Antagonistic potential of Moroccan entomopathogenic nematodes against root-knot nematodes, Meloidogyne javanica on tomato under greenhouse conditions

The root-knot nematode, Meloidogyne javanica is a devastating pest affecting tomato production worldwide. Entomopathogenic nematodes (EPNs) are considered very promising biocontrol agents that could be used to effectively manage plant-parasitic nematode. The antagonistic activity of five EPN strains isolated from different fields in Morocco was evaluated against juvenile (J2s) antagonism in soil, the number of egg masses, and the galling index of M. javanica and J2s reproduction in the root. In greenhouse experiments, Steinernema feltiae strains (EL45 and SF-MOR9), Steinernema sp. (EL30), and those of Heterorhabditis bacteriophora (HB-MOR7 and EL27) were applied to the soil alongside RKN J2s. There was a significant reduction in M. javanica densities in the soil and roots by EPNs treatments when compared to the positive control. The EPNs decreased both egg masses formation and galling index by 80% compared to the positive control. The application of EPNs at a rate of 50 and 75 infective juveniles (IJs) cm−2 gave significant control of all studied nematological parameters compared to the positive control, which confirmed the importance of the doses applied. The applied dose was significantly correlated with M. javanica parameters according to polynomial regression models. The results also showed that S. feltiae strain (EL45) significantly increased plant height and root length, while H. bacteriophora strain (HB-MOR7) only enhanced root fresh weight. Therefore, both indigenous EPN strains; EL45 and SF-MOR9 have eco-friendly biological potential against M. javanica in vegetable crops.


Scientific Reports
| (2022) 12:2915 | https://doi.org/10.1038/s41598-022-07039-0 www.nature.com/scientificreports/ via the toxins released by their symbiotic bacteria of Xenorhabdus and Photorhabdus in Steinernema and Heterorhabditis, respectively, into their hemocoel [12][13][14] . Bacteria produce some apoptosis or necrosis-induced substances (e.g., hemolysin, cytolysin, and toxins) in the host cells which trigger its death 15 . Management of PPNs using biological control agents (BCAs) is a promising alternative to the chemical alternatives 16,17 . The antagonistic activity exerted by EPNs on PPNs has been previously observed and reported 18 . The management of EPNs against different nematode species, such as Criconemoides spp., Rotylenchulus reniformis 19 , Globodera rostochiensis 20 , Belonolaimus longicaudatus 21 , and Meloidogyne spp. [22][23][24][25] has been proven under both field and greenhouse conditions. The most significant levels of control of PPNs have been observed against RKNs 2 . The application of EPN infective juveniles (IJs) from different strains has significantly controlled Meloidogyne spp., both in the number of eggs 27 , egg masses 24 , and the infectivity of J2s inside root matrix 28 . Furthermore, the use of symbiotic bacteria and/or their metabolites alone significantly reduced RKNs J2s in vitro 29,30 as well as decreased host infectivity in greenhouse conditions 24,31 . In a study carried out by Vyas et al. 31 , indicated that the level of control was reported to be comparable to some of the chemical treatments used. In addition, Caccia et al. 25 reported a significant nematicidal effect of three Argentinean EPN isolates against M. hapla using the bacterial supernatant of Photorhabdus luminescens and Xenorhabdus spp.
Recently, several Moroccan strains of S. feltiae and H. bacteriophora were isolated by Benseddik et al. 32 . However, their efficacy has not been yet evaluated against Meloidogyne species. Therefore, the main objective of this study is to investigate the antagonistic activity of native EPN strains against M. javanica under greenhouse conditions in Morocco.

Results
Antagonistic activity of Moroccan EPNs toward M. javanica. The influence of the EPNs isolates on the nematological parameters is shown in Fig. 1. The number of M. javanica infective juveniles (J2s) in the soil was significantly reduced across the different treatments when compared to the positive control. Steinernema strains significantly reduced J2s densities per 250 g of soil (Fig. 1A) and per 10 g of roots (Fig. 1C) compared to H. bacteriophora strains. For both parameters, S. feltiae (EL45) strain reduced J2s densities in the soil by 95% and in the root by 90% (F index = 36.4; df = 2; P < 0.05). SF-MOR9 strain caused significant reductions in J2s numbers when applied at 50 and 75 IJ cm −2 of soil. This reduction was more or less similar to the effect obtained by Oxamyl and Garlic extract treatments. On the other hand, H. bacteriophora strains (EL27 and HB-MOR7) demonstrated less antagonistic activity toward J2s densities in the soil and root system when compared to the positive control (F index = 22.4; df = 1; P < 0.05). The egg-masses formation was also decreased by EPN application at both doses (Fig. 1B).
To confirm the relationship between M. javanica parameters and EPN concentrations, polynomial regression analyses were performed (Fig. 3). For the J2s densities, significant regression models were obtained for Steinernema strains in both soil and root matrices (R 2 = 0.59 and R 2 = 0.27; P < 0.05), respectively, (Fig. 3A,B). Heterorhabditis bacteriophora strains were not significantly affiliated with the applied doses (R 2 = 0.14 and R 2 = 0.18; P > 0.05), respectively. Similarly, the same findings were obtained between EPNs and the number of egg masses (Fig. 3C), and the galling index (Fig. 3D).
The EPNs effect on plant growth parameter co-treated with M. javanica. EPN strains led to significant differences in plant height (F index = 78.5; df = 9; P < 0.05) (Fig. 4). The greatest increases in plant height were observed when S. feltiae strain EL45 applied at 50 IJ cm −2 (120.2 ± 5.54 cm), this increase was similar to those recorded by the garlic extract and Oxamyl product (Fig. 4A). On the other hand, H. bacteriophora strains (EL27 and HB-MOR7) gave the lowest plant height values (90.04 ± 6.10 cm) but still slightly more than the positive control (86.8 ± 4.32 cm). Root length was also significantly affected by EPN strains (F index = 56.4; df = 9; P < 0.05), with the strain EL45 effectively promoting root growth at every dose applied (23.8 ± 3.70 cm), when compared to the reference controls (Fig. 4B). An opposite trend was observed with H. bacteriophora strains (EL27 and HB-MOR7) with a maximum rate of 17.4 ± 2.60 cm in root growth. Root fresh weights were more pronounced with H. bacteriophora strain (HB-MOR7) that gave the highest values (14.08 ± 0.87 g) compared to the S. feltiae strains (F index = 19.34; df = 9; P < 0.05) (Fig. 4C). The Steinernema strain EL30 was the most effective treatment in increasing root growth compared to the positive control.

Discussion
The interaction between PPNs and EPNs resulted in distinct antagonistic patterns that could have implications for implementation in vegetable production and attempts to reduce the use of chemical methods. In addition, EPNs are commercially available for the management of various insect pests 33 and could be used simultaneously for nematode and insect pest management. In this study, the antagonistic activity of five native EPNs (Steinernema sp., S. feltiae, and H. bacteriophora) were assessed against the RKN, M. javanica in tomato plants under greenhouse conditions. Both nematode infection and development, as well as their effects on tomato growth parameters, were measured over a period of 2 months. The J2 M. javanica densities in the soil and root were moderate to highly affected by the antagonistic effects of the different EPN treatments applied to the soil. All EPN treatments were able to reduce nematode final population densities in both the soil and root matrices when compared to the positive control. In addition, the EPNs strains decreased both egg-mass number and gall formation. Steinernema feltiae strains EL45 and MOR9 were significantly more effective in reducing nematode impact when compared to H. bacteriophora strains and the other strain of Steinernema sp. in which different mechanisms could have interfered. To our knowledge, this is the first investigation of the applicability of Moroccan EPNs to Previous studies indicated that the direct application of EPN IJs has shown an antagonistic effect on different PPN species 28,30,34,35 Which are in a agreement with our findings in the current study. The observed antagonism in our study could be due either to the competitive patterns (most likely over space) between both nematode trophic levels or to the production of special chemical substances that could diverge the harmful effects of PPN. However, EPNs may not be active against all PPNs and depends on the species and crop aspects 26 . Thus, a decrease in symptoms may not always lead to significant control efficiency, especially under field conditons 36,37 . Our experiment depicted the effect of S. feltiae strains on RKN indices to be significantly reduced compared to the positive control confirming their potential efficacy, but not to the level of Oxamyl and garlic extract products. In addition, the sensitivity of nematode parameters could be the main reason behind this antagonistic response. Similarly, in cucumber plants, Sayedain et al. 38 reported that S. carpocapsae and H. bacteriophora were shown to decrease all the pathogenicity indices (number of galls, eggs, and egg masses) of M. javanica in both growth chamber and greenhouse conditions. The same findings were obtained by Fallon et al. 36 , indicating that applying both S. feltiae MG-14 and S. feltiae SN strains significantly minimized M. javanica invasion on soybean 3 days after treatment but did not affect M. javanica egg formation in tomato plants after 30 days. Pérez and Lewis 27 had applied H. bacteriophora and S. feltiae (25 IJs cm −2 ) before and after inoculation of M. incognita. They observed that these EPNs were able to inhibit the penetration of this RKN and decrease the production of eggs on tomato plants. However, the effect was not proficient enough against M. javanica 27 . In the same context, S. feltiae was confirmed to be ineffective against M. javanica on cucumber 38   The antagonistic effects of EPNs toward Meloidogyne spp. are closely associated with the application time frame, inoculum density, host plant, and the species of both the PPN and EPN 40 . In our study, both 50 and 75 IJs cm −2 EPN doses were shown to be effective in the control of M. javanica compared to the positive control. Sayedain et al. 38 found that applying densities of 125 IJs cm −2 (19.1 IJs cm −3 ) significantly increased the biocontrol of M. javanica. Similarly, Pérez and Lewis 27 confirmed that using 125 IJs cm −2 of H. bacteriophora co-inoculated with M. hapla in peanut reduced the egg production, while the same dose of S. riobrave did not inhibit the J2s of M. hapla. The observed antagonistic activity of EPNs against M. javanica in this study might be highly related to allelochemicals and ammonium production by the associated symbiotic bacteria 41 , plants systemic resistance 42 , competitive patterns EPN-RKN, and EPNs attraction toward exudates emitted by the plant root system 40 . In our study, the reduction of M. javanica by S. feltiae strains (EL45 and SF-MOR9) may have been due to metabolites produced by its mutualistic bacteria (Xenorhabdus spp.). Therefore, the interactions involved are complex and due to the interference of a tripartite system (host plant, PPN, and EPN) 43 .
In the current study, the tested EPNs has significant role on plants' growth parameters. That the results indicated that the S. feltiae strain EL45 caused significant increase in both plant height and root length when compared to the positive control, and this increase was similar to that obtained by the Oxamyl and garlic extract. while the H. bacteriophora strains EL27 and HB-MOR7 have only enhanced tomato fresh root weight. Conversely, Sayedain et al. 38 reported significant increases in root fresh weight of cucumber plants when S. carpocapsae was applied. However, previous studies have reported inconsistent effects of different EPNs IJs on plant dry weight 36,44 as well as their potential for promoting plant biomass 29,45 . In our study, this effect may imply the nullifying characteristics of EPN species towards PPN occurrence in the soil and thus investing in plants' growth and development.
In conclusion, this study provides insights into the practical usage of EPNs as biological control agents against the root-knot nematode M. javanica on tomato plants. Our results demonstrated that inoculation of S. feltiae strains (EL45 and SF-MOR9) with M. javanica J2s leads to significant levels of antagonistic activity. Heterorhabditis bacteriophora strains (HB-MOR7 and EL27) gave inconsistent results in terms of M. javanica infection and plant growth parameters. Further studies are required to evaluate the effectiveness of these endemic EPN strains in commercial greenhouses and optimize their inputs on plants yielding aspects. The interaction between EPNs and PPN needs to be further studied to determine if one EPN inoculation is sufficient or multiple treatments over a longer period of time are required. In addition, further studies are needed to determine whether the EPNs   (Daykin and Hussey, 1985) and then the total number of egg masses were counted per root system under a stereomicroscope. The root of each sample was gently washed in tap water to free adhered soil particles, cut into pieces (ca 0.5 cm), and then J2s M. javanica were extracted from 10 g subsample using Baermann tray method 50 . Root galling caused by M. javanica was indexed on each tomato root using a 0 to 5 scale 52 as follows: 0 = no galls, 1 = 1-2 galls, 2 = 3-10 galls, 3 = 11-30 galls, 4 = 31-100 galls and 5 = > 100 galls.

Statistical analysis.
Meloidogyne javanica and plant parameters were subjected to ANOVA procedure using the XLSTAT software, ver. 2016.02.28451 (Addinsoft, New York, USA). Datasets were normalized using the Anderson-Darling normality test 53 . Each trial was independently repeated twice. A two-way ANOVA test was performed to examine sources of variation in the observed variables. Significant differences among variables were tested using protected least significant difference and Fisher's protected least significant difference (LSD) test at P < 0.05. Differences obtained at levels of P < 0.05 were considered significant. Polynomial regres-