The root microbiome: techniques for exploration and agricultural applications

Modern agricultural practices are under pressure from a growing population, climate change, land degradation and limited resources. To meet population demand, agricultural production would need to increase 70% from its current level by 2050 [1]. However, recent studies estimate that over the next 25 years, global food production will decrease by 12% [2].

In response, agricultural scientists have shifted their attention to a solution just below the surface – the root microbiome. The root microbiome consists of the community of microbes living in association with a plant's root system that plays a crucial role in plant health and productivity [3]. The microbiome broadens plants' functional systems by participating in various processes like growth promotion, nutrient absorption and stress resistance [3]. As a result, plant phenotypes are influenced by both genotype and microbiome. Therefore, a better understanding of the molecular mechanisms of host–microbiota interactions could later enable us to manipulate the microbiome to increase agricultural production, reduce disease, decrease greenhouse gas emissions and minimize chemical use [4–7].

Here, we explore the interactions between host plants and their microbiota, highlight recent advancements in our studies of root microbiomes and look at how these findings could contribute to the development of sustainable agricultural practices.

Root microbiota community structure

Root-resident microbes can be separated into three niches: bulk soil, the rhizosphere and endosphere microbiota. Bulk soil is regarded as the ‘seed bank’ for root microbial communities. The microbes within each niche are derived from the bulk soil through horizontal transfer. Root exudates, substances secreted by the root, recruit microbes from the bulk soil to the rhizosphere. These exudates include primary metabolites that are mostly responsible for attracting the microbes, and secondary metabolites that screen the recruited microbes. Entry of microbes from the rhizosphere to the endosphere is controlled by the hosts' immune system and therefore depends on their genotype [3].

Other factors also shape the microbial communities formed in each niche, including the environment, plant–microbe interactions and microbe–microbe interactions.

Factors affecting root microbiota

Environment

Environmental factors and soil physiochemical properties (climate, moisture, nutrient availability and soil organic matter content) affect root bacterial community diversity. This consequently affects plant growth and development [3].

Phosphorus (Pi), for instance, is a vital macronutrient for plants. Phosphorus ions easily form complexes with other ions in soil that would otherwise be difficult for plants to absorb. In response to phosphorus restrictions, plants undergo the phosphate starvation response (PSR), triggering changes in morphology, such as an increase in root to shoot ratio [8], physiology and metabolism, like the upregulation of Pi transporters of the plasma membrane [9]. PHR1, a key transcription factor, regulates both PSR and plant immunity, inhibiting microbe-induced immune responses during phosphorus deficiency [10,11]. Therefore, PSR affects root microbiota structure, leading to an atypical composition. For instance, Pi-stressed plants are susceptible to colonization by latent opportunistic competitors within their microbiome, exacerbating their Pi starvation [12].

Host

The species of plant influences the structure of rhizosphere communities through differences in root morphology and metabolite exudation [13]. These biological, chemical and physical changes that occur in the soil, known as rhizosphere effects, also regulate root microbiota composition [3].

Nutrient mobilization and the host plants' immune system and responses to stress influence microbial community structure [3]. Plants have a sophisticated immune system that allows them to permit commensals and mutualists to colonize their roots whilst excluding harmful microbes. The immune system achieves this by actively overlooking colonization by beneficial microbes. These microbes will also evade or suppress the host immune system [14].

As the name suggests, mutualists establish mutualistic structures with plants; using plants to obtain carbon sources whilst enabling them to absorb nutrients. Pathogens, on the other hand, deprive hosts of nutrients, disturb physiological processes, inflict tissue damage and arrest growth by producing toxins, cell wall-degrading enzymes and virulence proteins [15].

Most plant-associated microbes, however, are commensals. Commensals are loosely host dependent, but in certain situations, can be recruited to influence plant growth and development through beneficial, harmful, or neutral interactions [3]. Therefore, some commensals can act as mutualists and pathogens in different scenarios.

Plants have a wide variety of cell-surface and intracellular immune receptors to detect immunogenic signals associated with pathogen infection [15]. This interaction subsequently activates defense mechanisms that operate through two major layers of plant immunity. The first takes place at the cells' surface and is known as pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI). When pathogens bind to cell-surface pattern recognition receptors (PRRs), this mechanism becomes activated. In plants, cell-surface receptors consist of receptor-like proteins (RLPs) and receptor kinases (RKs) that are conceptually analogous to Toll-like receptors in animals. Activation of these receptors leads to early resistance responses, such as the production of reactive oxygen species (ROS), stomatal closure and activation of mitogen-activated protein kinase (MAPK) [15].

The second line of defense, effector-triggered immunity (ETI), is intracellular and is activated by the recognition of pathogen type III secreted effectors (T3SES) with plant intracellular resistance (R) proteins. Intracellular immune receptors in plants are nucleotide-binding, leucine-rich repeat receptors (NLRs), which also exist in animals. This mechanism is associated with programmed cell death following infection and is known as the hypersensitive response (HR) [15].

Effects of root microbiota on hosts

Enhance performance under stress

The function of the microbial community is closely associated with the host's regulation. The root microbiota can directly, through specific members of the microbial community, or indirectly, through microbe–microbe interactions, improve plants' resistance to abiotic stresses [16–18]. For instance, they can enhance host plants' drought stress adaptation, which has become a pressing issue within the agricultural industry.

In a 2021 study by Santos-Medellín et al. it was found that short-term drought conditions altered root microbiome structure of rice plants; however, its pre-stress structure was largely regained following watering. Conversely, long-term drought caused severe and irreversible effects on the endosphere community, even after watering [19].

Another global threat to agriculture is soil salinization. As a result of improper fertilizer and water management, the area of saline-alkali soil is increasing; however, manipulation of the microbial community could be used to improve hosts' ability to cope with salt stress [3]. For example, in Arabidopsis, inoculation with a fungal isolate of Hypoxylon spp., boosts salt tolerance [20].

Facilitate mineral element uptake

Root microbiota directly promote plants' ability to absorb mineral elements, which are essential for plants to carry out a myriad of vital biological processes, including photosynthesis and respiration [21,22]. These microbes also indirectly help maintain plants' mineral balance by regulating the development of diffusion barriers in the root endodermis [3].

Improve disease resistance

Commensals support plants' ability to resist pathogen invasion. This can occur through direct pathogen suppression mechanisms like niche competition and antibiotic secretion, or indirectly through increasing the physical or chemical barrier protecting the host, known as induced systemic resistance (ISR) [23,24].

Regulate plant growth & development

Based on the above evidence, there is no doubt that microbial communities regulate host growth and development. Case in point, modifying host ethylene levels alters plant phenotypes; however, the effect depends on the subsequent growth-defense trade-offs [25]. Pseudomonas putida UW4 reduces ethylene levels, promoting plant growth, but results in hypersensitivity to stress [26].

In Arabidopsis, specific soil microbes can alter flowering time; a result of nitrification by rhizosphere microbes that increase and prolong nitrogen bioavailability, delaying flowering and promoting vegetative growth [27].

Techniques for studying the root microbiome

Research into the root microbiome mainly consists of culture-dependent and -independent approaches (Figure 1).

Culture-dependent approaches

Classic microbiology relies on isolating, culturing and purifying microbes from a sample using nutrient media growth conditions that vary depending on the target species. Initial studies of the root microbiome using this approach were able to identify the cultured species and their abundances based on their morphology and numbers. However, replicating natural conditions suitable for the microbe growth in a laboratory is challenging. In addition, a significant proportion of microbes are unculturable, leading to a lack of information on the actual microbial diversity within a sample. Nonetheless, culture-dependent approaches remain important for research and are increasingly moving toward microbiome functionality, e.g., the reductionist synthetic community (SynCom) approach [3].

The SynCom approach is an emerging technique that aims to mimic the structure and function of a microbiome by co-culturing multiple taxa under well-defined and controlled conditions. It incorporates a synthetic biology approach with knowledge obtained from microbial community analyses, metagenomics and bioinformatics. The goal is to simplify the complexity of the original microbial community while maintaining crucial interactions between the microbes and their hosts. By deciphering the dynamic interactions within microbial ecosystems, we can enhance stability in engineered microbial consortia through predictable and synergistic interactions [29].

Despite its potential, the SynCom approach faces challenges when it comes to large-scale application; with hundreds of microbes, it is impractical due to technological limitations. One solution is to construct SynComs with inoculants that possess multiple beneficial traits for and carry out an array of synergistic interactions with the plant of interest. Inoculants are added to either soil or plant to improve crop productivity and health. However, maintaining multiple species is challenging due to population fluctuations caused by the influence of medium composition and stochastic events [29].

Most studies have been conducted in controlled systems, which differ from the diversity of natural environments [29]. Evaluating the stability and performance of SynComs under field conditions is crucial. Several studies have reported that SynCom applications enhanced plant growth under greenhouse conditions as well as field conditions [30–33]. However, extensive field studies across various climates would be necessary to confirm the applied inoculant.

Culture-independent

Culture-independent methods have revolutionized our ability to explore the diversity of plant microbiome species. By bypassing the need for culturing, these methods enable us to identify a greater variety of species and have allowed us to systematically explore the functional aspects of root microbiomes [3]. By integrating them with traditional culture-based approaches, we have the potential to elevate our understanding of plant–microbe interactions to a systems level.

Metagenomics

Metagenomic studies perform DNA extraction of total DNA from environmental samples using DNA extraction kits [34].

To study prokaryotes, PCR amplification of the ubiquitous 16S ribosomal RNA (rRNA) gene is most used. By sequencing the variable regions, we enable precise taxonomic identification at species and strain level. Therefore, high-throughput sequencing technologies have been widely adopted as they allow scientists to identify thousands to millions of sequences in a sample, revealing the abundance of even rare microbial species [35].

In contrast, eukaryotes present a challenge as the equivalent rRNA gene (18S) does not provide enough taxonomic discrimination because it is not sufficiently polymorphic to enable the classification of groups at a low taxonomic level. To overcome this limitation, researchers often use the hypervariable internal transcribed spacers 1 or 2 (ITS 1/2) [34]. The ITS sequences are sections of nonfunctional DNA located between the genes that code for the subunits of ribosomal RNAs. The issue with this approach is that primer design inherently biases PCR amplification of genomic DNA and generally only identifies the target organisms [35]. Due to the intricacies of the interactions within the root microbiome, it is important to capture as much diversity as possible. This is where global analyses such as metagenomics, metatranscriptomics and metaproteomics, which allow the simultaneous assessment of microbial populations across all domains, come in.

Next-generation sequencing (NGS) techniques, such as Illumina (CA, USA) HiSeq and MiSeq sequencing, have proven instrumental in detecting microbiome genomic characteristics involved in plant–microbial interactions, including responses to abiotic and biotic stresses, metabolite production and utilization [34]. NGS also facilitates the identification of microbial functional traits essential for plant growth and sustainable agriculture. Recent studies have highlighted the utility of Illumina platforms in studying and identifying microorganisms in farmland soil [37–39].

While Illumina remains the most used NGS system, attention has recently shifted toward third-generation sequencing technologies, namely the MinION sequencer from Oxford Nanopore (UK) and PacBio's (CA, USA) HiFI sequencing technologies [40]. A key advantage of third-generation sequencing is that reads are completed in 4–6 hours rather than in days [34]. This technique is also more affordable, can be connected to a laptop, does not require special computer or equipment training for data analysis, and can obtain ultra-long reads, up to thousands of bases. They also provide a high taxonomic resolution by generating a bacterial and archaeal population profile derived from the full length of the 16S rRNA gene sequence [41–44]. Although metagenomics has allowed us to profile plant-growth promoting (PGP) rhizobacteria and gain a comprehensive understanding of their functional characteristics and the roles of plant microbial communities in soil, fully elucidating the associations between plants and microorganisms requires studying these traits in situ. This can be achieved by combining metagenomics with other techniques like transcriptomics and proteomics, offering a more holistic perspective on plant-microbe interactions and their underlying mechanisms [34].

Metaproteomics

The aim of metaproteomic techniques is to identify and quantify the total protein content of a sample [45]. The process involves protein extraction, trypsin proteolysis, detection of the resulting peptides by tandem mass spectrometry, interpretation of MS/MS spectra to assign peptide identities and higher-level interpretation in terms of taxonomy and function. However, metaproteomic methods need to be developed to suit individual soil types as the humic acids and contaminants can interfere with protein extraction [46].

The acquired MS/MS spectra are subsequently analyzed by comparing them to a database containing the sequences of all proteins potentially present in the sample. To construct this database, the most effective approach involves conducting metagenomics or metatranscriptomics on the same sample [32]. Another option is to compile protein sequences from organisms identified in similar samples to complement the metagenomic information. Alternatively, a specific database can be created based on organisms identified through 16S rRNA amplicon sequencing and taxonomical assignment, or those potentially present in the habitat the sample was taken from [47–49].

However, soil metaproteomics remains challenging due to the large proportion of organisms within the samples that have yet to be taxonomically characterized, and the small fraction of reference genome sequences that are available in public data repositories [45]. Additionally, the community structure of these samples can vary over time and space. Nonetheless, pioneering studies have been conducted on soils from diverse environments such as forests, agricultural areas, mining drainage, permafrost and arid environments [50–56]. However, the extent to which specific soil gene catalogs contribute to the enhancement of metaproteomic interpretation has not yet been estimated [45].

Kingdom-level changes in crop–plant rhizosphere microbiomes have been uncovered using metaproteomics. Notably, it was found that the relative abundance of eukaryotes in the rhizospheres of peas and oats was five times higher when compared to plant-free soil [57].

Metatranscriptomics

Metatranscriptomics enables the simultaneous study of gene expression and abundance of microorganisms in an environmental sample, offering valuable insights into the active community members and metabolic pathways [58].

As the RNA-seq workflow is very complicated, it is easy to produce bias through the processing steps. This may damage the quality of the RNA-seq dataset and lead to an incorrect interpretation of the true microbial diversity within samples [59]. However, the eukaryotic mRNAs used in metatranscriptomic studies have a poly(A) tail that enables cDNA synthesis from mRNA templates in total RNA pools with selective primers, mitigating some of this bias.

A challenge of metatranscriptomic studies is the enrichment of mRNA. Total RNA is dominated by rRNA transcripts like prokaryotic 16S and 23S rRNAs and eukaryotic 18S and 28S rRNAs and only a small fraction, 1 to 5%, is comprised of mRNA [58]. Therefore, scientists are testing several strategies to enrich prokaryotic mRNA and reduce the rRNA fraction of metatranscriptomes. These strategies include polyadenylation of mRNA, selective nuclease degradation of rRNA and rRNA depletion by capture with commercial kits and sample specific probes [60–62]. Combining stable isotope probing (SIP) with metatranscriptomics enhances the detection and identification of mRNA from target microorganisms [63].

While challenging, the dominance of rRNA in a metatranscriptomic sample allows for robust assessment of the entire microbiome without prior selection of the taxonomic groups of interest [57]. This approach is less challenging than enrichment of mRNA, avoids PCR bias and can be carried out on multiple samples. However, like with metaproteomics, the co-extraction of enzyme-inhibiting compounds such as humic and fulvic acids and phenolic compounds from samples can be challenging [58]. This is a particular issue within metatranscriptomics as large quantities of RNA are needed for double stranded cDNA synthesis prior to sequencing. However, several studies have proposed strategies to remove these inhibitory substances, including using Sephadex spin columns and polyethylene glycol precipitation of nucleic acids [64,65].

Another limitation of metatranscriptomics is the short half-lives of mRNA molecules, which vary among species and different genes [61]. Consequently, changes in transcript patterns can occur due to alterations in soil conditions at the time of sample retrieval. RNA isolation from soils presents specific challenges due to the inaccessibility of cells situated on and within soil particles, inefficient cell lysis, the binding of RNA to soil particles and the presence of RNases [64]. However, it was found that rapid flash freezing of peat soil samples in liquid nitrogen immediately after sampling counteracted changes in transcript patterns [58].

Despite these challenges, metatranscriptomics has been successfully applied in studies of the gene expression of plant growth-promoting rhizobacteria in different plant systems. For example, in a study conducted on Mexican maize roots, metatranscriptomics was used to investigate the interactions between PGPRs [67]. The researchers specifically examined the gene expression of Rhizobium phaseoli in the presence of other PGPRs and observed several upregulated genes associated with nitrogen fixation proteins, nitrogenase cofactor biosynthesis and nitrogenase stabilization.

Overall, metatranscriptomics offers valuable insights into gene expression patterns in microbial communities. By studying the expression patterns of PGPRs in different plant systems, researchers can uncover the intricate interactions between microorganisms and plants, ultimately contributing to the development of improved agricultural practices based on microbiome interactions.

Sustainable agriculture applications

Despite the progress that has been made in our understanding of the root microbiome, using this knowledge to improve crop growth and development is still in early stages partially because microbes with PGP traits tend to perform poorly in natural environments [68].

For inoculants to be successful, they need to be able to compete with other microbes, efficiently colonize and establish a stable association with plants throughout the growing season [69]. This requires an in-depth knowledge of microbial abundance, diversity and plant–microbe interactions to be able to predict functionality.

Traditionally, beneficial microbes with PGP traits are selected based on in vitro screening with limited assessment under controlled environmental conditions [70]. Current inoculants are also formulated with pure isolates. Therefore, under field conditions, these microbes tend to yield inconsistent results and often fail to compete with indigenous soil microbes when exposed to different climactic conditions, soil types and other environmental factors [71,72].

The positive impact of root microbiota on plant health and productivity is evident. However, effectively managing the interactions between multiple organisms with crops under diverse conditions is complex. Whilst significant technological advancements have been made, more research is required to gain a comprehensive understanding of plant–microbe interactions at a systems level and how they can be manipulated to achieve desired traits in crops. This will likely require the integration of available technologies and also a long-term multidisciplinary approach involving microbiologists, plant biologists and agronomists.

In coming years, it is likely that the root microbiome will play a more active role in the development of sustainable agricultural practices.