Root-knot nematodes are one of the most economically important pest causing severe damages and losses in a wide variety of crops worldwide [1]. Estimates of nematode damage of tomato yield worldwide ranged from 28 to 68%. However, the control of plant parasitic nematodes still essentially relies on crop rotation and chemical nematicides. Unfortunately, chemical nematicides are costly and very toxic compounds which pose a hazard to human health hazards and the environment. Therefore, safe and environmentally friendly alternatives are becoming increasingly important [2,4]. Microorganisms with the ability to antagonize nematodes represent realistic alternatives to chemical nematicides. Certain fungi have been proposed for the biocontrol of plant-parasitic nematodes [3,5]. Unfortunately, these fungi are usually unable to compete effectively with the resident soil microorganisms. Also fungal growth is slower than that of bacteria, so biological control by using bacteria is preferable [6,8]. The bacterium Pasteuria penetrans which is an obligate parasite, can control the spread of root knot nematodes in soil, but is difficult to grow under laboratory conditions and does not survive well in soil [9]. Recently, research has focused on other bacteria for control of plant parasitic nematodes. Because of their catabolic versatility, their excellent root colonizing ability and their capacity to produce a wide range of antiphytopathogenic metabolites, pseudomonads have received particular attention [1,9,12]. However, the available information on nematicidal activity of Egyptian bacterial isolates, particularly rhizofluorescent pseudomonads, is scant. Therefore, the main objective of this study was to find isolates with effective nematicidal activity among a collection of Egyptian rhizosphere fluorescent pseudomonad (ERFP) isolates. To achieve this objective, (i) 52 ERFP isolates were screened in vitro for nematicidal activity, (ii) cultures and the cell-free culture supernatants of the best isolates were evaluated for nematicidal activity and their effects on plant growth in glasshouse pot experiment in comparison with a chemical nematicide were determined and (iii) phylogenetic relationships of effective ERFP isolates based on phenotypic and genotypic characterization were established.

Phenotypic characterization of Pseudomonad isolates

A total of 52 fluorescent pseudomonad isolates were screened in a previous study by Gamal-Eldin et al. [13] for production of indoleacetic acid (IAA), siderophores, cyanide, proteinase, chitinase, antagonism towards plant-pathogenic fungi and zinc and phosphate solubilization. In the present study, all of these isolates were screened for their nematicidal activity and the top five isolates were identified by using the API 20 NE identification kit (Bio Mérieux, France) and by comparing the results with the known species present in the API database.

PCR amplification and sequence analysis of 16S rDNA gene

For isolation of the bacterial genomic DNA, each individual Pseudomonas isolate was grown in liquid Kings B medium at 30°C with shaking at 100 rpm for 24 h and the template DNA was prepared according to Johnsen et al. [14]. DNA 16S region amplification was performed using the primer set 16SF-16SR [15]. The 16S rDNA was sequenced as described by Sanger et al. [16]. The 16S rDNA sequences were initially analyzed by using the program Blast (National Center Biotechnology Information, http//:www.ncbi.nml.nih.gov). Sequencing data obtained from different primers were collected together using the CAP program (Contig Assembly and Genomic Expression programs). The consensus sequence from the isolates and sequences of strains belonging to the same phylogenetic group and of other representatives of Pseudomonas strains (retrieved from the NCBI database) were aligned using the computer-program ClustalX [17]. The resulting trees were displayed with Tree View [18]. The phylogenetic tree was calculated using the neighbor-joining method [19] and Acinetobacter calcoaceticus and E. coli K12 were used as outgroups.

Preparation of bacterial inocula and their cell –free culture supernatants (CFS)

The pseudomonad isolates were grown separately for 36 h in King’s B medium broth at 28°C on a rotary shaker at 150 rpm. Part of the cultures was kept without centrifugation for use at 4°C. The bacterial cultures mixture was prepared by mixing equal volumes of cell suspension of individual isolates. For preparation of cell-free culture filtrate, the other part of the bacterial cultures were centrifuged at 5,000 x g for 30 min; pellets were discarded and the supernatants were filtered through bacterial filter (0.22µm pore size) and kept for use at 4°C.

Preparation of nematode inocula

A population of the root-knot nematode Meloidogyne incognita was routinely maintained on the susceptible tomato cultivar Castle Rock in a glasshouse at 27°C ±5°C in a box filled with sandy loam soil. Nematode eggs were extracted from heavily infested tomato roots using the extraction technique described by Hussey and Barker [20]. The eggs were washed from the 25µm sieve into a 300 ml Erlenmeyer flask with sterile tap water. To promote the development of eggs and the hatching of J₂, the nematode eggs/water suspension was kept in darkness at 24°C and aerated with an aquarium pump.

Bioassay of egg hatching and mortality of J₂ of Meloidogyne incognita

To study the effect of the bacterial isolates cultures or their supernatants on the egg hatching, three egg masses containing approximately 300 eggs per egg mass were mixed with 1 ml of the bacterial culture (2.5x10⁸ cfu/ml) or their supernatant in 1.5 ml Eppendorf tubes in three replicates and incubated at 28°C for 48 h. Sterile half-strength King’s B medium containing the same number of egg masses were used as controls. The numbers of hatched J₂ were recorded after 48 h. Eggs and J₂ larvae were counted under microscope using a Hawksley counting slide. To study the effect of the bacterial isolate cultures or their supernatants on the viability of J₂ larvae, 500µl sterile King’s B medium containing 1000 of the J₂ larvae were mixed with 1 ml of each individual bacterial culture or supernatant or their mixtures in 1.5ml Eppendorf tubes in three replicates and incubated at 28°C for 48 h. Sterile half-strength King’s B medium containing the same number of J₂ larvae were used as control. The numbers of dead J₂ larvae were recorded after 48 h by microscope.

Pot experiment

Pot experiments were carried out in a glasshouse at the Faculty of Agriculture, Fayoum University. Tomato (Lycopersicon esculentum Mill. cv Castle Rock) 21 day-old seedlings were immersed in bacterial cultures or supernatants mixed with 10 % gum arabic. The experiments were conducted in plastic pots (11 cm diameter) filled with one kilogram of a sterilized 2:1 mixture of clay and sand. One week after transplanting, soil around the roots was carefully removed without damaging the roots and 5 ml tap water containing 1000 J₂ larvae was poured around the roots before replacing the soil. Control plants were given 5 ml half-strength King’s B medium. Treatments were arranged in a randomized complete block with four replicate pots. Plants were irrigated and fertilized with recommended dosage as needed. The pots were maintained in a glasshouse under natural light and day/ night temperatures of approx. 26/15°C. Three weeks after infection, plants of three replicates were harvested and the following parameters were measured: shoot height(cm), shoot dry weight plant-1 (g), root size plant^-1 (cm³), leaves number plant^-1, leaves area plant^-1 (cm²), total nitrogen content (mg/g dry weight). Total indoles (mg g^-1) were determined in fresh shoot by using P-dimethylaminobenzaldehyde reagent according to Larson et al. [21]. Chlorophyll content of leaves (mg/g fresh weight) was determined according to Graan and Ort [22]. Nematode galls on the plant roots were counted under a low power (x10) microscope and larvae were extracted from 100 cm³ soil by a centrifugal flotation method [23]. Extracted nematodes were counted under the microscope using a Hawksley counting slide. The fourth replicate was harvested at 60 days old to follow the development of root galling. For histological investigation, the root sections were prepared according to Sayan et al. [24] and the digital micrographs were taken using a JVC Camera (model No. TK 890E) attached to an OLYMPUS microscope (OLYMPUS BH2, OLYMPUS OPTICAL Co. LTD No.106105 Japan). The giant cells were detected as a syncytium containing many nuclei formed around the head of nematode females as a result of secretions from the esophageal glands.

Determination of phenolic content plant defense- related enzymes

For determination of the plant defense-related enzymes, one gram of fresh tomato tissue was washed in running tap water immediately after sampling and homogenized using a mortar and pestle. The homogenized tissue was extracted with 2 ml of 10 mM 0.1 M sodium phosphate buffer (pH 7.0) at 4°C and centrifuged at 10,000 rpm for 15 min at 4°C. The total protein was determined according to Bradford [25] and the supernatant served as enzyme sources.

For determination of phenylalanine ammonia-lyase (PAL), the assay mixture containing 100µl of enzyme extract 500µl of 50 mM Tris HCl (pH 8.8) and 600µl of mM L-phenylalanine was incubated for 60 minutes. The reaction was stopped by adding 2 N HCl. Transcinnamic acid was extracted with 1.5 ml toluene and measured spectrophotometrically at 290 nm against the blank of toluene. Standard curves were determined with graded amounts of cinnamic acid in toluene. PAL activities were determined from a standard curve of cinammic acid and expressed as nmol cinammic acid min^-1 mg protein^-1 [26].

For determination of Peroxidase (PO), the reaction mixture consisted of 1.5 ml of 0.05 M pyrogallol, 0.5 ml of enzyme extract and 0.5 ml of 1 % H₂O₂. The reaction mixture was incubated at room temperature (28 ± 2°C).The change in absorbance at 420 nm was recorded at 30 sec intervals for 3 min. The boiled enzyme preparation served as blank and the enzyme activity was expressed as change in the absorbance of the reaction mixture min^-1g^-1 protein [27].

For determination of polyphenoloxidase (PPO), the reaction mixture consisted of 1.5 ml of 0.1 M sodium phosphate buffer (pH 7.0) and 200µl of the enzyme extract. To start the reaction, 200µl of 0.01 M catechol were added and the activity was expressed as change in absorbance min^-1g^-1 protein [28]. For determination of phenolics content, one gram of tissue was homogenized in 10 ml of 80 % methanol and agitated for 15 minutes at 70°C. One ml of the methanolic extract was added to 5 ml of distilled water and 250µl of 1N Folin-Ciocalteau reagent and the solution was kept at 25°C. The absorbance was measured at 725 nm. The content of total soluble phenols was calculated according to a standard curve obtained from a Folin-Ciocalteau reagent with a phenol solution (C₆H₆O) and the amount of phenolics was expressed as µg catechol mg^-1 fresh weight of tissue [29]. The data were analyzed for significant variation by Duncan’s New Multiple Range test and treatment means were separated by the least significant difference (LSD) test of P ≤ 0.05 [30].

In this study two in vitro and one glasshouse pot experiments were conducted. In the first in vitro experiment, the ERFP isolates were screened for nematicidal activity in terms of nematode egg hatching reduction. Exposure of the M. incognita eggs to culture of each isolate revealed that all the 52 ERFP isolates showed nematicidal activity, although to differing degrees ranged from 57 % to 100 % (data not shown).

In the second in vitro experiment, cultures and cell-free supernatants (CFS) either individually or in combination of the top five ERFP isolates; Ps 14, Ps 21, Ps 22, Ps 36 and Ps 54, that recorded the highest inhibitory effect on hatching of M. incognita eggs in the first in vitro screening experiment were evaluated in vitro for their effects on egg hatching and viability of the second-stage juveniles (J₂) in comparison with effect of the chemical nematicide Videt^©. The results in (Table 1) revealed that four out the five selected isolates completely inhibited egg hatching and killed 100 % of J₂, with the exception of isolate Ps 14 which reduced hatching by 94 %. The chemical nematicide showed effects similar to those of ERFP isolates, whereas the control (King’s B medium) caused 14.7 % reduction in egg hatching and 9.0 % reduction in viable J₂.

Table 1: Nematicidal activity of selected ERFP isolates against Meloidogyne incognita in vitro and pot experiment.

Effect of ERFP isolates on root galling and nematode population in soil

In the glasshouse pot experiments, pre-inoculated, nematode infected tomato seedlings with cultures or CFS of the selected ERFP isolates, either individually or in mixtures caused reductions in root galling and nematode multiplication in soil (Table 1). Effects of bacterial cultures ranged from 81.0 to 95.4 % reduction in root galling and 91.9 to 95.7 % reduction in nematode multiplication. The effects of CFS ranged from 61.3 to 84.5 % root galling reduction and from 83.5 to 84.5 % reduction in nematode multiplication. Application of the chemical nematicide led to 97.3 % and 96.5 % reduction on root galling and nematode multiplication in soil, respectively, compared to positive control (nematode only). Furthermore, examination of root sections (Figure 1) showed that the size of giant cells in root sections of the positive control (nematodes only) were 3049µm², while the size of the giant cells in root sections of Ps-inoculated, nematode infected roots reduced to 144µm².

Figure 1: (Photos 1, 2 & 3) Plant growth, formation of galls on roots and microscopic examination of giant cells as affected by: (a) control+ nematode, (b) control without nematode or ERFP (c) treatment infected with nematode in the presence of ERFP- mixture isolates, (d) treatment with ERFP- mixture in the absence of nematode.

Effect of ERFP isolates on growth of tomato seedlings infected with nematodes

Pre-inoculated, nematode-infected tomato plants treated with cultures or CFS showed significant improvement in plant growth characters compared to both chemical nematicide-treated plants and the untreated controls (Table 2). Cultures of ERFP isolates showed significant increases in plant height (23-44.5 %), root size (21.8-47.3 %), shoot dry weight (47.6-90.5 %), leaves area (13.7-24 %), total chlorophyll (3.2-23.4 %), total indole (35-60 %) and phenols contents (14.4-35.5 %) compared to the control without nematode and bacteria.

Table 2: Effect of inoculation with selected ERFP isolates on growth of tomato seedlings (Lycopersion esculentum) infected with nematodes.

Regarding the effects on activities of the plant defense-related enzymes, inoculation with ERFP isolates showed significant increase in activities of PO (25-55 %), PPO (12.4-44.5 %) and PLA (5-31.5%) in tomato leaves (Table 3). On the other hand, as expected plants infected with nematodes only showed significant decreases in growth characters and activities of PO, PPO and PLA compared to the uninfected controls.

Table 3: Effects of inoculation with selected ERFP isolates on total indol, total phenols and activities of the plant defense-related enzymes in leaf tissues of tomato plants infected or uninfected with nematodes.

Phenotypic and genotypic characterization

Based on their superior nematicidal activity, the five isolates Ps14, Ps21, Ps22, Ps36 and Ps54 were selected among the 52 isolates for further phenotypic and genotypic characterization. The results of the phenotypic characterization (Table 4) of these isolates, based on morphological, physiological and biochemical tests, show the features that differentiate the isolates studied compared to the reference strains. All five isolates were Gram-negative, motile and aerobic rods. They were positive for fluorescent pigment, oxidase, catalase, arginine dihydrolase, urease, chitinase and gelatin hydrolysis. In addition, they were positive for indoleacetic acid and siderophore production and could utilize a wide range of carbon sources. However, isolates Ps 14, Ps 21and Ps 22 differed from isolate Ps 36 and Ps 54 by their ability to utilize N-acetyl-glucosamine and adipate and their inability to hydrolyze aesculin and production of cyanide. According to the API analytical profile index, the previous results permitted the identification of isolate Ps 14, Ps 21 and Ps 22 as Pseudomonas aeruginosa and isolates Ps 36 and Ps 54 as Pseudomonas putida. Analysis of the partial 16S rDNA sequences placed all of the selected isolates firmly within the Pseudomonas genus and a comparison to the partial 16S rDNA sequences of other PGPR fluorescent pseudomonad strains in this genus further elucidated their phylogenetic positions (Figure 2). Based on this genotypic characterization, two major clusters were observed; strains Ps 36 (Accession number GU168529) and Ps 54 (Accession number GU168528) appeared to fall within the variability range of P. putida, while Ps 21 (Accession number GU168531), Ps 22 (Accession number GU168530) and Ps 14 (Accession number GU168532) are likely to be strains of P. aeruginosa. The distinctness of Ps 36 and Ps 54 from Ps 14, Ps 21 and Ps 22 was also supported by their physiological characteristics, such as the ability of Ps 14 and Ps 21 and Ps 22 to utilize mannitol, N-acetyl-glucosamine and adipate.

Table 4: Biochemical and physiological characteristics of selected ERFP isolates.

Figure 2: Neighbor-joining tree showing the estimated phylogenetic relationships of the five newly-isolated strains (shown in grey) and other closely-related strains of the genus Pseudomonas. Bootstrap values out of 1000 are given at the nodes.

The plant-parasitic nematode M. incognita, tomato plants and pseudomonads were used in this investigation. Studies concerning the distribution of the plant–parasitic nematodes in soil indicated that M. incognita is the most dominant, representing approximately 64% of the total population of Meloidogyne spp. [31]. Tomato was chosen as a test plant in pot experiment as it is an excellent host for M. incognita and an important vegetable crop of global interest [32]. Pseudomonads, particularly the fluorescent pseudomonads, are among the most prevalent bacteria inhabiting the rhizosphere. In the present study, screening of a collection of 52 ERFP isolates showed high percentage of nematicidal activity. This result indicates that nematicidal activity seems to be widespread among ERFP isolates. In the literature, the proportions of rhizosphere bacteria with nematicidal activity was lower than that found in the present study and were 58-74 % [1], 70 % [33], 65 % [34]

Based on the screening results, the five isolates which showed the highest nematicidal activity (Ps 54, Ps 36, Ps 22, Ps 21 and Ps 14) were selected for further investigation. In the in vitro experiment, cultures and cell-free supernatants of the five isolates showed a significant reduction in the number of eggs hatching and a significant increase in mortality of J₂ of M. incognita. Similar results were reported by Bin et al. [35] who found that whole cultures and culture filtrates of some rhizobacteria showed nematicidal effects on J₂ of M. javanica that ranged from 62-64 % and 62-70 % respectively. This finding provides evidence that the five chosen ERFP isolates produce nematicidal compounds which may be used to control M. incognita. Interestingly, microscopic examination of the unhatched eggs indicated that a large proportion of the unhatched eggs was severely damaged. However, this phenomenon was not observed in unhatched eggs treated with the chemical nematicide. This phenomenon is presumably due to lytic enzymes secreted by the tested isolates. In this context, Khan et al. [36] found that M. javanica eggs treated with chitinase or protease in a liquid culture of Paecilomyces lilacinus displayed large vacuoles in the chitin layer and the vitelline layer was split and had lost its integrity. In the present study, the tested ERFP isolates were positive for chitinase and protease production (Table 1). However, the potential of the cultures was more pronounced than that of their cell-free supernatants, suggesting that there are other mechanisms beside production of nematicidal metabolites which may be involved in control of M. incognita by these isolates. One of these mechanisms may be the induction of systemic resistance (ISR) in plants by the presence of live bacteria, which could lead to enhanced resistance against M. incognita. As indicated in (Table 3), activities of the plant-defense related enzymes PO, PPO and PAL were found to be significantly higher in leaf tissues of tomato plants which had been inoculated with ERFP isolates compared to noninoculated plants, either in presence or absence of nematodes. These results are in agreement with those of Kavitha and Jonathan [37] who found that the inoculation of tomato plants with FP strains caused significant increase in PO, PPO and PAL. Recent investigations of the mechanisms of biological control by plant-growth promoting fluorescent pseudomonads revealed that PGPR strains protect plants from pathogen attack through systemic protection by strengthening the epidermal and cortical cell walls with deposition of newly formed barriers beyond infection sites including callose, lignin and phenolics [38] and by activating defense genes encoding chitinase, PO, PPO and PLA and enzymes which are involved in the synthesis of phytoalexins [12]. PLA plays an important role in the biosynthesis of phenolics that are effective chemical barriers against pathogen infection. PO and PPO catalyze the last step in the biosynthesis of lignin and other oxidative phenols [38]. Biological control using introduced bacteria, particularly rhizosphere pseudomonads, with the capacity to elicit induced systemic resistance (ISR) against plant diseases has been extensively studied [39,40] but little research has been conducted to determine whether biological control bacteria can elicit ISR against nematodes. However, further studies are needed to conclusively show the induction of systemic resistance induced by Ps in the control of nematodes. In addition, it is well known that RFP extensively colonize plant roots [41]. This could contribute to the prevention or at least reduction of nematode attack of plant roots. Therefore, the results of this research strongly indicate that the ERFP isolates tested produce nematicidal metabolites, whereas suggesting but not proving that competition for penetration/ colonization of the roots and the induction of ISR against nematode are mechanisms for controlling nematodes by inoculation with those ERFP isolates.

Results regarding the effect of the ERFP isolates on tomato growth (Table 2) indicate significant increases in almost all measured plant growth parameters in plants inoculated with the ERFP isolates either alone or in the presence of nematodes. This could be due to the plant growth promoting traits of the tested ERFP isolates such as synthesize several plant hormones and enzymes that can modulate plant hormone levels. It was reported that phytohormones produced by PGPR play a major role in growth promotion and many bacteria have the ability to produce auxins, gibberellins, cytokine’s and ethylene [42]. In pot experiments, the cultures were more effective than cellfree supernatants, which could be due to dilution of metabolites by irrigation water. Inoculation with the cultures guarantees a continuous supply of nematicidal metabolites.

It is worthy of mention that inoculation with the chosen ERFP isolates not only masked the deleterious effects of nematode on plant growth, but also increased growth of plants infected with nematode over the uninfected control plants. These results indicate that these isolates can be characterized as the most efficient strains and should be used as nematicidal bio-agent and as PGPR. Based on phenotypic and genotypic characterizations of the five chosen ERFP isolates, two major clusters were observed; strains Ps 36 and Ps 54 appeared to fall within the variability range of P. putida, while Ps 21, Ps 22 and Ps 14 are likely to be strains of P. aeruginosa. The distinctness of these was, however, also supported by their physiological characteristics, such as the ability of Ps 14 and Ps 21 and Ps 22 to utilize mannitol, N-acetyl-glucosamine and adipate [43]. Strains of both species are commonly encountered in the plant rhizosphere and many of these strains have been frequently reported as PGPR and as biological control agents [44]. It is worth mentioning that Ali et al. [33] pointed to some health risks may be involved in the application of P. aeruginosa to control plant pathogens. In the light of data available about the distribution of the P. aeruginosa in plant rhizospheres and their consideration by a number of researches as plant growth promoters and biocontrol agents against a lot of plant pathogens, the question arises if these strains are human pathogens or if they are completely different from human pathogenic strains. To answer such questions, further work should be conducted. Fortunately, isolate Ps 54, which showed the highest nematicidal activity against Meloidogyne incognita and also showed significant promotion of most plant growth parameters than any other isolate, clearly falls within P. putida group.

In conclusion, cultures and cell-free supernatants of some selected ERFP isolates exhibited a pronounced nematicidal activity either in vitro or in pot experiments. This is indicated by the significant reduction in nematode egg hatching, J₂ viability and nematode population in soil and numbers of galls on roots. Furthermore, they showed results comparable with that of the chemical nematicide Videt. In addition, these isolates considerably promoted the growth of tomato plants and previously also demonstrated in vitro antifungal activity [13]. Therefore, the application of these isolates in the control of nematodes could be advantageous for consumers, producers and the environment. However, further work is needed to evaluate effectiveness of these isolates under field conditions with different nematodes on different plants.

The authors would like to thank Dr. Martin Krehenbrink (Oxford University, UK) for helpful revision.