Local and Systemic Transcriptional Responses of Tomato to a Growth-Promoting Streptomyces Consortium

TL;DR

A Streptomyces consortium promotes tomato growth and suppresses disease by altering plant gene expression in roots and leaves.

Contribution

The study identifies molecular signatures of plant growth promotion and pathogen suppression by a defined Streptomyces consortium.

Findings

Application of the TOM consortium significantly enhanced root and shoot growth in tomato plants.

Transcriptional analysis revealed down-regulated defense responses and up-regulated growth-related genes in both roots and leaves.

The TOM consortium reduced disease severity caused by Fusarium oxysporum by approximately 60%.

Abstract

Members of the genus Streptomyces are prominent inhabitants of the plant rhizosphere and endosphere and are increasingly recognized for their roles in plant growth promotion and disease suppression. In this study, we isolated genetically distinct Streptomyces from the tomato (Solanum lycopersicum L.) rhizosphere, designated as TOM isolates, and assembled them into a defined 12-member TOM consortium. Application of the TOM consortium significantly promoted root and shoot growth in tomato. RNA-seq analysis revealed coordinated local and systemic transcriptional responses characterized by a predominance of down-regulated genes in both roots and leaves. In roots, differential gene expression reflected selective attenuation of defense- and cell wall-related processes, alongside increased expression of genes associated with phytoalexin biosynthesis, phosphate starvation responses, and hormonal regulation. In leaves, transcriptional reprogramming was dominated by reduced stress-related responses together with activation of metabolic and growth-associated functions. The TOM consortium also reduced disease severity caused by Fusarium oxysporum f. sp. radicis-lycopersici by approximately 60% compared to infected controls. To further characterize functional traits of individual consortium members, isolates were evaluated in vitro for antifungal activity and five strains displaying inhibition were selected for hybrid whole-genome sequencing. Genome analyses revealed diverse taxonomic affiliations and a rich repertoire of biosynthetic gene clusters, including clusters associated with known antimicrobial metabolites as well as numerous low-similarity clusters indicative of substantial unexplored biosynthetic potential. Collectively, this study provides new insights into plant interactions with beneficial Streptomyces, while revealing molecular signatures involved in Streptomyces-mediated plant growth promotion and pathogen suppression.

Figures

Full text

1. Introduction

Beneficial interactions between plants and microbes are widespread in natural and agricultural systems, improving host nutrition, promoting development, and enhancing resilience to biotic and abiotic constraints [1,2]. These associations are established and maintained through continuous molecular and chemical exchanges between the two partners, particularly in the rhizosphere, a highly competitive environment at the root–soil interface. Through the release and perception of diverse semiochemicals, including plant-derived exudates and microbial signaling compounds, plants and their associated beneficial microbes engage in a dialog that enables mutual recognition, coordination of activities and successful colonization [3,4].

From the host perspective, whole-genome transcriptomic analyses have been important in providing system-level insights into how plants respond to beneficial microbes. Such studies have revealed local transcriptional signatures in roots that differ from those triggered by microbial immune elicitors or pathogens and, in some cases, display additional layers of regulation, including cell type-specific expression, epigenetic modifications, or alternative splicing [5,6,7]. Beyond local root responses, root-associated microbes can also elicit systemic responses in aboveground plant tissues. Early work on rhizobacteria-induced systemic resistance (ISR) showed that effective root colonization primes defense responses in leaves without detectable steady-state changes in gene expression, demonstrating that systemic aboveground effects are not necessarily accompanied by widespread transcriptional reprogramming [8]. However, several other studies have shown that root-colonizing microbes elicit detectable or even prominent transcriptional responses in aboveground tissues [9,10,11].

In natural and agricultural environments, plants are exposed to complex microbial assemblages composed of phylogenetically and functionally diverse taxa. Such assemblages can exert emergent effects on host transcriptional programs and plant physiology that cannot be predicted from single-isolate studies, arising from functional redundancy, metabolic complementation or synergistic interactions among community members [12,13]. To capture these community-level properties under controlled conditions, synthetic and simplified microbial communities have been widely adopted as experimentally tractable models, enabling investigation of plant–microbe interactions at the consortium level while retaining key features of microbial complexity that are lost in single-strain approaches [14,15]. In many cases, such collective microbial assemblies confer functional advantages over individual strains, resulting in more consistent or robust effects on plant performance, including enhanced growth, nutrient acquisition, and disease suppression [13,16,17,18,19]. Despite the availability of such approaches, transcriptomic studies addressing plant responses to defined bacterial consortia remain limited.

Members of the genus Streptomyces are ubiquitous soil-dwelling Actinobacteria and constitute a prominent component of plant-associated microbial communities [20,21,22]. Beyond their well-established role as prolific producers of secondary metabolites, Streptomyces frequently colonize the rhizosphere [23,24,25] and, in many cases, the root endosphere [26,27] of diverse host plants. In our previous work, we identified Streptomyces as an evolutionarily conserved component of the tomato root microbiome, suggesting a long-standing association with potential functional relevance for the host [22]. Consistent with this, numerous studies have reported plant growth-promoting effects of individual Streptomyces species, involving mechanisms such as phosphate solubilization [28,29,30,31,32], siderophore production [29,33,34,35] and modulation of the root system architecture [36,37,38].

In addition to growth promotion, several Streptomyces protect the host against pathogens either by direct antagonism [24,26,31] or indirectly by inducing local [39,40,41] and systemic defense responses [42,43,44]. Notably, a recent study in Arabidopsis showed that the rhizosphere-associated Streptomyces strain AgN23 induces the accumulation of the phytoalexin camalexin, which is required for successful rhizosphere colonization, underscoring that the host defense-associated metabolism can play an active role in beneficial plant–Streptomyces interactions [25]. However, despite extensive evidence for these functional traits at the level of individual strains, how host plants respond transcriptionally to colonization by individual Streptomyces strains or Streptomyces-rich microbial consortia remain poorly characterized.

Tomato (Solanum lycopersicum L.) is an economically important crop that is threatened by several soil-borne pathogens. Understanding how tomato plants transcriptionally respond to colonization by growth-promoting Streptomyces is therefore of both fundamental and applied relevance. In this study, we isolated Streptomyces from the tomato rhizosphere and assembled a defined consortium, which significantly promoted tomato growth. Using RNA sequencing, we characterized local (root) and systemic (leaf) transcriptional responses induced upon root bacterization with this consortium. In parallel, we assessed the antagonistic activity of individual isolates against the pathogenic fungus Fusarium oxysporum f. sp. radicis-lycopersici (Foxrl), the causal agent of tomato crown and root rot, and investigated the genomic basis of antifungal potential through whole-genome sequencing and biosynthetic gene cluster mining. Our study provides an integrated view of how tomato-associated Streptomyces affect host transcriptional programs and highlights their potential for sustainable crop improvement and biological control.

2. Materials and Methods

2.1. Isolation of Streptomyces from the Tomato Rhizosphere

To isolate Streptomyces strains from the tomato rhizosphere, uniform 12-day-old seedlings (cv. Moneymaker) were transplanted into 2 L plastic pots containing a soil mixture composed of four parts sieved agricultural soil collected from a fallow field, two parts commercial potting soil (Potgrond P; Klasmann–Deilmann GmbH, Geeste, Germany), two parts sieved sand from the Evros (Maritsa) River, and one part perlite. Plants were cultivated in a greenhouse under controlled conditions. The natural photoperiod was approximately 12 h light/12 h dark, with mean day/night temperatures of 26–28 °C and 20–22 °C, respectively. Plants were irrigated as needed and fertilized once per week with a modified half-strength Hoagland nutrient solution containing 2.5 mM Ca(NO_3_)2, 2.5 mM KNO_3_, 1 mM MgSO_4_, 0.5 mM KH_2_PO_4_, micronutrients (25 µM B, 5 µM Mn, 0.5 µM Zn, 0.5 µM Cu, 0.25 µM Mo), and 25 µM Fe as FeNaEDTA [45]. The pH was adjusted to 6.0 with KOH.

Four weeks after transplantation, rhizosphere soil was collected from three independent biological replicates, each consisting of pooled soil tightly adhering to the roots of five tomato plants. For each replicate, 10 g of soil was transferred into sterile glass containers containing 90 mL of sterile 10 mM MgCl_2_ solution and homogenized by vigorous shaking at 200 rpm for 1 h at room temperature. The resulting suspensions were serially diluted (10^−1^) and aliquots were plated in triplicate onto ISP4 agar medium (Inorganic Salt Starch Agar), suitable for Streptomyces cultivation [46]. Plates were incubated at 28 °C and monitored for up to 5 days. The ISP4 medium contained (per liter): soluble starch (10 g), K_2_HPO_4_ (1 g), MgSO_4_ × 7H_2_O (1 g), NaCl (1 g), (NH_4_)2_SO_4 (2 g), CaCO_3_ (2 g), FeSO_4_ × 7H_2_O (0.001 g), MnCl_2_ × 4H_2_O (0.001 g), ZnSO_4_ × 7H_2_O (0.001 g), and agar (20 g). The final pH was adjusted to 7.2 ± 0.2 at 25 °C. To suppress fungal growth, cycloheximide was added at a final concentration of 100 µg mL^−1^. Colonies displaying Streptomyces-like morphology were selected and purified by repeated streaking on fresh ISP4 medium to obtain pure cultures.

2.2. 16S rRNA Gene Sequencing and Phylogenetic Analysis of Rhizosphere Bacterial Isolates

Genomic DNA was extracted from purified bacterial isolates using the Quick-DNA™ Fungal/Bacterial Miniprep Kit D6005 (Zymo Research, Irvine, CA, USA), following the manufacturer’s instructions. DNA concentration and purity were assessed using a NanoPhotometer™ P–Class P330 (IMPLEN GmbH, Munich, Germany). Amplification of the nearly full-length 16S rRNA gene was performed by polymerase chain reaction (PCR) in a final reaction volume of 20 μL using the universal bacterial primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-TACGGYTACCTTGTTACGACTT-3′) [47]. Each 20 μL PCR contained 25–50 ng template DNA, 0.2 mM of each dNTP, 0.5 μM of each primer, 1× DreamTaq Buffer, 5 U DreamTaq™ Hot Start DNA Polymerase, and nuclease-free water to volume. All PCR reagents were purchased from Thermo Fisher Scientific (Waltham, MA, USA). PCR amplification was carried out under the following conditions: initial denaturation at 95 °C for 3 min; 30 cycles of denaturation at 94 °C for 30 s, annealing at 58 °C for 45 s, and extension at 72 °C for 1 min; followed by a final extension step at 72 °C for 15 min. PCR products were purified using ExoSAP-IT™ Express (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions and subjected to bidirectional Sanger sequencing by Cemia (Larissa, Greece) using the same primers employed for PCR amplification. Forward and reverse sequences were quality-trimmed, aligned pairwise, and assembled into consensus sequences using BioEdit (v7.2.5).

Consensus 16S rRNA sequences were aligned using MAFFT (v7) [48] and subsequently curated by removing ambiguously aligned or low-information regions with BMGE (v1.12) [49], as implemented in the NGPhylogeny.fr web platform (https://ngphylogeny.fr/) [50]. The curated alignment was used to infer a maximum likelihood phylogeny using PhyML (v3.3) [51] under the GTR + Γ + I nucleotide substitution model, and the resulting phylogenetic tree was visualized using iTOL (https://itol.embl.de/) [52]. Based on clustering patterns in the phylogenetic tree, a single representative isolate per sequence cluster was retained, resulting in a final set of 12 non-redundant Streptomyces isolates. This collection, comprising isolates TOM–1 to TOM-12, is hereafter referred to as the Streptomyces TOM collection. The 16S rRNA consensus sequences of the TOM collection were compared against type strain sequences in the EzBioCloud database [53] to identify the closest phylogenetic relatives based on percentage sequence identity.

2.3. Spot-on-Lawn Compatibility Assays

To examine interactions among the TOM isolates, pairwise in vitro assays were performed using a modified spot-on-lawn confrontation assay described by Molina–Romero et al. [54]. Briefly, all twelve Streptomyces isolates were cultured to sporulation on ISP4 agar, and spore suspensions were prepared in sterile 10 mM MgCl_2_ to a final concentration of 1 × 10^7^ spores mL^−1^. For each assay, 150 µL of a single isolate was evenly spread onto ISP4 agar plates to establish a uniform lawn. Four different isolates were then spotted (15 µL each) at equidistant positions on the lawn surface. Plates were incubated at 28 °C and examined after 5 days. Interactions were classified as neutral (no evident alteration at the interaction interface), positive (enhanced peripheral growth at the interaction interface), or antagonistic (inhibition of lawn growth surrounding the spotted colony). Each interaction was assessed in two technical replicates.

2.4. Plant Growth Promotion Assays and Sample Collection for Transcriptomic Analysis

Surface-sterilized tomato seeds (cv. Moneymaker) were germinated and grown for 12 days in steam-sterilized propagation plugs (Eazy Plug^®^, Goirle, The Netherlands). The 12 non-redundant Streptomyces isolates selected based on 16S rRNA gene sequence analysis were cultured individually on ISP4 agar plates, and spores were harvested 12 days after plate inoculation and suspended in sterile 10 mM MgCl_2_.

A defined Streptomyces consortium composed of the TOM-1 to TOM-12 isolates (hereafter referred to as the Streptomyces TOM consortium) was prepared in sterile 10 mM MgCl_2_ by combining individual spore suspensions in appropriate volumes to obtain a mixed inoculum in which each isolate was present at a final concentration of 10^5^ spores mL^−1^ (Cons-H). This stock inoculum was subsequently diluted 10-fold and 100-fold to generate treatments in which each isolate was present at final concentrations of 10^4^ (Cons-M) or 10^3^ (Cons-L) spores mL^−1^, respectively.

Prior to transplantation, seedling plugs were immersed in the corresponding Streptomyces TOM inoculum. Control seedlings were treated identically but immersed in sterile 10 mM MgCl_2_. Seedlings were then transplanted into a steam-sterilized substrate consisting of perlite and vermiculite (1:1, v/v) and cultivated under greenhouse conditions as described above. Plant growth was assessed three weeks after transplantation by measuring fresh shoot and root weight. For each treatment, measurements were obtained from 10 individual plants (n = 10). Statistical analyses were performed in R (v4.4.2) using one-way ANOVA followed by Tukey’s HSD test (p < 0.05). Boxplots were generated in R using the ggplot2 package (v4.0.0).

2.5. RNA Extraction, Library Preparation, Sequencing, and RNA-Seq Analysis

For transcriptomic analyses, an independent experiment was conducted using the same experimental setup as the plant growth promotion assays, including seedling preparation, root bacterization, growth conditions, and harvest timing. Leaf and root tissues were collected three weeks after transplantation from plants treated with the highest consortium inoculum (Cons-H). Four biological replicates per treatment (n = 4) were generated, each consisting of pooled tissue from five individual plants. Samples were immediately flash-frozen in liquid nitrogen and stored at −80 °C until RNA extraction. To assess the potential presence of Streptomyces in leaves, additional leaf samples were collected, briefly rinsed in sterile distilled water, and homogenized in sterile 10 mM MgCl_2_. Aliquots were plated onto ISP4 agar medium supplemented with 100 µg mL^−1^ cycloheximide and incubated at 28 °C for 5 days.

Total RNA was extracted from frozen leaf and root tissues using the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA quality control, RNA-seq library construction, and sequencing were performed by a commercial sequencing service provider (BMKGENE, Münster, Germany). RNA quality was assessed using NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), Qubit 2.0 fluorometer (Thermo Fisher Scientific, Waltham, MA, USA), and an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Briefly, poly(A)+ mRNA was enriched from total RNA using oligo(dT)-coupled magnetic beads and fragmented prior to cDNA synthesis. First-strand cDNA was synthesized using random hexamer primers, followed by second-strand cDNA synthesis. Double-stranded cDNA was purified, end-repaired, A-tailed, and ligated to sequencing adapters. Libraries were size-selected to enrich fragments of approximately 300–400 bp and amplified by PCR. Paired-end sequencing (150 bp) was performed on an Illumina platform.

Samples were assigned to four experimental groups: leaf control (LfC1–LfC4), leaf treatment (LfT1–LfT4), root control (RtC1–RtC4), and root treatment (RtT1–RtT4). Raw sequencing reads were quality-filtered to remove adapter sequences, low-quality reads, and reads containing ambiguous bases, generating clean reads for downstream analyses. Clean reads were aligned to the tomato reference genome (Solanum lycopersicum ITAG4.0) using HISAT2 [55]. Transcript assembly and gene expression quantification were performed using StringTie [56,57] and expression levels were reported as fragments per kilobase of transcript per million mapped reads (FPKM).

Differential gene expression analysis was conducted using DESeq2 [58], taking biological replicates into account. Genes were considered differentially expressed when showing an absolute fold change ≥ 2 and a false discovery rate (FDR) < 0.05. Differential expression analyses were performed separately for leaf (LfC vs. LfT) and root (RtC vs. RtT) tissues. Functional interpretation of differentially expressed genes was carried out using Gene Ontology (GO) [59] enrichment analyses.

2.6. Greenhouse Biocontrol Assays

The greenhouse biocontrol assays were conducted against Foxrl. The Foxrl isolate used in this study was kindly provided by Prof. G. Karaoglanidis (Laboratory of Plant Pathology, Aristotle University of Thessaloniki) [60]. Tomato seedlings of the Foxrl-susceptible cultivar ACE55 were produced under the same conditions described for the growth-promotion assays. Prior to transplantation, root plugs were immersed for 10 min in the Streptomyces TOM consortium at the Cons-H concentration. Control plants were immersed in sterile 10 mM MgCl_2_. A conidial suspension of Foxrl was obtained from 7-day-old liquid cultures grown in half-strength potato dextrose broth (PDB) at 25 °C. Cultures were filtered through four layers of cheesecloth to remove mycelial fragments, and spore concentration was determined using a hemocytometer. Immediately after treatment, seedlings were transplanted individually into plastic pots containing a substrate consisting of perlite and vermiculite (1:1, v/v) artificially infested with Foxrl at a final concentration of 10^6^ spores g^−1^ substrate. Twenty plants were included per treatment (n = 20). Plants were maintained in a greenhouse under conditions similar to those described for the growth-promotion experiments.

Disease severity was assessed four weeks after transplantation using the four-class disease severity scale described by Taiebikhah et al. [61]: I = no signs of rot or stunting; II = mild rot at the collar or roots without stunting; III = significant rot at the collar and roots without stunting; IV = extensive rot at the collar and roots accompanied by severe stunting. Statistical differences between treatments were evaluated in R using the Mann–Whitney U test (two-tailed), and plots were generated using the ggplot2 package. In addition, the disease index percentage (DI%) was calculated as described by Myresiotis et al. [60]. Briefly, severity classes I–IV were converted to numerical values 0–3, respectively, and DI% was calculated as: DI% = [(0 × a) + (1 × b) + (2 × c) + (3 × d)]/(3 × n) × 100, where a–d represent the number of plants in severity classes I–IV and n is the total number of evaluated plants (n = a + b + c + d). Protection percentage was calculated as described by Taiebikhah et al. [61]: Protection (%) = (A − B)/A × 100, where A corresponds to the DI% of the infected control and B to that of the treated plants.

2.7. In Vitro Antifungal Activity Assay

The 12 Streptomyces isolates of the TOM collection were cultured individually on ISP4 agar plates and incubated at 28 °C for 12 days to allow sporulation. Foxrl was cultured on potato dextrose agar (PDA) plates at 25 °C for 5 days. Streptomyces spores were aseptically harvested from ISP4 plates, suspended in sterile 10 mM MgCl_2_, and the spore concentration was adjusted to 1 × 10^8^ spores mL^−1^. Antagonistic activity was evaluated using a dual culture assay on Mueller–Hinton agar plates (90 mm diameter), in which 20 μL of the Streptomyces spore suspension was spotted at one edge of the plate, while a 5 mm agar plug containing actively growing Foxrl mycelium was placed at the opposite edge of the same plate. Control plates were inoculated with Foxrl alone.

Fungal growth inhibition was quantified after 8 days of co-culture as the percentage reduction in radial growth relative to the control, using the following formula [62]: Inhibition (%) = (Rc − Rt)/Rc × 100, where Rc is the radial growth of Foxrl on control plates and Rt is the radial growth of Foxrl toward the Streptomyces inoculum. Statistical analyses were performed in R (v4.4.2) using one-way ANOVA followed by Tukey’s HSD test (p < 0.05). Boxplots were generated in R using the ggplot2 package.

2.8. Whole-Genome Sequencing, Assembly, and Annotation of Selected Streptomyces Isolates

Genomic DNA was extracted from selected Streptomyces isolates using the ZymoBIOMICS™ DNA Miniprep Kit D4300 (Zymo Research, Irvine, CA, USA) following the manufacturer’s instructions. DNA concentration and purity were initially assessed using a NanoPhotometer™ P-Class P330 (IMPLEN), and DNA integrity was evaluated by agarose gel electrophoresis prior to sample submission for sequencing.

For long-read sequencing, libraries were prepared using the Oxford Nanopore Native Barcoding Expansion kit according to the manufacturer’s recommendations. Library quality control was performed by the sequencing provider using a Qubit 3.0 fluorometer (Thermo Fisher Scientific, Waltham, MA, USA), and sequencing was carried out on an Oxford Nanopore platform. For short-read sequencing, libraries were prepared using the VAHTS Universal Plus DNA Library Prep Kit V4 (Vazyme, Nanjing, China). Library quality and fragment size distribution were assessed by the sequencing provider using a Qubit 3.0 fluorometer and a QSep-400 system (BiOptic, Taipei, China), respectively. Paired-end sequencing (150 bp) was performed on an Illumina NovaSeq platform.

De novo genome assemblies were generated from Oxford Nanopore long reads using Flye (v2.9) [63]. Raw Nanopore reads were mapped to the draft assemblies using minimap2 (v2.24) [64], and consensus polishing was performed with three iterative rounds of Racon [65], followed by an additional polishing step using Medaka (v1.7.2) to correct Nanopore-specific sequencing errors. Illumina paired-end reads were subsequently mapped to the Nanopore-polished assemblies using BWA-MEM2 (v2.2.1) [66,67], and two successive rounds of polishing were performed using Pilon (v1.24) [68] to correct residual base substitutions and small insertion–deletion errors. Assembly quality was assessed using QUAST (v5.2.0) [69] to obtain standard assembly metrics, including genome size and N50 values. Assembly graphs were visualized using Bandage (v0.8.1) [70] to evaluate structural continuity. Read coverage across the assembled genomes was calculated using SAMtools (v1.17) [71]. Genome-based taxonomic assignments were inferred using the Type (Strain) Genome Server (TYGS) (https://tygs.dsmz.de/) [72]. Biosynthetic gene clusters were predicted using antiSMASH (v7.0) [73].

3. Results

3.1. Recovery and Characterization of Streptomyces from the Tomato Rhizosphere

To recover Streptomyces from the tomato rhizosphere, soil tightly adhering to tomato roots was collected and plated on ISP4 medium. After incubation, colonies displaying typical Streptomyces morphology were selected and purified by repeated streaking. To estimate genetic redundancy within the isolate collection, the nearly full-length 16S rRNA gene was amplified and sequenced for all isolates. The consensus sequences were aligned and curated using MAFFT and BMGE, respectively, and sequence similarity-based clustering was used to identify groups of closely related isolates sharing identical 16S rRNA sequences. One representative isolate per sequence cluster was retained, resulting in a final set of 12 Streptomyces isolates (Figure 1A) with distinct 16S rRNA gene sequences (Figure 1B). These isolates were designated TOM-1 to TOM-12; the resulting collection is hereafter referred to as the Streptomyces TOM collection.

Taxonomic affiliation was inferred based on 16S rRNA sequence similarity using the EzBioCloud database. All TOM isolates were assigned to the genus Streptomyces, showing highest sequence similarity to described type strains (Table S1). Due to the limited discriminatory power of the 16S rRNA gene within the genus, species-level affiliations were regarded as tentative.

3.2. The Streptomyces TOM Consortium Promotes Tomato Growth

A defined Streptomyces TOM consortium, composed of isolates from the Streptomyces TOM collection, was evaluated for its ability to promote tomato growth. Spot-on-lawn assays revealed predominantly neutral and occasionally positive interactions among consortium members, whereas antagonistic effects were observed only in a limited number of specific combinations (Figure S1). No isolate exhibited broad inhibitory activity, nor was any consistently inhibited by the majority of consortium members, indicating the absence of pervasive competitive exclusion under the tested conditions. Before transplantation, tomato seedlings were root-treated with this consortium at three different inoculum levels, in which each TOM isolate was present at final concentrations of 10^3^ (Cons-L), 10^4^ (Cons-M), or 10^5^ (Cons-H) spores mL^−1^. Plant growth was assessed three weeks after transplantation. One-way ANOVA revealed a significant effect of consortium treatment on both shoot and root fresh weight (p < 0.001). Treatment with the lowest consortium concentration (Cons-L) did not result in detectable changes in shoot or root biomass compared to control plants. In contrast, seedlings treated with the Streptomyces TOM consortium at Cons-M and Cons-H exhibited increased plant growth (Figure 2A). Shoot fresh weight increased by approximately 33% and 41% in the Cons-M and Cons-H treatments, respectively, relative to control plants (Figure 2B). Similarly, root fresh weight increased by approximately 27% and 38% at the corresponding inoculum levels (Figure 2C).

3.3. Local and Systemic Tissues of Tomato Respond to Root Colonization by the Streptomyces TOM Consortium by Altering Gene Expression

To investigate local and systemic transcriptional responses of tomato plants to root colonization by the Streptomyces TOM consortium, RNA sequencing was performed on root (local) and leaf (systemic) tissues collected three weeks after treatment with the highest consortium concentration (Cons-H). To assess whether Streptomyces colonization extended to systemic tissues under the experimental conditions employed, additional leaf samples were collected and plated on ISP4 medium. No Streptomyces-like colonies were recovered after 5 days of incubation. Principal component analysis (PCA) separated root and leaf transcriptomes and distinguished treated and non-treated samples. Biological replicates clustered consistently within each tissue and treatment group, indicating good reproducibility of the dataset (Figure 3A).

Differential expression analysis using DESeq2 (|fold change| ≥ 2, FDR < 0.05) identified 262 differentially expressed genes (DEGs) in roots, including 86 up-regulated and 176 down-regulated genes (Figure 3B,C). In leaves, a broader transcriptional response was observed, with 458 DEGs detected, of which 134 were up-regulated and 324 were down-regulated (Figure 3B,D). Complete lists of DEGs identified in roots and leaves are provided in Tables S2 and S3. Twenty DEGs were shared between roots and leaves, displaying the same direction of regulation, with the exception of one gene (Table S4). Overall, root colonization by the Streptomyces TOM consortium elicited distinct transcriptional responses in roots and leaves, characterized by a predominance of down-regulated genes in both tissues.

3.3.1. The Transcriptome of Streptomyces-Colonized Roots Reveals Selective Attenuation of Defense-Related Pathways and Modulation of Nutrient-Associated Responses

In roots, Gene Ontology (GO) enrichment analysis of up-regulated genes revealed significant enrichment of biological processes primarily associated with phytoalexin biosynthesis and metabolism (Figure 4A). Although not captured by the GO enrichment analysis, several up-regulated genes in Streptomyces-colonized roots are characteristic of phosphate starvation and phosphate homeostasis responses. These include the induction of the tomato phosphate starvation-induced gene TPSI1 (Solyc10g085850.1), a well-established marker of Pi deficiency, as well as genes encoding purple acid phosphatases (PAPs; e.g., Solyc03g098010.3), enzymes implicated in organic phosphate mobilization under low-phosphate conditions [74,75]. In addition, multiple SPX domain-containing regulators (Solyc01g090890.3, Solyc01g091870.3), which function as key sensors and modulators of phosphate starvation signaling [76,77], were up-regulated. Finally, induction of a putative inorganic phosphate transporter (Solyc09g066410.3), belonging to the PHT1 family [78,79], further supports activation of phosphate acquisition pathways typically engaged under Pi–limiting conditions (Table S2). In addition, the up-regulated root DEGs included genes involved in nutrient transport, such as members of the NPF/NRT1–PTR family [80], a potassium transporter [81] and an amino acid permease [82], suggesting coordinated adjustments in nutrient uptake and redistribution (Table S2). Genes associated with hormone-related regulation were also induced in response to Streptomyces colonization, including GH3.6 (Solyc12g005310.2), which is involved in auxin conjugation and homeostasis [83], as well as cytokinin-related genes encoding cytokinin hydroxylase-like and cytokinin dehydrogenase activities [84] (Table S2).

Gene Ontology (GO) enrichment analysis of down-regulated genes in roots revealed a repression of defense- and cell wall-associated biological processes upon colonization by the Streptomyces TOM consortium. In particular, enriched GO terms included cell wall organization, cell wall biogenesis, xyloglucan metabolic processes, and jasmonic acid-mediated signaling, together indicating attenuation of structural defense reinforcement and hormone-associated immune responses in colonized roots (Figure 4B). Beyond these GO-level trends, a gene-centered inspection of the down-regulated root transcriptome revealed an additional transcriptional signal related to iron homeostasis that was not apparent from enriched GO categories. Specifically, key components of the iron uptake machinery, including a ferric-chelate reductase and the iron-regulated transporter IRT1 [85], were significantly repressed (Table S2), indicating attenuation of the iron-deficiency response in colonized roots.

3.3.2. Systemic Leaf Transcriptional Responses Reveal Attenuation of Stress-Related Pathways Alongside Growth-Promoting Signatures

In leaves, Gene Ontology (GO) enrichment analysis of up-regulated genes revealed a systemic transcriptional response that is predominantly nutrient- and metabolism-centered (Figure 4C). Enriched biological processes were strongly associated with phosphate starvation and phosphate homeostasis, including cellular response to phosphate starvation and phosphate ion transport, together with extensive reprogramming of carbon and energy metabolism, as reflected by enrichment of carbohydrate metabolic process, glucose import, tricarboxylic acid cycle, and glyoxylate cycle. Interestingly, several of the up-regulated DEGs in leaves involved in phosphate homeostasis were also found among the common DEGs shared between roots and shoots (Table S4), suggesting a whole-plant adjustment of phosphate status. In parallel, several GO terms related to lipid metabolism and lipid-mediated signaling—such as lipid metabolic process, inositol lipid-mediated signaling, and phosphatidic acid biosynthetic process—were enriched, indicating broad metabolic and signaling adjustments in aerial tissues. This functional profile was further complemented by enrichment of growth- and development-related processes, including basipetal auxin transport, positive regulation of organ growth, multidimensional cell growth, and positive regulation of cell population proliferation, suggesting that the systemic response extends beyond stress adaptation toward coordinated regulation of growth.

Conversely, GO enrichment analysis of down-regulated genes in leaves revealed a pronounced repression of cell wall remodeling and defense-related biological processes (Figure 4D). Strongly enriched down-regulated terms included cell wall organization, cell wall biogenesis, xyloglucan metabolic process, pectin catabolic process, and cellular glucan metabolic process, indicating attenuation of cell wall restructuring activities. In addition, multiple GO categories associated with defense and stress responses were suppressed, including response to wounding, response to bacterium, defense response to bacterium, response to external biotic stimulus, and defense response to other organisms. Down-regulation of the oxylipin biosynthetic process further supported repression of typical defense-associated signaling pathways. Together, these patterns indicate a transcriptional shift in leaves characterized by reduced expression of cell wall- and defense-related programs relative to nutrient- and metabolism-associated processes.

Analysis of the common DEGs shared between roots and shoots revealed that several up-regulated genes in both organs are associated with phosphate homeostasis and phosphate-related signaling, including a purple acid phosphatase, an SPX domain-containing protein, an inorganic phosphate transporter and an inorganic pyrophosphatase, suggesting a coordinated whole-plant adjustment of phosphate status. Conversely, the down-regulated genes were mainly related to cell wall-associated processes, including lignin-forming peroxidases and a xyloglucan endotransglucosylase (Table S4).

3.4. The TOM Consortium Reduces Disease Severity Caused by Foxrl

To determine whether the growth-promoting TOM consortium also confers protection against soil-borne pathogens, its efficacy was evaluated in a tomato–Foxrl pathosystem under controlled greenhouse conditions. To this end, tomato seedlings of the susceptible cultivar ACE55 were root-treated with the TOM consortium at the Cons-H concentration, while control seedlings were treated with sterile MgCl_2_. Treated and control plants were subsequently transplanted into Foxrl-infested substrate, and disease severity was assessed four weeks later using a four-class disease severity scale (I–IV). By the end of the experiment, control plants predominantly exhibited severe symptoms corresponding to classes III and IV, whereas TOM-treated plants were mainly classified within classes I and II. This shift in severity distribution resulted in significantly lower disease scores in treated plants compared with the infected control (Mann–Whitney U test, p < 0.001). Accordingly, the disease index percentage (DI%) decreased from 81.7% in the infected control to 33.3% in the TOM-treated plants, corresponding to 59.2% protection (Figure 5A).

To examine antifungal traits among consortium members, individual TOM isolates were tested against Foxrl in a dual culture assay. Of the TOM isolates, the TOM-2, TOM-3, TOM-4, TOM-9 and TOM-12 exhibited measurable inhibitory effects on fungal radial growth. The strongest suppression of Foxrl growth was observed for isolates TOM-4 and TOM-12, which reduced fungal growth in vitro by approximately 40% and 37%, respectively (Figure 5Β).

3.5. Genome Mining of Antagonistic TOM Isolates Highlights Biosynthetic Features Linked to Antifungal Potential

To further explore the genetic basis underlying the in vitro antagonistic activity observed in the dual culture assay (Figure 5), isolates TOM-2, TOM-3, TOM-4, TOM-9, and TOM-12 were selected for whole-genome sequencing (WGS) and downstream genetic analyses. Genome assemblies were generated from Oxford Nanopore long reads and subsequently polished using Illumina short-read data, resulting in high-quality hybrid assemblies for all five isolates. Final assemblies consisted of two contigs per genome, with total genome sizes ranging from 7.1 to 10.8 Mb and GC contents between 71.1% and 72.6%, values consistent with typical Streptomyces genomes. Assembly continuity was high for all isolates, with N50 values corresponding to the largest contig in each assembly and no ambiguous bases detected (Table 1). Detailed assembly metrics and sequencing statistics are provided in Table S5.

Genome-based taxonomic assignment using the Type (Strain) Genome Server (TYGS) revealed heterogeneous phylogenetic relationships among the five sequenced isolates. Two isolates (TOM-2 and TOM-3) showed digital DNA–DNA hybridization (dDDH) values exceeding the accepted 70% species delineation threshold, allowing their assignment to Streptomyces geysiriensis and Streptomyces rubrogriseus, respectively. Similarly, isolate TOM-12 clustered with Streptomyces melanosporofaciens, displaying dDDH values above 90%. In contrast, isolates TOM-4 and TOM-9 exhibited substantially lower dDDH values (<45% and <20%, respectively) relative to all available type strains, despite GC contents consistent with placement within the genus Streptomyces. These results indicate that TOM-4 and TOM-9 cannot be confidently assigned to any currently described species and likely represent distinct, previously uncharacterized Streptomyces lineages.

To explore the genetic basis underlying the antifungal activity of the selected isolates, biosynthetic gene clusters (BGCs) were predicted using antiSMASH. All five genomes encoded a diverse repertoire of secondary metabolite BGCs. The total number of predicted BGCs varied among isolates, ranging from 26 clusters in TOM-3, TOM-4, and TOM-9, to 33 clusters in TOM-2, and up to 51 clusters in TOM-12 (Figure 6). The predicted clusters spanned multiple biosynthetic classes, such as nonribosomal peptide synthetases (NRPS), type I polyketide synthases (PKS), hybrid NRPS–PKS clusters, terpenes, siderophores, ribosomally synthesized and post-translationally modified peptides (RiPPs). Several predicted BGCs across all five genomes were annotated as encoding metabolites previously associated with antimicrobial or antifungal activity, with confidence levels ranging from low to high according to antiSMASH classification (Table 2). A comprehensive annotation of all predicted BGCs is provided in Table S6. Notably, many predicted BGCs showed low similarity to characterized clusters in the MIBiG database, indicating the presence of potentially unexplored biosynthetic pathways. A comprehensive annotation of all predicted BGCs is provided in Table S6. Collectively, the combination of pronounced in vitro antifungal activity and extensive, partially uncharacterized biosynthetic potential provides a plausible genetic basis for the observed suppression of Foxrl.

4. Discussion

4.1. Growth Promotion by a Defined Streptomyces Consortium Coincides with Transcriptional Reprogramming in Local and Systemic Tissues

In this study, we investigated the effects of a defined consortium of tomato-associated Streptomyces on plant growth, host gene expression, and suppression of a major soil-borne pathogen. Rather than focusing on individual strains, we deliberately assessed the consortium as a defined multi-strain inoculum. This choice was motivated by growing evidence that, even within a single bacterial genus, cooperative or complementary interactions among closely related strains can be required to achieve consistent plant growth-promoting effects. For Streptomyces in particular, several studies have shown that combinations of strains can promote plant growth more effectively than the corresponding isolates applied individually, indicating emergent functional properties at the consortium level [86,87]. Similar principles have been demonstrated in mono-genus synthetic communities of Bacillus, where social interactions and strain compatibility within consortia strongly enhance plant growth and functional output compared to single-strain inoculants [17]. Building on this conceptual framework, we evaluated a defined Streptomyces consortium composed of phylogenetically distinct tomato-associated isolates, enabling us to capture collective effects on plant performance and host transcriptional responses that may not be apparent in single-strain experiments.

It is important to note, however, that we did not assess the relative representation of individual consortium members in the rhizosphere or root compartment at the end of the experimental period, as this would require the development of isolate-specific genomic markers for all consortium members. Consequently, the extent to which individual isolates contributed to the observed plant phenotypes cannot be resolved from the present data. Nevertheless, our in vitro compatibility assays revealed predominantly neutral interactions, with antagonistic effects confined to a limited number of specific combinations. Such localized inhibition does not preclude overall compatibility, as consortium design studies commonly retain strains exhibiting limited pairwise antagonism provided they remain compatible with most community members [54]. Collectively, these findings argue against pervasive mutual exclusion under the tested conditions and support their co-application as a largely compatible multi-member inoculum.

Growth promotion by the Streptomyces TOM consortium was observed only above a defined inoculum threshold, whereas lower spore densities did not result in detectable effects on tomato shoot or root biomass. Such threshold-dependent responses are consistent with the broader concept that beneficial root-associated microbial functions, including plant growth promotion and pathogen suppression, require sufficient microbial establishment and population density in the rhizosphere to achieve effective signaling and cumulative metabolic output [88,89,90]. In the context of a multi-member consortium, this behavior may further reflect the need for adequate population sizes to enable functional complementation or synergistic interactions among consortium members.

Beyond the phenotypic effects on plant growth, root colonization by the Streptomyces TOM consortium was associated with transcriptional reprogramming in both local (roots) and systemic (leaves) tissues three weeks after inoculation, indicating a stable host response rather than a transient reaction to microbial exposure. Notably, the dominant transcriptional signature in both organs was characterized by a prevalence of down-regulated genes, particularly those associated with defense- and cell wall-related processes. This can be interpreted as a host response that facilitates microbial accommodation and sustained colonization. In support of this interpretation, transcriptomic analyses in Arabidopsis have revealed distinct host transcriptional responses to beneficial root-associated microbes that differ from canonical immune activation and are consistent with host adaptation to compatible microbial colonization [5]. Consistent with this interpretation, attenuation of defense-associated gene expression may reflect regulatory trade-offs between growth, metabolism and immunity [91,92]. Alternatively, it is possible that Streptomyces-derived signals actively modulate host immune gene expression, as described for other beneficial plant–microbe interactions in which microbial colonization is accompanied by immune modulation and defense reprogramming [93,94].

Regardless of the underlying mechanism, attenuation on defense investment could, in principle, entail increased vulnerability under pathogen pressure. However, in roots, the Streptomyces TOM consortium acted protectively against Foxrl. This indicates that local immune attenuation does not necessarily compromise root protection, as the protective role may be fulfilled by the consortium itself through direct microbial antagonism. Whether similar protection extends to other soil-borne pathogens remains to be determined. In addition to direct antagonism, the induction of camalexin biosynthesis observed in Streptomyces-colonized roots suggests that, in local tissues, biocontrol may also involve targeted activation of host defense pathways. The situation in leaves is more ambiguous, as the consortium was not detected in aerial tissues. It is therefore possible that systemic defense attenuation may represent a growth–defense trade-off. However, the possibility that certain defense responses are primed in systemic tissues cannot be excluded, and this will require targeted infection assays with foliar pathogens representing contrasting lifestyles to determine whether induced systemic resistance (ISR) [95,96] is established in response to the Streptomyces TOM consortium.

4.2. Local Root Responses Suggest Modulation of Immunity, Nutrient Homeostasis, and Hormonal Homeostasis

At the local level in roots, and consistent with the broader transcriptional trends discussed above, Streptomyces colonization was associated with repression of cell wall- and jasmonic acid-related defense pathways, while genes involved in phytoalexin biosynthesis were up-regulated. This pattern indicates that immune modulation is selective rather than reflecting a global suppression of defense-related transcription. Notably, the enrichment of phytoalexin biosynthetic processes detected in our GO term analyses is consistent with recent findings in Arabidopsis thaliana, where colonization by the rhizosphere-associated Streptomyces strain AgN23 induces accumulation of the phytoalexin camalexin, a response shown to be required for successful bacterial colonization of the rhizosphere [25]. Together, these results support the concept that activation of defense-associated metabolic pathways can form part of a reciprocal communication process during beneficial plant–microbe interactions, rather than representing a strictly antagonistic response.

In parallel, Streptomyces-colonized roots exhibited transcriptional signatures indicative of altered nutrient homeostasis. Notably, genes associated with phosphate deficiency responses were induced, whereas components of the iron deficiency response were repressed. Induction of phosphate starvation-responsive genes may reflect competition between the consortium and the host for available phosphate or consortium-driven alterations in phosphate availability in the rhizosphere. Since components of the phosphate deficiency response were also activated in leaves, the observed transcriptional changes may reflect a systemic phosphate demand signal originating from the shoot. Alternatively, these responses may indicate a rewiring of host phosphate regulatory pathways that ultimately favor phosphate acquisition. Comparable modulation of nutrient-related host responses has been reported in the interaction of Arabidopsis with beneficial pseudomonads, where microbial activity enhances iron nutrition through activation of Strategy I responses [97,98]. Importantly, modulation of the phosphate deficiency response may contribute to host colonization by Streptomyces, as such responses have been shown to intersect with immune regulation during beneficial plant–microbe interactions [99].

In contrast, repression of iron deficiency-associated genes in Streptomyces-colonized roots may reflect modifications in rhizosphere iron dynamics mediated by Streptomyces siderophores. One possible explanation is that siderophore-mediated iron mobilization increases iron availability to the host, thereby reducing the requirement for host-driven Strategy I uptake mechanisms [100,101]. However, alternative, non-mutually exclusive scenarios should also be considered. For example, bacterial siderophores may sequester iron in forms that modify rhizosphere iron chemistry and influence host perception, or repression of iron uptake genes may represent a host-mediated regulatory adjustment in response to microbial colonization. Genome mining of the five sequenced TOM strains revealed multiple biosynthetic gene clusters predicted to encode siderophores, supporting the involvement of siderophore-related processes. Nevertheless, because iron concentrations were not quantified in plant tissues or in the rhizosphere, we cannot discriminate among increased host iron availability, microbial iron sequestration, or host-driven down-regulation of iron uptake as part of a colonization program.

Finally, local transcriptional changes affecting auxin- and cytokinin-associated pathways suggest adjustments in hormonal homeostasis that may contribute to the enhanced root growth observed at the phenotypic level. Collectively, our data indicate that Streptomyces colonization elicits coordinated local transcriptional reprogramming in roots, integrating selective immune reprogramming, nutrient-related transcriptional adjustments, and hormone-associated regulatory changes, thereby supporting both plant growth and effective root colonization.

4.3. Systemic Leaf Responses Reveal a Shift from Stress-Related Programs Toward Metabolic and Growth-Associated Functions

An interesting finding of our study was that the systemic transcriptional response in leaves was more extensive than the local root response, both in scale and in the diversity of affected biological processes. Under the experimental conditions employed in our study, no Streptomyces-like colonies were recovered from leaf tissues upon culturing on ISP4 medium. Although the presence of bacteria at very low abundance cannot be fully excluded without molecular detection approaches, these observations support the interpretation that the transcriptional changes detected in leaves reflect a systemic response triggered by root colonization. Such pronounced leaf reprogramming is consistent with previous reports showing that root-associated beneficial microbes can elicit stronger and more diverse systemic responses in aerial tissues than at the site of colonization [9,10,11], likely reflecting the role of leaves as central integrators of root-derived signals, where changes in nutrient status, hormonal balance, and growth demands are translated into broad metabolic and developmental transcriptional programs [102].

As in roots, the leaf transcriptome was dominated by the down-regulation of defense and general stress responses, indicating that root-associated microbial colonization does not elicit a stress response in the above-ground tissues. In parallel, leaves exhibited strong enrichment of metabolic and nutrient-associated processes, including carbohydrate metabolism, energy-generating pathways, and lipid metabolism, consistent with increased biosynthetic and energetic demands. Since components of the phosphate deficiency response were also activated in leaves, these metabolic adjustments may reflect, at least in part, an elevated phosphate demand at the whole-plant level. Importantly, the concurrent repression of defense-related pathways and activation of metabolic functions is consistent with the emergence of a coordinated systemic transcriptional state in which growth, nutrient management, and metabolic flexibility are prioritized, while the defense investment is reduced. As discussed above, bioassays with foliar pathogens will be necessary to determine whether systemic defense attenuation results in increased susceptibility or whether additional regulatory layers, not captured at the steady-state transcriptomic level, contribute to systemic protection.

A particularly informative aspect of the leaf response is that a subset of genes displayed coordinated regulation in both roots and leaves. Specifically, genes associated with phosphate deficiency and homeostasis were induced in both tissues, suggesting that Streptomyces-mediated activation of phosphate-responsive pathways in roots is accompanied by a corresponding adjustment at the whole-plant level. In parallel, genes related to cell wall-associated processes were consistently repressed in both organs, indicating coordinated structural reprogramming. These shared transcriptional changes likely represent a central signature embedded within the broader whole-plant reprogramming initiated in the roots, rather than a self-contained systemic signaling module. Importantly, the induction of phosphate-responsive genes and the repression of cell wall-associated processes may not be independent events. Phosphate status is closely linked to metabolic allocation and structural investment, and activation of phosphate-related pathways could be accompanied by reduced transcription of energetically demanding cell wall programs. Although direct causality cannot be inferred from the present dataset, the coordinated regulation of these gene categories supports the idea that nutrient signaling and structural remodeling are functionally interconnected components of the plant response to Streptomyces colonization.

4.4. Genomic Insights into the Biosynthetic Potential Underlying Streptomyces-Mediated Pathogen Suppression

In the greenhouse experiment, treatment with the TOM consortium significantly reduced disease severity caused by Foxrl. While this protective effect likely reflects emergent properties of the consortium, several individual TOM isolates also exhibited measurable antagonistic activity against Foxrl in dual culture assays, suggesting that direct microbial inhibition may contribute to the observed disease reduction. To explore the genetic basis of this antifungal potential, five Foxrl-suppressive isolates (TOM-2, TOM-3, TOM-4, TOM-9 and TOM-12) were subjected to whole-genome sequencing. Genome analyses revealed large, GC-rich chromosomes encoding extensive and partly uncharacterized repertoires of biosynthetic gene clusters, including clusters predicted to direct the production of antimicrobial secondary metabolites. Together, these features provide a plausible genetic foundation for antifungal traits within the consortium. Notably, genome-based taxonomic analyses indicated that two of the sequenced isolates (TOM-4 and TOM-9) exhibited digital DNA–DNA hybridization values well below the accepted species delineation threshold, suggesting that they may represent previously uncharacterized Streptomyces lineages at the species level.

It should be noted, however, that a substantial proportion of Streptomyces biosynthetic gene clusters remain cryptic and are not expressed under standard laboratory conditions, suggesting that additional antimicrobial activities encoded by these genomes may be activated only under specific environmental or host-associated cues [103]. Consequently, the antagonistic capacity of the TOM consortium under rhizosphere or host-associated conditions may extend beyond what is detectable in simplified greenhouse and laboratory assays.

5. Conclusions

Collectively, our study demonstrates that a defined consortium of tomato-associated Streptomyces promotes plant growth and induces coordinated local and systemic transcriptional reprogramming, characterized by selective attenuation of host defense responses alongside activation of nutrient- and growth-associated processes. These responses are not readily explained by classical immunity-based frameworks but instead support a model in which plants dynamically adjust immunity, metabolism, and development in response to beneficial microbes. The combined capacity of the Streptomyces TOM consortium to enhance plant growth and suppress soil-borne pathogens highlights its potential for sustainable crop management strategies.