Introduction

Tomato vascular wilt, caused by Fusarium oxysporum f. sp. lycopersici (Fol), is one of the most destructive diseases, causing qualitative and quantitative losses (Atif et al. 2021). Currently, several management strategies are available to control Fol, including the use of fungicides and plant-resistant varieties (Amini and Dzhalilov 2010). However, chemical treatment is only available as a protective barrier for seeds (Bodah 2017), and the treatment with agrochemicals to control Fol is avoided due to their persistence in soil and their toxicity to beneficial organisms (Yu et al. 2017). Consequently, recent research on its management has focused on alternative strategies such as disease-suppressive soils, beneficial microorganisms, and induction of host resistance (McGovern 2015; Mwangi et al. 2019).

The interest in bacteria for biological control has increased during the last years. These microorganisms exhibit biocontrol of plant diseases, inhibiting the growth of a wide range of plant pathogenic fungi, through the production of siderophores, antimicrobial compounds, and lytic enzymes, and by inducing plant defense responses (Moreno et al. 2018; Karthika et al. 2020). Following exposure to pathogens or beneficial microorganisms, plants activate defense responses both locally and systemically (Mauch-Mani et al. 2017). These responses are associated with Systemic Acquired Resistance (SAR) and Induced Systemic Resistance (ISR). ISR by beneficial microbes is commonly based on priming (Mauch-Mani et al. 2017). Priming is defined as a state of enhanced protection characterized by faster and stronger defense responses against environmental stresses (Martínez-Medina et al. 2016). During beneficial plant–microbe interaction, priming is expressed at the transcriptional level, for instance, elevated levels of expression of AP2/ERF family transcription factors have been observed. Several members of this family belong to the signaling cascade that is part of the Jasmonic Acid (JA) and Ethylene (ET)-mediated defense responses (Van der Ent et al. 2009).

Siderophores are iron-chelating agents produced by bacteria, fungi, and plants (Soares 2022). These molecules play a role in the solubilization of iron from minerals that are found as part of insoluble complexes. The excretion of these molecules allows bacteria to sequester the available iron and transport it into the cell. In the environment, the ferric form of iron is insoluble and physiologically unavailable. Microorganisms have evolved to synthesize siderophores that have a high affinity for ferric iron. These ferric iron-siderophore complexes are transported to the cytosol and then reduced to ferrous iron to be accessible to microorganisms (Saha et al. 2016). Siderophore production in Acinetobacter and other bacteria genera is regulated by the concentration of available iron in the growth medium. Maindad et al. (2014) showed that the synthesis of an acinetobactin-like siderophore produced by Acinetobacter calcoaceticus strain HIRFA32 decreases as iron concentration increases. Similarly, the catechol activity of Acinetobacter baumannii siderophores was found to reach 70% at a concentration of 20 µM of FeCl3 and decreased to 45% at 80 µM of FeCl3 (Modarresi et al. 2015). In Bacillus anthracis, while production of the siderophore bacillibactin was found to be highly regulated by the concentration of available iron, petrobactin synthesis is likely regulated by additional virulence-related factors (Lee et al. 2011). Dumas et al. (2013) described that Pseudomonas aeruginosa switches from the pyochelin production under highly iron-limiting conditions, pyoverdine production when a moderate iron concentration is available.

Besides siderophore's role as iron-bearers, these molecules have been reported to have an antifungal effect or to induce the activation of the plant defense responses, including the activation of ISR (Dellagi et al. 2009; Santoyo et al. 2010; Aznar and Dellagi 2015). Several studies have shown that different siderophore-producing Pseudomonas species, under low iron conditions, are associated with plant protection against pathogens, including, Botrytis cinerea, Colletotrichum lindemuthianum, Pythium splendens, Pseudomonas syringae, Magnaporthe oryzae and Fusarium oxysporum (Buysens et al. 1996; Meziane et al. 2005; De Boer et al. 2007; De Vleesschauwer et al. 2008; Aznar et al. 2014). In addition, results obtained by Yu et al. (2011) showed that treatment with bacillibactin, a catechol-like siderophore produced by Bacillus subtilis CAS15, resulted in a 50% disease reduction in pepper during infection with Fusarium oxysporum f. sp. capsica. These studies demonstrated that siderophores can (i) trigger ISR-mediated resistance and/or (ii) suppress the pathogen by uptake of iron from the environment. Recent reports indicate that siderophore-triggered ISR involves the accumulation of reactive oxygen species and phenolic compounds, and the activation of hormone signaling pathways particularly those mediated by ET and JA.

In view of the few studies that have conducted to explore the biological diversity and attributes of microorganisms adapted to megadiverse areas such as the Amazon rainforest (Guerra et al. 2020), this study aimed to characterize siderophore-producing bacteria isolated from this region and to identify promising candidates for the suppression of tomato vascular wilt caused by Fol. The expression level of gene markers from the JA, ET, and salicylic acid signaling pathways was measured to better understand the mechanism of antagonistic action exerted by siderophore-rich cell-free supernatant applied to plants.

Materials and methods

Bacteria isolates and physiological characterization

The fourteen bacterial isolates used in this study (CBIO021, CBIO024, CBIO086, CBIO101, CBIO117, CBIO118, CBIO120, CBIO121, CBIO123, CBIO127, CBIO133, CBIO142, CBIO149, and CBIO159) were previously isolated from rhizosphere and phyllosphere samples collected from wild cocoa trees (Theobroma subincanum Mart., Theobroma cacao L. and Herrania nitida (Poepp.) R.E. Shult.) from the Colombian Amazon. The isolates were deposited in the microbial germplasm bank of Agrosavia. The sampling coordinates are shown in Table S1. Isolates were characterized by Gram staining (Gram 1884). Lactose fermentation was visualized on MacConkey agar (Oxoid, MacConkey, 1985) where individual colonies were plated and incubated at 37 °C for 48 h in the dark. Hemolytic activity was determined qualitatively using blood agar culture medium (Oxoid) containing 5% v/v bovine blood (Lányi, 1987). After 24 h of incubation at 37 °C, the presence around the inoculum site of green or translucent halos, or the absence of halo was interpreted as alpha-hemolysis, beta-hemolysis or gamma-hemolysis, respectively. Experiments were performed in triplicate.

Quantification of siderophore production

Siderophore production was evaluated in iron-free modified minimal medium (MM) (Robertsen et al. 1981) containing: C5H8NNaO4, 2.42 g/L; C6H12O6, 10 g/L; supplemented with stock solution 10 mL/L of K2HPO4, 3 mg/mL; KH2PO4, mg/mL; CaCl2·6H2O, 0.5 mg/mL; NaCl, 0.5 mg/mL; MgSO4, 1,5 mg/mL; pH 7.0). The cell mass of a 24-h culture of each isolate was used to inoculate MM broth to an initial optical density (OD) of 0.2 at 600 nm. The cell culture was incubated at 28 °C under constant shaking at 117 rpm for 72 h with an oscillation diameter of 25 mm (Tecnal®, incubator shaker TE-421, Brazil). The cell-free supernatant was collected by centrifugation at 8273×g for 15 min at 4 °C (Eppendorf®, centrifuge 5430, Germany). Detection of siderophores was estimated by Chrome Azurol Sulfonate (CAS) colorimetric test, as described by Schwyn and Neilands, (1987). Briefly, 500 μL of the culture supernatant were mixed with 500 μL of CAS solution (As). After incubation for 1 h in the dark at room temperature, the OD was measured at 630 nm. MM broth without inoculum mixed with CAS solution was used as a control (Ar). Siderophore production was expressed as a color change of the reaction from blue to orange and was calculated by the following equation: \([(\mathrm{Ar }-\mathrm{ As}) /\mathrm{ Ar}] *100\). Where Ar represents the absorbance of the control and As represents the absorbance of the sample. To estimate the siderophore concentration, a standard calibration curve was performed using deferoxamine mesylate (DFOM) as an iron chelator following the methodology described by Radzki et al. (2013) and Mehnert et al. (2017). The concentration was expressed in µmol/L DFOM equivalents. According to Radzki et al. (2013) 1 mmol of DFOM captures 56 mg of Fe3+.

Taxonomic identification and phylogeny

The total deoxyribonucleic acid (DNA) from the fourteen selected isolates was extracted from grown cultures using the MoBio PowerLyzer® UltraClean® Microbial Kit (Carslbad, CA) following the manufacturer's instructions. The 16S V4 of the ribosomal ribonucleic acid (rRNA) gene was amplified by polymerase chain reaction (PCR). The PCR reaction mixture consisted of 5 µL of DNA template, 1 µL (10 µM) of the primers 515F (5ʹ-GTGYCAGCMGCCGCGGTAA-3ʹ) and 806R (5ʹ-GGACTACNVGGGTWTCTAAT-3ʹ) (Apprill et al. 2015; Parada et al. 2016), 10 µL of Go Taq® Green Master Mix (Promega) and 8 µL of nuclease-free water (Invitrogen™). Amplification was carried out by an initial denaturation step at 94 °C for 3 min, followed by 35 cycles of 45 s of denaturation at 94 °C, 60 s of annealing at 50 °C and 90 s of extension at 72 °C (Applied Biosystems SimpliAmp, Thermal Cycler, USA). Final extension was carried out at 72 °C for 10 min. The resulting amplification products were separated using a 1% w/v agarose gel. Sequencing of the amplified DNA fragments was performed by Corpogen Corporation (https://www.corpogen.org/, Bogotá. Colombia). To resolve ambiguous bases on the forward and reverse trace files, the resulting sequences were manually checked considering the PHRED scores received. Taxonomic identification of isolates to genus level was carried out using the NCBI GenBank database (https://blast.ncbi.nlm.nih.gov) and the basic local alignment search analysis tool (BLASTn). The resultant unrooted tree topology was evaluated by bootstrap analyses (Felsenstein 1985) of the neighbor-joining method based on 1000 bootstrap replicates. Trees were rooted using the partial sequence of the 16S rRNA gene of Lactococcus plantarum (NR_044358.1) as an outgroup.

Identification of Acinetobacter species by Vitek 2 system compact

The eight isolates identified as Acinetobacter sp. were cultured in Luria–Bertani media (LB) for 24 h at 35 °C. A sterile swab was used to transfer and suspend the cell mass in 3.0 mL of sterile saline solution (0.45% NaCl, pH 7.0) in a test tube. The turbidity of the cell suspension was measured and adjusted to the McFarland turbidity range between 0.50 and 0.63 using a DensiChekTM Plus (VITEK® 2, BioMereux, USA). The identification GN ID cards (Ref 21341, VITEK® 2 GN) were inoculated with each microorganism's cell suspension using a Vitek 2 Compact® (BioMerieux, Lyon, France). All cards were incubated at 35 °C for 18 h. Each test reaction was read every 15 min to measure turbidity and colored products of substrate metabolism. Interpretation of the results was carried out as described by Pincus (2006) where the GN card is based on established biochemical methods for the identification of isolates gram-negative bacilli.

Induction of siderophores production

The production of siderophores in five A. baumannii isolates that showed increased ability to excrete these molecules (CBIO024, CBIO086, CBIO117, CBIO123, and CBIO159) was evaluated in MM broth without iron, and in MM containing a low (22 µM) and a high (220 µM) concentration of FeCl3·6H2O. A bacterial inoculum was prepared by culturing the selected bacteria on nutrient agar for 24 h at 28 °C. The cell mass was then washed twice with NaCl (0.85% w/v) by centrifugation for 5 min at 4430×g (Hermle, centrifuge Z 326 k, Germany), removing the supernatant between washes. The resulting cell pellet was resuspended in MM broth supplemented with each of the FeCl3·6H2O concentrations and adjusted to a final OD of 2.0 at 600 nm (Thermo Fisher Scientific, Genesys 150 spectrophotometer, USA). Next, a 0.5 mL aliquot of each bacterial inoculum was added to Falcon tubes (15 mL) containing 4.5 mL of the MM broth without iron or supplemented with FeCl3·6H2O, until a final OD of 0.2 at 600 nm was reached. Cell cultures were incubated at 28 °C under constant shaking at 180 rpm with an oscillation diameter of 10 mm for 72 h (Heidolph, Shakers Unimax 1010, Germany). Siderophore production was quantified using CAS assay as described above. Three biological replicates were performed in a series of 4 samples.

Bioactivity assays against Fol

The bioactivity of the cell-free supernatant of the five A. baumannii isolates CBIO024, CBIO086, CBIO117, CBIO123, and CBIO159 was evaluated against Fol by in vivo experiments on tomato plants. For this purpose, the pathogen Fusarium oxysporum f. sp. lycopersici Race 2 identified as Fol59 was used in this study (Carmona et al. 2020; Yu et al. 2022).

Cell-free supernatant of each isolate with high content of siderophores (SodSid) was produced in MM without iron using the methodology described above. For this experiment, twenty-eight days old seedlings of the susceptible tomato cultivar Santa Cruz Kada grown in a growth chamber at a temperature of 28 °C, with a relative humidity of 60% and a photoperiod of 14 h of light and 10 h of darkness were used. Plants were treated by drenching with 10 mL of SodSid of each isolate. Forty-eight hours after SodSid treatment, plants were removed from the seedbed and infected with Fol59 by root immersion into a suspension of 1 × 105 microconidia/mL (Carmona et al. 2020). Subsequently, the plants were transplanted into 16 oz pots containing soil and sand in a 2:1 ratio. Treatments consisted of plants inoculated only with SodSid of A. baumannii isolates (CBIO024, CBIO086, CBIO117, CBIO123, and CBIO159) and plants inoculated with SodSid and infected with Fol59 (CBIO024 + Fol59, CBIO086 + Fol59, CBIO117 + Fol59, CBIO123 + Fol59, and CBIO159 + Fol59). Water-treated plants were considered as absolute control (Ab control), and plants infected with Fol59 (Fol59) were used as pathogen control. The experimental setup was a randomized complete factorial design with two factors. The first factor corresponded to the five bacterial supernatants, the second to plants with or without the pathogen infection, for a total of 10 treatments and two controls. Three biological replicates were performed. Each biological replicate comprised three technical replicates, and each technical replicate consisted of eleven plants. Selected parameters related to pathogenicity, plant growth, and regulation of ISR-associated genes were evaluated.

Disease incidence and severity were recorded periodically for 21 days. To assess disease severity, the scale proposed by Rongai et al. (2017) was used and modified as follows: absence of symptoms was recorded as level 0; slight yellowing appearance of one or two leaves was recorded as level 1; necrotic lesions formed both, at the base of the stem and on internodes in the center of the plant, with or without the yellowing appearance of basal and median leaves were recorded as level 2; total yellowing appearance of basal leaves, with some wilted leaves and necrotic lesions in the steam extending upwards was recorded as level 3; loss of turgor and wilting of the plant was recorded as level 4; a dead plant was recorded as level 5. The severity index (SI) and area under the disease progression curve (AUDPC) were calculated using the equations described by Chiang et al. (2017) and Pedroza-Sandoval and Gaxiola (2009). The plant height, dry weight, and chlorophyll content of plants were determined 16 days post-inoculation with Fol59 (dpi). The plant height was recorded using a tape measure. The weight of six plants per treatment was measured after drying in an oven at 60 °C for 72 h. Chlorophyll content was estimated in Soil Plant Analysis Development (SPAD) units using the Minolta® SPAD 502 chlorophyll meter. Leaf chlorophyll content was measured in the distal part of the adaxial side of an apical leaf (three measurements) (Hurtado et al. 2017), values shown are means ± SE (n = 11) plants, using the mean of three measurements per plant.

Quantification of the transcription of systemic resistance-related genes

The relative expression of three defense marker genes was calculated by quantitative Real-Time Reverse Transcription Polymerase Chain Reaction (qRT-PCR). The treatments evaluated were (i) non-inoculated plants treated with water (Ab control), (ii) plants only inoculated with CBIO117 SodSid (CBIO117), (iii) plants only infected with Fol59 (Fol59, pathogen control), and iv) plants inoculated with CBIO117 SodSid and infected with Fol59 (CBIO117 + Fol59). The differential expression of PR1, ERF1, and MYC2 genes associated with SA, ET, and Jasmonic acid (JA), respectively, was assessed at 24 and 48 h post infection with Fol (hpi). For this, specific primers previously reported for PR1, ERF1, and MYC2 were used, additionally, elongation factor expression (EF1a) was used as a reference (Table 1) (Martínez-Medina et al. 2013).

Table 1 Primer sequences for defense marker genes in qPCR analysis
Full size table

Total RNA extraction was performed following the protocol described by Yockteng et al. (2013). Complementary DNA (cDNA) synthesis was performed using the ProtoScript® II First Strand cDNA kit (New England BioLabs) following the manufacturer's recommended protocol. Quantitative RT-PCR reactions were performed using the IqTM SYBR® Green Supermix kit (Bio-Rad®) in a final volume of 10 μL per reaction. The reaction mixture consisted of 2 μL of cDNA, 0.4 µL of each primer (0.4 μM), 5 µL of 2X of IQ SYBR Green Supermix, and 2.2 μL of molecular grade water. Reactions were performed in duplicate for each of the three independent biological replicates. To analyze the relative gene expression data, the common-base method proposed by Ganger et al. (2017) was used according to the following equation: \(\Delta \mathrm{Ct}=({\left(\mathrm{Log}\right)}_{10}{\mathrm{E}}_{\mathrm{ref}}\times{\mathrm{Ct}}_{\mathrm{ref}} )-({\left(\mathrm{Log}\right)}_{10} {\mathrm{E}}_{\mathrm{Goi}}\times{\mathrm{Ct}}_{\mathrm{Goi}} )\) where Eref is the efficiency of the housekeeping gene EF1a, EGoi is the efficiency of the target gene, and Ct are threshold values of the cycle.

Antimicrobial activity test in dual-liquid culture assay

To test antimicrobial activity by the dual-liquid culture method, 50 mL of iron-deficient media containing glucose at 5% w/v (IDM, Muller et al. 1984) was taken in a 250-mL culture flask and inoculated with 50 mL of CBIO117 inoculum (107 cells/mL) and a 5 mm size PDA plug of freshly grown Fol59. The culture flasks were incubated at 28 °C for 72 h on a rotary shaker at 120 rpm with an oscillation diameter of 19 mm (MaxQ 4000, Barnstead/Lab-line, Switzerland). Fol59 cultures without CBIO117 inoculum were used as controls. After incubation, cell culture supernatants were filtered through a Whatman No. 1 filter paper (1001-055, Whatman) and dried at 60 °C. Fol59 biomass growth on both cultures, co-inoculated with CBIO117 and control (without CBIO117), was compared to assess the antagonistic activity of CBIO117. The percentage reduction in fungal biomass was calculated by the equation: \(Inhibition \left(\%\right)=\frac{\left({W}_{1 }{-W}_{2}\right)}{{w}_{1} \times 100}\) where W1 corresponds to the biomass (g) of Fol59 in the control culture, while W2 corresponds to the biomass of Fol59 grown in co-culture with CBIO117 (Basha and Ulaganathan 2002; Trivedi et al. 2008).

Statistical analysis

To evaluate the effect of treatments on disease severity, a linear mixed-effects model analysis was performed. Treatment, sampling day, biological replication, and interaction were considered fixed effects. The experimental unit (plant) was included as a random effect (Schandry 2017). Relative chlorophyll content was analyzed using linear mixed-effects models. Plant height and dry weight values were analyzed by ANOVA and Tukey test. Tukey tests for pairwise comparisons of means were performed at a significance level of P < 0.05. Relative expression levels were analyzed using cycle threshold values calculated by univariate analysis at each post-inoculation time point (Popović et al. 2021). All statistical analyses were carried out using R software (version 4.1.2).

Results

Phenotypic characterization of bacterial isolates

The fourteen isolates used in this study were previously selected for their ability to produce siderophores. For each isolate, geographic location, plant part for sample collection (rhizosphere or phyllosphere), and phenotypic characteristics are shown in Table 2. The isolates showed differential microscopic characteristics and ability to ferment lactose and lyse blood cells. Two isolates were Gram-positive bacilli, eight were Gram-negative coccobacilli, and four were Gram-negative bacilli. The lactose utilization assay showed that most isolates fermented lactose as observed by the appearance of a clear halo around colonies. The greatest diversity was found in the hemolytic activity test, in which isolates showing α, β, and γ activity were identified.

Table 2 Geographical location, plant part of sample collection (rhizosphere or phyllosphere) and phenotypic characteristics of isolates analyzed in this study
Full size table

Quantification of siderophore production

Comparison of siderophore production among 14 siderophore-producing isolates showed significant differences (P < 0.05, Fig. 1). The concentration of siderophores ranged from 2.7 µM ± 0.3 (CBIO101) to 129.9 µM ± 19.1 (CBIO123). Isolates CBIO024, CBIO086, CBIO123, CBIO117, and CBIO159 showed the highest siderophore production. For these isolates, siderophore concentrations were 99 ± 6.2, 91 ± 7.6, 129 ± 19.1, 107 ± 5.8, and 114 ± 11.1 µmol/L DFOM equivalents, respectively (Fig. 1). For isolates identified as Delftia and Pseudomonas (CBIO101 and CBIO142), low concentrations of siderophore (4.6 ± 0.5 µmol/L DFOM equivalents) were observed (Fig. 1 and Fig. 2).

Fig. 1
figure 1

Production of siderophores by CBIO-isolates (expressed in µmol/L equivalents of DFOM) after 72 h after incubation. Bacteria were grown in minimal media without iron. Values followed by the same letter do not differ significantly (Tukey test, P < 0.05). Values correspond to the mean (± ES) of three biological replicates with n = 3

Full size image
Fig. 2
figure 2

Phylogenetic-tree construction using the neighbor-joining method. The phylogenetic analysis was performed with partial 16S rRNA gene sequences. Bootstrap values, expressed as percentages of 1000 replications, are shown at branch points. The numbers above the branches indicate the percentage of consensus support. Bold letters highlight the CBIO-isolates characterized in this study. The tree was rooted using Lactococcus plantarum NR_044358.1 as an outgroup

Full size image

Taxonomic characterization

Comparison of the partial 16S rRNA gene sequence and phylogeny analysis allowed taxonomic identification of the isolates to genus level. These analyses classified the isolates into 14 genera, eight belonging to the Acinetobacter genus, two to the Bacillus genus, one to the Delftia genus, one to the genus Herbaspirillum, one to the genus Pseudomonas, and one to the genus Serratia (Fig. 2).

Vitek tests were performed to further characterize the taxonomic identity of the CBIO isolates classified within the genus Acinetobacter. Isolates CBIO024, CBIO086, CBIO117, CBIO123, CBIO159, CBIO120, CBIO127, and CBIO149 were classified within the A. baumannii complex with a 99% of probability. All these isolates showed five different metabolic patterns (Group 1, 2, 3, 4, and 5; Fig. 3). Most of them were able to grow using d-cellobiose, d-glucose, d-mannose, tyrosine arylamidase, l-lactate, succinate, and coumarate as carbon source. CBIO117 was the only isolate unable to grow using citrate as the sole carbon source, and inorganic ammonium salt as the sole source of nitrogen. Remarkably, CBIO117, CBIO123, and CBIO159 did not assimilate l-histidine and were the isolates with the highest siderophore production. In contrast, isolates CBIO120, CBIO127, and CBIO149 showed lower production of iron chelators and did not grow in the presence of urease (Fig. 3).

Fig. 3
figure 3

Metabolic profile using ID-GNB card of the VITEK 2 system for taxonomic identification of the Acinetobacter sp. isolates. Presence of growth (+), absence of growth (−). Ala-phe-pro-arylamidase (APPA), Adonitol (ADO), l-pyrrolydonyl-arylamidase (PyrA), l-arabitol (IARL), d-cellobiose (dCEL), Beta-galactosidase (BGAL), H2S production (H2S), Beta-N-acetyl-glucosaminidase (BNAG), Glutamyl arylamidase pNA (AGLTp), d-glucosa (dGLU), Gamma-glutamyl transferase (GGT), Fermentation of glucose (OFF), Beta-glucosidase (BGLU), d-maltose (dMAL), d-mannitol (dMAN), d-mannose (dMNE), Beta-xylosidase (BXYL), Beta-alanine arylamidase pNA (BAIap), l-proline-arylamidase (ProA), Lipase (LIP), Palatinose (PLE), Tyrosine arylamidase (TyrA), Urease (URE), d-sorbitol (dSOR), Saccharose/sucrose (SAC), d-tagatose (dTAG), d-trehalose (dTRE), Citrate (Sodium) (CIT), Malonate (MNT), 5-Keto-D-gluconate (5 KG), l-lactate alkalinization (ILATk), Alpha-glucosidase (AGLU), Succinate alkalinization (SUCT), Beta-N-acetyl galactosaminidase (NAGA), Alpha-galactosidase (AGAL), Phosphatase (PHOS), Glycine arylamidase (GlyA), Ornithine descarboxylase (ODC), Lysine decarboxylase (LDC), l-histidine assimilation (IHISa), Cumarate (CMT), Beta-glucuronidase (BGUR), O/129 Resistance (comp.vibrio.) (O129R), Glu-Gly-Arg-arylamidase (GGAA), l-malate assimilation (IMLTa), Ellman (ELLM), l-lactate assimilation (ILATa)

Full size image

A. baumannii isolates CBIO024, CBIO086, CBIO117, CBIO123, and CBIO159 were selected for further studies, based on their increased ability to produce siderophores and differences in metabolic profile compared to other isolates.

Effects of cell-free supernatants of A. baumanii isolates on the suppression of vascular wilt

We further confirmed the effect cell-free supernatants of A. baumanii on plant protection against Fol. To ensure that bioactivity assays were performed with siderophore-rich supernatants, siderophores production was evaluated in MM containing different concentrations of FeCl3·6H2O. The results showed that at higher iron concentrations, siderophore production decreased (Figure S1). When A. baumannii isolates CBIO024, CBIO086, CBIO117, CBIO123, and CBIO159 were culture in MM without iron, we found that the mean concentration of siderophores production was 43 µmol/L DFMO equivalents. Thus, activities against tomato Fusarium wilt were attempted using the cell-free supernatant of selected CBIO-isolates grown under iron-starved conditions (SodSid).

Preventive application of SodSid of A. baumanii isolates reduced disease incidence in plants infected with Fol59 by 31% (CBIO024 + Fol59), 11% (CBIO086 + Fol59), 33% (CBIO117 + Fol59), 24% (CBIO123 + Fol59) and 20% (CBIO159 + Fol59) relative to plants only infected with the pathogen, where 100% incidence was reached at 21 dpi. Disease severity index and AUDPC of plants pretreated with SodSid and subsequently infected with Fol59 (CBIO + Fol59) were lower compared to plants infected only with Fol59. However, at 21 dpi, the highest disease suppression was observed in CBIO117 + Fol59 treatment wheredisease severity was reduced up to 45% (Fig. 4B).

Fig. 4
figure 4

Effect of preventive application of cell-free supernatant from A. baumanii isolates on Fusarium wilt of tomato. A Disease severity index. B Effect of SodSid from five Acinetobacter sp. isolates on AUDPC severity at 21 days post inoculation. C Representative images of tomato plants treated with water (Ab control), infected with Fol59 (Fol59), and plants previously treated with CBIO117 SodSid, subsequently infected with Fol59 (CBIO117 + Fol59). Values correspond to the mean of three biological replicates with n = 15. Value means followed by the same letter do not differ significantly (Tukey test, P < 0.05)

Full size image

Growth parameters of tomato plants treated with cell-free supernatant of A. baumannii isolates

Next, we investigated whether A. baumannii SodSid affected plant development rate. The plant height at 16 dpi was significantly delayed (P < 0.05) in Fol59-infected plants compared to water-only treated plants (Ab control plants), which showed greater increase in height, and higher relative chlorophyll content (Table 3). Interestingly, plants treated only CBIO117 SodSid showed the greatest plant height, with significant differences with respect to the other treatments (P < 0.05). Plants treated with SodSid of isolate CBIO024, SodSid of CBIO123 and infected with Fol59 (CBIO123 + Fol59), and plants infected only with Fol59 showed lower amounts of total dry matter accumulation. As for the SPAD chlorophyll index, significantly higher SPAD units were observed in all plants treated with SodSid from CBIO-isolates and plants treated with SodSid of the CBIO-isolates and then infected with Fol59, compared to Ab and Fol59 control plants (P < 0.05).

Table 3 Growth parameters of tomato plants treated with cell-free supernatant of A. baumannii isolates
Full size table

Effect of SodSid on the expression of defense marker genes in Fol-infected tomato plants

To determine whether the reduction of wilt disease severity in tomato plants treated with SodSid from A. baumannii CBIO117 is related to the induction of plant defense genes, the relative expression of the PR1, ERF1, and MYC2 genes was evaluated in plants treated only with SodSid from CBIO117 isolate and plants pretreated with SodSid and subsequently infected with Fol59. Two separate time points were considered after infection with Fol59 (24- and 48-h post-infection (hpi)). At the earliest time (24 hpi), there was no significant difference in PR1 expression levels between treatments. However, at 48 hpi all plants, including those treated only with SodSid from A. baumannii CBIO117 showed a significant up-regulation of PR1 gene (P < 0.05) (Fig. 5A). These results indicate SodSid application of A. baumannii CBIO117 is able to induce plant defense responses, including PR1-based basal immunity (Fig. 5A).

Fig. 5
figure 5

Relative expression of defense genes to assess the effect of cell-free supernatant (SodSid) of A. baumannii CBIO117 on the activation of tomato defense responses. A Relative expression of PR1 gene. B Relative expression of the ERF1 gene, and C Relative expression of MYC2 gene. Treatments correspond to plants treated with water (Ab control), plants infected with Fol59 (Fol59), plants treated with SodSid of CBIO117 isolate (CBIO117), and plants treated with SodSid of CBIO117 and then infected with Fol59 (CBIO117 + Fol59). Values correspond to the mean (± ES) of two biological replicates with n = 2. Means of values followed by the same letter do not differ significantly (Tukey test, P < 0.05)

Full size image

Similarly, ERF1 gene expression levels increased in both conditions, after treatment CBIO117 SodSid alone and in plants pretreated with SodSid and subsequently infected with Fol59, compared with control plants. This means that the preventive application on tomato roots with SodSid of CBIO117 promoted ethylene-mediated defense responses in tomato plants (Fig. 5B).

Expression of the MYC2 transcription factor was not induced after preventive SodSid application of A. baumannii CBIO117 on tomato roots. The highest level of MYC2 expression occurred at 24 hpi in plants infected only with Fol59, followed by plants pretreated with SodSid and then infected with the pathogen (CBIO117 + Fol59). However, at 48 hpi, a decrease in MYC2 gene expression was detected in plants infected with Fol59, whereas the gene expression was maintained in plants previously stimulated with the SodSid and infected with the pathogen (CBIO117 + Fol59).

Antimicrobial activity assay in dual culture in liquid medium

To gain further insight into the antimicrobial activity of A. baumannii CBIO117, we performed an assay using the dual culture method in a liquid medium. It was observed that CBIO117 isolate was not able to inhibit the growth of Fol59. No significant differences (P < 0.05) were observed when comparing Fol59 biomass growth in either culture, co-inoculated with CBIO117 (0.20 ± 0.08 g/mL) or the control culture without CBIO117 (0.22 ± 0.06 g/mL).

Discussion

The tomato crop is threatened by wilt caused by Fusarium oxysporum f. sp. Lycopersici. Increasing attention is being paid to reducing the use of synthetic fungicides to control plant diseases, mainly because of their negative impact on production costs, environmental pollution, risks to human health, development of fungicide resistance, and post-harvest fruit quality. The use of beneficial microorganisms or their metabolites is a promising alternative for disease management. In this study, siderophore-producing bacteria from the rhizosphere and phyllosphere of wild cocoa trees collected in the Colombian Amazon were characterized. Furthermore, the bioactivity of the cell-free supernatants containing siderophores was tested against Fol.

Fourteen isolates showed high siderophore production, of which eight were identified as A. baumannii. Several structurally distinct siderophores have been described for this bacterium including acinetobactin, baumannoferrins A and B and, fimsbactins A–F. A large collection of well-characterized siderophores has been also described in the same bacterial genus where the remaining CBIO-isolates were classified. For instance, serobactin A, B, and C from Herbaspirillum (Rosconi et al. 2013; Tejman-Yarden et al. 2019), bacillibactin and petrobactin from Bacillus (Wilson et al. 2006), serratiochelin, chrysobactin, aerobactin, and enterobactin from Serratia (Ehlert et al. 1994; Weakland et al. 2020), and pyoverdine, ferrioxamine, piocheln, and ferribactin from Pseudomonas (Aguilar et al. 2021). The production of several structurally diverse iron chelators by the same bacteria has been associated with improved iron acquisition and prevention of iron deprivation under different environmental conditions, facilitating the uptake of other metals, and contributing to competitiveness (McRose et al. 2018). Given that the CBIO-isolates were isolated in iron-depleted media, it could be considered that the ability to produce structurally diverse siderophores favors the prevalence of the A. baumannii isolates, explaining why a higher representative number of isolates from this genus was obtained. To better understand the ability of CBIO-isolates to survive under iron-limited conditions, their growth and persistence under iron deficient and iron-abundant conditions need to be evaluated.

Plant protection activity of Acinetobacter sp. bacteria against phytopathogenic filamentous fungi and bacteria has been reported before (Xue et al. 2009; Safdarpour and Khodakaramian 2018; Foughalia et al. 2022; Khalil et al. 2021). Different molecules are associated with suppression of these fungi, including siderophores, gibberellic—and indole acetic acids, volatile compounds, or microbial enzymes such as chitinases and proteases. In this study, the application of cell-free supernatant reduced the incidence and severity of the Fusarium wilt of tomato. Further studies are needed to better understand the metabolites present in the supernatant of the CBIO-isolates that generated the plant protection activity. However, as these supernatants were rich in siderophores, it is possible that the antagonistic activity was at least partially generated by these molecules. Prashant et al. (2009) demonstrated the antifungal activity of catechol-like siderophores from Acinetobacter calcoaceticus strain SCW1 against Aspergillus flavus, Aspergillus niger, Colletotrichum capsicum, and F. oxysporum. Sayyed and Reddy (2011) reported the growth inhibitory potential of siderophore-rich cell-free supernatants obtained from Acinetobacter sp. SH-94B isolate against A. niger NCIM 1025, A. flavus NCIM 650, F. oxysporum NCIM 1281, Alternaria alternata ARI 715, Cercospora arachichola, Metarhizium anisopliae NCIM 1311 and Ralstonia solanacerum NCIM 5103. Bacterial siderophores have already shown to play a role in antagonizing plant pathogens through iron deprivation, but also by activating plant-induced systemic resistance (Höfte and Bakker 2007; Betoudji et al. 2020). A synthetic siderophore of fimsbactin-like structure produced by Acinetobacter sp. was reported as an iron-bearer under iron deprivation conditions and induce systemic priming in Arabidopsis thaliana (Betoudji et al. 2020). Since siderophore-rich cell-free supernatants from CBIO-isolates were used, the hypothesis regarding competition for iron as a plant protection mechanism seems unlikely. However, as the cell supernatant was applied before infection with Fol59, and gene expression analysis showed differences between plants inoculated with the supernatants alone (SodSid) with respect to absolute control plants, the bioactivity could be better explained by induction of systemic resistance. In addition, the absence of antimicrobial activity was observed when CBIO117 supernatant was directly confronted against Fol, indicating that disease suppression might not be associated with the action of antimicrobial compounds.

Plants respond to biotic stresses caused by pathogens and pests through a localized defense response at the site of infection or, sometimes, by a systemic response known as SAR. However, some beneficial microorganisms that mainly inhabit the rhizosphere can induce systemic resistance in the plant to counteract phytopathogen attack and strengthen physiological processes (Romera et al. 2019). Secretion of molecules such as siderophores by beneficial microorganisms has been shown to improve plant nutrition and trigger plant defense response (Verbon et al. 2017). In this study, SodSid CBIO117 activated SA and ET signaling pathways in tomato plants, as evidenced by changes in gene expression of PR1 and ERF1. These genes did not decrease in expression after infection with the pathogen. However, early expression of PR1 and ERF1 in plants only infected with the pathogen could be associated with PAMP-triggered immunity, in which PR1 protein accumulation could potentially restrain the pathogen and induce activation of defense-related pathways (Boccardo et al. 2019). These results suggest that the application of siderophore-rich cell-free supernatants prior to pathogen infection elicits a transient systemic response to counteract plant pathogen attack via the AS and ET signaling pathways. Interestingly, no changes were detected in the expression of the MYC2 gene marker of the jasmonic acid pathway upon SodSid CBIO117 application. However, when plants previously treated with SodSid were infected with Fol59, the MYC2 gene increased expression levels. This response may be associated with an induction of priming defense, reinforcing plant defenses against pathogen attack.

In conclusion, this study demonstrates that siderophore-rich cell-free supernatants of A. baumannii CBIO117 can activate systemic resistance in tomato plants and could generate a priming-type defense against Fol. This siderophore-producing bacterium could be, in the near future, a potential biocontrol agent to control plant diseases. To elucidate the hormonal signaling pathway involved in the plant protection response, evaluation of additional genes involved in systemic resistance is recommended.