Introduction

Utilization of plant growth promoting rhizobacteria (PGPR) is an eco-friendly method adopted for crop improvement and disease control in modern agricultural practices. The strains belonging to the genus of Bacillus and Pseudomonas have been mostly utilized for biomass improvement, yield enhancement and management of diseases in many agriculturally important crop plants (Compant et al. 2005; Cook 2007; Dutta et al. 2008; Dutta and Podile 2010: Bhattacharyya and Jha 2012; Perez et al. 2012) Production of secondary metabolites with antibiotic activity in several strains of fluorescent pseudomonads has been recognized as a major factor in suppression of root pathogens (Dwivedi and Johri 2003). Similarly, certain compounds produced by bacilli exhibited inhibitory activities against several plant pathogens (GaoFen et al. 2004; Chaurasia et al. 2005). The role of siderophores and phenazine type antibiotics produced by the fluorescent Pseudomonas species in growth promotion and/or disease control been reported (Defago and Haas 1990; Sharma and Johri 2003). However, the utilization of the metabolites produced by the PGPR strains has not attained the momentum, as it deserves.

Vascular wilt of pigeon pea, caused by Fusarium udum, is an agriculturally important disease in many parts of the world, particularly in Asian and African continents (Kiprop et al. 2002). Efficient high-yielding, disease resistant varieties are still lacking to tackle this problem (Reddy et al. 1990). The pathogen is soil borne and chemical control is impractical in established cases. Several cultural practices are recommended but the disease still poses a serious threat to this important pulse crop.

In this study, we examined the efficacy of three metabolites produced by a Pseudomonas aeruginosa strain RRLJ 04 and a Bacillus cereus strain BS 03 on growth promotion and fusarial wilt control in pigeon pea.

Materials and methods

Soil

The experiments were conducted on sandy–loam soil with pH 6.10, total nitrogen 0.015 %, organic carbon 0.268 % and have no previous history of any pesticide or synthetic agrochemical application.

Organisms

Two bacterial strains, P. aeruginosa RRLJ 04 (MTCC 7277) and a B. cereus strain BS 03 (MTCC 7278), which showed plant growth promotion and suppression of fusarial wilt in pigeon pea on our earlier investigation (Dutta et al. 2005), was selected for this study. The confirmatory identification of the strains was done from the Microbial Type Culture Collection and Gene Bank (MTCC) Division of Institute of Microbial Technology (CSIR), Chandigarh, India. The fungal pathogen F. udum was procured from International Crop Research Institute for Semi-Arid Tropics (ICRISAT), Hyderabad, India.

Extraction of bioactive metabolites from pseudomonad strain RRLJ O4

The bioactive metabolites produced by strain RRLJ 04 were extracted according to Dileep Kumar and Bezbaruah (1997). For this, a lawn of RRLJ 04 was grown in King’s B media for 7 days at 28 ± 2 °C. Then the bluish-green bacterial grown medium was cut into small pieces (≈1 cm2) and extracted with 80 % aqueous acetone till the medium turned into colourless. The extract was then filtered through a double layer of cheese cloth and the acetone was removed under vaccum pressure through a rota evaporator. The resultant aqueous part was then treated with sodium chloride (50 g l−1) and centrifuged at 11,000 rev min−1 for 20 min. The supernatant was collected and further extracted with diethyl ether (3:1 v/v) in a separating funnel. The resultant aqueous fraction then treated with chloroform (3:1 v/v) and the chloroform fraction was collected. The yellowish diethyl ether (BM 1) and the bluish green chloroform fractions (BM 2) were concentrated to get dried extracts.

Extraction of bioactive metabolites from bacilli strain BS 03

The bioactive compounds from BS 03 were extracted according to Kajimura et al. (1995). The strain was grown in potato dextrose broth (PDB) for 48 h was centrifuged at 8,000 rev min−1 and the supernatant was collected and extracted twice with equal volumes of n-butanol. The n-butanol extracts were combined and concentrated to give an oily residue. The residue was added to 20 ml of silica gel and slurried with methanol. After evaporating the methanol, the resulting powder was applied on top of a silica gel column (1.5 × 25 cm) that had been pre-packed with chloroform–methanol (3:1 v/v). The column was eluted stepwise with mixtures of chloroform–methanol from 3:1 to 1:1. Elutes thus obtained were combined (BM 3) and concentrated to get crude powder.

Characterization of bioactive metabolites

Crude extract of BM 1, BM 2 and BM 3 were further purified by thin layer chromatography (TLC) and the purified band were subjected to IR, 1H NMR, 13C NMR and mass spectrum studies for characterization studies.

In vitro antibiosis studies with bioactive metabolites

The experiments were conducted on three different media viz. Potato dextrose agar (PDA), Nutrient agar (NA) and King’s B medium (KB). For this, an actively growing mycelial disc (≈6 mm2) of F. udum was placed on the center of a Petri plate containing respective media. Sterilized filter paper discs (5 mm diameter) were impregnated with 50 μl of different concentrations of metabolites (0, 10, 20 mg bioactive metabolite in 100 ml−1 methanol) and two such discs were placed on either side of the mycelial disc at equal distances. Sterilized filter paper disc dipped in only methanol served as control. The plates were then incubated at 28 ± 2 °C and inhibition zone was measured as distance (in cm) between the filter paper disc and mycelium of the fungus after 7 days of growth.

Growth of the pathogen in media amended with different concentrations of bioactive metabolites

Stock solutions of BM 1, BM 2 and BM 3 were prepared to give 5 % solution in sterile distilled water. PDA and PDB amended with different concentrations of respective bioactive metabolites (0, 5, 10, 20, 30, 40 and 100 μl stock solution in 100 ml−1 media) were introduced with a mycelial disc (≈6 mm2) of F. udum and the radial growth in PDA and dry weight of mycelium in PDB were recorded after 7 days of growth.

Effect of bioactive metabolites under gnotobiotic condition

The gnotobiotic system consisted of a glass tube of 26 cm long and 1.5 cm in diameter, connected with a 100 ml flask at the bottom. The tube was filled with sand up to 12 cm by securing it with a piece of muslin cloth at the bottom of the tube. The tube was then fixed tightly to a conical flask containing Hoffland’s Plant Nutrient Solution (PNS) consisting 5 mM Ca(NO3)2; 5 mM KNO3; 2 mM MgSO4.7H20; 1 mM KH2PO4 and micronutrient (g l−1) MnSO4—0.61; ZnSO4—0.50; H3BO3—1.27; Na2MoO4—0.40; CuSO4—0.20. The bottom portion of the tube was immersed in the nutrient solution to keep the sand column moist. The whole system was autoclaved at 120 °C for 15 min before the use.

Optimization of the concentration of the metabolites was done by treating the seeds with 10, 20, 30, 40 and 50 μl of bioactive metabolite stock solution (20 mg 100 ml−1) and placing in Petri dishes containing sterile filter paper soaked with PNS. The concentration at which the seeds were able to survive and showed emergence of shoot and root were taken as the optimum concentration. Seeds with optimum concentration of respective treatments (BM 1, BM 2 and BM 3) were aseptically transferred approximately 5 mm below the surface of the sand column. For disease control studies, 2.0 ml of the conidial suspension (containing 1.0 × 107 conidia ml−1) was given in the sand column. The treatments taken were as below:

  1. 1.

    Control

  2. 2.

    BM 1

  3. 3.

    BM 2

  4. 4.

    BM 3

  5. 5.

    F. udum

  6. 6.

    BM 1 + F. udum

  7. 7.

    BM 2 + F. udum

  8. 8.

    BM 3 + F. udum

Shoot height, root length, fresh and dry weight of shoot and root were recorded after 15 days of growth.

Effect of bioactive metabolites under nursery condition

Pigeon pea seeds treated with optimum concentration of bioactive metabolites were sown in earthen pots (size 21 × 15 mm) filled with a mixture of soil and Farm Yard Manure (FYM) in 3:1 ratio. Treatments were initially optimized by placing surface sterilized seeds treated with different concentrations of bioactive metabolites in earthen pots (Size 21 × 15 mm) with sand. The concentrations used were 0, 10, 20, 30, 40, 50, 60, 70, 80 and 100 μl of bioactive metabolite stock solutions (20 mg 100 ml−1). The concentration in which the plants showed better growth as compared to only PNS (control) treated was taken as the optimum concentration in all the cases. For disease suppression studies, seeds were treated with different concentrations of bioactive metabolites, as given above, and placed in pathogen infested soil (2 ml containing 1.0 × 107 conidia ml−1). The concentration which showed better growth in comparison to only pathogen treatment and control was taken for further studies for each bioactive fraction. The treatments taken under gnotobiotic condition were also repeated in nursery condition. Observation for growth promotion and disease symptom development were noted up to 90 days of growth. Plants were carefully uprooted after 90 days of growth and shoot height, root length, fresh and dry weight of plants were recorded. Per cent disease incidence (PDI) and per cent disease control (PDC) were also calculated on 90 days after sowing (DAS).

Effect of bioactive metabolites on mycelium morphology and dry weight

For this, 1 ml of respective stock solution of bioactive metabolites (BM 1, BM 2 and BM 3) was introduced to the PDB medium growing F. udum at 0, 12 and 24 h. Dry weight of the mycelium was measured after 48 h of pathogen growth. Any morphological change of the mycelium was observed through a Leitz photomicroscope for 0–96 h at an interval of 6 h.

Statistical analyses

All the data obtained were subjected Duncan’s Multiple Range Test (DMRT) using a software programme with the help of a computer.

Results

Bioactive compounds from RRLJ 04 and BS 03

The strain RRLJ 04 produced approximately 0.15 g l−1 of crude BM 1. TLC studies showed that BM 1 consisted of three bands of which a yellowish compound was stable at room temperature. The IR spectrum of this compound showed absorption peaks at νmax 3,070, 2,959.5, 2,927.2 and 2,857.1 cm−1. It also showed strong absorption peaks at νmax 1,727.9, 1,275 cm−1 indicating the presence of ester group. 1HNMR (300 MHz): d 0.897–0.999 (m, 7′–11′–15′ CH3, m, 24H), 1.25–1.32 (m, 2′, 3′, 4′, 5′, 6′, 8′, 9′, 10′, 12′, 13′, 14′, 44H), 1.68 (m, 7′–11′–15′ CH, 6H), 4.2–4.23 (M, 1′ CH2, 4H), 7.5–7.7 (M, 6H) (Supplementary Fig. 1). Mass: m/z 804.0. The 1H NMR and mass spectrum studies of this compound confirmed that it is a symmetric phenazine 1, 6-dicarboxylate derivative (Fig. 1a).

Fig. 1
figure 1

Structure of metabolites [A] symmetric phenazine 1,6-dicarboxylate derivative and [B] ether derivative of 1-hydroxy phenazine

Full size image

The strain RRLJ 04 also produced approximately 0.50 g l−1 of crude BM 2 fraction. From the TLC studies of BM 2, it was found to be a mixture of three prominent compounds—yellow, green and blue colored of which blue compound was found to be unstable. The blue compound eventually decomposes into the yellow and green ones even at room temperature or in presence of light. The IR spectrum of yellow compound showed characteristic bands of phenazine at νmax 3,063.8, 2,960, 2,925.8 and 2,854.2 cm−1. 1HNMR (300 MHz): d 0.799–0.866 (m, 3′–7′–1′1 CH3, 12H), 1.172–1.259 (M, 2′–4′–5′–6′–8′–9′–10′, CH2, 14H), 4,12–4.16(M, 1′, CH2, 2H), 7.14–8.20 (m, 7H). 13CNMR (75 MHz):d 10.91, 14.2, 14.09, 22.64, 22.95, 23.67, 28.86, 29.33, 29.66, 30.14, 30.29, 31.38, 38.66, 68.1, 108.83, 119.98, 128.74, 129.61, 131.798, 141.15, 143.74, 144.09, 151.64 (Supplementary Fig 2–3). Mass: m/z 405.1. Considering the 1H NMR, 13C NMR and mass spectrum of this compound, it was confirmed to be an ether derivative of 1-hydroxy phenazine (Fig. 1b).

The strain BS 03 produced approximately 0.35 g l−1 of crude BM 3. Characterization of this compound was unsuccessful.

In vitro antibiosis studies with bioactive metabolites

The data recorded confirmed that the bioactive fractions obtained from both RRLJ 04 and BS 03 was able to control the growth of the test fungal pathogen in all three media tested. Among the metabolite fractions tested, BM 2 gave a better zone of inhibition (Table 1). PDA showed the best inhibition followed by NA and KB. In all cases, inhibition zone increased with increase in concentration of the bioactive metabolites.

Table 1 In vitro antibiosis of bioactive metabolites against F. udum
Full size table

Growth of the pathogen in media amended with different concentrations of bioactive metabolites

A gradual decrease in the mycelial growth, in terms of diameter in PDA and dry weight in PDB was observed with increase in concentration of respective bioactive metabolites. BM 2 gave the best inhibition as compared to other fractions at lower concentrations (Table 2). It is evident from the data that for BM 1, minimum of 10 μl 100 ml−1 media was required to inhibit the growth of F. udum whereas, inhibition was recorded at 5 μl 100 ml−1 media of BM 2. For BM 3, concentration below 30 μl 100 ml−1 media did not have any effect on the growth of the pathogen.

Table 2 Effect of different concentrations of bioactive metabolites on growth of F. udum in culture media
Full size table

Gnotobiotic studies with bioactive metabolites

The recorded data confirmed that high concentration of BM 1 and BM 2 were detrimental to the plant growth. Combination of both the fractions did not have significant effect on the growth. BM 1 (20 μl), BM 2 (10 μl) and BM 3 (30 μl) when applied to the seed showed emergence of shoot and root.

In soil without pathogen, highest shoot height and root length were given by BM 1 but it was statistically at par with BM 2 and BM 3 (Table 3; Fig. 2). Fresh weight was found to be statistically at par in BM 1 and BM 2 and dry weight was statistically higher in BM 1 than in BM 2.

Table 3 Effect of bioactive metabolites on growth under gnotobiotic condition (after 15 days)
Full size table
Fig. 2
figure 2

Effect of different concentrations of bioactive metabolites on growth under gnotobiotic condition (after 7 days)

Full size image

In pathogen infested soil, it was found that BM 2 was better than the other two fractions in terms of growth (Table 3). Lowest shoot height, root length, fresh and dry weight of plants were recorded in pathogen alone treated plants.

Optimization of concentration required for nursery studies

It was recorded that 40 μl seed−1 of BM 1 and BM 2, respectively and 60 μl seed−1 of BM 3 fraction were optimum for growth promotion (Table 4). Concentrations above this were found detrimental to the plants whereas concentrations below this did not have any significant effect on growth improvement. Among all the concentrations tested from all the fractions, highest shoot height, fresh and dry weight of plant were recorded in 40 μl seed−1 of BM 1 followed by 40 μl seed−1 of BM 2. Root length was highest in 40 μl seed−1 of BM 2 followed by 40 μl seed−1 of BM 1.

Table 4 Effect of different concentrations of bioactive metabolites on growth in small pots under laboratory condition (after 7 days)
Full size table

In pathogen infested soil, plants in the treatment 50 μl seed−1 of BM 1 and 30 μl seed−1 BM 2 gave better results than the other concentrations. Higher concentrations than these suppressed the effect of the pathogen but growth of the plants were also retarded. 70 μl seed−1 of BM 3 was found to be optimum for growth in pathogen infested soil for further studies (Table 5). Beyond this concentration, seeds did not germinate. Shoot height, fresh and dry weight of plants were found to be highest in control followed by 30 μl seed−1 of BM 2. Root length was found to be maximized in control but it was statistically at par with 50 μl seed−1 of BM 1, 30 and 40 μl seed−1 of BM 2, 40, 50, 60 and 70 μl seed−1 of BM 3.

Table 5 Effect of different concentrations of bioactive metabolites on suppression of pathogen in small under laboratory condition (after 7 days)
Full size table

Effect of bioactive metabolites on growth under nursery condition

Under nursery condition, it was recorded that plants treated with BM 1 gave the highest increase in growth in terms of shoot height, root length, fresh and dry weight of plants. In pathogen infested soil, plants with BM 2 treatment showed a better enhancement in growth as compared to other treatments (Table 6). But all the treatments were found to be statistically significant over control. In terms of disease control, it was observed that at 90 DAS, per cent disease control was 90 % in BM 2 treated plants whereas it was 87 and 83 % in BM 1 and BM 3 treated plants, respectively (Table 7).

Table 6 Effect of different bioactive metabolites on growth under nursery condition at 90 DAS
Full size table
Table 7 Effect of different treatments of bioactive metabolites on PDI and PDC at 90 DAS
Full size table

Effect of bioactive metabolites on mycelial morphology and dry weight of mycelium

Among the bioactive metabolites, lowest dry weight was recorded in F. udum + BM 2 (46.25) and highest dry weight in F. udum alone (126.79) which were found to be statistically at par with F. udum + BM 3 (Table 8).

Table 8 Effect of bioactive metabolite on dry weight of mycelium
Full size table

It was recorded that the dry weight of the mycelium was significantly reduced on introduction of BM 1 and BM 2 to the fungal culture media. BM 3 gave a statistically significant reduction in dry weight of the mycelium, when simultaneously introduced together with the fungus (0 h). Introduction of BM 3 after 12 or 24 h of fungal growth did not have any significantly effect on the dry weight of mycelium.

When BM 3 was simultaneously introduced with the fungus (0 h), chlamydospore-like structures developed in mycelium after 18 h of growth. Introduction of BM 1 and BM 2 gave no such structures at any growth period (Fig. 3).

Fig. 3
figure 3

Morphological changes in F. udum mycelium after treatment with bioactive metabolites. Arrow marks indicate the formation of chlamydospore-like structures

Full size image

Discussion

Production of antibiotics in several strains of fluorescent pseudomonads has been recognized as a key factor in suppression of root pathogens (Lugtenberg and Kamilova, 2009; Saharan and Nehra 2011; Perez et al. 2012). Bakker et al. (2003) reported that antibiotics do have direct effects on plants and, therefore, might induce systemic resistance. The bioactive fractions BM 1 and BM 2 isolated from P. aeruginosa strain RRLJ 04 showed inhibition of the pathogen in antibiosis studies. Earlier, we have reported the plant growth promoting and disease control properties of an antibiotic and siderophore producing fluorescent Pseudomonas strain in some important crop plants (Dileep Kumar and Bezbaruah 1997). Seeds treated with respective fractions showed an enhancement in growth besides suppressing the wilt disease in pathogen infested soil. The growth improvement and disease suppression due to the presence of the bioactive fractions was confirmed by the gnotobiotic studies. Analytical studies of the bioactive compounds showed that these are phenazine derivatives.

Growth enhancement and disease suppression in seeds treated with bioactive fraction from BS 03 was also observed in pathogen infested soil. It has been reported that the addition of B. coagulans to culture of fungi resulted in inhibition of ergosterol biosynthesis in mycelium (Czaczyk et al. 2002). Some of the antibiotics produced by Bacillus (iturin and surfactin) are able to modify bacterial surface hydrophobicity and, consequently, microbial adhesion to surfaces of mycelium (Ahimou et al. 2000). Moyne et al. (2001) identified an antifungal peptide Bacillomycin D from a B. subtilis strain AU195. Three isomers of iturin A, a cyclic lipopeptide antibiotic produced by B. subtilis, was also found to be produced by a B. amyloliquefaciens strain B94 (Yu et al. 2002). The strain was used as a biocontrol agent to suppress R. solani and other fungal plant pathogens. GaoFen et al. (2004) reported the production of antifungal substance from B. cereus strain, which exhibited inhibitory activities against F. oxy. f. sp. cucumerinum, F. oxy. f. sp. niveum and F. oxy. f. sp. vasinfectum. Sadfi et al. (2002) reported that a B. cereus strain X16 produced more than one antifungal metabolite.

An enhancement in growth as compared to control was observed when respective bioactive fraction treated seeds were sown in soil without any pathogen. This improvement in growth may be attributed to successful colonization of beneficial microbes to the host roots. Phenazine plays an important role in microbial competition in the rhizosphere, including survival and competence (Mazzola et al. 1992). Fernando and Pierson (1999) reported three phenazine antibiotics to be responsible for the ability of a P. aureofaciens strain 30–84 to compete with the indigenous microflora in the rhizosphere.

Growth enhancement may also result due to increase in nodulation by rhizobia thereby increasing nitrogenase activity (Alagawadi and Gaur 1988; Zhang et al. 1997). Rhizobium strains face stiff competition from the rhizosphere microorganisms, like deleterious microbes, to colonize the roots. Phenazine and other antibiotics produced by pseudomonads and bacilli may help in suppressing these deleterious microbes facilitating successful nodulation by the rhizobial strains (Kloepper et al. 1991).

It has been observed that two bioactive fractions from RRLJ 04 affected the plant growth differently in presence and absence of pathogen. While BM 1 showed better growth in absence of pathogen, BM 2 showed a slightly better result in terms of growth and disease control in presence of the pathogen. Several abiotic factors such as oxygen, temperature, specific carbon and nitrogen sources and microelements have been identified to influence antibiotic production by bacterial biological control agents. Nutritional inadequacy has been found to be the main constraint in antibiotic production and activity. Among the biotic factors that may play a determinative role in antibiotic production are the plant host, the pathogen, the indigenous microflora and the cell density of the producing strain (Raaijmakers et al. 2002). Moreover, in many biocontrol systems, multiple factors have been shown to play a role in disease suppression. Kumar et al. (2005) reported the production of phenazine-1-carboxamide and IAA by a saprophytic P. aeruginosa strain PUPa3 which also produced siderophores, growth hormones, protease and phosphatase and has fungicidal properties. The strain RRLJ 04 is also found to produce siderophore. Disease suppression is a multifunctional attribute and hence it is possible that none of the mechanisms are mutually exclusive.

From the studies with bioactive metabolites on mycelial morphology, it was observed that BM 3 induced production of chlamydospore-like structures in the mycelium when grown together. Presence of BM 1 and BM 2 although did not form any such structures, but mycelial dry weight was drastically reduced. Chlamydospores are resting spores which are produced by the pathogen to tide over unfavourable condition. Thus, it can be concluded that presence of BM 3 in the medium made the environment unsuitable for the normal growth of the pathogen.

This result also supports the work of Harish et al. (1998) who reported that in vitro interaction of F. udum and B. subtilis strain AF 1 showed that the fungus formed chlamydospore-like structures and increased vacuolation, when both cultures are simultaneously inoculated into PDB. Vasudeva et al. (1958) reported that in the presence of an antibiotic from Bacillus, a characteristic bulb formation was observed in the spores and hyphae of test fungi. The active principle had been shown to be thermolabile and the antibiotic was named ‘bulbiformin’ in view of its distinctive property of forming bulbs on the test fungi. This bulb formation takes place only in fungi having chitin in their cell wall. The mycolytic activity of antagonists could be mainly due to the lytic enzymes β-1,3-glucanase and chitinase (Srivastava et al. 2001). Development of fusarial wilt symptoms was prevented in pigeon pea seeds treated with B. brevis which produced extracellular substances (Bapat and Shah 2000). The extracellular antagonistic substances which induced swelling of the pathogen’s hyphal tips and cells were bulbous and swollen with shrunken and granulated cytoplasm. Cell free culture filtrates of B. subtilis strain were active against macroconidium germination and hyphal growth of F. graminearum, depending on the initial macroconidium density. It induced formation of swollen hyphal cells in liquid cultures of this fungus grown from macroconidia (Chan et al. 2003).

It can be concluded from the results of this study that the bioactive metabolites produced by PGPR strains play an important role in plant growth promotion and disease control of plants provided they are produced in adequate amount which is dependent upon the biotic and abiotic factors. The bioactive compounds produced by RRLJ 04 and BS 03 can be further exploited for growth enhancement and disease control of other crops.