Behavioral ecology of entomopathogenic nematodesIntroductionForaging strategiesInfective juvenile foraging mechanismsInteraction between foraging strategy and host-associated cuesEvolution of foraging strategiesRecognizing and entering a hostInfecting a hostResponses to hosts infected by conspecificsResponses to hosts infected by heterospecificsMales (or females) as colonizersPhased infectivityInteractions between extrinsic and intrinsic conditionsDefense of the cadaver: the ant-deterrent factorConclusionsReferencesBiological Control 38 (2006) 66–79www.elsevier.com/locate/ybcon1049-9644/$ - see front matter © 2005 Elsevier Inc. All rights reserved.doi:10.1016/j.biocontrol.2005.11.007Behavioral ecology of entomopathogenic nematodesEdwin E. Lewisa,b,¤, James Campbellc, Christine GriYnd, Harry Kayaa,b, Arne Petersea Department of Nematology, University of California, Davis, Davis, CA 95616, USAb Department of Entomology, University of California, Davis, Davis, CA 95616, USAc USDA ARS GMPRC, 1515 College Ave, Manhattan, KS 66502, USAd Department of Biology and Institute of Bioengineering and Agroecology, National University of Ireland, Maynooth, Co. Kildare, Irelande E-Nema GmbH, Klausdorfer Str. 28-36, D-24223 Raisdorf, GermanyReceived 9 June 2005; accepted 16 November 2005Available online 18 January 2006AbstractWe discuss the behavior and ecology of entomopathogenic nematodes in relation to their successes and failures as biological controlagents. Four categories of studies have been reviewed herein; infective juvenile foraging strategies, recognition and evaluation of the hostby infective juveniles, the actual behaviors of infective juveniles that result in infection, and the protective role of the symbiotic bacteriaduring nematode reproduction in the cadaver. This constitutes a chronological order of events. Two challenges are suggested. First, thedata on entomopathogenic nematode behavioral ecology, while very extensive for some species, are almost completely lacking for mostdescribed species. We need to approach future studies paying more attention to phylogenetic origins of the traits that we study. Second,there is little consensus on exactly what are the traits that are worth studying. By reviewing and synthesizing current work, we make somesuggestions about where future research should be directed.© 2005 Elsevier Inc. All rights reserved.Keywords: Behavior; Ecology; Entomopathogenic nematode; Infection; Host; Parasite; Insect; Steinernema; Heterorhabditis1. IntroductionThe behavior and ecology of entomopathogenic nema-todes (i.e., steinernematids and heterorhabditids) have beenstudied in the past in attempts to make them better biologi-cal control agents. Indeed, their utility as biological controlagents has spurred most of the research that has been con-ducted. Much of this work has been focused on their behav-ioral interactions with hosts. Several reasons have been putforward to explain the need for further understanding ofentomopathogenic nematode behavior and ecology. Manyof the papers published in this area, and grant proposals aswell, have begun with a sentence resembling the following:“Entomopathogenic nematodes have great potential as bio-logical control agents, but because of signiWcant gaps in ourknowledge of their foraging behavior (or infection strate-gies, host associations, etc.), this potential has not been real-ized.”Only relatively recently have entomopathogenic nema-todes been the subject of more basic studies, where entomo-pathogenic nematodes are viewed as model organisms withwhich to ask questions about parasite biology in general.Entomopathogenic nematodes are excellent models for thispurpose since they share the trait of having an active thirdstage infective juvenile (IJ) as the infective stage with manyother parasitic nematode species. They have distinct experi-mental advantages over many animal parasites; they areeasy and cheap to culture, they live from several weeks upto months in the infective stage and there is a large group ofspecies which facilitates comparative studies. However,entomopathogenic nematodes diVer from other parasiticnematodes because of their mutualistic association withbacteria. In this review, we address the basic studies thatattempt to examine the theoretical underpinnings of*Corresponding author.E-mail address: [email protected] (E.E. Lewis).E.E. Lewis et al. / Biological Control 38 (2006) 66–79 67entomopathogenic nematode behavior, and the potentialapplications of these Wndings to biological control of insectpests.We cannot know all biological aspects of all species ofentomopathogenic nematode. Therefore, we must pick andchoose our questions and experimental subjects well. Here,we ask which aspects of entomopathogenic nematode biol-ogy are most important to understand to improve their suc-cess as biological control agents. We divide this review intofour sections that describe the behavioral and ecologicalaspects of the infection process for entomopathogenic nem-atodes in chronological order. The Wrst section describeshost-Wnding behaviors and strategies. The second section,and next step in infection, is a mechanistic look at how IJsrecognize hosts and how they gain entrance into the hosthemocoel. Third is a section on infection strategies and theinteraction between nematodes inside and outside the host.Finally, we examine the defense of the host cadaver fromopportunistic competitors, speciWcally ants.2. Foraging strategiesHow organisms forage for resources can be describedusing diVerent conceptual models. These models were origi-nally developed for foragers other than entomopathogenicnematode infective stages, but are broadly applicable tomany taxa and life histories. Two broad categories of con-ceptual models are (1) those based on the behavioralresponses to sequentially encountered stimuli that vary inthe quality of information that they convey and (2) thosebased on how searchers move through their environment.Both types of models can be applied to infective stage juve-nile host foraging behavior. In the Wrst category of models,host search is divided into a hierarchical process of hosthabitat location, host location, host acceptance, and hostsuitability (Doutt, 1964; Laing, 1937; Salt, 1935). This con-ceptual model has been widely adopted and has proven use-ful for understanding parasitoid host search behavior(Godfray, 1994). This hierarchical set of behaviors is notnecessarily rigid (Vinson, 1981), and more