inoculation than with the CHA0-ASV1168 pair (Fig. 5B). This could be due to the displacement of ASV1173, thus increasing the abundance of P. protegens CHA0 (Fig. 5A) capable of exerting this growth-promoting effect.

The results obtained from these competition assays contrast with those in presence of a resident soil microbiome, where an order of magnitude higher normalized relative abundance was achieved by P. protegens CHA0 compared to ASV1168 in the wheat rhizosphere or bulk soil within the growing NatCom (Fig. 4C). However, in direct competition, in absence of the NatCom, the two strains established at similar numbers. Furthermore, ASV1168 showed a significantly reduced abundance when exposed to P. protegens CHA0 in the wheat rhizosphere (Fig. 4C). These differences can be explained by indirect fine-tuning interactions with other members of the microbiome [23, 29], or they might respond to different spatial colonization patterns within the plant roots [95], highlighting the importance of structurally complex environments, such as the wheat rhizosphere or the soil matrix for the prevalence of certain bacteria.

The proliferation of a plant-beneficial bacterial inoculant in rhizosphere microbiomes remains poorly understood due to the highly complex nature of the rhizosphere and soil environments, and fully reductionist approaches that focus on bacteria-bacteria interactions or synthetic communities only provide limited insight. Nonetheless, the investigation of the interplay of an inoculant with a species-rich microbiome and a key environmental factor such as nutrient availability as undertaken in the present study might broaden our understanding of inoculant establishment strategies.

We studied the effect of the plant-beneficial strain Pseudomonas protegens CHA0 when introduced into a soil natural, species-rich bacterial community (NatCom) [46] established in bulk soil and upon its assembly in the rhizosphere of wheat. We explored two community states, based on nutrient availability, a growing NatCom produced by community dilution and addition of new nutrients, and a stable, nutrient-limited NatCom in a stationary growth state. Our results are in line with the concept that access to nutrients is a major determinant of microbiome composition and structure, diversity and assembly processes [96, 97]. The P. protegens inoculant was able to establish within the microbiomes in all the conditions tested with a relative abundance of about 2% of the entire community. In the wheat rhizosphere under stable conditions, its relative abundance peaked up to an average of 14%, which also reduced the rhizosphere microbiome diversity. This supports the concept that in a nutrient-limited environment the plant rhizosphere can provide a niche available to specific taxa, due to root exuded nutrients and signalling compounds [34, 36], and thus supports proliferation by adapted inoculants. The trajectories followed by the microbiomes during the different sampling times revealed that under growing conditions, the environment and the inoculation pattern result in a divergence of microbiome successions, leading to different community compositions. Lack of convergence along sampling times evidenced an early microbiome assembly process [23, 81]. Whether microbiome convergence will be attained or if the inoculant will also persist for a longer period remains to be studied. In addition, the changes observed in the growing condition were mainly explained by homogenous selection as the major deterministic assembly process, regardless of environment (bulk soil or wheat rhizosphere) or inoculation pattern, emphasizing nutrient availability as the dominant force driving communities to the observed compositions. Conversely, we found microbiome divergence only in stable conditions in the wheat rhizosphere while observing no differences in the inoculation regime and a reduced speed of microbiome drift in the stable bulk soil. Furthermore, in stable, nutrient-limited conditions the lack of a common driving force fits a more complex scenario, where various processes may explain the observed microbiome composition. Notably, the rhizosphere effect with root exudates providing nutrients and signalling molecules might be amplified, being the only source of available nutrients in the stable community, in which otherwise other processes such as specialized scavenging mechanisms might be more prevalent [82]. Indeed, the rhizosphere environment produced changes in certain ASVs, mainly from the Actinobacteria, Bacteroidota, Firmicutes and Proteobacteria phyla. The numbers of ASVs enriched in the wheat rhizosphere compared to the bulk soil were higher in stable conditions. Association inference of taxa showed that communities are dominated by positive interactions, while modularity remained constant across samples. Nonetheless, the presence of different keystone taxa depending on the inoculation pattern suggests altered community networks. In fact, the introduction of P. protegens CHA0 changed a discrete number of ASVs in the rhizosphere of both, growing or stable conditions, or also in the bulk soil of the stable conditions. This implies that the introduction of the inoculant does not radically alter the rhizosphere or bulk soil microbiome composition, as also observed by other authors using different Pseudomonas inoculants [26, 27]. The changes observed involve nevertheless specific Pseudomonas ASVs, whose prevalence was diminished when exposed to the inoculant. Some of these Pseudomonas ASVs belong to the Pseudomonas fluorescens species complex, commonly associated with plant hosts [92, 93], making them likely competitors of the rhizosphere environment with the inoculant. Direct competition of the inoculant with two Pseudomonas strains isolated from the NatCom revealed that competition in bulk soil or the rhizosphere environment only partially explains the observations at the microbiome level. This suggests that a more intricate network of interactions across the microbiome members or a different spatial distribution across roots or the soil interface [19, 23] could play an important role in defining the plant-rhizosphere microbiome.

The approach followed in the present study, using a species-rich natural soil bacterial community under a structurally complex environment, allowed us to demonstrate that the niche created by the wheat rhizosphere allows the proliferation of P. protegens CHA0, which also efficiently competes with closely related bacteria.

Adj.: Adjusted, ASVs: amplicon sequence variants, CFUs: colony forming units, CHA0: Pseudomonas protegens CHA0, CI: competition index, CSS: cumulative sum scaling, dpi: days post inoculation, FC: fold change, fdr: false discovery rate, GAM: general additive model, Gm: gentamycin, LSD: Fisher’s least significant difference, ML: maximum-likelihood, MM: mineral medium, NA: nutrient agar, NatCom: soil natural bacterial community, NMDS: non-metric multidimensional scaling, NYB: nutrient yeast broth, OD: optical density, PD: phylogenetic diversity, PERMANOVA: permutational multivariate analyses of variance, RC_bray: Raup-Crick-based Bray-Curtis, Rhiz.: rhizosphere, Rif: rifampicin, SE: soil extract, SPIEC-EASI: sparse inverse covariance estimation for ecological association inference, βMNTD: beta-mean nearest taxon distance, βNTI: beta-nearest taxon index

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and material

All data generated and analysed in this study is publicly available in the NCBI Sequence Read Archive (RSA) under the BioProject accession number PRJNA948847, or included as supplementary information. The original R script used in this study, the metadata file, raw ASV sequences and the taxonomy table are publicly available in GitHub (https://github.com/dgarrs/https://github.com/dgarrs/Pprotegens_proliferation_NatComs).

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the Swiss National Centre in Competence Research (NCCR) Microbiomes (grant no. 51NF40_180575), and the Swiss National Science Foundation (SNSF, grant no. 310030_184666). Open access funding was provided by the University of Lausanne, Switzerland.

Author’s contributions

Conceptualization: DGS, SC, JRvdM and CK. Methodology: DGS, SC, JV, CMH and VS. Software: DGS. Validation: DGS. Formal analysis: DGS. Investigation: DGS and SC. Resources: all authors. Data Curation: DGS, JV and CK. Writing - original draft: DGS. Writing - review and editing: all authors. Visualisation: DGS. Supervision: JRvdM and CK. Project administration: JRvdM and CK. Funding acquisition: JRvdM and CK. All authors read and approved the final manuscript.

Acknowledgements

We thank Caterina Matasci (Delley seeds and plants Ltd, Switzerland), for providing us with the wheat seeds used in this study. We also thank the Lausanne Genomic Technologies Facility in Lausanne, Switzerland for sequencing the samples.

Rodríguez H, Fraga R, Gonzalez T, Bashan Y. Genetics of phosphate solubilization and its potential applications for improving plant growth-promoting bacteria. Plant and Soil. 2006;287(1):15-21.
Glick BR, Cheng Z, Czarny J, Duan J. Promotion of plant growth by ACC deaminase-producing soil bacteria. New Perspectives and Approaches in Plant Growth-Promoting Rhizobacteria Research. 2007:329-39.
Zahir Z, Munir A, Asghar H, Shaharoona B, Arshad M. Effectiveness of rhizobacteria containing ACC deaminase for growth promotion of peas (Pisum sativum) under drought conditions. Journal of Microbiology and Biotechnology. 2008;18(5):958-63.
Spaepen S, Vanderleyden J, Remans R. Indole-3-acetic acid in microbial and microorganism-plant signaling. FEMS Microbiology Reviews. 2007;31(4):425-48.
Haas D, Défago G. Biological control of soil-borne pathogens by fluorescent pseudomonads. Nature Reviews Microbiology. 2005;3(4):307-19.
Vacheron J, Péchy-Tarr M, Brochet S, Heiman CM, Stojiljkovic M, Maurhofer M, et al. T6SS contributes to gut microbiome invasion and killing of an herbivorous pest insect by plant-beneficial Pseudomonas protegens. The ISME Journal. 2019;13(5):1318-29.
Kupferschmied P, Maurhofer M, Keel C. Promise for plant pest control: root-associated pseudomonads with insecticidal activities. Frontiers in Plant Science. 2013;4:287.
Haskett TL, Tkacz A, Poole PS. Engineering rhizobacteria for sustainable agriculture. The ISME Journal. 2021;15(4):949-64.
Tabassum B, Khan A, Tariq M, Ramzan M, Khan MSI, Shahid N, et al. Bottlenecks in commercialisation and future prospects of PGPR. Applied Soil Ecology. 2017;121:102-17.
Badri DV, Weir TL, van der Lelie D, Vivanco JM. Rhizosphere chemical dialogues: plant–microbe interactions. Current Opinion in Biotechnology. 2009;20(6):642-50.
Liu F, Hewezi T, Lebeis SL, Pantalone V, Grewal PS, Staton ME. Soil indigenous microbiome and plant genotypes cooperatively modify soybean rhizosphere microbiome assembly. BMC Microbiology. 2019;19(1):1-19.
Lundberg DS, Lebeis SL, Paredes SH, Yourstone S, Gehring J, Malfatti S, et al. Defining the core Arabidopsis thaliana root microbiome. Nature. 2012;488(7409):86-90.
Philippot L, Raaijmakers JM, Lemanceau P, Van Der Putten WH. Going back to the roots: the microbial ecology of the rhizosphere. Nature Reviews Microbiology. 2013;11(11):789-99.
İnceoğlu Ö, Salles JF, van Overbeek L, van Elsas JD. Effects of plant genotype and growth stage on the betaproteobacterial communities associated with different potato cultivars in two fields. Applied and Environmental Microbiology. 2010;76(11):3675-84.
Peiffer JA, Spor A, Koren O, Jin Z, Tringe SG, Dangl JL, et al. Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proceedings of the National Academy of Sciences. 2013;110(16):6548-53.
Cotton TA, Pétriacq P, Cameron DD, Meselmani MA, Schwarzenbacher R, Rolfe SA, et al. Metabolic regulation of the maize rhizobiome by benzoxazinoids. The ISME Journal. 2019;13(7):1647-58.
Yu P, He X, Baer M, Beirinckx S, Tian T, Moya YA, et al. Plant flavones enrich rhizosphere Oxalobacteraceae to improve maize performance under nitrogen deprivation. Nature Plants. 2021;7(4):481-99.
Stassen MJ, Hsu S-H, Pieterse CM, Stringlis IA. Coumarin communication along the microbiome–root–shoot axis. Trends in Plant Science. 2021;26(2):169-83.
Klein M, Stewart JD, Porter SS, Weedon JT, Kiers ET. Evolution of manipulative microbial behaviors in the rhizosphere. Evolutionary Applications. 2022;15(10):1521-36.
Bernal P, Llamas MA, Filloux A. Type VI secretion systems in plant‐associated bacteria. Environmental Microbiology. 2018;20(1):1-15.
Tyc O, Song C, Dickschat JS, Vos M, Garbeva P. The ecological role of volatile and soluble secondary metabolites produced by soil bacteria. Trends in Microbiology. 2017;25(4):280-92.
Vacheron J, Heiman CM, Keel C. Live cell dynamics of production, explosive release and killing activity of phage tail-like weapons for Pseudomonas kin exclusion. Communications Biology. 2021;4(1):1-14.
Shi S, Nuccio EE, Shi ZJ, He Z, Zhou J, Firestone MK. The interconnected rhizosphere: high network complexity dominates rhizosphere assemblages. Ecology Letters. 2016;19(8):926-36.
Kiers ET, Denison RF. Sanctions, cooperation, and the stability of plant-rhizosphere mutualisms. Annual Review of Ecology, Evolution, and Systematics. 2008;39:215-36.
Zhalnina K, Dias R, de Quadros PD, Davis-Richardson A, Camargo FA, Clark IM, et al. Soil pH determines microbial diversity and composition in the park grass experiment. Microbial Ecology. 2015;69(2):395-406.
Buddrus-Schiemann K, Schmid M, Schreiner K, Welzl G, Hartmann A. Root colonization by Pseudomonas sp. DSMZ 13134 and impact on the indigenous rhizosphere bacterial community of barley. Microbial Ecology. 2010;60(2):381-93.
Yin D, Wang N, Xia F, Li Q, Wang W. Impact of biocontrol agents Pseudomonas fluorescens 2P24 and CPF10 on the bacterial community in the cucumber rhizosphere. European Journal of Soil Biology. 2013;59:36-42.
Xiong W, Jousset A, Li R, Delgado-Baquerizo M, Bahram M, Logares R, et al. A global overview of the trophic structure within microbiomes across ecosystems. Environment International. 2021;151:106438.
Rossmann M, Perez-Jaramillo JE, Kavamura VN, Chiaramonte JB, Dumack K, Fiore-Donno AM, et al. Multitrophic interactions in the rhizosphere microbiome of wheat: from bacteria and fungi to protists. FEMS Microbiology Ecology. 2020;96(4):fiaa032.
Thakur MP, Geisen S. Trophic regulations of the soil microbiome. Trends in Microbiology. 2019;27(9):771-80.
Gao Z, Karlsson I, Geisen S, Kowalchuk G, Jousset A. Protists: puppet masters of the rhizosphere microbiome. Trends in Plant Science. 2019;24(2):165-76.
Keel C, Ucurum Z, Michaux P, Adrian M, Haas D. Deleterious impact of a virulent bacteriophage on survival and biocontrol activity of Pseudomonas fluorescens strain CHA0 in natural soil. Molecular Plant-microbe Interactions. 2002;15(6):567-76.
Yang K, Wang X, Hou R, Lu C, Fan Z, Li J, et al. Rhizosphere phage communities drive soil suppressiveness to bacterial wilt disease. Microbiome. 2023;11(1):1-18.
Liu S, He F, Kuzyakov Y, Xiao H, Hoang DTT, Pu S, et al. Nutrients in the rhizosphere: A meta-analysis of content, availability, and influencing factors. Science of The Total Environment. 2022:153908.
Chaparro JM, Sheflin AM, Manter DK, Vivanco JM. Manipulating the soil microbiome to increase soil health and plant fertility. Biology and Fertility of Soils. 2012;48(5):489-99.
Bais HP, Weir TL, Perry LG, Gilroy S, Vivanco JM. The role of root exudates in rhizosphere interactions with plants and other organisms. Annual Review of Plant Biology. 2006;57(1):233-66.
Camenzind T, Hättenschwiler S, Treseder KK, Lehmann A, Rillig MC. Nutrient limitation of soil microbial processes in tropical forests. Ecological Monographs. 2018;88(1):4-21.
Bardgett R. The biology of soil: a community and ecosystem approach. Oxford University Press; 2005.
Durán D, Bernal P, Vazquez-Arias D, Blanco-Romero E, Garrido-Sanz D, Redondo-Nieto M, et al. Pseudomonas fluorescens F113 type VI secretion systems mediate bacterial killing and adaption to the rhizosphere microbiome. Scientific Reports. 2021;11(1):5772.
Gu Y, Dong K, Geisen S, Yang W, Yan Y, Gu D, et al. The effect of microbial inoculant origin on the rhizosphere bacterial community composition and plant growth-promotion. Plant and Soil. 2020;452:105-17.
Wang Z, Chen Z, Kowalchuk GA, Xu Z, Fu X, Kuramae EE. Succession of the resident soil microbial community in response to periodic inoculations. Applied and Environmental Microbiology. 2021;87(9):e00046-21.
Schmidt R, Köberl M, Mostafa A, Ramadan EM, Monschein M, Jensen KB, et al. Effects of bacterial inoculants on the indigenous microbiome and secondary metabolites of chamomile plants. Frontiers in Microbiology. 2014;5:64.
Fitzpatrick CR, Salas-González I, Conway JM, Finkel OM, Gilbert S, Russ D, et al. The plant microbiome: from ecology to reductionism and beyond. Annual Review of Microbiology. 2020;74:81-100.
Liu Y-X, Qin Y, Bai Y. Reductionist synthetic community approaches in root microbiome research. Current Opinion in Microbiology. 2019;49:97-102.
Schäfer M, Vogel CM, Bortfeld-Miller M, Mittelviefhaus M, Vorholt JA. Mapping phyllosphere microbiota interactions in planta to establish genotype–phenotype relationships. Nature Microbiology. 2022;7(6):856-67.
Čaušević S, Tackmann J, Sentchilo V, von Mering C, van der Meer JR. Reproducible Propagation of Species-Rich Soil Bacterial Communities Suggests Robust Underlying Deterministic Principles of Community Formation. mSystems. 2022;7(2):e00160-22.
Zhou X, Wang J, Liu F, Liang J, Zhao P, Tsui CK, et al. Cross-kingdom synthetic microbiota supports tomato suppression of Fusarium wilt disease. Nature Communications. 2022;13(1):7890.
Ramette A, Frapolli M, Fischer-Le Saux M, Gruffaz C, Meyer J-M, Défago G, et al. Pseudomonas protegens sp. nov., widespread plant-protecting bacteria producing the biocontrol compounds 2, 4-diacetylphloroglucinol and pyoluteorin. Systematic and Applied Microbiology. 2011;34(3):180-8.
Rochat L, Péchy-Tarr M, Baehler E, Maurhofer M, Keel C. Combination of fluorescent reporters for simultaneous monitoring of root colonization and antifungal gene expression by a biocontrol pseudomonad on cereals with flow cytometry. Molecular Plant-microbe Interactions. 2010;23(7):949-61.
Illumina (2013) 16S Metagenomic Sequencing Library Preparation, Part # 15044223 Rev. B. https://support.illumina.com/content/dam/illumina-support/documents/documentation/chemistry_documentation/16s/16s-metagenomic-library-prep-guide-15044223-b.pdf Accessed 1 Jan 2021.
Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884-i90.
Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nature Methods. 2016;13(7):581-3.
DADA2 (2020) Pipeline Tutorial (1.16). https://benjjneb.github.io/dada2/tutorial.html Accessed 1 Jan 2021.
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Research. 2012;41(D1):D590-D6.
Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Molecular Biology and Evolution. 2013;30(4):772-80.
Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, Von Haeseler A, et al. IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Molecular Biology and Evolution. 2020;37(5):1530-4.
McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PloS One. 2013;8(4):e61217.
Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Morlon H, Ackerly DD, et al. Picante: R tools for integrating phylogenies and ecology. Bioinformatics. 2010;26(11):1463-4.
Dixon P. VEGAN, a package of R functions for community ecology. Journal of Vegetation Science. 2003;14(6):927-30.
Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32(18):2847-9.
Stegen JC, Lin X, Fredrickson JK, Chen X, Kennedy DW, Murray CJ, et al. Quantifying community assembly processes and identifying features that impose them. The ISME Journal. 2013;7(11):2069-79.
Xu M, Huang Q, Xiong Z, Liao H, Lv Z, Chen W, et al. Distinct Responses of Rare and Abundant Microbial Taxa to In Situ Chemical Stabilization of Cadmium-Contaminated Soil. mSystems. 2021;6(5):e01040-21.
Stegen JC, Lin X, Fredrickson JK, Konopka AE. Estimating and mapping ecological processes influencing microbial community assembly. Frontiers in Microbiology. 2015;6:370.
Ning D, Yuan M, Wu L, Zhang Y, Guo X, Zhou X, et al. A quantitative framework reveals ecological drivers of grassland microbial community assembly in response to warming. Nature Communications. 2020;11(1):1-12.
Kurtz ZD, Müller CL, Miraldi ER, Littman DR, Blaser MJ, Bonneau RA. Sparse and compositionally robust inference of microbial ecological networks. PLoS Computational Biology. 2015;11(5):e1004226.
Meinshausen N, Bühlmann P. High-dimensional graphs and variable selection with the lasso. The Annals of Statistics. 2006;34(3):1436-62.
Kleinberg JM. Authoritative sources in a hyperlinked environment. Journal of the ACM (JACM). 1999;46(5):604-32.
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology. 2014;15(12):1-21.
Anders S, Huber W. Differential expression analysis for sequence count data. Nature Precedings. 2010:1-.
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. Journal of Molecular Biology. 1990;215(3):403-10.
Péchy‐Tarr M, Borel N, Kupferschmied P, Turner V, Binggeli O, Radovanovic D, et al. Control and host‐dependent activation of insect toxin expression in a root‐associated biocontrol pseudomonad. Environmental Microbiology. 2013;15(3):736-50.
Mendiburu Fd, Yaseen M: Agricolae: statistical procedures for agricultural research. R package version 1.4. 0. In.; 2020.
Paulson JN, Stine OC, Bravo HC, Pop M. Differential abundance analysis for microbial marker-gene surveys. Nature Methods. 2013;10(12):1200-2.
Wickham H. Data analysis. In: ggplot2. Springer; 2016. p. 189-201.
Delgado-Baquerizo M, Oliverio AM, Brewer TE, Benavent-González A, Eldridge DJ, Bardgett RD, et al. A global atlas of the dominant bacteria found in soil. Science. 2018;359(6373):320-5.
Bahram M, Hildebrand F, Forslund SK, Anderson JL, Soudzilovskaia NA, Bodegom PM, et al. Structure and function of the global topsoil microbiome. Nature. 2018;560(7717):233-7.
Berendsen RL, Pieterse CM, Bakker PA. The rhizosphere microbiome and plant health. Trends in Plant Science. 2012;17(8):478-86.
Meyer JB, Lutz MP, Frapolli M, Péchy-Tarr M, Rochat L, Keel C, et al. Interplay between wheat cultivars, biocontrol pseudomonads, and soil. Applied and Environmental Microbiology. 2010;76(18):6196-204.
de Werra P, Péchy-Tarr M, Keel C, Maurhofer M. Role of gluconic acid production in the regulation of biocontrol traits of Pseudomonas fluorescens CHA0. Applied and Environmental Microbiology. 2009;75(12):4162-74.
Beyeler M, Keel C, Michaux P, Haas D. Enhanced production of indole-3-acetic acid by a genetically modified strain of Pseudomonas fluorescens CHA0 affects root growth of cucumber, but does not improve protection of the plant against Pythium root rot. FEMS Microbiology Ecology. 1999;28(3):225-33.
Cordovez V, Rotoni C, Dini-Andreote F, Oyserman B, Carrión VJ, Raaijmakers JM. Successive plant growth amplifies genotype-specific assembly of the tomato rhizosphere microbiome. Science of The Total Environment. 2021;772:144825.
Kramer J, Özkaya Ö, Kümmerli R. Bacterial siderophores in community and host interactions. Nature Reviews Microbiology. 2020;18(3):152-63.
Dini-Andreote F, Stegen JC, Van Elsas JD, Salles JF. Disentangling mechanisms that mediate the balance between stochastic and deterministic processes in microbial succession. Proceedings of the National Academy of Sciences. 2015;112(11):E1326-E32.
Zhou J, Ning D. Stochastic community assembly: does it matter in microbial ecology? Microbiology and Molecular Biology Reviews. 2017;81(4):e00002-17.
Ghoul M, Mitri S. The ecology and evolution of microbial competition. Trends in Microbiology. 2016;24(10):833-45.
Palmer JD, Foster KR. Bacterial species rarely work together. Science. 2022;376(6593):581-2.
Kulichevskaya IS, Ivanova AO, Baulina OI, Bodelier PL, Damste JSS, Dedysh SN. Singulisphaera acidiphila gen. nov., sp. nov., a non-filamentous, Isosphaera-like planctomycete from acidic northern wetlands. International Journal of Systematic and Evolutionary Microbiology. 2008;58(5):1186-93.
Kavamura VN, Mendes R, Bargaz A, Mauchline TH. Defining the wheat microbiome: towards microbiome-facilitated crop production. Computational and Structural Biotechnology Journal. 2021;19:1200-13.
Simonin M, Dasilva C, Terzi V, Ngonkeu EL, Diouf D, Kane A, et al. Influence of plant genotype and soil on the wheat rhizosphere microbiome: evidences for a core microbiome across eight African and European soils. FEMS microbiology ecology. 2020;96(6):fiaa067.
Kour D, Rana KL, Yadav N, Yadav AN, Kumar A, Meena VS, et al. Rhizospheric microbiomes: biodiversity, mechanisms of plant growth promotion, and biotechnological applications for sustainable agriculture. In: Kumar A, Meena V, editors. Plant growth promoting rhizobacteria for agricultural sustainability: from theory to practices. Singapore: Springer; 2019. p. 19-65.
Qiao J, Yu X, Liang X, Liu Y, Borriss R, Liu Y. Addition of plant-growth-promoting Bacillus subtilis PTS-394 on tomato rhizosphere has no durable impact on composition of root microbiome. BMC Microbiology. 2017;17(1):1-12.
Garrido-Sanz D, Meier-Kolthoff JP, Göker M, Martín M, Rivilla R, Redondo-Nieto M. Genomic and genetic diversity within the Pseudomonas fluorescens complex. PloS One. 2016;11(2):e0150183.
Garrido-Sanz D, Redondo-Nieto M, Martin M, Rivilla R. Comparative genomics of the Pseudomonas corrugata subgroup reveals high species diversity and allows the description of Pseudomonas ogarae sp. nov. Microbial Genomics. 2021;7(6).
Heiman CM, Maurhofer M, Calderon S, Dupasquier M, Marquis J, Keel C, et al. Pivotal role of O-antigenic polysaccharide display in the sensitivity against phage tail-like particles in environmental Pseudomonas kin competition. The ISME Journal. 2022;16(7):1683-93.
Massalha H, Korenblum E, Malitsky S, Shapiro OH, Aharoni A. Live imaging of root–bacteria interactions in a microfluidics setup. Proceedings of the National Academy of Sciences. 2017;114(17):4549-54.
Trivedi P, Leach JE, Tringe SG, Sa T, Singh BK. Plant–microbiome interactions: from community assembly to plant health. Nature Reviews Microbiology. 2020;18(11):607-21.
Song C, Jin K, Raaijmakers JM. Designing a home for beneficial plant microbiomes. Current Opinion in Plant Biology. 2021;62:102025.

No competing interests reported.

Download PDF

Journal Publication

published 29 Sep, 2023

Read the published version in Microbiome →

Editorial decision: Major revision
14 Aug, 2023
Reviews received at journal
18 Jul, 2023
Reviews received at journal
11 Jun, 2023
Reviewers agreed at journal
22 May, 2023
Reviewers agreed at journal
20 May, 2023
Reviewers invited by journal
20 May, 2023
Editor assigned by journal
29 Mar, 2023
Submission checks completed at journal
28 Mar, 2023
First submitted to journal
28 Mar, 2023

You are reading this latest preprint version

Changes in structure and assembly of a species-rich soil natural community with contrasting nutrient availability upon establishment of a plant-beneficial Pseudomonas in the wheat rhizosphere

Status:

Journal Publication

Version 1

Abstract

Background:

Results:

Conclusions:

Figures

Background

Materials And Methods

Experimental design

Dna Extraction, Library Preparation, Sequencing And Asvs Identification

Diversity And Composition Analyses

Null Community Model For Assembly Processes

Ecological Association Inference And Differential Abundance

Competition Assays

Statistical Analyses

Results And Discussion

Assembly Processes Are Governed By Nutrient Availability But Do Not Impact Overall Association Networks

The Assembly Of The Natcom-derived Wheat Rhizosphere Microbiome Selects Specific Taxa

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1