Entomopathogenic potential of bacteria associated with soil-borne nematodes and insect immune responses to their infection

Abstract

Soil-borne nematodes establish close associations with several bacterial species. Whether they confer benefits to their hosts has been investigated in only a few nematode-bacteria systems. Their ecological function, therefore, remains poorly understood. In this study, we isolated several bacterial species from rhabditid nematodes, molecularly identified them, evaluated their entomopathogenic potential on Galleria mellonella larvae, and measured immune responses of G. mellonella larvae to their infection. Bacteria were isolated from Acrobeloides sp., A. bodenheimeri, Heterorhabditis bacteriophora, Oscheius tipulae, and Pristionchus maupasi nematodes. They were identified as Acinetobacter sp., Alcaligenes sp., Bacillus cereus, Enterobacter sp., Kaistia sp., Lysinibacillus fusiformis, Morganella morganii subsp. morganii, Klebsiella quasipneumoniae subsp. quasipneumoniae, and Pseudomonas aeruginosa. All bacterial strains were found to be highly entomopathogenic as they killed at least 53.33% G. mellonella larvae within 72h post-infection, at a dose of 106 CFU/larvae. Among them, Lysinibacillus fusiformis, Enterobacter sp., Acinetobacter sp., and K. quasipneumoniae subsp. quasipneumoniae were the most entomopathogenic bacteria. Insects strongly responded to bacterial infection. However, their responses were apparently little effective to counteract bacterial infection. Our study, therefore, shows that bacteria associated with soil-borne nematodes have entomopathogenic capacities. From an applied perspective, our study motivates more research to determine the potential of these bacterial strains as biocontrol agents in environmentally friendly and sustainable agriculture.

Fig 1. Entomopathogenic potential of different bacterial strains against G. mellonella larvae.

Insects were injected with three bacterial concentrations 10², 10⁴, or 10⁶ CFU/larva of the following bacterial strains. A) Acinetobacter sp. A_IN1. B) Alcaligenes sp. A_TC9. C) Bacillus. cereus Bc_IN18. D) Bacillus cereus Bc_TG3A. E) Bacillus cereus Bc_IN7. F) Enterobacter sp. E_TC7. G) Klebsiella quasipneumoniae subsp. quasipneumoniae Kqq_KG18. H) Kaistia sp. K_TC2. I) Lysinibacillus fusiformis Lf_OC2. J) Morganella morganii subsp. morganii. Mmm_EN01. K) Pseudomonas aeruginosa Pa_HWO. Bars correspond to confidence intervals (P = 0.05).

Fig 2. Total Hemocyte Counts (THC) in bacteria-infected insects.

Mean hemocyte number per ml of hemolymph of G. mellonella larvae injected with different bacteria or with buffer as controls at 4h, 24h and 48h post-treatment. Bars correspond to confidence intervals (P = <0.05). Different letters indicate significant differences between the number of hemocytes across all treatments within each time point (Duncan test, Kruskall, and Wallis test, P<0.05). Asterisks indicate significant differences in the number of hemocytes of control larvae and bacteria-injected larvae within each time point (Dunnet test, P<0.05). Arrows indicate that the insects died.

Fig 3. Lysozyme activity in the hemolymph of bacteria-infected insects.

Mean lysozyme activity in the hemolymph of G. mellonella larvae injected with different bacteria or with buffer as controls at 4h, 24h and 48h post-treatment. Bars show confidence intervals (P = 0.05). Different letters indicate significant differences in lysozyme activity across all treatments within each time point (Duncan test, Kruskall and Wallis test, P<0.05). Asterisks indicate significant differences in lysozyme activity between control larvae and bacteria-injected larvae within each time point (Dunnet test, P<0.05). Arrows indicate that the insects died.

Fig 4. Relative phenoloxidase activity in the hemolymph of bacteria-infected insects.

Mean relative phenoloxidase activity per ml of hemolymph of G. mellonella larvae injected with different bacteria or with buffer controls at 4h, 24h, and 48h post-treatment. Absorbance measurements were taken 30 min after hemolymph incubation. Bars correspond to the confidence intervals at P = 0.05. Different letters indicate significant differences in phenoloxidase activity across all treatments within each time point (Duncan test, Kruskall and Wallis test, P<0.05). Asterisks indicate significant differences in phenoloxidase activity between control larvae and bacteria-injected larvae within each time point (Dunnet test, P<0.05). Arrows indicate that the insects died.

Fig 5. Relationship between insect immune responses to bacterial infection and bacteria pathogenesis.

Correlation between insect mortality caused by bacterial infection 72h post-infection and: A) Mean hemocyte number per ml of hemolymph of G. mellonella larvae injected with different bacteria 4h post-infection; B) Mean lysozyme activity in the hemolymph of G. mellonella larvae injected with different bacteria 4h post-infection; C) Mean relative phenoloxidase activity per ml of hemolymph of G. mellonella larvae injected with different bacteria 4h post-treatment; D) Mean hemocyte number per ml of hemolymph of G. mellonella larvae injected with different bacteria 24h post-infection; E) Mean lysozyme activity in the hemolymph of G. mellonella larvae injected with different bacteria 24h post-infection; and F) Mean relative phenoloxidase activity per ml of hemolymph of G. mellonella larvae injected with different bacteria 24h post-treatment. Correlations were statistically assessed by person test and by independent phylogenetic contrast (PIC). n.s.: Statistically not significant.