The entomopathogenic nematode Heterorhabditis megidis: host searching behaviour, infectivity and reproduction

Entomopathogenic nematodes in the families Heterorhabditidae and Steinernematidae have considerable potential as biological control agents of soil-inhabiting insect pests. Attributes making these nematodes ideal biological control agents include their broad host range, high virulence, safety to non target organisms, ability to search for hosts, high efficacy in favourable habitats, high reproductive potential, ease of mass production, ease of application, and compatibility with other control strategies. In chapter 1 , I give a short review on biology and some of the important biotic and abiotic factors that affect the infectivity and dispersal ability of entomopathogenic nematodes, state the research aims and present the outline of the thesis.Density-dependent factors within a host can have an important influence on the population dynamics of entomopathogenic nematodes. In chapter 2 the effects of increasing Heterorhabditis megidis (strain NLH-E87.3) density in Galleria mellonella larvae were compared. Although the number of nematodes that established in the host increased with increasing dose, the percentage of invasion decreased. The number of progeny produced per host initially increased with dose, but the highest production of infective juveniles (IJs) per cadaver was reached at a dose of 300 IJs per host, when about 62 IJs were established per cadaver. The smallest infective juveniles were produced at a dose of 1000 IJs per host and the largest at a dose of 300 IJs per host. Time to first emergence of juveniles was generally shorter when the number of IJs inoculated was large. Effects of high density appear to result from competition for limited nutrients within the host. For the success of entomopathogenic nematodes in the field, the knowledge on density effects of each nematode species or strain should be taken into account.Between production and application, the infective juveniles are exposed to environmental stress, especially during storage and shipment. Among the factors hampering the performance of entomopathogenic nematodes, as biocontrol agents are time and the temperature conditions of storage. In chapter3 the effect of storage conditions, temperature and time, on the survival, infectivity and development of IJs of H. megidis was investigated. Infective juveniles were stored at 5, 10, 15 and 20ºC for a period of up to 70 days (10 weeks). Infectivity and reproduction after each storage time and temperature were measured in bioassays with G. mellonella larvae exposed to a dose of one or 30 nematodes. The results show that independent of the time of storage, IJs performed best when stored at a temperature of 10 or 15ºC. An increase of the storage time caused a decrease of "quality" of the nematodes stored at 5 and 20ºC. The low storage temperature induced most of the nematodes into a state in which they lose their ability to parasitise a host and the temperature of 20ºC directly affected the nematode survival. Time to first emergence was affected more by the inoculum size than by the storage conditions. Infective juveniles emerged earlier from cadavers exposed to a dose of 30 nematodes than from those infected by only one nematode. This result shows that the reproduction time inside a host is more tightly regulated by density-dependent constraints than by the storage conditions of the infective juveniles.Entomopathogenic nematodes are able to invade and, in most the cases, to kill a large number of insect species. However, a close relationship between the nematode and the host is generally observed and this suggests a particular susceptibility of the insect, variable also within its different developmental stages, and an intrinsic virulence of the nematode species or strain. In chapter 4 I report the results of investigations on the ability of IJs of H. megidis to cause infection and reproduce in differently sized larvae of G. mellonella and Otiorhynchus sulcatus . Larvae of both insect hosts were weighed, divided in groups of small, medium and large, and exposed to a dose of one or 30 IJs. The number of invading IJs increased with host size while the host mortality at a dose of one IJ decreased with the increase of host size. However, IJs showed to be able to invade and kill each size group of larvae of both insect hosts tested. At a dose of 30 IJs, larvae of G. mellonella show to be significantly more susceptible than O. sulcatus larvae, whereas at a dose of one IJ, O. sulcatus larvae were more susceptible . In general, time to first emergence was longer at the lowest IJs inoculum and increased with the increase of host size in both insect species. The production of progeny differed between host species, host sizes and doses of nematodes. G. mellonella larvae produced more nematodes than O. sulcatus when the production from larvae of the same size was compared. The total progeny production per larva increased with the increasing larval size but no progeny production was observed in small larvae of O. sulcatus exposed to a dose of one IJ. Generally, the IJs body size increased with an increasing host size and the longest infective juveniles were produced at the lowest IJ doses.Using IJs of H. megidis originating from small, medium or large larvae of G. mellonella, and from medium or large larvae of O. sulcatus, previously exposed to a different IJs dose, I attempted to answer the question as to whether infective juveniles obtained from the different larval size of an insect host are capable of invading and killing larvae of the same host where they originate from and/or a new host. A series of infectivity tests were done ( chapter 5 ). Independent of the size of the larvae from which IJs were originating they were capable of infecting larvae of all sizes of its own host and also larvae of a new host. In general IJs originating from small cadavers of both host insects showed to be more infective than those originating from the medium and large cadavers. When tested at a dose of one IJ per larva, IJs originating from medium O. sulcatus cadavers were more infective against G. mellonella than against O. sulcatus larvae. Large G. mellonella larvae were in general less susceptible to all IJ batches than medium and small larvae.In chapter 6 an agar-based assay was used to assess the effect of nematode density, nematode age, incubation time and the presence of insect hosts on the dispersal behaviour of IJs of H. megidis . Infective juveniles dispersed faster and further at high densities than at low densities. Dispersal was also influenced by the age of the IJs. Nematodes stored for a period of 1.5 and 4.5 weeks showed to be more active than those stored for 2.5 and 3.5 weeks. The presence of a host insect enhanced the dispersion of nematodes. After 90 minutes IJs had responded positively to cues from G. mellonella but poorly to cues from O. sulcatus larvae.In chapter 7 the host-finding and dispersion behaviour of H. megidis in the presence of G. mellonella or O. sulcatus larvae and strawberry ( Fragaria x ananassa Duch.) roots alone, in the presence of, or under attack by O. sulcatus larvae was studied. Bioassays were conducted in Petri dishes (19 cm diameter) filled with moist sand and incubated at 15ºC over 24 hours. Infective juveniles responded positively to the presence of G. mellonella larvae, to roots of a single strawberry plant, and to O. sulcatus larvae in direct contact with roots of a single strawberry plant. A neutral or negative response was observed when infective juveniles were presented with only O. sulcatus larvae or a combination of several strawberry plants with O. sulcatus larvae, either in contact or not in contact with the roots. Infective juveniles responded strongly to the combination of plant roots and feeding larvae indicating that the tritrophic interaction formed by infective juveniles, O. sulcatus larvae, and strawberry plants may be an infochemical-mediated interaction.In chapter 8 and 9 a newly developed Y-tube olfactometer filled with sand and incubated at 15 ºC was used to test the host-searching behaviour of .H. megidis. Within an incubation period of 24 hours, IJs were significantly attracted to living G. mellonella larvae and caused 100% larval mortality. O. sulcatus larvae, however, did not elicit host-oriented movement of IJs and no larval mortality was observed. Roots of strawberry plants induced IJs movement but caused IJs to move away from the plant roots. The combination of strawberry roots and O. sulcatus larvae, however, strongly attracted IJs leading to 37% host mortality ( chapter 8 ). The results in chapter 9 showed that IJs were activated by the presence of intact roots of both strawberry ( Fragaria x ananassa Duch.) and thuja ( Thuja occidentalis L.). Some nematodes aggregated in the compartments with roots but most moved away from the roots to the opposite side. Given a choice, IJs showed preference for strawberry roots alone above O. sulcatus larvae. No difference in preference was observed between thuja roots and O. sulcatus larvae. The combination of strawberry roots with O. sulcatus larvae was preferred above strawberry roots alone. In the assays with the combination of thuja roots plus O. sulcatus larvae versus thuja roots alone, IJs were stimulated to move but away from both roots plus larvae or only roots arm. When challenged with insect damaged roots and mechanically damaged roots IJs were most attracted by thuja roots damaged by larvae, whereas in the case of strawberry IJs showed a clear preference for the mechanically damaged roots above insect damaged roots. A preference for strawberry roots, alone or in combination with O. sulcatus larvae, over thuja roots, in the same condition was always observed. It was also shown that the Y-tube choice arena used to perform the assays is a useful tool in studying the searching behaviour of entomopathogenic nematodes in a semi-natural habitat.In chapter 10 the most important research findings and the contribution of the results to the existing knowledge and the supposed applicability of the findings are discussed.