Culturomics of Andropogon gerardii rhizobiome revealed nitrogen transforming capabilities of stress-tolerant Pseudomonas under drought conditions

Background Climate change will result in more frequent droughts that impact soil-inhabiting microbiomes in the agriculturally vital North American perennial grasslands. In this study, we used the combination of culturomics and high-resolution genomic sequencing of microbial consortia isolated from the rhizosphere of a tallgrass prairie foundation grass, Andropogon gerardii. We cultivated the plant host-associated microbes under artificial drought-induced conditions and identified the microbe(s) that might play a significant role in the rhizobiome of Andropogon gerardii under drought conditions. Results Phylogenetic analysis of the non-redundant metagenome-assembled genomes (MAGs) identified the bacterial population of interest – MAG-Pseudomonas. Further metabolic pathway and pangenome analyses detected genes and pathways related to nitrogen transformation and stress responses in MAG-Pseudomonas. Conclusions Our data indicate that the metagenome-assembled MAG-Pseudomonas has the functional potential to contribute to the plant host’s growth during stressful conditions. This study provided insights into optimizing plant productivity under drought conditions.


Background
Global climate change is a serious concern, resulting in soil degradation, soil erosion, and impacts on soil health [1]. Climate change has severe impacts worldwide including in the USA, resulting in more frequent and prolonged droughts [2], and is gradually degrading the plant diversity and ecosystem functions [3]. The rhizobiome, microbial communities that are intimately associated with the plant rhizosphere [4,5], is one of the key factors in maintaining ecosystem function, soil quality and plant health [6,7]. The plant rhizosphere is the primary site for plant-microbe and microbe-microbe interactions, governed primarily by root exudates [8]. Microbes in the rhizobiome can facilitate plant host nutrient and water uptake, element cycling (carbon, nitrogen, phosphorus) and other processes that are beneficial to plants [9][10][11].
Rhizobiomes are also instrumental in enhancing plant hosts' resistance and resilience against abiotic stresses such as drought, salinity, and heavy metal exposure [12].
Therefore, with the more frequent and more extreme droughts events predicted in the global climate change scenarios in the future, it is even more urgent to provide insights into the mechanisms of how the rhizobiome may promote plant host resilience and response to stress. Although there are various studies that have dissected how climate change impacts the rhizobiome [13][14][15][16], more concerted efforts are needed to provide insights into the mechanisms of how the rhizobiome can enhance the plant host resilience during drought-induced experimental stress.
Previous studies have reported a clear contribution from plant-associated microbial members to plant growth and resilience during drought conditions [17][18][19]. Plant growth-promoting bacteria (PGPB) reportedly enhance plant growth during drought [20,21], an observation attributed to the microbial nitrogen cycling and transformation in soil [22]. Therefore, candidate microbes capable of nitrogen transformation and increase nitrogen availability in the rhizosphere have been the key targets in a growing number of experimental and observational studies that focus on the assembly of plant health promoting Synthetic Communities (SynCom) [23,24]. SynComs have been successfully deployed to alter the plant phenotype, to enhance plant disease resistance and productivity [25,26]. However, it is challenging and tedious to select optimal members of SynComs because of the lack of knowledge of the microorganisms that could impart favorable functions under stressful conditions [27]. Therefore, in identifying candidate microbes for SynComs, it may be more expedient to identify specific microbial functions and mechanisms rather than to depend solely on taxonomy.
Our long-term, ongoing research on the microbiome of dominant Great Plains prairie grass Andropogon gerardii (Big Bluestem) provided an excellent opportunity to acquire deeper insights into the microbial functional potential under abiotic stress [28][29][30]

MAGs analysis, phylogenetic analysis, and identification of MAG-Pseudomonas
We cultured the A. gerardii rhizosphere microbial populations using the following samples and media -dry ecotype in R2A, dry ecotype in R2A with PEG, wet ecotype in R2A, and wet ecotype in R2A with PEG. We expected that the PEG-amended media would yield bacterial populations enriched with drought resistant gene functions. We recovered a total of 125 MAGs and generated a total of 63 non-redundant MAGs from the four conditions ( Figure 1A, Supplementary  [54][55][56]. Pseudomonas thivervalensis were isolated from the roots of Brassica napus and Arabidopsis thaliana [57], and is a functionally significant member of soil microbial communities [55]. Pseudomonas have also been implicated to be a plant growth-promoting rhizobacteria (PGPR) and have been associated with plant growth, control of pathogenicity [55] and aid in plant resilience under drought-stressed conditions [54,56].  [58]. In our study, the amendment of PEG in R2A for cultivation not only mimicked the drought conditions, but also generated moderate levels of osmotic shock for the cells [59], the conditions that often accompany drought conditions [60]. Furthermore, we identified Ves proteins during drought/ osmotic stress in MAG-Pseudomonas. Our results suggest that MAG-Pseudomonas had the potential to be more resilient and tolerant against drought stress, demonstrating moderate tolerance to dehydration and water limiting conditions [61][62][63][64]. Under stress, most microbes will restructure their metabolism and specifically activate various stress pathways [17]. Pseudomonas can utilize a range of mechanisms such as alginate [65] and trehalose production [66] to be more resilient during drought conditions. Besides the 14 stress related gene functions, we also identified DNA-binding transcriptional regulator YbjK, (Figure 2; Supplementary Table S2) which is involved in the regulation of stress response [67]. Drought conditions often result in the generation of oxidative stresses in both plants and their associated microbes [68,69]. There is established documentation about the beneficial role of Pseudomonas to alleviate the oxidative stress [70,71]. Our results illustrated that our MAG-Pseudomonas had several regulatory responses that might enhance its drought stress resilience and fitness.  Table S2). These genes have been previously reported in Pseudomonas spp [72][73][74]. All the nitrogen transformation gene functions that were detected in our MAG-Pseudomonas can be essential in helping to fulfill the plant host's need for nitrogen, especially in N-depleted soils [75][76][77][78]. Bacterial Nif genes transform the atmospheric nitrogen to the form that can be utilized by the plants [79,80], whereas NtrY encodes for the sensory kinase of the two-component regulatory system of NtrY/NtrX associated with nitrogen metabolism [81]. NtrB also plays a role in nitrogen metabolism and can regulate the nitrogen dynamics under nitrogen-deprived and enriched environments [82]. NtrC is another nitrogen metabolism regulator that contributes to nitrogen assimilation [73]. Similarly, nitrogen regulatory protein PII (GlnK) and nitrogen PTS system EIIA components are also involved in regulating nitrogen metabolism [83]. Besides the Nif genes, we further detected gene functions in the MAG-Pseudomonas that corresponded to nitrogen assimilation and nitrogen dissimilation (nitrification and denitrification) (Figure 2, Supplementary Table S2), which contributes to the nitrogen cycle [84]. Assimilatory nitrate reductase catalytic subunit was identified in this study that catalyzes the process from nitrate to nitrite [85]. NAD(P)H-nitrite reductase, a large subunit (NirB) was detected that can catalyze nitrite reduction, and forms ammonia [86]. We also detected nitrite reductase (NADH) large and small subunits that can carry out similar processes and contribute to the nitrogen cycle [86].

Nitrogen transformation potential of MAG-Pseudomonas could enhance
We used the comparative pathway tool in PATRIC, and identified 138 potential pathways of MAG-Pseudomonas based on genomic information from 3 Pseudomonas  Table S3). We further analyzed the differential occurrence of the genes in MAG-Pseudomonas and the 3 Pseudomonas genomes, and observed that there was a higher occurrence of nitrate reductase, nitrate reductase aspargine synthase (glutamine-hydrolyzing) (n=1) ( Figure 3B). We observed a high number of genes that were related to nitrate reductase, suggesting its importance to the nitrogen metabolism in our MAG-Pseudomonas. The nitrate reductase is critical in reducing nitrate to nitrite for several crop plants, as this reaction leads to the production of proteins that are necessary to maintain plant health [88]. Glutamate synthases are actively involved in ammonia assimilation pathways in bacteria [89], while glutamate dehydrogenase has a prominent role in nitrogen assimilation and is capable of maintaining the balance of carbon and nitrogen [90]. Putting it all together, our resolved MAG-Pseudomonas with its potential in microbial-driven nitrogen transformation processes could play a critical role in the regulation of primary productivity of its plant host, A. gerardii, even during times of drought-induced stress. Desiccation stress tolerance proteins with LEA/WHy domain (LEA) is suggested to confer a broad range of stress response function to bacteria such as Escherichia coli [91], while genes corresponding to a WHy protein homologue have been identified in both archaea and bacteria including Pseudomonas [92,93] although the specific function in Pseudomonas is still incomprehensible. Our findings in MAG-Pseudomonas and 6-closely related genomes provided insights into potential gene functions in

MAG-Pseudomonas is
Pseudomonas that could be instrumental in providing resilience against drought induced stress. We also identified a set of universal stress proteins (UspA, UspE), which belonged to bacterial universal stress proteins, and were produced under stressful conditions [94]. We also identified other genes -YaaA (oxidative stress); TypA/BipA (general stress-response regulator, [95]; BolA (family transcriptional regulator, [96]; and Ribosomal protein L25 (general stress protein Ctc) (RplY) [97], that demonstrated the potentiality of Pseudomonas to elicit one or more microbial mechanisms to become more resilient when subjected to abiotic stresses. Similar to stress response genes, our findings, which identified numerous nitrogen transformation gene functions (Figure 4, Besides the core-clusters gene functions, we also observed genes related to chemotaxis in the MAG-Pseudomonas accessory gene-clusters. We detected genes corresponding to methyl-accepting chemotaxis protein and chemotaxis protein CheD ( Figure 4, Supplementary Table S4). Our resolved MAG-Pseudomonas might show chemotaxis towards certain amino acids by using methyl-accepting chemotaxis proteins [98], as these bacterial cells are known to methylate the methyl-accepting chemotaxis proteins when adapting to environmental repellents and attractants [99]. Similarly, CheD chemotaxis proteins might be used by MAG-Pseudomonas to attract or evade various environmental stimuli [100][101][102]. Our MAG-Pseudomonas also had a gene corresponding to insecticidal toxin complex protein TccC. These proteins exhibit toxicity to a wide range of insects that could be utilized in designing strategies for crop protection [103]. Interestingly, we also identified the LuxR family transcriptional regulator, quorum-sensing system regulator ExpR in MAG-Pseudomonas, suggesting that MAG-Pseudomonas might use the LuxR proteins to communicate with neighboring bacteria [104] involving Acyl homoserine lactone (AHL)-dependent Quorum Sensing mechanism [105]. MAG-Pseudomonas, therefore, has a plethora of gene functions that might enable the microbe to show phenomena such as chemotaxis and quorum sensing.
Tailoring SynComs is an important approach to provide insights into plant host-microbe interactions. Understanding the mechanisms and functions of host-associated microbial populations is particularly relevant in the construction of these plant-associated SynComs. Our study showed that MAG-Pseudomonas not only possessed the resiliency to survive in drought-induced conditions, but were also able to perform essential microbial functions for generating products related to the nitrogen cycle [106], which could be exploited by plant host and other host-associated microbes [107]. A SynCom consisting of six Pseudomonas strains isolated from the garlic rhizosphere has been reported to promote plant growth [108]. Apart from the potential to contribute to the plant host's well-being, our MAG-Pseudomonas might also be able to influence and interact with other bacteria, [109], contributing to its role as an important member of the core rhizobiome along with other members such as Streptomyces, Rhizobium, Burkholderia, Nitrosomonas, Nitrospira, Azospirillum, Bradyrhizobium, and Azotobacter [110]. Overall, our study emphasized that the understanding of the MAG-Pseudomonas mechanism and functional potential might contribute to the successful construction of a SynCom that can benefit the plant-host during drought-induced stress [27].

Conclusion
In this study, we used culturomics and metagenomic strategy to identify bacterial populations in the A. gerardii rhizobiome, and identified MAG-Pseudomonas as the candidate microbe that had significant functional potential in nitrogen transformation and stress response. In support of other studies, our study verified the abundance of MAG-Pseudomonas in the rhizobiome and suggested its potential pivotal role under drought conditions. In a continuing effort to understand the contributions of different microbiota in the plant rhizobiome, it is important to remember that identity and relative abundance alone may not truly reflect the relative functional importance of the bacterial population.
Instead, understanding the functional role of the microbe during host-microbe and microbe-microbe interactions might provide more insights. The functional potential of our resolved MAG-Pseudomonas, resulting from a combination of conventional culturing and high-throughput analysis, showed the immense potential to inform and refine our efforts to dissect the mechanistic interaction taking place in the rhizobiome.

Plot design, sampling, and cultivation of rhizosphere communities from soil samples
We collected Andropogon gerardii rhizosphere samples from a common garden in Colby at the Kansas State University Agricultural Research Center located in Thomas County (39°23′N, 101°04′W). Further information on the experimental layout, ecotypes, and sampling collections has been described previously [30]. In this comparative study, we selected rhizosphere samples from native dry (Hays, Kansas) and wet (Carbondale, Illinois) ecotypes for microbial cultivation. We separated bulk soil from the soil attached to the rhizosphere by handshaking the roots gently. We dissolved 0.1 g of the rhizosphere samples in 0.9 ml of Phosphate-Buffered Saline (PBS) [pH 7] buffer, serially diluted the samples (10 -1 -10 -6 ), and spread 100µl solution onto the Petri plates.
We designed two culture conditions -0.315% R2A media (Teknova, USA) [31] and 0.315% R2A media amended with a 36% Polyethylene Glycol 8000 (PEG) (Ψ = -1.54 MPa) to alter the media osmotic potential and to mimic absence and presence of water limitation, respectively [32,33]. A similar range of PEG concentrations has been used to simulate dry environments in other studies [34,35]. DNA Flex for library preparation and S1 flow cell. We used the program 'iu-filer-qualityminoche' [38] to process the short metagenomic reads, and the quality-filtered short reads were assembled into longer contiguous sequences (contigs) using MetaHit [39] with a minimum contig length of 1000 bp. We then identified open reading frames (ORFs) in the contigs, and recruited metagenomic short reads to the contigs. We used CONCOCT [40] to bin the metagenomes, and used anvi'o ver 7.0 [41] to manually curate the bins into metagenome-assembled genomes (MAGs) that satisfied the conditions of >70% completion and <10% redundancy based on single copy genes.
NCBI's Cluster of Orthologous Groups (COGs) [42] was used to assign functions to the ORFs. The MAGs were assigned to taxa using the single-copy core genes of bacteria and archaea. We further compared the resolved MAGs using Average Nucleotide Identity (ANI) [43] to identify non-redundant MAGs based on 95% ANI [44].

Phylogenetic, pathway, and pangenomic analyses
Among the resolved MAGs, there was a MAG of interest for this study: MAG-

Pseudomonas. The selected non-redundant MAG was analyzed by Similar Genome
Finder service that uses the MinHash on the Pathosystems Resource Integration Center (PATRIC) web portal [45,46]. Similar genomes deposited in public databases were obtained and used to estimate the genome distances to the MAG-Pseudomonas. We constructed a phylogenetic tree for the selected non-redundant MAG and 40 closely related genomes. The workflow used the PATRIC Codon Tree Service which used the amino acid and nucleotide sequences from a well-defined database of global protein families [47]. Then, we used the RAxML program [48] to construct a tree based on the pairwise differences between the aligned protein families of the selected sequences.
We used the comparative pathway tool of PATRIC to predict the metabolic pathways in our selected MAG. To compare the pathways, we selected Pseudomonas genomes from rhizospheres of cotton and soybean. KEGG maps and heat maps of the nitrogen metabolism pathway were generated in the PATRIC portal.
We downloaded 6 closely-related Pseudomonas genomes from NCBI RefSeq [49] and performed pangenomic analyses using anvi'o workflow [41,50]. The workflow uses BLASTP [51] to compute amino acid level similarities between all possible ORF pairs.
We then used Markov Cluster Algorithm (MCL) [52] to group ORFs into homologous gene clusters and aligned amino acid sequences in each gene cluster using MUSCLE for visualization [53]. We determined the core gene clusters of the MAG-Pseudomonas and the 6 additional, available Pseudomonas genomes, as well as the accessory gene cluster of MAG-Pseudomonas.

Availability of data
The raw data used in this study are publicly available at NCBI under the project     analyzed the data. M.G., A.J., and L.J. reviewed the writing for the manuscript. S.L.
conceptualized, performed the data analysis, supervised, was responsible for resource acquisitions, and wrote the manuscript. All the authors contributed to the article and approved the submitted version.

Ethics approval and consent to participate
Not Applicable.