Designing Microbial Inoculants for Agroecosystems: Integrating Soil and Plant Context

2. Soil Fertility and the Microbiome: The Relationship Between Plant Yield and Soil Type with the Use of Inoculants

2.1. The Influence of Abiotic Factors on the Effectiveness of Microbial Inoculants

The effectiveness of microbial inoculants depends on soil type, fertility and climate conditions. This has been confirmed by numerous studies, which have also shown that all these factors are closely related. A meta-analysis has shown that the introduction of nitrogen-fixing microorganisms can contribute to an increase in soil organic carbon (SOC) reserves. However, this effect is not uniform and is influenced by climatic conditions (Sun et al. 2025). The greatest efficiency of nitrogen fixation in SOC accumulation is observed in warmer and drier regions. A positive correlation with average annual temperature and a negative correlation with precipitation indicates the maximum advantage of these microorganisms under stressful conditions. Thus, the successful use of nitrogen-fixing inoculants, such as those based on rhizobia or free-living bacteria, is highly dependent on the specific region. The greatest benefits are expected not in optimal conditions, but in marginal conditions, where the ability of these inoculants to provide nitrogen for plants is fully realized, leading to increased soil fertility. This shifts the strategy from searching for "universal" strains to regionally adapted solutions.

2.2. The Combined Influence of Climate and Soil Conditions

A comprehensive study on wild strawberries demonstrates that the formation of the plant microbiome is determined by a stable set of abiotic factors, the influence of which is replicated across continents (Mittelstrass et al. 2021). This conclusion is supported by a pan-European study of Arabidopsis thaliana's phyllosphere, which revealed two distinct microbiome types geographically confined to Northern and Southern Europe. The key factor explaining this variability is the Palmer Drought Stress Index (PDSI), which is the best predictor of microbiome composition surpassing other climatic parameters and plant phenotypic traits (Karasov et al. 2024; Mohkam-Singh et al. 2025; Semenov et al. 2025). Taken together, these data indicate that the natural assembly of the plant microbiome is predetermined by a complex of abiotic factors, primarily moisture availability. This creates a fundamental basis for regional variability in soil and phyllosphere communities, into which artificially introduced inoculants must integrate.

2.3. The Role of Soil Type and Fertility Gradients

The key parameters that determine the type and quality of soil are SOC and pH, which directly shape the environment for the microbiome and serve as the main abiotic filters (Mohkam-Singh, Nunes, 2025). The direct influence of soil type on rhizobiome formation is supported by a fundamental study on Arabidopsis thaliana, where it was shown that soil type is a primary factor significantly influencing the bacterial community in the rhizosphere and root endosphere (Lundberg et al. 2012).

A meta-analysis provides a quantitative assessment of the feasibility of using bioinoculants to increase SOC. Although the average increase was +0.44 grams C/kg, this effect was highly context-dependent. It was statistically significant and greatest (+0.97 grams C/ kg) in tropical humid climates, where soils had low initial SOC levels and high mineralization rates. In contrast, the effect was not significant in arid climates or on soils with high initial levels of SOC (Just et al. 2024).

A recent study has shown that the relationship between SOC and the agronomic effectiveness of bioinoculants follows an inverted U-shaped curve. The highest level of effectiveness is achieved at moderate levels of SOC (~1.5%) (Huang et al. 2025). This phenomenon can be explained by the concept of the "SOC-immune-microbiome axis". A moderate level of SOC represents a "window of opportunity", where the resource base is sufficient to support the activity of inoculants, but the immune and competitive barriers in the rhizosphere have not become insurmountable yet.

Experimental confirmation of the power of this factor comes from a study in which the authors created a soil fertility gradient by mixing soils with high and low fertility. The gradient acted as an ecological filter, forming statistically different bacterial communities at each point (Qiao et al. 2024). This result not only confirms the importance of fertility, but also provides a methodology: using these gradients as a tool to stress test and select the most competitive strains for creating synthetic consortia (Qiao et al. 2024).

Data analysis shows that the effectiveness of bioinoculants (their contribution to soil organic carbon accumulation and colonization success) depends non-linearly on the soil context, primarily on initial soil organic carbon content and pH. The maximum increase in SOC is observed in conditions that promote rapid organic matter turnover, such as warm and humid regions. Soil fertility itself acts as a powerful filter, forming a resident community that contributes to the overall carbon cycle. This requires a shift from generic solutions to more precise ones: mandatory pre-plant soil testing, the use of soil fertility gradients to select stress-tolerant crop varieties, and the regional adaptation of fertilizer formulations. Therefore, the use of bioinoculants should be based on a precise consideration of soil conditions and the target function, rather than on the assumption of their universal effectiveness.

3. Secondary Antimicrobial Metabolites of Bioinoculants are a Factor in the Formation of the Plant Rhizosphere and Their Impact on Crop Yield and Fertility

3.1. From Antagonism to Signaling: A Paradigm Shift in Understanding the Role of Antibiotics

Until the beginning of the 21st century, antibiotics produced by microorganisms were mainly viewed in the context of interspecies competition, as a tool for direct antagonism to inhibit the growth of competing organisms (Waksman 1947) (Figure 2A). This property has long been seen as a key factor in determining the effectiveness of bioinoculants. However, with the advent of omics technology, a new perspective has emerged that sees antibiotics as signaling and regulatory molecules (Davies 1990; Linares et al. 2006; Yim et al. 2007; Fajardo, Martínez 2008; Romero et al. 2011; Spagnolo et al. 2021). In this capacity, antimicrobial metabolites of bioinoculants act as ecosystem engineers, capable of deliberately altering not only the taxonomic composition, but also the functional state of the entire microbial community (Figure 2B).

A significant argument in favor of this concept is direct concentrations of antibiotics in natural environments. A meta-analysis of 887 measurements showed that their content in soils and sediments typically ranges from 1/4 to 1/230 of clinically relevant doses (Chow et al. 2021). This suggests that their primary ecological role is likely not lethal antagonism, but rather sublethal regulatory effects.

3.2. In Situ Evidence Suggests that Bioinoculants Create a Stable "Chemical Environment" in the Rhizosphere

Direct quantitative measurements using High-Performance Liquid Chromatography-Mass Spectrometry confirm that bioinoculant strains actively synthesize antibiotics directly in the rhizosphere (Raaijmakers, Mazzola 2012). For example, in cucumber inoculated with Bacillus subtilis, surfactin and iturin A concentrations reach 33 and 630 μg/g fresh weight of roots, respectively, by the 78th day of vegetation, even under competitive pressure in highly fertile soil (Kinsella et al. 2009). Other studies have found 59 micrograms of surfactin and 1 milligram of iturin per gram of cucumber roots (Kinsella et al., 2009) and 0.3 micrograms of surfactin and 0.02 micrograms of fengycin per gram of tomato roots (Debois et al. 2015). Similarly, on rice roots, B. subtilis NH-100 produced up to 200 mg/L surfactin (Sarwar et al. 2018). Pseudomonas, in turn, produces 1.65-2.77 μg/g of 2,4-diacetylphloroglucinol (2,4-DAPG) in the wheat rhizosphere, depending on the strain (Raaijmakers, Mazzola 2012).

Taken together, these data suggest that bioinoculants create a stable and growing chemical background of biologically active compounds in the rhizosphere. This background inevitably exerts selective pressure on the composition and structure of the microbial community.

3.3. Mechanisms of Influence: Molecular Interference and Chemical Communication

Bioinoculants alter the functional state of the microbiome not only through direct antimicrobial action but also through complex molecular interference (Figure 2B). Studies have shown that secondary metabolites can cause a significant change in the metabolome profile of cells in response to chemical signals from pathogens. For instance, Bacillus amyloliquefaciens alters its spectrum of synthesized macrolactins and polyketides when exposed to metabolites from Ralstonia solanacearum (Onuh et al. 2025). A key finding was that the production of protective lipopeptides in B. velezensis was dramatically increased in the presence of the pathogen, which acted as a metabolic trigger (Cao et al. 2018). In parallel, B. velezensis metabolites modulate the Phytophthora sojae transcriptome by suppressing virulence genes (pectate lyases, components of the MAPK/PKA signaling pathways).

Secondary metabolites of Pseudomonas also play a role in interspecies and interpopulation communication. In the rhizosphere, 2,4-DAPG and phloroglucinol act as signals. 2,4-DAPG increases the expression of the phlA gene in other strains, while phloroglucinol regulates the production of pyoluteorin in neighboring bacteria (Maurhofer et al. 2004; Clifford et al. 2016). 2,4-DAPG from P. kilonensis F113 enhances the phytostimulating effect of Azospirillum brasilense on wheat (Combes-Meynet et al. 2011). Treatment with 2,4-DAPG of Neurospora crassa conidia also causes a transient increase in intracellular Ca^2+, suggesting its role in signaling pathways (Troppens et al. 2013).

These data suggest that bioinoculants may be involved in a chemical arms race with pathogens, dynamically influencing their physiology and reducing their pathogenic potential through mechanisms other than simple antagonism (Figure 2B). Genomic studies confirm this ecological specialization, showing that rhizobacteria, particularly those belonging to the genus Pseudomonas, have an increased capacity for secondary metabolite synthesis compared to soil isolates (Stringlis et al., 2018).

3.4. Antibiotics as "Ecosystem Engineers": from Cells to Metabolites

The question arises: if antibiotic production is a key feature of bioinoculants and living cells have a limited lifespan, can purified metabolites be used? Although research in this area is still limited, existing data suggests that such a strategy may be promising, but there is a critical caveat: Only metabolites that act at environmentally relevant concentrations and in the context of rhizosphere interactions are effective. A meta-analysis of the effects of tetracyclines, quinolones and sulfonamides shows that their use leads to a significant reduction in microbial biomass: total - by 17%, bacterial - by 20%, fungal - by 30%. This confirms their fundamentally different, destructive impact on soil ecosystems (Chen et al. 2023). Combined stressors, such as conventional antibiotics and elevated temperature, disrupt microbial communities and nutrient cycling, further causing destabilization of microbial networks (Jane et al. 2021). This demonstrates a fundamental difference between the toxicological effects of extra-systemic chemical stressors and the regulatory function of autochthonous signaling molecules in their natural ecological niche.

3.5. Selective Remodeling of the Rhizobiome by Specific Metabolites

A pioneering 1998 study showed that the production of trifolitoxin antibiotic by Rhizobium etli on beans selectively reduces the diversity of susceptible proteobacteria, without affecting the majority of the microbial community. This provides early evidence for antibiotics as selective ecosystem engineers, rather than agents of total suppression (Robleto et al. 2008).

Modern research not only specifies, but also significantly expands this concept. It demonstrates that secondary antimicrobial metabolites are capable not only of restructuring the composition of a community, but also of stimulating its functional activity. A striking example is gliotoxin, a metabolite of fungi of the genus Trichoderma and Aspergillus. New data show that this compound acts as a "driving factor" for stress, to which microbes respond not by suppression, but by adaptation and a sharp increase in functional activity (Teslya et al. 2024a). Its influence significantly increases microbial respiration (mineralization activity) and enhances the activity of enzymes key to the carbon, nitrogen, and phosphorus cycles. This demonstrates that antibiotics can serve as stimulants, forcing communities to restructure themselves to perform ecological functions more efficiently (Teslya et al. 2024b). Other metabolites exhibit a similar yet specific complex effect. For example, cyclic lipopeptides produced by B. velezensis selectively alter the taxonomic composition of the microbiome, increasing fungal diversity and activating enzymes involved in carbon, nitrogen, and phosphorus cycling (Vasilchenko et al. 2025).

The most detailed picture of ecological engineering is revealed by metagenomic studies, demonstrating how different antimicrobial metabolites of PGP bacteria specifically restructure the functional landscape of soil communities. Metagenomic profiling has shown that macrolactin A from B. velezensis acts as a highly specific environmental filter, suppressing a range of Gram-positive bacteria while significantly increasing the relative abundance of key symbiotic genera of Pseudomonadota, such as Bradyrhizobium and Mesorhizobium, which play a crucial role in nitrogen fixation (Yuan et al. 2016; Poshvina et al. 2025). Functional gene analysis revealed that exposure to macrolactin A triggers profound microbiome restructuring, the nature of which is critically dependent on dose. Low concentrations of macrolactin A act as a signaling modulator, enriching genes involved in chemotaxis, intercellular communication, and efflux pump function, indicating adaptive community mobilization. At high doses, macrolactin A acts as a selective factor, shifting the metabolic profile towards increased biosynthesis of secondary metabolites and ion transport.

A similar dose-dependent principle is demonstrated by 2,4-DAPG, which, by suppressing Mucoromycota and Actinomycetota, causes enrichment of the phylum Pseudomonadota and dose-dependent activation of enzymes of the carbon and nitrogen cycles (Teslya et al. 2025). Functional profiling revealed that exposure to high doses of 2,4-DAPG causes targeted suppression of metabolic pathways associated with maintaining cellular vitality in the microbiome. Simultaneously, there is a sharp activation of adaptive systems designed to respond to stress. The greatest enrichment is observed in gene categories responsible for xenobiotic resistance and stabilization of cellular macromolecules. This indicates a selective pressure that selects strains with enhanced mechanisms for detoxification and maintaining homeostasis under antimicrobial stress.

3.6. Towards a Strategy for Targeted Selection of Bioinoculants

The analysis of the presented data leads to a fundamental conclusion: antimicrobial metabolites act as "engineers", carrying out targeted remodeling of the rhizobiome through a universal mechanism called ecological dualism, - "signal-stressor". At low, ecologically relevant concentrations, these compounds serve as signaling molecules, mobilizing communities (as in the case of macrolactin A). However, at higher doses, they act as selective agents that radically alter the metabolic landscape, favoring adapted and functionally beneficial taxa, as demonstrated by 2,4-DAPG.

Therefore, when developing bioinoculants and selecting strains, it is strategically important to shift the focus from searching for producers of antibiotics with broad "killing power" to selecting strains that synthesize compounds that act as subtle "ecosystem engineers." The key criterion should be their ability not to completely suppress the rhizobiome, but to precisely modulate its structure and function. Ideal candidates, such as cyclic lipopeptides, 2,4-DAPG, and gliotoxin, are effective precisely because their action goes beyond simple antagonism.

4. Practical Recommendations for the Creation of Bioinnoculants

A modern trend in agriculture is the transition from empirical screening to targeted design of bio-products based on ecological principles. The data presented in previous chapters allow us to formulate a systematic strategy aimed at overcoming key barriers to field efficiency.

A key paradigm shift involves moving away from the focus on strains with the maximum expression of specific PGP traits (in vitro production of phytohormones and antibiotics) and towards selection based on "ecological competence". A promising candidate strain should be able to not only grow well in a Petri dish, but also possess a range of adaptations for survival and success in the highly competitive environment of the rhizosphere. These include:

1. The ability to withstand the environmental challenges of the target agricultural ecosystem, such as low pH and osmotic stress (Figure 3A).

2. The capacity to resist the effects of specific secondary metabolites produced by the host plant, such as the benzoxazinoid APO in cereals (Figure 3B).

3. The presence of social interaction mechanisms: the capacity for chemotaxis, quorum sensing, and the formation of cooperative biofilms with other rhizobacteria (Figure 3B).

A striking illustration of this approach is the B. velezensis SQR9 strain, whose effectiveness is due to its outstanding ability to restructure microbial networks (Liu et al. 2025). It acts as a "social hub". By exhibiting selective antagonism, it suppresses part of the native community, while simultaneously creating conditions for enrichment with moderately related cooperative Bacillus strains. This leads to the formation of a stable and functionally active consortium. Precisely this "network competence" should become the priority criterion for selection (Figure 3C).

Critical progress has been made in implementing the principle of selecting based on ecological competence with the development of the rhizoSMASH computational tool (Li et al. 2025). This algorithm, based on genomic synteny, predicts the presence of catabolic gene clusters associated with rhizosphere competence (rCGCs) in bacteria, that is, their genetically determined ability to utilize specific components of root exudates.

The presence of these clusters has been shown to be highly correlated with successful colonization of the rhizosphere, which has been confirmed in two independent validation studies. RhizoSMASH therefore provides a missing link between genomic data and field performance predictions, enabling the targeted selection of strains with the potential for successful colonization in a specific plant host during genomic screening, before the need for labor-intensive seeding experiments.

4.2. Consortium Design: the Principle of Synergy Between Moderately Related Communities

The data convincingly demonstrate the benefits of polymicrobial drugs. However, an effective consortium is not just a collection of strong producers, but rather a synergistic community. Studies show that artificial communities of moderately related strains exhibit higher synergy in plant growth promotion and better plant biomass production compared to communities of highly related strains or antagonistic mixtures (Liu et al. 2025).

When forming a consortium, conscious selection of secondary metabolite producers is critical. It's crucial strategically to select not the strains with the greatest "killing power," but rather those that produce compounds acting as "ecosystem engineers": cyclic lipopeptides (surfactin, iturin), 2,4-DAPG, and gliotoxin, and other compounds capable of dose-dependent modulation of the community.

4.3. Taking into Account the Soil-Climatic Context: From Universality to Precision

The effectiveness of an inoculant is not inherent; it only manifests itself in suitable soil and climatic conditions. Therefore, development and application must begin with diagnostics (Figure 3A).

1. SOC and pH are the primary diagnostic parameters. Maximum agronomic performance is expected in soils with moderate SOC percentage, where there is a "window of opportunity" for colonization.

2. It is necessary to select and test strains that are tolerant to stressors prevalent in the region, such as drought and salinity. The greatest economic and environmental benefits of inoculants, especially those that fix nitrogen, are expected in marginal and degraded agroecosystems where they act as triggers for fertility restoration.

3. The SOC gradient methodology is a powerful tool for stress testing and selecting the most competitive strains adapted to a specific level of fertility.

4. Compatibility with the host plant's chemical strategy. The plant actively shapes its microbiome through secondary metabolites. Ignoring this chemical dialogue guarantees inoculation failure.

5. Formulation technologies for inoculant protection are necessary, but not sufficient. To overcome the survivability barrier, encapsulation technologies in biodegradable polymers and the creation of combined carriers based on clays and humic substances are being actively developed to provide protection against abiotic stress and controlled cell release (Monica et al 2025). However, these technologies are of secondary importance. Without taking into account the ecological principles listed above, even a perfectly protected but environmentally incompetent strain will quickly be eliminated from the rhizosphere.