Introduction

The rhizosphere is a major site for a diversity of microbial interactions with plants. Members of the microbial community in the rhizosphere may be beneficial or non-beneficial for plant growth and development1,2,3. Some microbes (soil-borne pathogens) have deleterious effects on plant health through infection and competition with beneficial microbes and the plant for nutrients. Other microbes like plant growth promoting bacteria (PGPB) and some mycorrhizae-forming fungi promote plant growth by mitigating stresses caused by biotic and abiotic factors, promoting nutrient mobilization and enhancing yield4,5. Specific compounds in root exudates mediate signaling interactions with microbes. Metabolites including sugars, amino acids and organic acids, as well as various other carbon compounds, are delivered to the rhizosphere to support microbial growth6,7,8.

Root-colonizing rhizobacteria are reported to confer beneficial impacts on plants through direct mechanisms such as ammonia production and regulation of phytohormone levels, as well as nutrient solubilization, but indirect mechanisms like production of HCN, siderophores and antibiotics are also involved9. Many bacterial species, including those belonging to the genera Bacillus, Azotobacter, Azospirillum, Klebsiella, Pseudomonas, Alcaligenes, Burkholdeira, Arthrobacter, Serratia and Enterobacter, are reported to promote plant growth and can be categorized as plant growth-promoting bacteria (PGPB)7,9,10,11. In tomato, rhizobacteria including Bacillus spp., Azotobacter spp., Pseudomonas fluorescens, Micromonospora spp. and Serratia spp.12,13,14,15 are reported to influence plant-pathogen interactions and stimulate plant growth.

One of the most common fungal pathogens of tomato is Alternaria alternata, which causes disease characterized by the presence of early blight, stem and fruit canker16. Early blight caused by A. alternata has great economic importance globally. This disease reduces the photosynthetic surface of the leaves and, in severe cases, leads to complete plant defoliation17. According to a field survey conducted in 2016–2017 in different regions of Pakistan, A. alternata was found to be the second most common pathogen isolated from leaves and fruits of tomato, with disease incidence ranging from 9 to 75%18. The main meteorological components contributing to pathogen invasion and deterioration of a tomato crop are temperature and relative humidity. Strictly speaking, A. alternata is a necrotrophic pathogen but, in the case of severe damage to the plant, it can also be transmitted to the seeds, thereby affecting the next generation19. A. alternata is reported to produce specific fungal toxins called AALs (an abbreviation for A. alternata f. sp. lycopersici)20. These host-selective toxins (HST) lead to apoptosis followed by necrotic spots on leaf, stem and fruit21.

Biological control of plant infectious diseases prevents collateral damage to beneficial microbes caused by the use of synthetic fungicides22. Biofertilizers and biopesticides are both environmentally friendly and potent for fungal control and are reported to promote the availability of soil nutrients to plants. In this respect, the use of plant growth-promoting rhizobacteria is considered one of the most effective approaches to control pathogens23. Bacterial strains from the order Bacillales have been particularly useful24,25,26 and are commonly commercialized for these applications27. Strains of the Bacillus genus are recognized for their antagonist activity against tomato pathogens28,29. The biocontrol activity of these bacteria has been attributed to their ability to produce antibiotic compounds and to compete for space and nutrients in the rhizosphere30. Chowdappa et al.31 reported that strains of B. subtilis inhibited the growth of mycelium from A. alternata, A. solani and P. infestans in vitro. In tomato plants, the inoculants induced the production of auxins and gibberellins and increased the activities of peroxidase, polyphenol oxidase and superoxide dismutase, which contributed to systemic resistance in tomato plants against early and late blight.

In the present study, the rhizosphere of the quinoa plant (Chenopodium quinoa Willd), which belongs to the large family Amaranthaceae, was explored to identify potent biocontrol agents. Quinoa is a native crop of the Andean Plateau in west-central South America, and was introduced to Pakistan recently for its revenue-generating potential, its higher adaptability to climatic conditions and potential for improvement in the cropping system32,33,34,35. Quinoa is renowned for its ability to withstand drought and salinity36 and for its remarkable nutritional value37.

The objectives of this research were to (1) isolate and identify beneficial rhizobacteria associated with C. quinoa; (2) evaluate the growth-promoting and fungal pathogen-antagonizing potential of these rhizobacteria; (3) evaluate the biological control ability of the rhizobacteria against A. alternata in vitro and in vivo, and (4) investigate the effects of exogenous inoculations on the antioxidant defense system of tomato in the presence of A. alternata.

Materials and methods

Crop location, growth conditions and rhizosphere sampling

Cultivated Chenopodium quinoa plants along with roots and rhizospheric soil were collected and identified by the National Herbarium, Department of Plant Sciences, Quaid-i-Azam University, Islamabad. Plant collection was done from two districts of Pakistan, Khanewal (30.2864° N, 71.9320° E) and Faisalabad (31.4504° N, 73.1350° E), which are considered irrigated climatic zones because of their inclusion in the sugarcane belt. Annual precipitation of the sampling year was recorded as 202.78 mm and 347.2 mm for Khanewal and Faisalabad, respectively. The soil in the rhizosphere was characterized as a clay loam. Samples from the rhizosphere of quinoa plants were obtained at a depth of 20–30 cm. Soil and root samples weighing approximately 50 g were collected from the individual area of the field, stored in sterile plastic bags on ice, taken to the laboratory immediately, and stored at 4 °C.

The plant material collected and used in this study completely complies with institutional, national, and international guidelines and legislation regarding this type of experiment. Moreover, permissions were obtained from agricultural farms based in Khanewal and Faisalabad, for collecting and using the plant in this study.

Bacterial isolation and differentiation

To isolate rhizosphere bacteria, the soil in the immediate vicinity of the quinoa roots was carefully isolated, crushed and sieved under aseptic conditions. Serial dilutions of the soil were made using sterile 0.8% saline solutions. Aliquots (100 μL) of each serial dilution were spread on LB as well as on nutrient agar media plates and incubated at 28 ± 2 °C for 24–48 h. Colony forming units were counted on the following day and representative colonies exhibiting different morphological aspects were differentiated, spotted and isolated to form pure cultures. Bacterial cultures were differentiated and purified by observing their color, texture, margins, optical density, size and shape. Overall, 68 different colonies were isolated based on their colony morphology and were then subjected to gram staining (Supplementary Table 1).

In vitro screening for antagonistic activity

All the isolated bacterial strains were subjected to in vitro testing for their antagonistic activity against Alternaria alternata cultures obtained from the Applied Microbiology and Biotechnology Lab of COMSAT University Islamabad. The bacteria were grown for 24 h in LB and/or nutrient broth according to their isolation media at 28 ± 2 °C. An aliquot (70 µL) was taken from the overnight-grown bacterial cultures with 108 CFU/mL and spread uniformly on PDA plates. A 5-mm plug of 7 days-old A. alternata culture was placed in the middle of the plate and incubated at 30 °C for 7 days. Control plates without bacterial inoculation were also prepared. The percentage growth inhibition was measured by comparing mycelial growth with control plates and calculated using the following formula:

$$\% \;{\text{inhibition}} = \left[ {({\text{C}} - {\text{T}}) \times {1}00)/{\text{C}}} \right]$$

where C is the growth of fungal mycelia in control plate, and T is the growth of fungal mycelia in bacteria-inoculated plates.

Production of extracellular enzymes and siderophores

The selected strains based on their antagonistic potential were further assessed for their potential to produce major antifungal enzymes and siderophores. Cellulase production was detected by testing bacteria on carboxymethyl cellulose (sodium salt) (CMC) agar38, whereas detection of siderophore production was facilitated using chrome azurole sulphur (CAS) agar39. The CMC and CAS agar plates were inoculated with 24 h freshly grown cultures by the spot inoculation method and incubated at 28–30 °C for 4–7 days. Orange-colored halo zones around the bacterial colony were observed in CAS agar plates and their diameters measured in millimeters. Cellulase production was observed by staining with 0.1% congo red solution and destaining with 1 M NaCl solution.

For pectinase and amylase activities, plates containing pectin40 and 1% soluble starch41 were prepared and inoculated with 24-h bacterial cultures. After incubation for 4–7 days, the plates were flooded with 1% lugol solution (10 g KI, 5 g iodine, 100 mL distilled H2O). Decolorized clear halo zones were observed and their diameters measured in millimeters. All experiments were performed in triplicate.

In vitro screening for plant growth-promoting traits

Phosphate, potassium and zinc solubilization assays

Selected bacterial strains were further screened using nutrient solubilization assays to assess their nutrient solubilization potential. The selected strains were tested for phosphorus, zinc and potassium solubilizing assays using Pikovskaya agar42, Burnt and Rovira agar43 and Aleksandrove agar44, respectively. All the selected bacterial cultures were streaked on the respective media to obtain 24 h fresh cultures. These bacteria were then inoculated on media plates by the spot method and incubated at 28–30 °C for 4–7 days. The presence of a halo zone around the bacterial colony in the respective agar medium showed their potential to solubilize the nutrient. The zone and colony dimensions were measured in millimeters. The solubilization index was calculated using the following formula:

$${\text{Solubilizing}}\;{\text{Index}} = \left( {{\text{Colony}}\;{\text{diameter}} + {\text{Halo}}\;{\text{zone}}\;{\text{diameter}}} \right)/{\text{Colony}}\;{\text{diameter}}$$

Petri plate experiment

In vivo growth promotion activity was estimated via a petri plate experiment. Tomato seeds (Rio grande) were obtained from the National Agriculture Research Centre (NARC) in Pakistan. The selected bacterial cultures were grown in LB broth or nutrient broth according to their isolation medium. The OD of 24 h grown bacterial cultures was maintained at 1.0. The cultured broth was centrifuged at 10,000 rpm for 5 min and bacterial pellets were resuspended and dissolved in autoclaved distilled water. Tomato seeds, which were surface-sterilized with 3% sodium hypochlorite (NaOCl) solution and 70% ethanol for 1 min, separately, and washed 4–5 times in autoclaved distilled water45, were soaked in bacterial suspension for 3 h. The bacteria-treated seeds were sown in sterile petri plates fitted with a layer of cotton and sterile filter paper. The experiment was conducted in triplicate and seeds treated with distilled water were characterized as the control. Germinated seeds were observed and counted on a daily basis up to 15 days from sowing to calculate the germination percentage, promptness index, emergence index and vigor index, using the formulas below46. Difference in seedling length and fresh weight was also measured. All treatments were performed in triplicate.

$$\begin{aligned} & {\text{Germination}}\;{\text{percentage}} = {\text{Number}}\;{\text{of}}\;{\text{seeds}}\;{\text{germinated}}/{\text{Total}}\;{\text{number}}\;{\text{of}}\;{\text{seeds}} \times {1}00 \\ & {\text{Promptness}}\;{\text{index}} = {\text{nd}}_{{2}} \left( {{1}.00} \right) + {\text{nd}}_{{4}} \left( {0.{75}} \right) + {\text{nd}}_{{6}} \left( {0.{5}0} \right) + {\text{nd}}_{{8}} \left( {0.{25}} \right) \\ \end{aligned}$$

where nd2, nd4, nd6 and nd8 is the number of seeds germinated on 2nd, 4th, 6th and 8th day, respectively.

$$\begin{aligned} {\text{Emergence}}\;{\text{index}} & = \left( {{\text{number}}\;{\text{of}}\;{\text{seeds}}\;{\text{germinated}}/{\text{days}}\;{\text{of}}\;{\text{first}}\;{\text{count}}} \right) + \cdots \\ & \quad + \left( {{\text{number}}\;{\text{of}}\;{\text{seeds}}\;{\text{germinated}}/{\text{days}}\;{\text{of}}\;{\text{final}}\;{\text{count}}} \right) \\ & {\text{Vigor}}\;{\text{index}} = {\text{S}} \times \Sigma \left( {{\text{Gt}}/{\text{Dt}}} \right) \\ \end{aligned}$$

where S is the length of the seedling measured on the 7th day after germination, Gt is the overall number of germinated seeds on the ‘t’ day, and Dt is the cumulative number of days from the initial to the ‘t’ day.

Identification and phylogeny of selected PGPR

Genomic DNA of selected strains was extracted using the CTAB method47. Universal primers P1 (5′–CGGGATCCAGAGTTTGATCCTGGTCAGAACGAACGCT–3′) forward and P6 (5′–CGGGATCCTACGGCTACCTTGTTACGACTTCACCCC–3′) reverse48 were used to amplify the 16S ribosomal RNA gene. The PCR products were purified utilizing a PCR purification kit (Thermo Scientific) and sent to MCLAB (San Francisco, USA) for sequencing. The forward and reverse sequences were aligned using CLUSTALW. The aligned consensus sequences were subjected to BLAST on NCBI and EZ taxon49,50. The bacteria were identified based on their maximum sequence homology to the type strains51. After identification to check their phylogenetic relationship with other closely related strains, a phylogenetic tree was constructed using MEGA X.

Pot experiment

The properties of selected PGPRs were further investigated in a pot experiment in the shade house at COMSAT University, Islamabad. Pots (18 cm diameter, 22 cm height) were filled with autoclaved sand, soil and peat moss in a ratio of 1:2:1. Tomato seeds were surface-sterilized with 3% NaOCl and 70% ethanol and subsequently washed (3×) with sterile distilled water. The seeds were then soaked in bacterial suspension maintained at 1.5 × 108 CFU mL−1 for 3 h. Ten inoculated tomato seeds were sown in each pot. The experiment consisted of nine sets of treatments and two sets of positive and negative controls. The experiment was laid out in completely randomized design and conducted in triplicate.

Pathogen inoculation

After 45 days of emergence, the tomato plants were inoculated with the A. alternata pathogen. For this purpose, a culture of A. alternata was grown in potato dextrose broth for 7 days at 28 °C in a shaking incubator. The mycelial sheet was separated by filtration and a spore suspension maintained at 1 × 105 spores/mL by utilizing a hemocytometer. Tween 80 was added to the spore suspension at a final concentration 1% and the suspension used to both drench the soil near the plant root and spray onto the leaves52,53. The pathogen-inoculated plants were covered with polythene sheets for 24 h to maintain humidity54 and allowed to grow in a shade house at 26–28 °C during the day and 22–24 °C during the night.

Sample collection

Leaf samples were collected on 3, 6 and 9 days after inoculation (dai), preserved in liquid nitrogen and stored at − 70 °C for further analysis.

Disease scale and disease severity index

After 14 dai, plants were evaluated from all the replicates in random order to evaluate the degree of infection and the effectiveness of selected PGPR inoculations on the biocontrol of A. alternata. A disease score 0–5 was employed for the assessment of infection degree. A score of 0–1 indicated very early stage of infection with few symptoms; 1–2 indicated the appearance of brown spots and discolored regions around the spots; 2–3 indicated brown necrotic spots combined to form concentric rings, 3–4 represented an increase in the damaged area more than 50%, and 4–5 corresponded to severe damage when concentric rings join to form large patches of necrosis leading to wilting, discoloration and defoliation. Plants were evaluated according to this scale to calculate the disease severity index and percent disease reduction or disease protection for each treatment and control using the data from five replicates. The following formula was used to calculate the disease severity index as a percentage55:

$$\begin{aligned} {\text{DSI}}\;\left( \% \right) & = \left[ {{\text{sum}}\;\left( {{\text{class}}\;{\text{frequency}} \times {\text{score}}\;{\text{of}}\;{\text{rating}}\;{\text{class}}} \right)} \right]/\left[ {\left( {{\text{total}}\;{\text{number}}\;{\text{of}}} \right.} \right. \\ & \quad \left. {\left. {{\text{plants}}} \right) \times \left( {{\text{maximal}}\;{\text{disease}}\;{\text{index}}} \right)} \right] \times {1}00 \\ \end{aligned}$$

Total soluble protein content

The total soluble protein content of leaves was determined for all sampled stages (3, 6 and 9 dai)56. Leaf tissue (0.1 g) was homogenized using 1 mL of sodium phosphate buffer (pH 7.5). The homogenate was centrifuged at 4000 rpm for 10 min and 0.1 mL of the supernatant transferred to a test tube to make the volume up to 1 mL with distilled water. The test tubes were then kept on a shaker for 10 min after adding 1 mL of CuSO4 reagent. An aliquot (100 µL) of Folin & Ciocalteu phenol reagent was added and the solutions allowed to incubate at room temperature for 30 min. The absorbance at 650 nm was recorded on a UV–visible spectrophotometer against sodium phosphate buffer as blank. The concentration of protein was determined by comparison to a bovine serum albumin (BSA) standard curve.

Antioxidant enzyme activities

Leaf samples were homogenized at 4 °C to measure antioxidant enzyme activities using 0.1 M phosphate buffer (pH 6.8) for polyphenol oxidase (PPO) and peroxidase (POD), while borate buffer (pH 8.8) was used for PAL. The samples were centrifugated at 17,000g for 20 min and the supernatant characterized as enzyme extract, which was used for further reaction mixture preparations for each respective enzyme activity.

PPO and POD activities were determined using the method of Kar and Mishra57. A reaction mixture (total volume 1.5 mL) was prepared using 1.2 mL 25 mM phosphate buffer (pH 6.8), 250 µL 100 mM pyrogallol and 50 µL enzyme extract for the determination of PPO activity, while for POD activity 100 µM H2O2 was also included. The absorbance at 420 nm of triplicates from each treatment for POD and PPO activities was measured.

PAL activity was evaluated using the protocol of Peioxto58. A reaction mixture was made consisting of 200 µL 0.2 M borate buffer (pH 8.8), 200 µL of 50 mM phenylalanine and 20 µL enzyme extract. This mixture was incubated at 39 °C for 1 h. The reaction was stopped using 20 µL 6 N HCl and the absorbance at 290 nm determined.

RNA extraction, cDNA synthesis and RT-qPCR

Total RNA was extracted from leaves using trizol reagent (Invitrogen). DNA contamination was removed using RNase-free DNase (TURBO DNAfree kit, Ambion, USA). RNA was quantified using a Colibro microvolume spectrophotometer (Titertek, Brethhold). cDNA synthesis was performed using an Oligo dT Maxima H Minus Double-Stranded cDNA Synthesis Kit (Thermoscientific) with single-stranded RNA as a template following the manufacturer’s instructions. Confirmation of cDNA synthesis was checked using conventional PCR with primers for the housekeeping gene Actin. Details of primers used for PR-1, β-1,3 glucanase, chitinase and PAL genes, as the internal control, actin, are given in Supplementary Table 1. All experiments were performed in triplicate with a negative control. Quantitative real time PCR (qRT-PCR) was performed using Fast SYBR™ Green Master Mix (Thermoscientific). The reaction mixture contained 2 µL of cDNA, 1 µL of forward and reverse primer (100 pm), 5 µL of Fast SYBR™ Green Master Mix and 1 µL of nuclease free water. PCR conditions used were 95 °C for 10 min for initial denaturation and 39 cycles of 95 °C for 15 s, 58 °C for 30 s, and 72 °C for 30 s.

Statistical analysis

Data were expressed as mean ± standard error (n = 3) for each treatment. The values obtained for physiological and biochemical activities were subjected to analysis of variance (ANOVA) using Statistix v.8.1. Comparison between mean values of treatments to test significant differences at P ≤ 0.05 was made by least significant difference (Gomez and Gomez, 1984). Statistix v.8.1 was used to calculate Pearson’s correlation coefficients to determine the relationship between disease reduction and expression of each PR genes.

Results

Isolation and characterization of rhizosphere bacteria associated with C. quinoa antagonistic activity

The 68 bacteria strains isolated from the C. quinoa rhizosphere were individually tested and screened for their antagonistic ability against A. alternata in vitro. All bacterial isolates were found to be antagonistic to the tomato pathogen, exhibiting values for percent inhibition ranging from 14 to 100% (Table 1). The isolated bacteria CQ5, CQ6, CQ7 and CQ9 exhibited inhibition of A. alternata growth by 100%, followed by CQ14 and CQ17 by 98%. The isolates CQ11, CQ2, CQ3, CQ4, CQ10 and CQ1 showed 88, 87.5, 87.5, 86, 80 and 77% inhibition, respectively, against A. alternata (Table 1, Fig. 1, Supplementary Table 2).

Table 1 Screening of rhizobacteria associated with Chenopodium quinoa.
Full size table
Figure 1
figure 1

Antagonistic potential of selected bacterial strains against Alternaria alternata. Bacterial strains isolated from the rhizosphere of C. quinoa were subjected to antagonistic plate assay against A. alternata. The strains shown in this picture displayed maximum inhibition potential in addition to growth promoting activity and production of extracellular enzyme production, nutrient solubilization and siderophore production.After identification these strains were checked in pot experiment of tomato plants against A. alternata infection.

Full size image

Nutrient solubilization

Twelve rhizospheric bacteria isolates were selected to assess their plant beneficial traits. Ten isolates displayed phosphate solubilization activity. The highest solubilization index was exhibited by CQ1 (3.04) while bacterial strains CQ3, CQ5, CQ6, CQ7, CQ10, CQ11, CQ14 and CQ5 exhibited a moderate solubilization index. In contrast, the bacteria CQ9 and CQ2 displayed no phosphate solubilization.

Among all the selected bacterial isolates, CQ10 (9.33) exhibited maximum zinc solubilizing ability, while CQ7, CQ9, CQ1, CQ14, CQ3 and CQ11 exhibited moderate to low zinc solubilization (Table 1). CQ2, CQ4, CQ5 CQ6 and CQ17 displayed no zinc solubilizing ability. None of the selected strains displayed potassium solubilization activity.

Production of extracellular enzymes

The bacterial isolates were further examined for the production of extracellular enzymes (cellulase, pectinase, amylase) and siderophores. The solubilization index for cellulase enzyme production ranged from 3.0 to 6.13 for 11 of the strains while no cellulase production was detected using CQ9 (Table 1).

The solubilization index for pectinase ranged from 2.61 to 5.62 for 10 of the strains, whereas CQ9 and CQ10 displayed no pectinase activity (Table 1).

The solubilization index for amylase production ranged from 2.5 to 4.7 for CQ11, CQ5, CQ2, CQ3, CQ6 and CQ17. In contrast, no amylase production was observed in CQ1, CQ7, CQ9, CQ10 and CQ14. The solubilization index for siderophore production ranged from 2.56 to 7.56 for 11 of the strains, whereas no siderophore production was observed in CQ11 (Table 1).

Effects of selected bacterial strains on tomato plant growth in Petri plate experiment under axenic conditions

Tomato seeds inoculated with selected PGPR strains showed a higher germination percentage compared to uninoculated control seeds (Supplementary Figs. 1 and 2). The most prominent enhancements were recorded for CQ6 (36%) and CQ5 (26%), respectively, while other strains displayed increases in germination ranging from 4 to 23%, compared to control seeds (Fig. 2A).

Figure 2
figure 2

Growth promotion activity of rhizobacteria inoculated tomato seedlings in Petri plate experiment under axenic conditions (A) seedling length, (B) seedling fresh weight, (C) germination percentage, (D) seedling vigor index, (E) promptness index, (F) emergence index. Values are means of three replicates. Values with different letters are significantly different from each other at p ≤ 0.05.

Full size image

Applications of the selected PGPR significantly increased the seedling vigor index (SVI), promptness index (PI) and emergence index (EI). Bacterial isolate CQ6 exhibited the maximum increase, by 72% in SVI, 54% in PI and by 28% in EI, respectively, in comparison with the control. Eight other bacterial strains displayed 23–51% increases in SVI, 19–40% in PI and 11–20% in EI (Fig. 2B–D).

Applications of the selected isolates lead to a significant increase in tomato seedling length and fresh weight compared to the untreated control. CQ6 displayed the maximum enhancement, by 36%, in seedling length and 190% in seedling fresh weight, respectively. The other bacterial strains displayed increases by 14–29% in seedling length and 31–169% in fresh weight (Fig. 2E,F).

Identification of selected efficient bacterial strains

Based on nucleotide identity and phylogenetic analysis of near-complete 16S rRNA gene sequences, the nine most efficient antagonistic isolates were identified, of which eight belonged to the Bacillus genus and one from Enterobacter (Table 2, Fig. 3). The isolated bacteria were identified as follows (with a similarity index percentage to their respective type strains in parentheses): CQ2, B. paralicheniformis (99.6); CQ3, B. subtilis subsp. stercois (99.9); CQ4, B. glycinifermentans (99.6); CQ5, B. pumilus (99.7); CQ6, B. licheniformis (99.7); CQ10, E. hormachei subsp. hoffmanii (99.6); CQ11, B. tequilensis (100); CQ14, B. subtilis subsp. spizinenii (100), and CQ17, B. sonorensis (99.7).

Table 2 Homology of rhizobacterial strains isolated.
Full size table
Figure 3
figure 3

Phylogenetic tree of selected nine bacterial strains inferred using the Neighbor-Joining method. The optimal tree with the sum of branch lengths = 0.35042803 is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches. The evolutionary distances were computed using the Maximum Composite Likelihood method and are in the units of the number of base substitutions per site. This analysis involved 32 nucleotide sequences. All ambiguous positions were removed for each sequence pair (pairwise deletion option). There were a total of 1469 positions in the final dataset. Evolutionary analyses were conducted in MEGA X The red colored names are the strains isolated from the rhizosphere of C. quinoa in this experiment.

Full size image

Effects of selected bacterial strains on growth of tomato plants infected with A. alternata in pot experiment

From the in vitro experiment, nine bacterial strains were selected based on their antagonistic and plant growth-promoting potential for an in-planta experiment. In this pot experiment, A. alternata infection significantly reduced all the growth parameters of the tomato plants compared to the uninfected healthy controls. In contrast, plants inoculated with selected bacterial strains along with pathogen infection showed significantly higher root and shoot lengths, as well as root and shoot fresh and dry weights compared to the plants inoculated with A. alternata alone (Table 3). A. alternata infection reduced tomato root and shoot lengths by 28 and 17%, respectively, compared to the uninfected untreated plants. However, all the plants with bacterial inoculations exhibited significant enhancements in root length ranging from 10 to 263% compared to the uninoculated infected control. Among all the selected PGPRs, B. licheniformis, B. pumilus and B. subtilis subsp. spizizenii displayed maximum increases in root length by 263, 245 and 245% and in shoot length by 34, 28 and 30%, respectively, compared to the infected plants without bacterial treatment (Table 3).

Table 3 Effect of selected bacterial strains on tomato plant growth following A. alternata infection.
Full size table

Alternaria alternata infections significantly reduced root fresh weight (RFW) and root dry weight (RDW) by 243 and 283% and shoot fresh weight (SFW) and shoot dry weight (SDW) by 42 and 146%, respectively, compared to the uninfected control. All the selected PGPRs combined with A. alternata infections significantly increased root fresh and dry weights ranging from 51 to 87% and shoot fresh and dry weights ranging from 10 to 85% compared to the untreated infected control. Among all PGPR inoculations, B. licheniformis, B. pumilus and B. subtilis subsp. spizizenii exhibited maximum increases by 85, 84 and 83% in RFW, 84, 83 and 83% increase in RDW, 80, 79 and 73% increase in SFW and 87, 85 and 84% increase in SDW, respectively (Table 3).

Disease severity index and disease protection

All the selected PGPR strains significantly reduced disease severity index and showed significant disease protection against A. alternata infection (Fig. 4A,B). Minimum disease severity index and maximum disease reduction/protection was observed in B. licheniformis (92%) and B. pumilus (90%) followed by B. subtilis subsp. spizizenii (83%), B. sonorensis (76%), B. paralicheniformis (66%), B. subtilis subsp. stercois (61%), B. glycinifermentans (56%), B. tequilensis (56%) and E. hormachie (51%) compared to the untreated infected control.

Figure 4
figure 4

Rhizobacteria-mediated disease protection and induction of antioxidant enzymes in tomato infected with Alternaria alternata, (A) Disease severity index, (B) Disease protection, (C) Total soluble protein content (D) Polyphenol oxidase (PPO) activity, (E) Peroxidase (POD) activity, (F) Phenylalanine ammonia lyase (PAL) activity. All values are mean of three replicates. Values with different letters are significantly different from each other at p ≤ 0.05.

Full size image

Total soluble protein content

Fungal infection enhanced total soluble protein content by 51% at 6 dai compared to the uninfected control. All the PGPR-inoculated plants exhibited increases in protein content compared to the untreated A. alternata-infected control. For the PGPR treatments the maximum increases in protein content were observed in B. licheniformis and B. pumilus by 53 and 43%, respectively, while B. subtilis subsp. spizizenii, B. sonorensis, B. paralicheniformis, B. subtilis subsp. stercois, B. glycinifermentans, B. tequilensis and E. hormachie subsp. hoffmanii exhibited 5–28% increases (Fig. 4C).

Induction of antioxidant defense enzymes

Application of selected PGPR strains induced antioxidant defense enzyme activities in A. alternata-infected tomato plants compared to the untreated infected control.

Polyphenol oxidase (PPO) was higher in fungal infected plants by 16% compared to the uninfected control. The selected PGPR strains exhibited greater PPO activity compared to untreated infected control (Fig. 4D). The maximum increase in PPO activity was observed in B. licheniformis (17%).

An increase in peroxidase (POD) activity by 45% was observed in A. alternata-infected plants as compared to the uninfected control while PGPR applications exhibited even greater enhancement of POD activity compared to untreated fungal infected plants. The maximum increases were observed in plants treated with B. licheniformis and B. pumilus (59 and 58%, respectively), followed by B. tequilensis and E. hormachie (40 and 39%, respectively) compared to the untreated infected control (Fig. 4E).

Alternaria alternata infections enhanced PAL activity by 8% as compared to healthy plants while PGPR applications following A. alternata infections increased PAL activity compared to the fungal infected control. Maximum increase in PAL activity was observed in B. licheniformis, as well as B. pumilus and B. subtilis subsp. spizizenii by 95 and 94%, respectively, compared to the untreated infected control (Fig. 4F).

Differential expression of defense-related genes

Upregulation of all the selected defense related genes after fungal inoculation was observed as compared to healthy uninoculated control; however, bacterial application further increased the expression of defense-related genes. The expression level of the acidic chitinase gene was enhanced at 24 h after inoculation and reached a maximum at 48 h with 1.5-fold change with a slight decrease at 72 h. All bacterial treatments further enhanced expression of the acidic chitinase gene with a maximum value of 3.7 after 48 h of infection and maximum fold change was observed by B. pumilus and B. licheniformis (Fig. 5A).

Figure 5
figure 5

Rhizobacteria-mediated differential expression of defense-related genes in tomato infected with Alternaria alternata. All values are means of three replicates.

Full size image

PR-1 and PAL genes were upregulated by disease induction with a fold change of 1.6 and 1.0 after 24 h of infection. At 48 h maximum expression was observed with a slight decrease at 72 h; however, most of bacterial treatments further enhanced the expression of these genes. Maximum expression was observed by the application of B. pumilus and B. licheniformis with a fold change of 6.2, 7.3 in PR-1 expression and 7.5, 8.1 in PAL expression (Fig. 5C,D).

Alternaria alternata infection also upregulated the expression of the β-1,3 glucanase gene with a fold change of 1.53 at 24 h with a subsequent decrease in expression at 48 and 72 h. B. pumilus and B. paralicheniformis further enhanced the expression of the β-1,3 glucanase gene with a maximum value of fold change of 4.4 and 3.9 at 24 h post inoculation. A subsequent decrease in expression of the β-1,3 glucanase gene was observed at 48 and 72 h in all the treatments (Fig. 5B).

Gene expression analysis of chitinase, PR-1 and PAL was greatly upregulated at 48 h of A. alternata infection in the PGPR inoculated tomato plants while the expression of β-1,3 glucanase was found to be maximum at 24 h of infection. Our results revealed a positive correlation between the expression of these particular genes at 24, 48 and 72 h of infection and significant reduction in A. alternata disease incidence (Table 4).

Table 4 Pearson correlation coefficients for A. alternata disease suppression in tomato plant and expression of defense related genes. Means labeled with ***showed significant positive correlation at p ≤ 0.001, with ** at p ≤ 0.01, and with * at p ≤ 0.05.
Full size table

Discussion

The use of rhizosphere bacteria to control pathogens and promote plant growth is advocated in many reports. PGPR help host plants to suppress pathogens as well as absorb mineral nutrients by making them bioavailable59,60. PGPR can strengthen plant defense by initiation of induced systemic resistance61. The present study focused on the isolation, screening and characterization of rhizospheric bacteria isolated from quinoa (Chenopodium quinoa). This ancient grain crop is differentiated from other plants in terms of abiotic stress resistance due to its ability to grow in saline, arid and water-deficient conditions, as well as the regeneration potential of its broken seeds62. Microbes belonging to the genus Bacillus are known for their vertical transmission from one generation to the other; thus, it can be assumed that the presence of specific microbial diversity associated with quinoa may be a factor in its survival.

In this study, bacterial isolates were screened based on their antifungal potential and nutrient solubilization for plant growth promotion. The screened bacteria were able to solubilize phosphorus and zinc but no potassium solubilization potential was observed in any strain. PGPR are reported to solubilize minerals by liberating organic acids; i.e., gluconic acid, lactic acid, oxalic acid and citric acid63,64. These organic acids are responsible for decreasing soil pH, which favors the solubilization of nutrients like zinc and phosphorus, but makes potassium less available65,66,67. The reason for the lack of potassium solubilization observed here may be due to an inate inability of the bacteria to solubilize this nutrient or a pH effect, but further studies are needed to understand this in detail.

Twelve bacteria among the isolated strains effectively inhibited fungal mycelial growth, which may have been due to production of hydrolytic enzymes and siderophores by these strains. Pectinase, cellulase, amylase and siderophore production was observed in these selected strains, which likely enabled the bacteria to inhibit the radial growth of the A. alternata mycelia. Previous studies have reported that biocontrol agents are considered good PGPR68,69. Proteases, cellulases and other bacterial enzymes secreted exogenously have a role in promoting soil fertility and control of phytopathogen invasion by the mechanism of disintegration of the pathogen membrane (fungal cell wall). Siderophores are key compounds produced by a diverse group of antagonistic PGPR such as Bacillus, Azotobactor, Rhizobium and Pseudomonas70,71,72. They impede the proliferation of phytopathogens by sequestering Fe3+ in the rhizosphere, making it unavailable to pathogens15. Studies have reported that the siderophores produced by biocontrol Bacillus stains control diseases caused by R. solani, F. oxysporum, S. rolfsii, F. solani and G. graminis var. tritici and stimulate plant growth73,74,75,76.

Results obtained from in vitro experiments under axenic conditions revealed the growth-promoting properties of the 12 strains selected above. These bacteria improved the germination percentage, seedling vigor index and emergence index of tomato plants as compared to the control (Fig. 1). Similar findings were reported by different researchers in various crops such as maize77,78, wheat79,80, canola81 and sorghum82. The improved growth parameters associated with the bacterial seed treatment may have been due to increased production of phytohormones, such as gibberellins, which promote the synthesis of specific enzymes like α-amylase, which is partly responsible for starch breakdown and plays a role in early growth promotion77. Similarly, seedling vigor index improvement may be due to an increase in auxin synthesis77,83.

PGPRs have been isolated from the rhizosphere of several plants, including maize84, wheat85 and rice86. However, only a few studies have reported on the PGPR associated with quinoa roots87,88. Here, we isolated native PGPR associated with quinoa roots from soils in Pakistan and assessed their role in tomato plant growth and disease resistance. Out of 68 isolates, nine strains (CQ2, CQ3, CQ4, CQ5, CQ6, CQ10, CQ11, CQ14 and CQ17) were characterized for their potential PGPR properties. Genotyping revealed that these strains corresponded to B. paralicheniformis, B. subtilis subsp. stercois, B. glycinifermentans, B. pumilus, B. licheniformis, E. hormachei subsp. hoffmanii, B. tequilensis, B. subtilis subsp. spizinenii and B. sonorensis, respectively.

To resist fungal infections, plants naturally trigger several physiological responses including phytohormone synthesis, nutrient uptake regulation and the production of enzymes for antioxidant defense. Rhizosphere bacteria are known to confer resistance and protection against various plant pathogens89; however, very few studies have reported biocontrol agents having antagonistic potential along with plant growth promotion ability. The present study was, therefore, designed to explore the potential of rhizosphere bacteria to control the A. alternata pathogen with concurrent stimulatory effects on plant growth. A biocontrol agent having broad-spectrum antagonistic as well as growth promotion ability is more promising under field conditions90,91.

Results obtained from the pot experiment here validated the innate biocontrol and biofertilizer potential of selected nine bacterial strains identified. In planta evaluation showed that tomato plants infected with A. alternata and treated with B. licheniformis and B. pumilus showed the greatest disease protection and biomass. This response may be attributed to the multifarious potential of both the bacterial agents that combined disease control with plant growth promotion. PGPR application triggered plant immunity through induced systemic resistance (ISR) as this successfully controlled leaf blight disease caused by A. alternata in the tomato plants. Disease control activity of these strains may be attributed to the production of antifungal metabolites and the collective action of lytic enzymes and siderophore production. Increased production of these protective molecules by PGPR can result in an enhanced nutrient uptake by host plants infected with a fungal pathogen69,71.

B. licheniformis and B. pumilus inoculations exhibited increases in total soluble protein content and antioxidant defense enzyme activities including POD, PPO and PAL. Increased production of protein supports the increased activity of plant defense mechanisms in generating stress-related enzymes92. Increases in PPO and POD activities triggers the biosynthesis of phenolic compounds, which leads to the strengthening of plant defense against pathogen attack, while PAL promotes the activation of phenylpropanoid mechanism to induce the production of phytoalexins, resulting in cell wall lignification and suppression of pathogen attack93. Many researchers have reported the ISR potential of PGPRs through increased activities of defense-related enzymes against fungal attack69,92,93,94.

Alternaria alternata infection in tomato plants induced the expression of pathogenesis-related proteins including chitinase, β-1,3 glucanase, PR-1 and PAL genes. We found a significant increase in expression of all studied genes after fungal infection at 24 h post inoculation. Transcript abundance continued to increase at 48 h (with the exception β-1,3 glucanase), while after 72 h the expression of all genes started to decline. The PR proteins are small, mostly acidic, resistant to breakdown and most commonly found in intercellular spaces. The activation of these genes is reported to prevent the progress of damage in the plant tissues95. The biochemical products of these genes result in activation of defense-related pathways as well as degradation of pathogen cell walls96. However, overexpression of the PAL gene is also linked to the strengthening of the plant defense barrier by lignification. Our results showed that tomato plants inoculated with the bacteria B. pumilus and B. paralicheniformis significantly enhanced the gene expression of chitinase, β-1,3 glucanase, PR-1 and PAL genes at 24, 48 and 72 h post inoculation as compared to untreated infected control. The expression of β-1,3 glucanase has been reported to be maximum at 24 h in wheat infected with Puccinia triticina96 while maximum expression of PR-proteins, chitinase and PAL has been reported maximum at 48 h post inoculation97.

In conclusion, this study adds support to the application of PGPR as a valuable strategy for the improvement of agricultural productivity through enhancement of plant defense against fungal pathogens.