Microbial Inoculants as Regulators of Plant Acclimation to Abiotic Stress: Hormonal, Redox, Osmotic, Ionic, and Hydraulic Mechanisms

1. Introduction

Abiotic stresses are among the principal constraints on crop productivity and yield stability worldwide. Drought, salinity, temperature extremes, flooding, and exposure to toxic elements disrupt plant water relations, photosynthetic carbon assimilation, membrane integrity, nutrient homeostasis, and reproductive development, thereby reducing biomass accumulation and harvestable yield [1]. Climate change is increasing crop exposure to warming, altered precipitation regimes, and extreme weather events. Global assessments indicate that anthropogenic climate change has already slowed agricultural productivity growth and that substantial production losses may persist even when agricultural adaptation is considered [1,2,3]. Moreover, crops grown under field conditions are frequently exposed to sequential or simultaneous stresses rather than to a single isolated environmental constraint. Combined stresses can generate physiological, metabolic, and transcriptional responses that differ qualitatively from those induced by each component stress applied individually [4,5]. Consequently, improving crop resilience requires strategies capable of maintaining plant function and yield stability under complex and fluctuating environmental conditions.

In this context, microbial inoculants have attracted increasing attention as biologically based tools for supporting plant growth under both favourable and stressful conditions. Their agricultural development is driven by the need to reduce dependence on intensive chemical inputs and establish more sustainable and climate-resilient production systems [6,7,8,9]. This approach builds on naturally occurring plant–microorganism associations, including biological nitrogen fixation by rhizobia, symbioses with arbuscular mycorrhizal fungi (AMF), and interactions with plant growth-promoting rhizobacteria (PGPR) and beneficial endophytes [8,9,10,11]. Frequently investigated bacterial genera include Azospirillum, Bacillus, Methylobacterium, Pseudomonas, and Serratia, although their functional traits and agronomic effectiveness depend strongly on the microbial strain, host plant, soil properties, resident microbiome, and environmental conditions. Recent studies illustrate this functional diversity through the application of Azospirillum brasilense with biochar in maize, endophytic Bacillus megaterium strains in maize, plant-associated Methylobacterium spp., co-inoculation of Azospirillum and Pseudomonas in chilli, and application of Serratia sp. KF23 to salt-stressed soybean [12,13,14,15,16]. Beyond biological nitrogen fixation and mycorrhizal nutrient transfer, selected microorganisms can increase phosphorus availability, mobilise potassium from K-bearing minerals, and facilitate iron acquisition through siderophore production [17,18,19].

The benefits of microbial inoculants extend beyond nutrient acquisition. Beneficial microorganisms can modulate physiological, biochemical, and molecular processes involved in plant acclimation to abiotic stress, including phytohormonal signalling, cellular redox homeostasis, antioxidant metabolism, ion transport, osmotic adjustment, water relations, and stress-responsive gene expression [6,7,8,9]. Selected microorganisms can also produce or influence bioactive compounds such as phytohormones, osmoprotectants, extracellular polysaccharides, siderophores, and microbial volatile organic compounds [6,7,17,18,19,20,21]. Through interactions among these mechanisms, microbial inoculation can preserve photosynthetic function, improve plant water status, restrict ionic and oxidative injury, and support growth or yield under adverse conditions [7,16,21]. Nevertheless, detection of a potentially beneficial trait in axenic microbial culture does not establish that the trait is expressed at a physiologically relevant level during plant colonisation or that it is responsible for the observed plant response.

The plant-holobiont concept provides a broader framework for considering the plant, its associated microbiota, and the surrounding environment as an interacting biological system [22]. Within this framework, beneficial microorganisms can influence both plant physiology and rhizosphere processes, including nutrient transformations, soil aggregation, water retention, and chemical signalling [8,9,22]. Plant-associated microbiomes are dynamic rather than fixed: their composition and functional activity are shaped by host genotype, developmental stage, soil properties, agricultural management, interactions among microbial taxa, and the nature, intensity, and duration of environmental stress [8,9,22]. Consequently, microbial inoculants should not be regarded merely as nutrient suppliers or universally effective products but as context-dependent biological modulators whose effects emerge from interactions among the microorganism, host plant, resident microbiome, and environment.

Accordingly, this review critically examines the physiological, biochemical, and molecular mechanisms through which plant growth-promoting microorganisms modulate plant acclimation and tolerance to abiotic stress. Particular emphasis is placed on phytohormonal regulation, redox homeostasis, osmotic adjustment, ion homeostasis, plant water relations, rhizosphere hydraulic properties, and stress-responsive signalling networks. The review also distinguishes correlative evidence from experiments providing stronger mechanistic support and identifies the methodological requirements for translating promising microbial effects from controlled conditions into reproducible agricultural applications.

1.1. Scope and Literature Selection

This narrative review was developed through structured searches of the Web of Science Core Collection, Scopus, PubMed, and Google Scholar databases, with the literature updated to July 2026. Search terms were combined using Boolean operators and included different combinations of “plant growth-promoting microorganisms”, “microbial inoculants”, “biofertilizers”, “plant growth-promoting rhizobacteria”, “arbuscular mycorrhizal fungi”, “endophytes”, “abiotic stress”, “drought”, “salinity”, “heat stress”, “cold stress”, “phytohormones”, “abscisic acid”, “auxin”, “cytokinin”, “ethylene”, “ACC deaminase”, “reactive oxygen species”, “antioxidant defence”, “osmotic adjustment”, “osmolytes”, “proline”, “trehalose”, “glycine betaine”, “ion homeostasis”, “aquaporins”, “root hydraulic conductivity”, and “water-use efficiency”. Additional relevant studies were identified by examining the reference lists of the selected articles and recent reviews.

Priority was given to peer-reviewed original research published from 2022 onwards, particularly studies that provided physiological, biochemical, molecular, or agronomic evidence of microbially mediated plant responses to abiotic stress. Earlier publications were retained when they represented foundational contributions, introduced key concepts, or provided particularly strong mechanistic evidence. Recent reviews were used to establish the conceptual framework and identify emerging research directions, whereas specific mechanistic conclusions were supported, whenever possible, by primary experimental studies.

The literature selection focused on beneficial bacteria, arbuscular mycorrhizal fungi, endophytes, and microbial consortia applied to crop plants or experimentally relevant model species exposed to drought, salinity, temperature extremes, or closely related osmotic and ionic constraints. Studies conducted under controlled, greenhouse, pot, and field conditions were considered. Articles dealing exclusively with nutrient acquisition under non-stress conditions were included only when they contributed directly to understanding the broader physiological functions of microbial inoculants or provided essential context for their agricultural use.

Particular attention was given to the strength of the evidence supporting each proposed mechanism. Studies based solely on changes in biomass, metabolite concentrations, antioxidant enzyme activities, or transcript abundance were distinguished from those combining these responses with direct measurements of plant water status, gas exchange, ion accumulation or fluxes, root hydraulic properties, microbial colonization, protein activity, or crop yield. Greater mechanistic weight was assigned to investigations using microbial or plant mutants, functional complementation, hormone-deficient or signalling-impaired genotypes, inhibitors, metabolite tracing, or other approaches capable of establishing causal relationships. Owing to the substantial variability among microbial strains, plant genotypes, soil properties, stress intensities, and experimental systems, contradictory findings were retained and critically discussed rather than excluded solely because they differed from the predominant response.

2. Microbial Modulation of Hormonal Signalling Under Stress

2.1. Plant Hormones as Central Nodes of Stress Acclimation

Plant responses to abiotic stress are coordinated through interconnected hormonal networks that integrate environmental perception with growth, development, metabolism, and resource allocation. Abscisic acid (ABA), auxins, cytokinins, ethylene, gibberellins, jasmonates, salicylic acid, brassinosteroids, and strigolactones participate in plant acclimation to drought, salinity, temperature extremes, flooding, and nutrient limitation [23,24,25,26,27,28]. Stress exposure can modify hormone biosynthesis, degradation, conjugation, transport, receptor abundance, and downstream signalling. These changes regulate processes including stomatal behaviour, root-system development, vascular transport, osmotic adjustment, antioxidant metabolism, senescence, and the balance between continued growth and resource conservation [23,24,25,26,27,28]. The effects of individual hormones are strongly dependent on concentration, tissue, developmental stage, stress intensity, and exposure time. ABA accumulation can restrict transpirational water loss and promote stress-responsive gene expression, but prolonged ABA signalling may limit carbon assimilation and growth. Auxin redistribution can remodel root architecture, whereas excessive auxin may inhibit primary-root elongation. Cytokinins can preserve chloroplast function and delay leaf senescence but may favour shoot maintenance over root exploration. Ethylene can participate in stress perception and acclimation, while excessive or sustained ethylene accumulation commonly restricts cell expansion and accelerates senescence [23,24,25,26,27,28]. Hormonal responses should therefore not be classified simply as beneficial or detrimental; their physiological value depends on their spatial and temporal coordination. Hormonal pathways also interact extensively. ABA-dependent stomatal regulation is influenced by cytokinins, auxins, ethylene, jasmonates, and redox signals, while auxin-mediated root development is connected with cytokinin distribution and ethylene production. Jasmonic acid and salicylic acid interact with ABA, ethylene, and reactive oxygen species to coordinate growth, defence, and cellular protection. Gibberellins and brassinosteroids influence the balance between active growth and stress-induced growth restriction, whereas strigolactones connect root and shoot development with nutrient availability and symbiotic signalling [23,24,25,26,27,28]. Consequently, changes in the concentration or signalling activity of one hormone can produce secondary effects throughout the wider regulatory network.

Plant-associated microorganisms can influence these networks by producing hormone-like compounds, metabolising hormone precursors, modifying plant hormone biosynthesis or degradation, and altering hormone transport, perception, or downstream signalling [27,28,29,30]. Rhizospheric and endophytic bacteria can produce compounds such as indole-3-acetic acid ((IAA), cytokinins, gibberellins, or ABA-related metabolites, whereas ACC deaminase-producing microorganisms can decrease the amount of 1-aminocyclopropane-1-carboxylate available for plant ethylene biosynthesis. Beneficial bacteria and fungi may also induce endogenous hormonal responses without producing the corresponding hormone themselves [27,28,29,30]. Microbial hormonal effects are therefore not equivalent to the external application of a purified plant-growth regulator. Their physiological outcome depends on microbial colonisation, spatial proximity to responsive tissues, metabolite production under plant-associated conditions, and interaction with endogenous plant metabolism [27,28,29,30]. Improved nutrition, altered root development, changes in plant water status, and reduced cellular injury can themselves modify hormone concentrations or signalling. A change in plant hormone content following inoculation consequently does not establish whether the microorganism acted directly on that hormonal pathway or whether the change was secondary to broader stress alleviation. Robust attribution requires integration of hormone quantification with microbial colonisation, tissue-specific and time-resolved analyses, microbial biosynthetic mutants, functional complementation, and plant hormone-deficient or signalling-impaired genotypes. Measurements of downstream processes, including root architecture, stomatal conductance, hydraulic function, photosynthesis, water status, oxidative injury, growth and yield, are also necessary to determine the physiological importance of the hormonal response [27,28,29,30]. The following subsections examine the evidence for microbial regulation of ABA, auxins, cytokinins, ethylene, and additional hormonal pathways under abiotic stress.

2.2. Microbial Modulation of ABA Homeostasis and Signalling

Abscisic acid is a central regulator of plant responses to water deficit and salinity. Stress-induced ABA accumulation promotes stomatal closure, restricts transpirational water loss, modifies root and shoot growth, and regulates genes involved in osmotic adjustment, antioxidant metabolism, ion transport, and cellular protection [23,24,27,28,31]. Nevertheless, increased ABA concentration is not invariably beneficial. Prolonged or excessive ABA signalling can restrict stomatal conductance, photosynthetic carbon assimilation, leaf expansion, and biomass production. The physiological significance of microbial modulation of ABA therefore depends on hormone concentration, tissue distribution, exposure time, developmental stage, and the balance between water conservation and continued carbon gain [27,28,31]. Plant-associated microorganisms may influence ABA responses through several routes. Some bacteria can produce ABA or ABA-related metabolites in culture, whereas others modify plant ABA biosynthesis, catabolism, transport, or signalling without necessarily synthesising the hormone themselves [27,28,31]. Microbial effects on plant hydration, root development, mineral nutrition, and stress severity can also produce secondary changes in ABA accumulation. Consequently, an inoculation-induced change in plant ABA concentration does not, by itself, demonstrate direct microbial ABA production or identify ABA as the principal mechanism responsible for stress alleviation.

In salt-stressed wheat, inoculation with Bacillus subtilis IB-22 and Pseudomonas mandelii IB-Ki14 altered ABA accumulation and distribution while improving plant growth and water relations [29,32]. Both strains increased ABA concentration in roots, whereas shoot ABA responses depended on the bacterial treatment and duration of salinity exposure. The cytokinin-producing strain IB-22 produced particularly strong effects on transpiration, root hydraulic conductivity, and shoot growth [29]. These responses indicate that bacterial regulation of ABA operated together with changes in cytokinins, auxin-related activity, and root water transport rather than as an isolated hormonal mechanism [29,32]. Moreover, because the strains differed in several functional traits, their contrasting plant effects cannot be attributed exclusively to ABA.

Drought and salinity modify root exudation and the composition of root-associated microbial communities [33]. These community changes may alter the abundance of microorganisms capable of influencing ABA metabolism or plant water relations. However, stress-associated microbiome restructuring is primarily ecological and correlative evidence and does not establish ABA-mediated recruitment or reciprocal ABA regulation without direct experimental tests [33]. Demonstrating such a feedback mechanism would require manipulation of plant ABA biosynthesis or signalling combined with quantitative microbiome analysis and functional testing of the recruited microorganisms.

In Arabidopsis thaliana exposed to drought, inoculation with A. brasilense increased plant ABA concentrations and improved survival, leaf water status, and biomass retention [34]. The weaker protective response observed in an ABA-deficient plant background supported the participation of ABA in the inoculation phenotype. Nevertheless, the bacterium also influenced root growth and other physiological processes, and the experiment did not demonstrate that bacterially synthesised ABA was transferred directly to the plant. The evidence therefore supports microbial stimulation of plant ABA-dependent acclimation rather than simple external provision of ABA [34].

ABA-related regulation has also been reported under salinity in tomato. Inoculation with the halotolerant strains Bacillus aryabhattai H19-1 and Bacillus mesonae H20-5 improved growth, chlorophyll retention, and stress tolerance [35]. B. mesonae H20-5 increased expression of the ABA-biosynthetic gene NCED1 and the ABA-responsive transcription factor AREB1, providing molecular evidence that inoculation influenced ABA biosynthesis or signalling [35]. However, transcript abundance was measured for a limited number of pathway components, and neither ABA-deficient nor ABA-insensitive tomato plants were used. The results therefore support involvement of ABA-responsive regulation but do not establish that ABA was the sole cause of the protective phenotype.

Beneficial microorganisms can also regulate stomata independently of canonical ABA signalling. Root colonisation of Arabidopsis by Pseudomonas chlororaphis O6 promoted stomatal closure and increased tolerance to drought and salinity [36]. Stomatal closure was maintained in ABA-deficient and ABA-insensitive plant backgrounds, indicating that the response did not require conventional ABA biosynthesis or perception [36]. This result is important because improved plant water status after inoculation should not automatically be interpreted as evidence of enhanced ABA accumulation. Microbial volatile compounds, other hormonal pathways, ROS, Ca²⁺ signalling, or changes in guard-cell regulation may produce similar physiological outcomes.

2.3. Microbial Auxins and and Root-System Plasticity Under Adverse Conditions

Auxins regulate cell division and elongation, vascular differentiation, apical dominance, tropic responses, and root-system architecture. IAA is the principal naturally occurring auxin and can be synthesised by taxonomically diverse plant-associated bacteria through several tryptophan-dependent and tryptophan-independent pathways [37,38]. Microbial IAA production is nevertheless strain specific and strongly influenced by substrate availability, root exudates, microbial growth conditions, and interactions with the host plant. Detection of IAA in axenic culture therefore does not establish that physiologically relevant concentrations are produced during plant colonisation or that bacterial IAA explains the resulting plant phenotype [37,38].

Plant responses to auxin depend not only on hormone concentration but also on transport, tissue sensitivity, receptor stability, and the abundance of transcriptional regulators. In A. thaliana, HSP90 stabilised the auxin co-receptor TIR1 and thereby contributed to temperature-dependent seedling growth, demonstrating that environmental conditions can modify auxin perception independently of changes in auxin concentration [39]. Auxin-sensitive Aux/IAA transcriptional repressors are also required for plant tolerance to several environmental stresses, linking canonical auxin signalling with stress-responsive growth regulation [40]. These studies establish the physiological framework within which microbial auxins may operate, but they do not constitute direct evidence of microbial regulation because no microbial inoculants were evaluated [39,40].

Microbial IAA can modify primary-root elongation, lateral-root initiation, adventitious rooting, and root-hair development, thereby changing the volume of soil explored and the absorptive surface available for water and nutrient acquisition [37,38]. Moderate modification of auxin homeostasis may promote root branching and improve resource acquisition, whereas excessive auxin exposure can inhibit primary-root elongation or produce an architecture that is poorly suited to the prevailing stress. The resulting response depends on the concentration and location of bacterial IAA production, endogenous plant auxin status, developmental stage, and interactions with ethylene, cytokinins, ABA, and nutrient-signalling pathways [37,38]. Consequently, increased bacterial IAA production should not automatically be interpreted as greater abiotic-stress tolerance.

Direct genetic evidence for the participation of bacterial IAA in root development was obtained with Pseudomonas putida GR12-2. Comparison of the wild-type bacterium with an ipdC-disrupted derivative deficient in indolepyruvate-pathway IAA biosynthesis showed that bacterial IAA contributed to alterations in the host root system [41]. The mutant comparison provides stronger causal support than correlations between IAA production in culture and plant growth. Nevertheless, the study primarily demonstrated regulation of root development and did not establish that the IAA-dependent phenotype improved plant performance under drought, salinity, or another abiotic stress [41].

Manipulation of the bacterial ipdC pathway can also produce effects extending beyond the inoculated plant. In wheat, derivatives of A. brasilense with genetically modified ipdC-dependent auxin biosynthesis altered aspects of the indigenous rhizosphere microbial community [42]. This finding indicates that microbial auxin metabolism can influence rhizosphere ecology, potentially through changes in bacterial physiology, root development, or root-derived substrates. However, the study did not demonstrate that modification of the rhizosphere community resulted in enhanced abiotic-stress tolerance, and its results should not be presented as direct evidence of drought or salinity alleviation [42].

Microbial auxin effects must therefore be evaluated as part of an integrated plant–microorganism–soil response. Changes in root length or branching may improve access to water and nutrients, but they can also alter carbon allocation, hydraulic architecture, and interactions with the resident microbiome. Robust attribution requires microbial mutants deficient in defined IAA-biosynthetic pathways and their complemented derivatives, quantitative determination of IAA during plant colonisation, assessment of endogenous plant auxin metabolism and signalling, and measurements of root architecture, microbial colonisation, water and nutrient acquisition, and whole-plant performance under stress [37,38,41,42].

2.4. Microbial Cytokinins and Maintenance of Shoot Function Under Stress

Cytokinins regulate cell division, meristem activity, chloroplast differentiation, source–sink relationships, stomatal behaviour, and leaf senescence. Although these hormones are synthesised in several plant organs, root-derived cytokinins can act as long-distance signals that influence shoot growth and leaf function. Root-specific upregulation of cytokinin biosynthesis in tomato delayed leaf senescence and maintained photosynthetic activity, supporting the existence of a cytokinin-mediated root-to-shoot communication network [43]. Nevertheless, cytokinin responses are strongly dependent on hormone concentration, tissue, developmental stage, and environmental conditions. Elevated cytokinin signalling generally promotes shoot growth, whereas it may restrict primary-root elongation or modify lateral-root development. Consequently, maintaining cytokinin activity during drought can preserve photosynthetically active tissues but may also interfere with stress-induced resource allocation towards root growth.

Selected plant-associated bacteria can synthesise cytokinins or alter cytokinin biosynthesis, degradation, transport, and signalling in their host plants. Genera in which cytokinin production has been reported include Bacillus, Methylobacterium, Paenibacillus, and Pseudomonas, although this capacity varies substantially among species and strains. In lettuce seedlings exposed to drying soil, inoculation with cytokinin-producing B. subtilis increased shoot cytokinin concentrations and shoot biomass but shortened the roots and decreased the root-to-shoot ratio [44]. Despite the capacity of cytokinins to promote stomatal opening, inoculation did not increase stomatal conductance because shoot ABA also accumulated. These findings illustrate that microbial cytokinins do not operate independently but interact with plant-derived ABA and other hormonal signals to determine the balance among shoot maintenance, water conservation, and root exploration.

Further evidence of this interaction has been obtained in durum wheat. As described above, the cytokinin-producing strain B. subtilis IB-22 increased plant cytokinin concentrations, chlorophyll accumulation, transpiration, and root hydraulic conductivity more strongly than the auxin-producing strain P. mandelii IB-Ki14 [29]. Importantly, the greater transpiration of plants inoculated with IB-22 was not accompanied by reduced tissue water content or turgor, suggesting that enhanced root hydraulic capacity compensated for the increased transpirational demand. Thus, the beneficial effect of microbial cytokinins under limited water availability may depend on whether the plant can sustain the additional water requirement associated with shoot growth and gas exchange.

Stronger causal evidence has been obtained with the cytokinin-producing strain Pseudomonas fluorescens G20-18. Inoculation of tomato plants with this bacterium promoted growth and, under progressive drought, maintained higher chlorophyll concentrations, altered ABA accumulation and stomatal regulation, enhanced antioxidant and carbohydrate-metabolism responses, and induced extensive transcriptional reprogramming [45]. Comparisons with the cytokinin-deficient bacterial mutants CNT1 and CNT2 demonstrated that bacterial cytokinin biosynthesis contributed to several of these responses. Nevertheless, cytokinin production did not explain the complete inoculation phenotype, indicating that additional bacterial traits and interactions with endogenous plant-signalling pathways also participated in drought acclimation.

Members of Methylobacterium are particularly relevant because they are frequent colonisers of plant surfaces and can produce diverse cytokinin forms. A comparative analysis of 46 Methylobacterium representatives detected 16 cytokinin metabolites, including trans-zeatin, but revealed large quantitative differences among strains [46]. Cytokinin secretion was also affected by carbon availability, demonstrating that hormone production is environmentally regulated rather than a fixed taxonomic property. These findings support the potential of Methylobacterium for microbial-inoculant development [10,46], but they also reinforce the need for caution: cytokinin detection in axenic microbial cultures does not demonstrate that physiologically relevant amounts are produced during plant colonisation.

2.5. Microbial Regulation of Stress-Induced Ethylene and ACC Metabolism

Ethylene is a gaseous phytohormone that regulates seed germination, root and shoot development, senescence, abscission, fruit ripening, and plant responses to environmental stress. Abiotic stresses such as drought, salinity, flooding, heavy-metal exposure, and temperature extremes frequently stimulate ethylene biosynthesis. Transient and spatially controlled increases in ethylene can contribute to stress perception and acclimation, whereas excessive or prolonged ethylene accumulation commonly restricts cell expansion, root elongation, photosynthetic activity, and biomass production and may accelerate senescence [47]. Microbial regulation of ethylene should therefore not be interpreted as the complete suppression of this hormone but as the maintenance of ethylene signalling within a range that permits stress acclimation without imposing sustained growth inhibition.

Beneficial microorganisms can influence ethylene responses by modifying plant hormone biosynthesis and signalling or by metabolising 1-aminocyclopropane-1-carboxylate (ACC), the immediate precursor of ethylene. In salt-stressed wheat, inoculation with the halophilic Bacillus strains NMCN1 and LLCG23 enhanced root-system development, biomass accumulation, photosynthetic activity, and stomatal conductance. These responses were accompanied by increased expression of genes associated with auxin responses (ARF), cytokinin metabolism (CKX2), and cell-wall expansion (expA1), together with reduced expression of an ethylene-responsive factor gene (ERF) [48]. This transcriptional pattern is consistent with attenuation of ethylene-associated growth inhibition, but it does not directly demonstrate that bacterial inoculation reduced ethylene biosynthesis or concentration. ERF genes encode transcription factors involved in ethylene-responsive signalling rather than enzymes that synthesise ethylene, and individual ERFs may act as positive or negative regulators depending on the gene, tissue, and environmental context.

The best-characterised microbial mechanism affecting stress-induced ethylene involves ACC deaminase, which is encoded by the bacterial acdS gene. After plant-derived ACC becomes available to colonising microorganisms, ACC deaminase cleaves it into α-ketobutyrate and ammonia, thereby decreasing the amount of ACC available for conversion into ethylene by plant ACC oxidase [47,49]. ACC deaminase activity has been identified in taxonomically diverse rhizospheric and endophytic bacteria, including strains belonging to Pseudomonas, Bacillus, Enterobacter, Azospirillum, Methylobacterium, Paraburkholderia, and Herbaspirillum. Nevertheless, this trait is strain specific and should not be inferred solely from taxonomic identity.

Experimental evidence with Methylobacterium fujisawaense provides direct support for this mechanism. Inoculation of canola seedlings with ACC deaminase-producing strains decreased plant ACC concentrations, reduced ACC oxidase activity and ethylene production, and alleviated ethylene-mediated inhibition of root elongation [50]. The bacteria also increased plant indole-3-acetic acid concentrations and ACC synthase activity, illustrating the functional connection between microbial auxin production and ethylene metabolism. Bacterial IAA may stimulate plant growth but can also induce ACC synthesis; ACC deaminase can subsequently prevent the resulting ACC pool from generating excessive ethylene. Thus, microbial auxin and ACC deaminase activities may act in a coordinated rather than independent manner.

Particularly strong causal evidence has been obtained using Pseudomonas sp. UW4 and bacterial mutants deficient in ACC deaminase or trehalose biosynthesis. Under salt stress, the wild-type strain improved tomato root and shoot growth, dry biomass, and chlorophyll content, whereas the single acdS and treS mutants provided substantially weaker protection. The double acdS/treS mutant was even less effective, demonstrating that ACC deaminase and trehalose accumulation made complementary contributions to microbially mediated salinity tolerance [51]. This study also shows that ACC deaminase should not be considered an isolated mechanism because its effect can depend on additional microbial traits involved in osmoprotection, colonisation, and stress survival.

In pea plants subjected to osmotic and salinity stresses, inoculation with the ACC deaminase-producing bacterial isolates GKP KS2_7 and MBD 133 decreased stress-induced ethylene accumulation and improved seedling growth, chlorophyll content, carbohydrate and protein concentrations, and antioxidant enzyme activities [52]. These findings indicate that microbial regulation of ethylene can have downstream effects on redox metabolism and cellular protection. However, because the strains possessed several plant growth-promoting traits, the observed physiological responses cannot be attributed exclusively to ACC deaminase without comparisons involving acdS-deficient mutants.

More recent screening of bacteria associated with Brachiaria identified ACC deaminase activity in strains belonging to several genera, including Paraburkholderia, Azospirillum, Herbaspirillum, Gluconacetobacter, and Pseudomonas. Selected strains increased plant height and biomass under polyethylene-glycol-induced water deficit, and root colonisation by Paraburkholderia silvatlantica NRB142 was confirmed by microscopy [53]. Importantly, growth on medium containing ACC as the sole nitrogen source did not consistently predict quantitatively measurable ACC deaminase activity, and enzyme activity in vitro was not directly proportional to plant growth promotion. These observations reinforce the need to combine acdS detection with enzymatic assays, colonisation measurements, plant ACC and ethylene quantification, and physiological evaluation under realistic stress conditions.

2.6. Additional Hormonal Pathways and Microbial-Induced Crosstalk

In addition to ABA, auxins, cytokinins, and ethylene, plant-associated microorganisms can influence gibberellins, jasmonates, strigolactones, brassinosteroids, and salicylic acid. These hormonal pathways regulate overlapping aspects of plant growth, defence, carbon allocation, root development, senescence, and environmental acclimation. However, microbial effects on these hormones remain less comprehensively characterised than those involving IAA, ABA, cytokinins, or ACC metabolism, and their participation is frequently inferred from changes in a limited number of transcripts or from hormone concentrations measured at a single sampling time [27,28,30]. Their contribution should therefore be interpreted within the broader context of hormonal crosstalk rather than as a series of independent microbial mechanisms.

Gibberellins regulate seed germination, cell elongation, shoot and root development, leaf expansion, flowering, and the balance between active growth and stress-induced growth restriction. Several rhizospheric and endophytic bacteria can synthesise gibberellins or modify endogenous gibberellin metabolism in their host plants [27]. For example, the coastal-soil isolates Priestia aryabhattai KW05 and Pseudomonas frederiksbergensis KW07 produced GA₁, GA₃, GA₄, and GA₇ and improved germination and early seedling growth in mallow and broccoli under salinity [54]. Both isolates also produced IAA and displayed additional plant growth-promoting properties, including siderophore production, phosphate solubilisation, and exopolysaccharide production. Consequently, although their gibberellin production provides a plausible contribution to the observed response, the improvement in salt-stressed seedlings cannot be attributed exclusively to this hormonal mechanism. Moreover, improved germination and early seedling establishment do not necessarily demonstrate greater tolerance during vegetative or reproductive development.

Jasmonic acid (JA) and its bioactive conjugates participate in the coordination of growth, defence, redox regulation, root development, and acclimation to drought and salinity. Beneficial microorganisms may alter jasmonate biosynthesis or prime plants for a faster and more effectively regulated response after stress exposure [28,55]. In A. thaliana, inoculation with Bradyrhizobium japonicum IRAT FA3 induced an early increase in jasmonates before salt exposure and subsequently enhanced shoot biomass, restricted Na⁺ accumulation, and reduced lipid peroxidation under salinity [55]. The involvement of JA signalling was reinforced by experiments showing that the JA-related components MYC2 and JAR1 were required for bacterially induced shoot growth. Thus, the treatment did not merely maintain constitutively elevated JA concentrations but modified the timing and downstream regulation of the jasmonate response, supporting its interpretation as microbial priming. This distinction is physiologically important because prolonged activation of JA signalling can restrict growth, whereas transient or stress-responsive activation may improve acclimation.

Strigolactones occupy a distinctive position in plant–microbe interactions because they function both as endogenous plant hormones and as rhizosphere signals involved in the establishment of arbuscular mycorrhizal symbiosis. They regulate shoot branching, root-system architecture, resource allocation, and plant responses to nutrient deficiency and water stress. Under drought, arbuscular mycorrhizal colonisation increased strigolactone biosynthesis in lettuce and tomato and was associated with improved plant growth, photosynthetic performance, water status, and drought tolerance [56]. The relationships among drought intensity, mycorrhizal colonisation, and strigolactone accumulation suggest a dynamic feedback mechanism through which stressed plants may reinforce their interaction with AMF. However, the responses differed between plant species and among individual strigolactone compounds, indicating that the AMF–strigolactone relationship is neither uniform nor solely determined by stress severity.

Brassinosteroids regulate cell expansion, vascular differentiation, photosynthetic activity, membrane stability, antioxidant metabolism, ion homeostasis, and the balance between plant growth and stress acclimation. Direct evidence that a beneficial microorganism can modulate this pathway was obtained using Bacillus endophyticus J13 in salt-stressed A. thaliana [57]. Bacterial inoculation increased plant biomass, relative water content, and chlorophyll concentration while reducing membrane injury and lipid peroxidation. These physiological responses were accompanied by upregulation of the positive brassinosteroid-signalling regulators BZR1 and BES2 and downregulation of the negative regulator BIN2. Importantly, a functional BES1 gene was required for J13-mediated alleviation of salinity stress, providing stronger evidence for the participation of brassinosteroid signalling than studies based exclusively on transcriptional correlations [57]. Nevertheless, J13 also modified polyamine metabolism, ethylene biosynthesis, and proline homeostasis, indicating that brassinosteroid signalling formed part of a broader regulatory network rather than acting as the sole protective mechanism.

Salicylic acid (SA) contributes to redox regulation, antioxidant metabolism, photosynthetic protection, and stress-responsive gene expression, but its effects depend on hormone concentration, tissue, developmental stage, and interactions with ABA, JA, ethylene, and ROS [27,28]. In pepper plants exposed to drought, salinity, or Cd stress, inoculation with Bacillus amyloliquefaciens B11 increased leaf SA concentrations while improving survival, chlorophyll retention, and plant growth and reducing H₂O₂ accumulation [58]. The bacterium also possessed several additional plant growth-promoting traits, including ACC deaminase activity, IAA production, nutrient solubilisation, nitrogen fixation, and siderophore production. Therefore, the contribution of increased SA could not be separated from its other functional properties.

Similarly, inoculation of chickpea with P. putida MTCC5279, also designated strain RA, modified the expression of the SA-responsive marker PR1, together with genes involved in JA and ethylene signalling, transcriptional regulation, cellular protection, and ROS scavenging during drought and recovery [59]. These molecular responses were accompanied by improvements in germination, root-system development, plant water status, membrane integrity, osmolyte metabolism, and antioxidant regulation. However, PR1 expression represents evidence of altered SA-responsive signalling rather than direct proof of changes in SA concentration. Because neither this study nor the experiment with B. amyloliquefaciens B11 used SA-deficient or signalling-impaired plants, the specific contribution of SA remains physiologically plausible but not conclusively isolated from other microbial mechanisms.

Hormonal responses induced by microbial inoculation are unlikely to result from a single compound acting independently. Microbial IAA can stimulate plant ACC biosynthesis, which may subsequently be moderated by bacterial ACC deaminase [47,50]; cytokinin-mediated maintenance of shoot activity can interact with ABA-dependent stomatal regulation and changes in root hydraulic conductivity [29,44,45]; and JA, SA, ethylene, brassinosteroids, strigolactones, and ROS can jointly regulate growth–defence trade-offs and acclimation to stress [55,56,57,58,59]. Time-resolved hormone profiling, tissue-specific analyses, microbial biosynthetic mutants, plant hormone-deficient or signalling-impaired genotypes, and measurements of downstream physiological functions are therefore required to distinguish direct microbial hormonal effects from secondary consequences of improved nutrition, colonisation, water status, or general stress alleviation.

The Supplementary Table S1 presents selected examples that summarize the studies reviewed in this section, including the hormonal pathway involved, the responsible microorganism, the type of abiotic stress, the plant response, and the underlying mechanism.

3. Microbial Inoculants and Regulation of Cellular Redox Homeostasis

3.1. Abiotic Stress as a Redox Perturbation

Abiotic stresses such as drought, salinity, temperature extremes, flooding, and exposure to toxic elements disrupt cellular metabolism and alter the redox balance of plant cells. These conditions frequently enhance the production of reactive oxygen species (ROS), including superoxide anion (O₂•⁻), hydrogen peroxide (H₂O₂), hydroxyl radical (•OH), and singlet oxygen (¹O₂). Excessive ROS accumulation can oxidise membrane lipids, proteins, pigments, and nucleic acids, impairing membrane integrity, photosynthesis, enzyme activity, and cellular viability. However, ROS are not merely toxic metabolic by-products. When their production is spatially and temporally regulated, they act as signalling molecules that connect stress perception with local and systemic acclimation responses. The NADPH oxidase RBOHD, for example, is required for the rapid propagation of ROS-dependent systemic signals in response to several environmental stimuli [60].

Reactive nitrogen species (RNS), particularly nitric oxide (NO) and NO-derived molecules such as peroxynitrite (ONOO⁻) and S-nitrosoglutathione, interact closely with ROS metabolism. These compounds can modify the activity, stability, localisation, or interactions of target proteins through post-translational modifications such as S-nitrosylation and tyrosine nitration. Under salinity, NO-derived modifications differentially regulated enzymes of the ascorbate–glutathione cycle in pea: nitration and S-nitrosylation inhibited monodehydroascorbate reductase, whereas cytosolic and chloroplastic glutathione reductase activities were comparatively preserved [61]. This example illustrates that nitro-oxidative signalling does not uniformly activate or inhibit antioxidant metabolism but selectively modifies individual components of the redox network.

The physiological effect of ROS therefore depends on their chemical identity, concentration, subcellular origin, duration, and interaction with antioxidant and hormonal networks. H₂O₂ is sufficiently stable to participate in cellular and intercellular signalling, whereas the hydroxyl radical is extremely reactive and generally causes damage close to its site of production. Local ROS production can also be propagated through successive activation of NADPH oxidases, generating a systemic ROS wave that coordinates responses in tissues not directly exposed to the initial stress [60]. In A. thaliana, local heat stress and wounding triggered ROS signals that travelled through vascular tissues and, under specific conditions, through mesophyll cells, demonstrating that systemic redox communication can follow both vascular and non-vascular routes [62].

Plant growth-promoting microorganisms can influence this balance by restricting excessive ROS generation, stimulating enzymatic and non-enzymatic antioxidant systems, modifying redox-sensitive signalling pathways, or priming plants for a faster and more controlled oxidative response after stress exposure. Reductions in H₂O₂ accumulation, lipid peroxidation, electrolyte leakage, and membrane injury have been reported following microbial inoculation under drought and salinity [12,35,51,52,57,58,59]. Nevertheless, lower ROS concentrations in inoculated plants may result either from enhanced antioxidant removal or from reduced stress perception and ROS production. Conversely, increased antioxidant activity may indicate greater protective capacity, but it may also reflect a more severe oxidative challenge. Microbial effects on redox homeostasis should therefore be interpreted by integrating ROS and RNS measurements with antioxidant activity, redox-metabolite ratios, membrane damage, plant water status, photosynthetic performance, and growth or yield.

3.2. Activation of Enzymatic and Non-Enzymatic Antioxidant Systems

One of the most frequently reported effects of microbial inoculation under abiotic stress is the modulation of endogenous antioxidant defence systems. The enzymatic antioxidant network includes superoxide dismutase (SOD), which catalyses the dismutation of O₂•⁻ to H₂O₂, and several enzymes involved in H₂O₂ removal, including catalase (CAT), ascorbate peroxidase (APX), and other peroxidases. Glutathione reductase (GR), monodehydroascorbate reductase (MDHAR), and dehydroascorbate reductase (DHAR) contribute to the regeneration of reduced glutathione and ascorbate and thereby sustain cellular redox buffering. Plant-associated microorganisms can modify the activity or expression of these enzymes under heat, drought, and salinity, although the direction and magnitude of the response depend on the microbial strain, host genotype, stress intensity, tissue, and sampling time [61,63,64,65].

Microbial regulation of redox homeostasis also involves non-enzymatic antioxidants. Ascorbate and glutathione constitute major water-soluble redox buffers, whereas carotenoids, tocopherols, flavonoids, anthocyanins, and other phenolic compounds contribute to the protection of membranes, photosynthetic structures, and cellular macromolecules. The ascorbate–glutathione cycle connects APX-mediated H₂O₂ reduction with the regeneration of reduced ascorbate and glutathione through MDHAR, DHAR, and GR. Measurements of the reduced and oxidised forms of these metabolites, particularly the ascorbate/dehydroascorbate and GSH/GSSG ratios, can provide more informative indicators of cellular redox status than total ascorbate or glutathione concentrations alone [61,63].

Recent evidence under heat stress illustrates the broad metabolic scope of microbial regulation. Inoculation of two wheat cultivars with Micromonospora sp. enhanced the activities of antioxidant enzymes and components of the ascorbate–glutathione pathway, increased ascorbate and glutathione concentrations, and stimulated the accumulation of tocopherols, polyphenols, flavonoids, and anthocyanin-related metabolites [63]. These changes were accompanied by lower H₂O₂ and malondialdehyde accumulation and improved photosynthesis, RuBisCO activity, chlorophyll retention, and biomass production. The effects were considerably stronger in the heat-sensitive cultivar than in the tolerant cultivar, demonstrating that the response to a microbial inoculant is conditioned by the pre-existing stress tolerance and metabolic capacity of the host genotype [63].

Under salinity, inoculation of wheat with selected strains of Bacillus cereus, Serratia marcescens, and Pseudomonas aeruginosa increased antioxidant enzyme activities while reducing O₂•⁻ and H₂O₂ accumulation [64]. These biochemical responses were associated with greater membrane stability, leaf water content, photosynthetic pigment concentrations, gas exchange, K⁺ and Ca²⁺ accumulation, K⁺/Na⁺ ratios, growth, and yield. The results support a coordinated relationship among antioxidant regulation, water status, ion homeostasis, and productivity rather than an isolated effect on ROS detoxification [64]. Nevertheless, the taxonomic identity and biosafety of candidate strains must be carefully assessed before experimental isolates are proposed for agricultural formulation, particularly when the species concerned includes opportunistic pathogens.

A similarly complex response was observed in salt-stressed rice inoculated with three rhizospheric Bacillus isolates. Inoculation increased CAT, POD, SOD, and APX activities and enhanced the expression of OsCATA and OsAPX1, while decreasing superoxide accumulation and lipid peroxidation [65]. However, reduced glutathione concentrations were lower in inoculated plants than in non-inoculated salt-stressed plants, despite their superior growth, chlorophyll content, ion balance, and membrane protection [65]. This result demonstrates that successful microbial alleviation of oxidative stress does not require every antioxidant component to increase simultaneously. The lower glutathione concentration may reflect a reduced oxidative burden, altered glutathione turnover, or redistribution among redox and metabolic pathways rather than a weaker protective response [65].

Antioxidant enzyme activity must therefore be interpreted cautiously. Higher SOD, CAT, APX, or GR activity may indicate enhanced defensive capacity, but it may also reflect greater ROS production and a stronger oxidative challenge. Conversely, lower enzyme activity or lower concentrations of proline, glutathione, or phenolic compounds in inoculated plants may indicate that microbial treatment reduced stress severity and, consequently, the requirement for compensatory defence. This distinction is evident across studies in which microbial inoculation reduced H₂O₂ accumulation, lipid peroxidation, electrolyte leakage, or membrane injury while producing contrasting responses among individual antioxidant components [12,35,52,58,59,63,64,65]. Reliable assessment should therefore combine antioxidant enzyme activities with ROS measurements, redox-metabolite pools and ratios, lipid peroxidation, electrolyte leakage, photosynthetic performance, tissue water status, and plant growth or yield [61,63,64,65]. Whenever possible, enzyme activity should also be complemented by transcript and protein abundance because transcriptional changes do not necessarily predict enzyme abundance, subcellular localisation, post-translational regulation, or catalytic activity [61,65].

3.3. Redox Signalling and Transcriptional Reprogramming

ROS and RNS participate in signalling networks that connect environmental perception with calcium-dependent responses, protein phosphorylation, hormonal regulation, and transcriptional control. Controlled changes in H₂O₂ and NO can activate or modify protein kinases, phosphatases, ion channels, metabolic enzymes, and redox-sensitive regulatory proteins. These signals interact with mitogen-activated protein kinase cascades and hormonal pathways to regulate genes involved in antioxidant metabolism, osmotic adjustment, ion transport, cellular protection, and growth [30,55,60,61,62]. Plant-associated microorganisms can influence this network by modifying the magnitude, timing, duration, or spatial distribution of redox signals rather than simply suppressing ROS accumulation.

Transcriptomic evidence demonstrates that microbial colonisation can establish a distinct regulatory state before stress exposure and subsequently alter the plant response to stress. In A. thaliana, root colonisation by P. chlororaphis O6 modified the expression of genes associated with defence, ROS responses, auxin, jasmonic acid, ethylene, and ABA signalling [66]. Under well-watered conditions, colonised plants showed increased expression of defence- and ROS-responsive genes and several auxin- and jasmonate-related genes, whereas some transcriptional regulators associated with ethylene and ABA signalling were downregulated. Following drought exposure, the expression pattern of colonised plants differed from that of non-colonised stressed plants, indicating that microbial treatment reconfigured the subsequent response rather than merely amplifying the conventional drought programme [66].

The response induced by P. chlororaphis O6 also illustrates the close interaction between redox and hormonal signalling. Expression of the jasmonate-responsive markers VSP1 and PDF1.2, the salicylic-acid-responsive gene PR1, and the ethylene-responsive gene HEL was modified by bacterial colonisation, although their responses differed between well-watered and drought-stressed conditions [66]. These results support a priming-like mechanism in which microbial colonisation modifies the basal transcriptional state and the kinetics of the subsequent stress response. However, changes in marker-gene expression do not establish that a single hormonal or redox pathway is responsible for the resulting phenotype, particularly because ABA-, JA-, SA-, ethylene-, and ROS-related pathways interact extensively [55,58,59,66].

Proteomic and targeted gene-expression analyses provide complementary evidence of microbially induced reprogramming. Pepper plants inoculated with the auxin- and ACC-deaminase-producing strain Bacillus licheniformis K11 survived prolonged water deprivation more effectively than non-inoculated plants and displayed distinct protein and transcript profiles under drought [67]. Two-dimensional electrophoresis identified differential accumulation of proteins related to dehydration protection, energy and ion transport, protein stabilisation, defence, and primary metabolism, including a dehydrin-like protein, vacuolar H⁺-ATPase, small heat-shock protein, pathogenesis-related protein 10, adenosine kinase, early nodulin, and S-adenosylmethionine synthetase [67]. The increased expression of Cadhn, VA, sHSP, and CaPR-10 further indicated that bacterial treatment influenced several protective pathways rather than a single antioxidant mechanism [67].

Specific experimental studies further demonstrate that microbial inoculation can alter regulatory genes located at the interface of hormonal and redox responses. Under salinity, Bacillus mesonae H20-5 increased the expression of the ABA-biosynthetic gene NCED1 and the ABA-responsive transcription factor AREB1 in tomato [35]. Salt-tolerant Bacillus strains modulated ARF, CKX2, expA1, and ERF expression in wheat, connecting auxin, cytokinin, cell-wall expansion, and ethylene-responsive signalling [48]. In Arabidopsis, B. endophyticus J13 altered the brassinosteroid regulators BZR1, BES2, and BIN2 [57], whereas P. putida RA modified hormone-, defence-, and oxidative-stress-related gene expression in drought-stressed chickpea [59]. These examples show that microbial regulation frequently converges on transcriptional nodes integrating growth, hormonal signalling, and cellular protection [35,48,57,59].

The interpretation of transcriptional responses nevertheless requires caution. Increased expression of an antioxidant, heat-shock, dehydrin, or defence-related gene may indicate enhanced preparedness, but it may also reflect stronger stress perception. Conversely, reduced expression of classical stress markers in inoculated plants may result from effective stress alleviation rather than suppression of protective capacity [65,66,67]. Transcript abundance also does not necessarily predict protein accumulation, subcellular localisation, post-translational modification, or enzymatic activity. Robust mechanistic studies should therefore combine time-resolved transcriptomics with proteomics, enzyme or transporter activity, hormone and redox profiling, microbial colonisation measurements, and physiological traits such as water status, gas exchange, ion balance, oxidative damage, growth, and yield [59,65,66,67].

4. Microbial Regulation of Osmotic and Ionic Adjustment

4.1. Osmotic Adjustment as a Key Adaptive Strategy

Drought and salinity restrict plant water acquisition by decreasing the water potential of the root environment. The resulting water-potential gradient promotes cellular dehydration, reduces turgor, restricts cell expansion, and alters stomatal conductance, photosynthetic carbon assimilation, nutrient transport, and metabolic activity. Plants can partially counteract these effects by accumulating osmotically active solutes that decrease cellular osmotic potential and help maintain water uptake and turgor under declining external water availability [68,69,70]. Osmotic regulation is particularly important in actively growing tissues, where positive turgor must be maintained to sustain cell expansion and development. The relative contributions of organic compounds and inorganic ions to this response vary among plant species, genotypes, organs, developmental stages, and stress intensities [68,69,70].

A distinction should be made between osmoregulation and osmotic adjustment. Osmoregulation broadly encompasses the processes through which cells regulate their osmotic status, whereas osmotic adjustment refers more specifically to the active net accumulation of solutes that lowers osmotic potential in response to water deficit or salinity [68,69]. An increase in proline, soluble sugars, amino acids, or inorganic ions expressed per unit fresh mass does not necessarily demonstrate osmotic adjustment because the apparent increase may partly result from tissue dehydration and the passive concentration of pre-existing solutes. More rigorous assessment requires determination of osmotic potential at comparable tissue water contents or at full turgor, preferably combined with pressure–volume analysis and measurements of relative water content, water potential, and turgor [68,69,70]. Quantification of the relative contributions of individual organic and inorganic solutes can further clarify whether the measured metabolite accumulation is sufficient to account for the observed decrease in osmotic potential [70].

Plant-associated microorganisms can influence osmotic regulation through several interconnected mechanisms. They may stimulate compatible-solute biosynthesis, modify carbon and nitrogen metabolism, improve photosynthetic performance and mineral nutrition, or reduce cellular dehydration through changes in root development, stomatal behaviour, and hydraulic function [29,34,51,52,59,71,72,73]. In drought-stressed maize, inoculation with drought-tolerant Pseudomonas strains increased relative water content, leaf water potential, proline, soluble sugars, and free amino acids while reducing leaf water loss and electrolyte leakage [71]. In groundnut, the osmolyte-producing strains Acinetobacter calcoaceticus AC06 and B. amyloliquefaciens BA01 accumulated proline, trehalose, and glycine betaine under osmotic stress and increased plant relative water content, proline, soluble sugars, photosynthetic pigments, and biomass under moderate and severe drought [72]. Similarly, Azospirillum lipoferum FK1 enhanced osmolyte accumulation, photosynthetic performance, antioxidant metabolism, and stress-related gene expression in salt-stressed chickpea [73].

The synthesis of compatible organic solutes requires the allocation of carbon, nitrogen, reducing power, and metabolic energy that could otherwise support growth or reproduction [68,70]. Microbial inoculation may partially offset this cost by preserving photosynthetic carbon assimilation, improving nutrient acquisition, and maintaining plant water status [71,72,73]. Moreover, organic and inorganic solutes do not contribute equally to the energetic cost of osmotic adjustment. Accumulation of inorganic ions is generally metabolically less expensive than de novo synthesis of organic osmolytes, although potentially toxic ions must be maintained outside the cytosol or compartmentalised within vacuoles [70]. Beneficial microorganisms may therefore support osmotic adjustment not only by increasing compatible-solute accumulation but also by coordinating carbon metabolism, ion homeostasis, and cellular compartmentalisation [71,72,73].

Changes in osmolyte concentrations must nevertheless be interpreted cautiously. Higher proline, soluble-sugar, or glycine-betaine concentrations in inoculated plants may contribute to turgor maintenance, membrane stabilisation, redox buffering, and macromolecular protection, as reported in maize, groundnut, chickpea, and pea [52,71,72,73]. However, lower osmolyte accumulation can also accompany successful microbial alleviation of stress. In maize exposed to salinity, inoculation with ACC-deaminase-producing rhizobacteria maintained higher relative water content, chlorophyll concentration, and K⁺/Na⁺ ratios while reducing proline accumulation [74]. Likewise, soybean inoculated with Serratia sp. KF23 accumulated 20.1% and 28.5% less proline at 150 and 300 mM NaCl, respectively, while showing lower H₂O₂ and malondialdehyde concentrations and improved growth under salinity [12]. These contrasting results demonstrate that increased osmolyte concentration may indicate enhanced adaptive metabolism, but it may also reflect stronger stress perception; conversely, lower accumulation may indicate reduced cellular dehydration or injury [12,71,72,73,74].

Stronger evidence for a functional contribution of microbial osmoprotectant metabolism has been obtained through genetic manipulation of the microbial partner. In common bean, inoculation with Rhizobium etli engineered to overexpress a trehalose-6-phosphate synthase gene increased nodule number, nitrogenase activity, plant biomass, drought recovery, and grain yield compared with the wild-type strain [75]. In maize, an A. brasilense strain engineered to accumulate more trehalose improved leaf water potential and increased root and shoot biomass under drought [76]. More recently, the role of bacterial trehalose was examined using the trehalose-deficient derivative Az19F of Azospirillum argentinense Az19. Although the wild-type and mutant strains colonised maize roots and promoted growth similarly under adequate irrigation, Az19F failed to improve plant height, shoot fresh weight, or relative water content under drought and induced a stronger oxidative response than the wild-type strain [77]. Together with evidence obtained using Pseudomonas mutants affected in trehalose synthesis and ACC deaminase activity [51], these studies provide substantially stronger mechanistic support than correlations between inoculation and plant osmolyte concentrations alone.

Even in these genetically informed studies, the physiological contribution of microbial osmolytes should not be interpreted simply as their direct transfer to plant tissues in quantities sufficient to determine plant osmotic potential. Trehalose and other compatible solutes can increase microbial survival under osmotic stress, preserve membrane and protein function, improve rhizosphere competence, and sustain the expression of additional plant-beneficial traits [72,75,76,77]. Their effects on the host may therefore arise from improved microbial persistence, symbiotic functionality, signalling, or reprogramming of plant metabolism. Future investigations should combine microbial mutants and functional complementation with metabolite tracing, quantitative colonisation, plant osmotic-potential measurements, tissue-specific metabolomics, and physiological assessment throughout stress and recovery [75,76,77].

4.2. Compatible Solutes and Related Osmoprotective Metabolites

Compatible solutes are low-molecular-weight compounds that can accumulate at relatively high intracellular concentrations without substantially interfering with normal cellular metabolism. Besides contributing to osmotic balance, these metabolites can stabilise proteins and membranes, protect photosynthetic structures, buffer cellular redox status, and provide carbon or nitrogen reserves during stress and subsequent recovery. Proline, glycine betaine, trehalose, soluble sugars, free amino acids, and polyamines are among the osmoprotective compounds most frequently associated with microbially mediated responses to drought and salinity [68,69,71,72,73]. However, their functions extend beyond a direct contribution to osmotic potential and include metabolic, signalling, antioxidant, and macromolecular-protective effects.

Proline is one of the most extensively investigated stress-associated metabolites in plants treated with beneficial microorganisms. Its concentration is determined by the balance among biosynthesis, degradation, transport, and incorporation into proteins, and its accumulation may contribute to osmotic regulation, membrane and protein stabilisation, redox buffering, and nitrogen storage [68,71,73]. In salt-stressed rice, inoculation with the seed-borne endophyte B. amyloliquefaciens RWL-1 increased the accumulation of several amino acids, including glutamate, aspartate, phenylalanine, cysteine, and proline, while improving growth and modulating endogenous ABA and salicylic acid concentrations [78]. Because RWL-1 also produced ABA and influenced multiple metabolic pathways, the higher proline concentration cannot be interpreted as an isolated mechanism but rather as one component of broader hormonal and metabolic reprogramming induced by the endophyte.

Glycine betaine protects membrane organisation, photosynthetic protein complexes, and enzyme structure and can contribute to osmotic and redox regulation. Certain stress-tolerant bacteria can accumulate glycine betaine themselves and may also influence its concentration in the host plant, as reported for the osmolyte-producing strains A. calcoaceticus AC06 and B. amyloliquefaciens BA01 in drought-stressed groundnut [72]. However, studies combining bacterial inoculation with externally applied glycine betaine must be distinguished from evidence of microbially induced glycine-betaine biosynthesis. In maize, the biofilm-producing strains B. subtilis SRJ4 and Curtobacterium citreum MJ1 improved plant growth, chlorophyll content, relative water content, and antioxidant responses during water deficit, while exogenous glycine betaine provided an additional benefit when applied together with the bacterial inoculants [79]. This experiment demonstrates complementarity between microbial treatment and an externally supplied osmoprotectant, but it does not establish that the bacteria induced endogenous glycine-betaine production in maize.

Arbuscular mycorrhizal fungi can also modify the accumulation of compatible solutes, although the direction of the response is not uniform. In drought-stressed basil, AMF inoculation improved soluble-carbohydrate concentrations and plant growth. However, proline, glycine betaine, and free amino acids accumulated to lower levels in mycorrhizal drought-stressed plants than in non-inoculated plants exposed to the same water deficit [80]. This response suggests that the symbiosis reduced the physiological severity of drought and consequently the requirement for high compensatory osmolyte accumulation. By contrast, biopriming mustard seeds with the salt-tolerant strain Bacillus flexus increased proline, total soluble sugars, soluble proteins, photosynthetic pigments, germination, and seedling biomass under salinity [81]. These contrasting examples reinforce that the direction of osmolyte responses depends on the plant–microorganism combination and cannot be interpreted independently of plant hydration, oxidative damage, and growth [80,81].

Mycorrhizal effects on compatible solutes may also interact with other nitrogenous osmoprotectants. In soybean exposed to severe drought, AMF inoculation increased leaf proline and glycine-betaine concentrations and modified the expression of genes involved in polyamine biosynthesis, together with improvements in relative water content, gas exchange, antioxidant regulation, and plant growth [82]. In contrast, another study in drought-stressed soybean showed that inoculation with AMF, Bradyrhizobium japonicum, or their combination reduced proline concentration and downregulated genes involved in proline biosynthesis and degradation while simultaneously enhancing antioxidant activity, nodulation, biomass, and grain yield [83]. The latter result provides a particularly clear example of why lower proline accumulation may indicate effective stress alleviation rather than impaired osmotic protection.

Trehalose requires a distinct interpretation because its functions differ between microorganisms and higher plants. In many bacteria and fungi, trehalose can accumulate as a quantitatively important osmoprotectant that preserves proteins and membranes and supports survival during dehydration or salinity. In most higher plants, by contrast, trehalose is present at comparatively low concentrations, while its phosphorylated precursor, trehalose-6-phosphate, acts primarily as a signal linking sucrose availability with carbon allocation, growth, and development [84]. Consequently, enhanced microbial trehalose metabolism should not automatically be interpreted as the direct transfer of sufficient trehalose to determine plant osmotic potential. Genetic studies using trehalose-overproducing and trehalose-deficient microbial strains instead indicate that bacterial trehalose supports microbial survival, root colonisation, symbiotic performance, and the induction of plant acclimation responses [51,75,76,77].

The interpretation of compatible-solute responses therefore requires simultaneous consideration of metabolite concentration, tissue hydration, osmotic potential, oxidative injury, photosynthesis, and growth. Measurements expressed only on a fresh-mass basis may be affected by differences in tissue water content, while total metabolite concentrations provide no information about subcellular distribution or metabolic turnover [68,69,70]. Moreover, the accumulation of a specific osmolyte does not establish causality because microbial inoculants frequently influence several hormonal, nutritional, antioxidant, and hydraulic mechanisms simultaneously [72,78,79,80,81,82,83]. Stronger evidence can be obtained by combining targeted and untargeted metabolomics with microbial biosynthetic mutants, plant metabolic or transport mutants, isotopic tracing, measurements of osmotic potential at full turgor, and physiological evaluation during both stress exposure and recovery [75,76,77,84].

4.3. Microbial Regulation of Ion Homeostasis Under Salinity

Salinity imposes an initial osmotic constraint that limits water uptake, followed by an ionic phase associated with the progressive accumulation of Na⁺ and Cl⁻ in plant tissues. Excess Na⁺ disrupts membrane potential, competes with K⁺ for transport and enzyme-binding sites, impairs nutrient acquisition, and alters photosynthesis and cellular metabolism. Maintenance of a favourable K⁺/Na⁺ ratio therefore constitutes an important component of salinity tolerance, although total tissue concentrations do not necessarily reflect the cytosolic ion environment because Na⁺ can be excluded, retrieved from the xylem, recirculated, or compartmentalised within vacuoles [68,69,70,73,74]. Beneficial microorganisms can influence these processes by modifying root growth, ion uptake and translocation, transporter-related gene expression, proton-pump activity, hormonal signalling, and rhizosphere ion availability. In the halophytic grass Puccinellia tenuiflora, exposure to B. subtilis GB03 improved growth and selective K⁺ acquisition over Na⁺ under salinity while reducing Na⁺ transport to the shoot [85].

Microbially induced ion homeostasis frequently involves components of the Salt Overly Sensitive (SOS), High-Affinity K⁺ Transporter (HKT), and Na⁺/H⁺ Exchanger (NHX) systems. Plasma-membrane SOS1 participates in Na⁺ efflux and long-distance Na⁺ distribution, HKT-type transporters influence Na⁺ and K⁺ movement and can retrieve Na⁺ from the transpiration stream, and tonoplast NHX proteins contribute to vacuolar cation compartmentalisation and cellular osmotic regulation. The electrochemical gradients required for these transport processes are generated principally by H⁺-ATPases and H⁺-pyrophosphatases, while Ca²⁺-dependent signalling participates in the perception of ionic disequilibrium and activation of downstream transport responses [85,86,87]. However, transporter function depends on the isoform, tissue, membrane localisation, plant species, and stress stage. Consequently, increased transcript abundance of an HKT, NHX, or SOS gene cannot be interpreted automatically as increased Na⁺ exclusion or sequestration without measurements of ion distribution, protein abundance, localisation, or transport activity [85,86,87].

Transcriptomic analysis of rice roots inoculated with A. brasilense Sp245 revealed broad reprogramming under 200 mM NaCl, including differential expression of genes associated with Na⁺, K⁺, and nitrate transport, Ca²⁺ signalling, ABA metabolism, antioxidant responses, and nutrient acquisition [86]. Inoculated rice plants maintained greater root and total biomass under 100 and 200 mM NaCl, supporting a connection between transcriptional reprogramming and improved plant performance. Nevertheless, the study did not directly measure ion fluxes or demonstrate that any individual transporter was essential for the phenotype. The results therefore support a multigenic and correlative model rather than a single transporter-mediated mechanism [86].

A clearer relationship between ion distribution and transporter-related expression was observed in A. thaliana inoculated with the PGPR strain PVr_9, which is closely related to Beijerinckia fluminensis. Under 150 mM NaCl, inoculated plants accumulated less Na⁺ in leaves and more in roots, indicating altered root-to-shoot distribution rather than a simple reduction in whole-plant Na⁺ uptake [87]. These changes were accompanied by upregulation of SOS1 and NHX1, lower ROS and oxidative DNA damage, reduced ABA and osmolyte accumulation, and improved root and leaf growth. The lower stress-marker accumulation alongside improved performance suggests that PVr_9 reduced the physiological severity of salinity rather than merely stimulating stronger compensatory responses [87].

In tomato, colonisation by P. aeruginosa HG28-5 decreased Na⁺ concentrations in both roots and leaves, increased K⁺ accumulation, and enhanced the expression of SOS1, SOS2, HKT1;2, and NHX1 under salinity [88]. The treatment also improved photosynthetic performance, antioxidant capacity, biomass, and fruit yield. Importantly, the protective effect was strongly reduced in the ABA-deficient notabilis mutant, linking ABA signalling with microbial regulation of Na⁺/K⁺ homeostasis and oxidative protection [88]. A complementary tomato study showed that inoculation with Pseudarthrobacter oxydans and Staphylococcus pasteuri, applied individually or as a consortium, increased the expression of the tonoplast antiporter gene NHX3, together with antioxidant genes and biomass, under 150 mM NaCl [89]. However, because ion concentrations and fluxes were not quantified in the latter study, NHX3 upregulation should be considered molecular-correlative evidence rather than direct demonstration of enhanced vacuolar Na⁺ sequestration [89].

Arbuscular mycorrhizal fungi can also modify cation homeostasis in a host-, cultivar-, and fungal-isolate-dependent manner. In two lettuce cultivars exposed to 60 mM NaCl, inoculation with Funneliformis mosseae or Claroideoglomus lamellosum increased plant growth, relative water content, photosynthetic performance, and root K⁺/Na⁺ ratios [90]. Mycorrhizal colonisation differentially regulated LsaHKT1;1, LsaHKT1;6, LsaNHX2, LsaNHX4, LsaNHX6, and LsaNHX8, while also increasing phosphorylated and non-phosphorylated PIP2 aquaporins. The response differed between lettuce cultivars and fungal inocula, demonstrating that “AMF inoculation” should not be treated as a uniform intervention and that ion transport and plant hydraulic regulation can be modified simultaneously [90].

Multilevel physiological and molecular evidence has also been obtained with Mucilaginibacter sp. K in maize. Under 100 mM NaCl, inoculated seedlings accumulated approximately 50% less Na⁺ in shoots and 60% less in roots than non-inoculated plants, whereas K⁺ concentrations were not significantly altered [91]. The bacterial treatment increased the expression of HKT1, NHX1, and the Ca²⁺ sensor gene SOS3, alongside improvements in growth, chlorophyll content, antioxidant activity, soluble-sugar accumulation, and ABA concentration. These results are consistent with coordinated regulation of Na⁺ retrieval, vacuolar compartmentalisation, Ca²⁺-dependent signalling, and stress acclimation. Nevertheless, the study did not use transporter-deficient plants, and the authors recognised that the possible contribution of bacterial exopolysaccharides to extracellular Na⁺ retention could not be separated from plant transporter regulation [91]. Cl⁻ accumulation was also not evaluated, highlighting a common limitation of studies that equate salinity tolerance exclusively with Na⁺/K⁺ homeostasis [91].

Evidence obtained under both controlled and agriculturally relevant conditions is provided by the endophytic actinomycete Streptomyces albidoflavus OsiLf-2. Inoculation increased rice growth, photosynthetic efficiency, compatible-solute accumulation, and antioxidant protection under drought and salinity while modifying the expression of stress-responsive and ion-transport-related genes [92]. Importantly, the effects were evaluated not only in hydroponic experiments but also during reproductive development and in saline–alkaline soil, where inoculation improved rice yield. This broader experimental design strengthens the agronomic relevance of the response, although the contribution of individual ion transporters remains unresolved because the bacterium simultaneously influenced osmolyte metabolism, redox regulation, photosynthesis, and several stress-responsive pathways [92].

Interpretation of microbial effects on ion homeostasis therefore requires more than transporter-gene expression or total shoot Na⁺ concentration. A lower shoot Na⁺ content may result from reduced root uptake, enhanced root retention, xylem retrieval, phloem recirculation, increased efflux, vacuolar sequestration, or extracellular binding within microbial biofilms and exopolysaccharides [85,86,87,88,89,90,91,92]. Conversely, unchanged total Na⁺ may coexist with greater tolerance if cytosolic Na⁺ is redistributed into vacuoles or less metabolically active tissues. Robust mechanistic assessment should combine tissue- and cell-specific ion quantification, K⁺/Na⁺ and Ca²⁺ balance, Cl⁻ determination, xylem- and phloem-sap analysis, non-invasive ion-flux measurements, transporter protein abundance and localisation, proton-pump activity, microbial colonisation, and the use of transporter- or signalling-deficient plant genotypes [85,86,87,88,89,90,91,92].

5. Microbial Modulation of Plant Water Relations

5.1. Water Availability and the Soil–Plant–Atmosphere Continuum Under Abiotic Stress

Water availability is a major determinant of plant growth and crop productivity because cell expansion, stomatal conductance, photosynthetic carbon assimilation, nutrient transport, and evaporative cooling depend on the maintenance of an adequate plant water status. Drought reduces the amount and mobility of water in the soil, salinity decreases the osmotic potential of the soil solution, and high temperature or elevated atmospheric vapour-pressure deficit increases transpirational demand. Although these constraints arise from different environmental conditions, they can all generate an imbalance between water supply to the roots and water loss from the shoot [68,69,70,93,94,95,96]. The physiological consequences of this imbalance depend on stress intensity, duration, rate of development, timing within the crop cycle, soil hydraulic properties, rooting depth, atmospheric demand, and plant developmental stage [93,94,95,96]. Salinity additionally combines restricted water uptake with progressive Na⁺ and Cl⁻ accumulation, requiring coordinated osmotic, ionic, and hydraulic acclimation [96].

Water movement should be considered within the soil–plant–atmosphere continuum, in which water flows along gradients in water potential from the bulk soil, across the rhizosphere and root tissues, through the xylem and leaves, and finally to the atmosphere [93,94,95]. This flow is constrained by hydraulic resistances located in the soil, at the soil–root interface, within radial and axial root pathways, in the xylem, and across leaf tissues. During progressive soil drying, the marked reduction in unsaturated soil hydraulic conductivity and the development of steep water-potential gradients around absorbing roots can become dominant limitations to water uptake [93,94]. Under comparatively moist soil and high atmospheric demand, plant hydraulic conductance and stomatal regulation may instead exert greater control over transpiration [94,95]. In mature apricot trees, the integration of sap-flow measurements with water-potential gradients permitted estimation of hydraulic conductance and its temporal variation under orchard conditions, illustrating the value of studying water transport at the whole-plant scale [97]. Soil drying and atmospheric drying should therefore not be treated as physiologically equivalent, even when both ultimately reduce stomatal conductance and carbon assimilation [94,95,97].

Water-use efficiency is a scale- and method-dependent concept that must be defined precisely in studies involving microbial inoculation. At leaf level, intrinsic water-use efficiency usually refers to the ratio between net CO₂ assimilation and stomatal conductance, whereas instantaneous water-use efficiency commonly relates assimilation to transpiration. At whole-plant or crop scale, water-use efficiency may be expressed as biomass or grain yield per unit of water transpired, evapotranspired, supplied through irrigation, or received through irrigation and rainfall [94,95,98]. These variables are related but not interchangeable. An inoculant may increase instantaneous intrinsic water-use efficiency through stronger stomatal restriction without improving seasonal biomass or yield. Conversely, increased crop water productivity may result from sustained carbon assimilation, reduced non-productive soil evaporation, altered phenology, greater rooting depth, or improved access to stored soil water [95,98]. Claims of enhanced water-use efficiency should therefore identify the spatial scale, temporal interval, numerator, and water-use denominator employed [94,95,98].

Plant-associated microorganisms can influence water relations at several interconnected points within the soil–plant–atmosphere continuum. Modifications of root-system architecture can change the volume and depth of soil explored, whereas microbial regulation of ABA, cytokinins, auxins, ethylene, and aquaporins can affect stomatal behaviour, root growth, and root hydraulic properties [29,32,34,44,47,90]. Fungal hyphae may extend beyond the water-depletion zone surrounding roots, while bacterial exopolysaccharides, biofilms, and microbially induced changes in soil aggregation can affect rhizosphere structure, root–soil contact, and local water retention [99]. Microbial modulation of osmotic adjustment and ion homeostasis may additionally help preserve cellular water uptake under drought or salinity [71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92]. These mechanisms are strongly interconnected because improved nutrient acquisition can support root growth and photosynthesis, hormonal signalling can modify water loss, and rhizosphere properties can alter the hydraulic conditions under which plant responses occur [29,32,90,99].

Direct hydraulic evidence has been obtained in plants associated with arbuscular mycorrhizal fungi. During progressive soil drying, tomato plants colonised by Rhizophagus irregularis maintained less negative leaf water potentials and greater soil–plant hydraulic conductance than plants in which mycorrhizal colonisation was strongly reduced [100]. The fungal hyphal network was proposed to extend the effective water-absorbing radius of the root system, moderate the decrease in matric potential at the soil–root interface, and help preserve hydraulic continuity as the soil dried [100]. In Viburnum tinus, AMF inoculation was assessed under irrigation with treated saline wastewater using stem water potential, gas exchange, leaf hydraulic conductance, leaf structural traits, and a subsequent recovery phase [101]. In soybean, inoculation with Rhizophagus clarus moderated drought-induced changes in leaf water potential, photochemical performance, carbon assimilation, and growth, although the magnitude of the response differed between cultivars [102]. These studies demonstrate that mycorrhizal effects on plant water relations depend on successful colonisation, host genotype, soil conditions, irrigation-water quality, and the hydraulic component evaluated [100,101,102].

Field studies in perennial crops provide an important physiological framework for interpreting stomatal and hydraulic responses under realistic irrigation conditions. In grapefruit and mandarin trees subjected to deficit irrigation and irrigation with reclaimed water, the two species displayed contrasting water-regulation strategies [103]. Grapefruit exhibited more isohydric behaviour, maintaining tighter control of plant water potential through stomatal regulation, whereas mandarin showed a comparatively less isohydric response. Variations in stomatal conductance were associated with changes in hydraulic conductance along the root-to-stem and canopy pathways, indicating that stomatal behaviour cannot be interpreted independently of whole-plant hydraulic limitations [103]. These observations complement evidence from adult apricot trees, in which sap flow and water-potential gradients were integrated to quantify hydraulic conductance under field conditions [97,103].

Microbial benefits are not universal, and experiments reporting weak or conditional effects are essential for identifying the factors controlling inoculant performance. In Cistus albidus subjected to water deficit, inoculation with the ectomycorrhizal fungus Pisolithus tinctorius improved plant water status and moderated some physiological effects of drought despite relatively low root colonisation [104]. The treatment also altered ABA, salicylic acid, and ethylene-related responses, suggesting that the observed effects integrated water relations and hormonal regulation [104]. By contrast, AMF colonisation remained low in tomato plants irrigated with saline reclaimed water under commercial greenhouse conditions, and inoculation did not significantly improve most water-relation variables or yield [105]. Substrate composition exerted a stronger influence on water and nutrient availability, although some effects on gas exchange became evident late in the experiment [105]. These contrasting outcomes show that inoculation cannot be expected to overcome limitations associated with inadequate colonisation, unsuitable substrates, salinity, or irrigation management [104,105].

Bacterial inoculants can also improve plant hydration during water deficit, although direct measurements of hydraulic conductance remain less common than measurements of relative water content, growth, or membrane stability. In drought-stressed wheat, desiccation-tolerant strains of Enterobacter cloacae, B. cereus, and B. megaterium increased relative water content, root and shoot growth, photosynthetic pigments, compatible solutes, and antioxidant protection [106]. The isolates produced exopolysaccharides under osmotic stress, suggesting that bacterial survival and modification of the rhizosphere may have contributed to plant performance [106]. In barley, combined application of PGPR and biochar improved germination, root and shoot development, chlorophyll concentration, nutrient acquisition, and antioxidant metabolism under drought [107]. Nevertheless, higher relative water content or biomass does not by itself demonstrate increased hydraulic conductance or water-use efficiency because water flow, hydraulic gradients, transpiration, and total water consumption were not directly quantified [106,107].

The interpretation of microbial effects on plant water relations therefore requires discrimination among improved hydration, altered stomatal regulation, delayed dehydration, increased access to soil water, and genuine enhancement of hydraulic transport. Higher relative water content may result from increased root water uptake, reduced transpirational loss, stronger osmotic adjustment, or differences in plant size and developmental stage [93,94,95]. Similarly, greater stomatal conductance may sustain photosynthesis during moderate stress but accelerate soil-water depletion, whereas stronger stomatal closure may increase instantaneous intrinsic water-use efficiency while limiting carbon gain and yield [94,95,98]. The agronomic value of a response consequently depends on whether water deficit is intermittent, progressive, terminal, or chronic and on the relationship among water availability, atmospheric demand, rooting volume, and crop phenology [93,94,95,98].

Translation from controlled experiments to agricultural conditions remains a major challenge. Microbial establishment and physiological effectiveness are influenced by soil texture, soil water availability, nutrient status, temperature, indigenous microbial communities, host genotype, inoculum formulation, irrigation strategy, and crop management [100,101,102,103,104,105,106,107,108]. Across 54 maize fields, growth responses to R. irregularis inoculation varied substantially among locations and were predicted more effectively when soil characteristics and resident microbiome composition were considered together [108]. Although the study was not specifically designed to investigate drought hydraulics, it demonstrates that inoculant performance cannot be extrapolated solely from microbial traits identified under controlled conditions. Effects on crop water relations must therefore be validated across representative soils, genotypes, climates, and irrigation regimes [100,101,102,103,104,105,106,107,108].

Experimental assessment of microbially mediated water relations should integrate soil water content and soil water potential with predawn and midday leaf or stem water potential, relative water content, stomatal conductance, transpiration, photosynthesis, sap flow, root and whole-plant hydraulic conductance, rooting distribution, biomass, and yield [93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108]. Measurements should be collected throughout stress development and recovery because a single endpoint cannot distinguish delayed dehydration from sustained hydraulic acclimation [93,101]. Gravimetric water loss should be normalised for plant size and corrected for soil evaporation, while pot experiments should consider restricted rooting volume and artificial soil-water distributions [93,94]. At crop scale, seasonal evapotranspiration, irrigation input, rainfall, biomass, grain yield, and harvest index are required to determine whether inoculation improves water productivity rather than merely modifying instantaneous gas exchange [95,98,103,108].

5.2. Microbial Regulation of Root Hydraulic Conductivity and Aquaporins

Root hydraulic properties determine the capacity of the root system to absorb water from the soil and transport it towards the xylem. Root hydraulic conductance refers to water transport through the entire root system, whereas root hydraulic conductivity is generally normalised by root surface area, root dry mass, or another measurement of root size [109,110,111]. These variables depend on the water-potential gradient between the root medium and the xylem and on the combined resistance of radial and axial transport pathways. Drought, salinity, nutrient limitation, low root-zone temperature, and oxygen deficiency can modify these resistances through changes in root anatomy, suberisation, membrane permeability, xylem development, aquaporin activity, and root–soil contact [96,99,109,110,111]. Greater root biomass, relative water content, or shoot water potential cannot therefore be considered direct evidence of increased root hydraulic conductivity unless both water flow and its driving force are quantified [93,94,109,110,111].

Radial water transport across roots occurs through apoplastic, symplastic, and transcellular pathways, whose relative contributions vary with root developmental stage, anatomy, environmental conditions, and the magnitude of the water-potential gradient [109,110,111]. Aquaporins facilitate the membrane component of cell-to-cell water transport, with plasma membrane intrinsic proteins and tonoplast intrinsic proteins playing important roles in root water uptake and intracellular water redistribution [110,111]. Aquaporin activity is regulated at several levels, including transcription, translation, protein abundance, trafficking, phosphorylation, heteromerisation, gating, cytosolic pH, Ca²⁺ concentration, ROS signalling, and membrane tension [110,111,112,113]. Considerable discrepancies can occur between aquaporin transcript abundance and the abundance, localisation, or functional state of the corresponding proteins under drought and salinity [113]. Increased expression of a PIP or TIP gene should consequently not be interpreted automatically as greater membrane water permeability or higher root hydraulic conductivity [110,111,112,113].

Plant-associated microorganisms may influence root hydraulics through changes in root-system architecture, cortical development, suberin deposition, nutrient availability, hormonal status, aquaporin abundance and gating, fungal water transport, and hydraulic continuity at the soil–root interface [32,90,99,111,112,113]. Inoculation may also alter the relative contribution of apoplastic and cell-to-cell transport as soil water availability changes [111,112]. However, beneficial colonisation does not uniformly increase aquaporin expression. Individual isoforms may be induced, repressed, or unaffected, and the direction of regulation may differ between well-watered and stressed plants [111,112,113]. Such responses may reflect functional optimisation rather than inconsistent regulation because maximal membrane water permeability is not necessarily advantageous during severe or rapidly developing water deficit [111,112,113].

In maize exposed to drought, inoculation with autochthonous AMF, Bacillus thuringiensis, or their combination modified root hydraulic conductivity and the expression of several plant aquaporins while improving nutritional status, physiological performance, and antioxidant protection [114]. Individual and combined inoculations did not affect every aquaporin or hydraulic variable in the same direction, indicating that the response depended on inoculum composition and plant physiological status [114]. The simultaneous modification of nutrient acquisition, redox regulation, root physiology, and water transport demonstrates that improved drought performance could not be attributed exclusively to aquaporin regulation [114].

Host genotype strongly conditions microbially induced hydraulic responses. In two maize cultivars differing in inherent drought tolerance, colonisation by R. irregularis improved root hydraulic conductivity during water deficit, but the magnitude of the response and the associated aquaporin-expression profiles differed markedly between genotypes [115]. In the drought-sensitive cultivar, the reduction in hydraulic conductivity was substantially greater in non-mycorrhizal than in mycorrhizal plants. Mycorrhizal regulation encompassed several plant PIP, TIP, NIP, and SIP aquaporins, together with fungal aquaporins, but individual genes were variously induced, repressed, or unaffected [115]. Hydraulic improvement can therefore result from coordinated reorganisation of multiple transport components without requiring generalised aquaporin upregulation [113,115].

Interactions between aquaporins and hormonal pathways further increase the complexity of microbial hydraulic regulation. In tomato, comparison of the jasmonate-deficient def-1 mutant with the wild type demonstrated that jasmonate status influenced stomatal conductance, root hydraulic conductivity, and the expression of both plant and fungal aquaporins during association with R. irregularis [116]. In maize, AM symbiosis increased root hydraulic conductivity under drought, whereas exogenous salicylic acid modified hydrostatic and osmotic conductivity and changed the relative contributions of apoplastic and cell-to-cell water transport [117]. These responses were accompanied by differential regulation of ZmPIP2;4, ZmTIP1;1, and other aquaporins and by changes in the balance among salicylic acid, ABA, and jasmonates [117]. Microbial regulation of aquaporins should therefore be interpreted within an integrated hormonal and hydraulic network rather than as an isolated transcriptional mechanism [116,117].

The presence of a mycorrhizal fungus does not necessarily account for every hormonal or aquaporin response observed in colonised plants. In split-root poplar seedlings exposed to mild localised drought, the drying root compartment accumulated more ABA, contained less trans-zeatin riboside, and exhibited altered expression of several PIP aquaporins [118]. Colonisation by the ectomycorrhizal fungus Laccaria bicolor modified the expression of other aquaporins and reduced PIP2 protein abundance, but the principal local responses to water deficit were predominantly governed by root ABA/cytokinin status independently of fungal colonisation [118]. These results highlight the need to distinguish microorganism-induced effects from intrinsic plant responses to spatially heterogeneous soil drying [118].

Hydraulic regulation may also emerge from interactions involving several microbial partners. In the tripartite association among Helianthemum almeriense, the desert truffle Terfezia claveryi, and the mycorrhizal helper bacterium P. mandelii #29, bacterial inoculation increased fungal colonisation, nutrient acquisition, and plant performance under drought [119]. Water deficit reduced root hydraulic conductivity, but this reduction was moderated in the presence of the helper bacterium. The response was accompanied by changes in root aquaporin expression and hormone concentrations, including ABA [119]. The bacterium therefore appeared to improve plant hydraulics both indirectly, by facilitating the mycorrhizal association, and through modification of plant regulatory processes [119]. This system illustrates that inoculant effectiveness may depend on interactions with resident or co-inoculated microorganisms rather than solely on direct bacterium–plant activity.

Particularly strong evidence for an ABA–aquaporin–hydraulic connection was obtained using B. subtilis IB-22 and an ABA-deficient barley mutant. The mutant exhibited lower root hydraulic conductivity and impaired shoot growth compared with its parental genotype, whereas bacterial inoculation increased leaf water potential, shoot growth, and root hydraulic conductivity [120]. Immunolocalisation showed that inoculation increased PIP2;1 and PIP2;2 abundance in the mutant, supporting their involvement in bacterial restoration of root water transport [120]. In the parental genotype, however, the bacterium increased PIP2;5 abundance without significantly increasing root hydraulic conductivity. This genotype-dependent response demonstrates that greater abundance of a particular aquaporin is not necessarily sufficient to modify whole-root water transport and that microbial effectiveness depends strongly on the initial hormonal and hydraulic status of the host [120].

Mechanistic evaluation of microbial effects on root hydraulics should combine direct measurements of root hydraulic conductance or conductivity with root surface area, root anatomy, suberisation, xylem development, soil and xylem water potentials, and whole-plant transpiration [109,110,111,112,113,114,115,116,117,118,119,120]. Aquaporin analyses should integrate transcript abundance, protein abundance, phosphorylation state, membrane localisation, and, where possible, membrane water permeability or pharmacological and genetic tests of channel function [110,111,112,113,114,115,116,117,118,119,120]. Hormonal profiling, quantitative microbial colonisation, soil hydraulic properties, and temporal measurements during stress development and recovery are also required to distinguish increased water uptake from restricted water loss, delayed dehydration, or plant-size effects [93,100,109,110,111,112,113,114,115,116,117,118,119,120]. The functional significance of microbial hydraulic regulation can only be established when these measurements are connected with photosynthetic performance, plant water status, biomass, and yield [93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120].

5.3. Microbial Modification of Rhizosphere Hydraulic Properties

The rhizosphere is not a passive boundary between roots and bulk soil but a dynamic hydraulic domain whose physical properties are modified by root activity and microbial metabolism. Extracellular polysaccharides, biofilm matrices, microbial cells, fungal hyphae, and microbially promoted soil aggregation can alter the amount of soil adhering to roots, pore-size distribution, aggregate stability, wettability, and the spatial distribution of water around absorbing surfaces [121,122,123,124]. These changes may moderate rapid fluctuations in rhizosphere water content and help maintain root–soil contact during drying. Nevertheless, greater water retention should not automatically be interpreted as greater water availability to the plant because water may be retained at matric potentials that restrict root extraction, while changes in pore geometry can produce contrasting effects on unsaturated water flow [121,122,123,124].

Experimental evidence supports a functional relationship among microbial extracellular-polysaccharide production, rhizosheath formation, and plant performance under water deficit. In sunflower, inoculation with the exopolysaccharide-producing Rhizobium sp. strain YAS34 increased the amount of root-adhering soil relative to root dry mass under both normal and restricted water supply [121]. The inoculated plants also showed improved root development and shoot biomass, linking microbial polysaccharide production with rhizosphere aggregation and plant growth. Similarly, the drought-tolerant P. putida strain GAP-P45 produced high amounts of extracellular polysaccharides under water stress and increased sunflower survival, root and shoot dry biomass, root-adhering soil, aggregate stability, and aggregate mean weight diameter during drought [122]. These responses are consistent with the formation of a more cohesive rhizosheath, although both strains possessed additional plant growth-promoting properties and the measured plant benefits cannot be attributed exclusively to the quantity of polysaccharide produced [121,122].

Field-scale observations provide additional evidence that extracellular polysaccharides can influence soil hydrophysical behaviour. In a field experiment with two tomato cultivars exposed to full or deficit irrigation, differences in rhizosphere microbial activity and extracellular-polysaccharide concentrations were associated with changes in soil water retention and contrasting crop responses to water deficit [123]. This study is particularly valuable because the measurements were conducted in rhizosheath and bulk soil under field conditions rather than exclusively in artificial substrates. In a separate agricultural-soil experiment, long-term organic amendments and irrigation regime modified extracellular-polysaccharide content, microbial biomass, water-stable aggregation, and microbial-community composition [124]. However, these studies also illustrate an important attribution problem: extracellular-polysaccharide concentrations respond to root genotype, irrigation, organic amendments, microbial biomass, and community composition, making it difficult to isolate the contribution of an individual inoculated microorganism without appropriate sterile, non-inoculated, and substrate-matched controls [123,124].

Biofilm formation represents a related but broader microbial strategy because it integrates surface attachment, cell aggregation, extracellular-matrix production, and protection against desiccation. Genetic evidence was obtained using Paenibacillus polymyxa A26, in which inactivation of an Sfp-type phosphopantetheinyl transferase increased biofilm formation relative to the wild-type strain [125]. Wheat inoculated with the mutant showed substantially greater survival and biomass under severe drought, while complementation restored biofilm formation towards the wild-type phenotype, providing stronger mechanistic support than studies based solely on correlations between inoculation and plant hydration [125]. More recently, biofilm-forming bacterial isolates obtained from arid grassland rhizospheres increased wheat relative water content and harvest index under moderate and severe water deficit and modified root traits and antioxidant responses [126]. Nevertheless, most isolates also produced auxins, expressed ACC deaminase activity, or solubilised mineral nutrients. In vitro biofilm formation therefore identifies a potentially useful microbial trait, but it does not demonstrate that biofilm production was the sole or predominant mechanism operating on the root surface under soil conditions [125,126].

Arbuscular mycorrhizal fungi can modify rhizosphere structure through extraradical hyphae, fungal-derived organic compounds, and interactions with plant roots and other soil microorganisms. Across five grassland plant species, water-stable aggregation was associated with root length, plant cover, AMF hyphal abundance, and the soil fraction operationally quantified as glomalin, with path analysis indicating a strong relationship between this fraction and aggregate stability [127]. However, such covariance does not by itself establish that a single chemically defined fungal protein caused aggregation, and measurements of glomalin-related soil proteins should be interpreted operationally rather than as direct quantification of one exclusive AMF product. In a root-free experimental system, AMF mycelium alone was sufficient to maintain water-stable aggregates but also increased soil water repellency [128]. This latter response is especially important because microbial or fungal promotion of aggregation is not invariably equivalent to improved infiltration or hydraulic access to water [127,128].

The effects of AMF on soil hydraulic properties are strongly dependent on soil texture. In root-free compartments populated by R. irregularis, fungal ingrowth reduced water retention and increased unsaturated hydraulic conductivity in loam, whereas it increased water retention and tended to reduce conductivity in quartz sand, particularly at low water contents [129]. These opposing responses were attributed to texture-dependent changes in pore-size distribution: fungal activity appeared to increase the relative contribution of larger, more conductive pores in loam but promoted smaller water-retaining pores or partial pore constriction in sand [129]. Thus, the same fungus can potentially facilitate drainage and water movement in a fine-textured soil while increasing water storage but restricting flow in a coarse substrate. Irrigating mycorrhizal and non-mycorrhizal treatments to the same volumetric water content may therefore expose them to different water potentials and hydraulic limitations [129].

In addition to altering soil structure, extraradical fungal hyphae may provide a pathway for water movement across soil volumes that are poorly connected by liquid water films. Direct evidence was obtained using isotopically labelled water placed in root-free compartments separated from plant roots by air gaps. Labelled water was detected in host plants only when intact AMF hyphae connected the compartments, demonstrating water transport through the fungal pathway [130]. A subsequent compartmentalised experiment also quantified hyphal contributions to plant water acquisition and showed that AMF-mediated transfer remained detectable under contrasting soil-moisture conditions [131]. Nevertheless, the relative importance of this pathway depends on the amount and hydraulic continuity of water already accessible to the roots, the dimensions of the hyphal network, soil texture, the degree of hydraulic disconnection among pores, and the experimental ratio between hyphal and root compartments [129,130,131]. Air-gap systems provide strong evidence that hyphae can transport water, but they may overestimate the proportional contribution of this route in continuous field soils where matrix flow and root access operate simultaneously [129,130,131].

Claims that microbial inoculation improves rhizosphere water availability should therefore be supported by direct measurements rather than inferred exclusively from bacterial polysaccharide production, in vitro biofilm formation, soil moisture content, or plant relative water content. Mechanistic assessment should combine quantitative microbial colonisation with extracellular-polysaccharide extraction and characterisation, rhizosheath mass, aggregate-size distribution and stability, soil wettability, water-repellency measurements, soil water-retention curves, unsaturated hydraulic conductivity, rhizosphere and bulk-soil water potentials, and root–soil contact imaging [121,122,123,124,125,126,127,128,129]. Isotopic tracing and hydraulically isolated compartments can determine whether fungal hyphae transport water directly, whereas measurements in continuous soils are required to establish the importance of that pathway relative to matrix flow [129,130,131]. These analyses should ultimately be connected with root and whole-plant hydraulic conductance, transpiration, gas exchange, plant water status, biomass, and yield under realistic drying–rewetting cycles and field irrigation regimes [121,122,123,124,125,126,127,128,129,130,131].

6. Conclusions and Future Perspectives

Abiotic stresses increasingly restrict crop establishment, vegetative growth, reproductive development, yield stability, and agricultural water productivity. The evidence examined in this review shows that plant-associated bacteria and fungi can contribute to stress acclimation through mechanisms extending well beyond nutrient acquisition. Microbial inoculation can modify hormonal homeostasis, redox signalling, antioxidant metabolism, osmotic adjustment, ion distribution, root-system development, plant hydraulics, aquaporin regulation, and the physical properties of the rhizosphere [6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131]. These processes are strongly interconnected and should not be interpreted as independent protective modules. For example, microbial regulation of ABA and cytokinins can simultaneously affect stomatal conductance, root hydraulic conductivity, aquaporin abundance, shoot growth, and antioxidant demand, whereas changes in root architecture and rhizosphere structure can influence nutrient acquisition, soil-water accessibility, and ion transport [29,32,44,90,100,111,112,113,114,115,116,117,118,119,120].

The available evidence supports viewing microbial inoculants primarily as regulators of plant and rhizosphere function rather than as simple external suppliers of hormones, antioxidants, or osmoprotectants. Microorganisms may produce phytohormones, ACC deaminase, extracellular polysaccharides, trehalose, siderophores, or other bioactive compounds, but their effects on the host generally emerge through interactions with endogenous plant signalling and metabolism [27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,121,122,123,124,125,126,127,128,129,130,131]. A microbial metabolite detected under in vitro conditions should therefore not automatically be considered the mechanism responsible for plant protection. Demonstrating physiological relevance requires evidence that the microorganism produces or modifies the compound under plant-associated conditions, that the response occurs in the appropriate tissue and developmental stage, and that disruption of the microbial or plant pathway weakens the protective phenotype [34,35,36,41,45,50,51,57,75,76,77,120,125].

A recurring interpretative difficulty is that many commonly measured stress markers can reflect either enhanced tolerance or greater stress exposure. Increased antioxidant enzyme activity, proline concentration, ABA accumulation, or expression of stress-responsive genes may indicate improved defensive capacity, but it may also reveal a stronger oxidative, osmotic, or ionic challenge [12,52,58,59,63,64,65,66,67,71,72,73,74,75,76,77,78,79,80,81,82,83,84]. Conversely, inoculated plants may accumulate less proline, ABA, ROS, or antioxidant metabolites because microbial treatment reduced the severity of stress and the requirement for compensatory responses [12,65,74,80,83,87]. Similar caution applies to transporter and aquaporin transcripts, which do not necessarily predict protein abundance, membrane localisation, post-translational regulation, or functional transport activity [85,86,87,88,89,90,91,92,110,111,112,113,114,115,116,117,118,119,120]. Mechanistic interpretation should consequently rely on coordinated physiological, biochemical, molecular, and microbiological evidence rather than on isolated biomarkers.

The strength of the evidence also varies substantially among the mechanisms considered. Correlations between inoculation and changes in metabolites, transcripts, or plant performance provide useful initial support but do not establish causality. More convincing evidence has been obtained through microbial biosynthetic mutants and complemented strains, hormone-deficient or signalling-impaired plant genotypes, transporter or regulatory mutants, protein localisation, direct hydraulic measurements, isotopic tracing, and experiments separating fungal hyphal transport from root water uptake [34,35,36,41,45,50,51,57,75,76,77,88,116,117,118,119,120,125,130,131]. Wider adoption of these approaches is necessary to determine whether microbial protection results from direct regulation of a particular plant pathway, improved microbial persistence, altered rhizosphere conditions, or the cumulative action of several interacting mechanisms.

The physiological outcome of inoculation is strongly dependent on the microbial strain, host species and genotype, developmental stage, stress type and intensity, soil properties, nutrient availability, irrigation regime, atmospheric demand, resident microbiome, and capacity of the inoculant to colonise and persist in the target environment [6,7,8,9,10,11,12,21,22,43,44,45,46,80,81,82,83,90,97,98,99,100,101,102,103,104,105,106,107,108,114,115]. Results obtained in sterile substrates, hydroponics, agar plates, or small pots are therefore insufficient to predict agronomic performance. Controlled experiments remain essential for identifying mechanisms, but promising treatments must subsequently be evaluated in non-sterile soils, under realistic root volumes, fluctuating environmental conditions, repeated drying–rewetting cycles, and relevant crop developmental stages. Field validation should include yield and quality traits, seasonal water use, nutrient-use efficiency, colonisation dynamics, persistence of the introduced microorganism, effects on native microbial communities, and performance across contrasting soils and climates [92,97,98,99,100,101,102,103,104,105,106,107,108].

Water relations represent a particularly important area in which methodological precision is still required. Higher relative water content, biomass, or leaf water potential does not by itself demonstrate enhanced water uptake, root hydraulic conductivity, or water-use efficiency [93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120]. Similarly, increased soil moisture after microbial inoculation does not necessarily indicate that the retained water is hydraulically available to plants [121,122,123,124,125,126,127,128,129]. Future studies should integrate soil water-retention and conductivity measurements with soil, root, stem, and leaf water potentials; sap flow; gas exchange; root and whole-plant hydraulic conductance; aquaporin protein regulation; root-system distribution; and seasonal evapotranspiration. Isotopic tracing and spatially resolved measurements can further distinguish direct hyphal water transfer, changes in matrix flow, improved root–soil contact, and plant-level regulation of water loss [100,121,122,123,124,125,126,127,128,129,130,131].

Multi-omics approaches can substantially improve understanding of plant–microorganism interactions, but their value depends on integration with functional measurements. Transcriptomics, proteomics, metabolomics, ionomics, and microbiome profiling can identify candidate pathways and regulatory networks, yet omics associations alone cannot demonstrate physiological function [59,63,65,66,67,78,86,92]. Time-resolved and tissue-specific analyses should be connected with microbial colonisation, hormone concentrations, redox state, enzyme activity, ion fluxes, osmotic potential, hydraulic transport, photosynthesis, growth, and yield. Experiments should also include appropriate microbial mutants, plant genetic resources, complemented strains, and uninoculated controls exposed to equivalent water and nutrient conditions. Such designs will allow direct microbial regulation to be distinguished from secondary responses caused by improved nutrition, delayed stress development, or differences in plant size.

The development of microbial consortia may offer broader functional capacity than single-strain inoculants, particularly when complementary traits such as nutrient mobilisation, ACC deaminase activity, hormonal regulation, osmoprotection, ion homeostasis, biofilm formation, and mycorrhizal facilitation are combined [8,9,10,11,12,46,89,106,114,119]. However, greater taxonomic complexity does not necessarily produce greater effectiveness. Microbial partners may compete for resources, interfere with colonisation, alter one another’s metabolite production, or perform inconsistently across soils and host genotypes. Rational consortium design should therefore be based on demonstrated compatibility, functional complementarity, ecological persistence, biosafety, and reproducible plant benefits rather than on the simple accumulation of strains possessing desirable traits in vitro [8,9,10,11,12,22,64,89,101,108].

Future microbial products will also require formulations capable of preserving cell viability, functional activity, shelf life, and colonisation competence during storage and after application. Carrier material, inoculum density, application timing, compatibility with fertilisers and crop-protection products, soil moisture, temperature, and irrigation management can all determine whether a microorganism establishes successfully in the rhizosphere or plant tissues [6,7,8,9,10,11,12,21,22,101,105,108]. Biosafety assessment is particularly important when candidate isolates belong to taxa that include opportunistic plant, animal, or human pathogens [64]. Regulatory evaluation and commercial development should consequently integrate microbial identity at strain level, genomic assessment of virulence and antimicrobial-resistance determinants, environmental persistence, non-target effects, and consistent agronomic performance.

Microbial inoculants therefore represent a valuable component of integrated strategies for improving crop resilience, but they should not be presented as universal substitutes for appropriate irrigation, soil management, fertilisation, crop selection, or genetic improvement. Their greatest potential lies in complementing these practices by increasing the efficiency and plasticity of plant responses and by modifying the rhizosphere environment in ways that reduce the physiological cost of stress acclimation [6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,121,122,123,124,125,126,127,128,129,130,131]. Progress towards reliable field application will depend on moving from descriptive demonstrations of growth promotion to experimentally supported mechanisms, clearly defined environmental targets, reproducible formulations, and validation of yield, resource-use efficiency, and economic benefit under realistic agricultural conditions.