Behavioral Modifications in Insects Induced by Parasites and Pathogens

Abstract

	Behavioral modifications have been observed in many insects as a result of parasitism or 
pathogens.  Altered behaviors can include changes in activities which result in the insect being more 
conspicuous to predation.  The insect may also alter its body temperature, foraging or oviposition time 
and location, or respond to environmental or mate stimuli differently.  These alterations in behavior 
may represent a wide range of underlying mechanisms with many possible outcomes for parasite and 
host (Moore and Gotelli 1990).  Induced behavioral alterations could have arisen for the following 
reasons as given by Horton and Moore (1993):  the modifications could be natural selection benefiting 
the host, natural selection benefiting the parasite or pathogen, or a consequence of pathology benefiting 
neither the host nor the parasite or pathogen.

	Many hypotheses have been introduced to explain how the modified behaviors evolved.  These 
theories include kin selected-host suicide, enhanced parasite transmission and survival, and host 
defense.   Kin selected-host suicide, which may be a behavior of bumblebee workers infected with 
parasitoids, results in behavior changes of the host that increase the host's predation risk with the hope 
of protecting uninfected kin.   The insects rely on inclusive fitness, that is fitness gained through the 
replication of copies of an individual's genes carried in others through the result of their own actions.  
Enhanced parasite transmission is thought to consist of behaviors induced by the parasite on the host 
which increase the chances of the parasite entering its final host.  This type of behavior may be 
characteristic of several cockroach species whose behavior becomes more conspicuous to predation 
when infected.  Parasites may also alter host behavior in order to increase the parasite's chances of 
survival, as in aposematic caterpillars and aphids infected with their respective parasitoids.   That is, 
the behavioral change may cause the host to be less susceptible to predation, thereby increasing the 
chances of parasitoid development and survival.   Host defense is best demonstrated by those insects who 
exhibit behavioral fever in response to a pathogen.  By choosing microhabitats which are warmer, the 
host is able to effectively protect itself from the pathogen by inducing physiological fever.  Examples of 
behavioral fever have been documented in the grasshopper, the cricket, the Madagascar cockroach, the 
tenebrionid beetle, and the house fly. 

	Although Poulin (1995) warns against defining these behaviors as adaptations, many of the 
behaviors do appear to have some adaptive value.  The following is a review of several articles 
demonstrating behavioral modifications of insects to parasitic and pathogenic infections.

Introduction

	Parasites are known to alter the behavior of the animals in which they live.  This behavioral 
alteration can make the host more conspicuous to predators than uninfected individuals are.  Altered 
behaviors can also include choosing different microhabitats, dietary items, and displaying different 
social status, competitive ability, attractiveness to mates, and activity levels from uninfected 
conspecifics (Moore and Gotelli 1990).  Changes in microhabitat preference in parasitized insects that 
constitute an altered behavior include seeking higher elevations, seeking exposed locations, seeking 
concealed locations, changes in reaction to light, nocturnal insects displaying diurnal activity, changes 
in temperature regime, or changes in foraging or oviposition sites (Horton and Moore 1993).

	There are many hypotheses as to why parasitized animals would exhibit behavioral changes.  The 
behaviors could be described as (1) beneficial to the parasite,  making the intermediate host more 
susceptible to predation and allowing the parasite to be transmitted to the definitive host.  This 
susceptibility to predation is only advantageous however, if the parasite is in an intermediate host 
which is preyed upon by the final host, and if the parasite has developed to its infective stage.  "By far 
the majority of documented parasite-induced changes in host behavior thought to be parasite 
adaptations are believed to enhance parasite transmission  from host to host" (Poulin 1995).  The 
behaviors may also be beneficial to the pathogen.  The changes may cause better dispersal of an air-
borne pathogen or enhance the pathogens growth rate.  (2) The modified behavior may be beneficial to 
the host.    From the intermediate host's perspective behaviors such as choosing different locations or 
lighting regimes could constitute induced physiological fever (behavioral fever) and exist as an 
mechanism to fight off the parasite.   Such situations would have arisen by natural selection to benefit 
the host.  Behavioral fever has  been observed for several insects infected with protozoans, endotoxins, 
or bacterial pyrogens.  This behavior would be considered host defense (Horton and Moore 1993).  
Other explanations for altered behavior due to parasitism include kin selected-host suicide.  In host 
suicide the host behaves in such a way as to increase the probability of death by predation in order to 
lower the risk of parasite infection for other members of the host species (Smith Trail 1980).  (3) 
Other modifications of behavior could not be adaptive to the host or the parasite, but rather the 
response to the pathological effects of the parasite (Robb and Reid 1996).

	Changes in host behavior are often reported in scientific literature and are attributed as being 
adaptive for the parasite or for the host.  Poulin (1995) suggests however that defining an adaptation 
deserves more rigorous criteria than what has been presented in the past.  Accordingly, alterations in 
host behavior following infection can be defined as adaptations only if they meet certain criteria: "(1) 
they must be complex; (2) they must show signs of a purposive design; (3) they are more likely to be 
adaptations if they have arisen independently in several lineages of hosts or parasites; and (4) they 
must be shown to increase the fitness of either the host or the parasite" (Poulin 1995).  Poulin 
(1995) indicates that few host-parasite relationships display all of these criteria.  Although many 
show a purposive design, few are complex, and the fitness benefits for most are still ambiguous.  

	The following is a review of the more recent works in the area of behavioral alterations 
observed in insects.  This review has been split into two sections: modified behaviors induced by 
parasitism and modified behaviors induced by endotoxins.  In each section examples of behavioral 
modifications are included for a variety of insects.  Also included is a hypothesis on the adaptive value of 
the behavior: beneficial to host (host defense, and kin selected-host suicide),  beneficial to parasite, or 
a result of pathology with no adaptive value to host or parasite.

Discussion

Modified behaviors induced by parasitism

	Stamp (1981) observed parasitized and nonparasitized aposematic caterpillars (the Baltimore 
checkerspot Euphydryas phaeton Drury).  She examined the level of mortality of the parasitoids 
(Apanteles euphydryidis Muesebeck) in order to test the hypothesis of host suicide, that parasitized 
caterpillars advertise themselves to their predators which increases the survivorship of their 
nonparasitized siblings.  For the host suicide hypothesis to hold, certain restrictions would have to be 
met: the parasitized host should be unable to reproduce, kin should be gregarious, possibilities should 
exist that the developing parasite would eventually parasitize the kin, and the changes in the host 
behavior should decrease the hosts survivorship (Horton and Moore 1993).  Her observations did not 
support a kinship protection-host suicide hypothesis.  Rather, the behavior of the caterpillar seemed to 
be induced by and beneficial to the parasitoid.   The caterpillars, by acting differently, were actually 
increasing their survivorship as well as the survivorship of the parasitoid by allowing the parasitoid 
to escape predation and hyperparasitism.

	The parasitic wasp Aphidius nigripes, an endoparasitoid of the potato aphid Macrosiphum 
euphorbiae, completes its pupal development within the mummified aphid host (Brodeur and McNeil 
1989).  Brodeur and McNeil (1989) not only showed that parasitized hosts behaved differently than 
unparasitized hosts did, but that the behavior depended on the developmental stage of the parasitoid.  
Aphids containing diapausing parasitoids tended to mummify in concealed areas, while aphids containing 
nondiaposing parasitoids remained on the leaf near food and other possible aphid hosts.  The aphid hosts 
of diaposing parasitoids, which were preparing to overwinter,  sought out protected areas, away from 
possible physiological and mechanical damage.  The diaposing parasitoids were also able to avoid adverse 
climatic conditions and reduce the chances of hyperparasitism by concealing themselves and their host.  
This shows one of the more clear examples of a parasite altering the host's behavior in order to 
increase its survivorship.

	The altered behaviors observed in several cockroach species also seem to support the parasite 
survival hypothesis.  Parasitism affected substrate use and activity in the cockroach species studied 
(Moore et al. 1994).  Infected male brown cockroaches, Periplaneta brunnea, spent more time on 
white horizontal surfaces than did uninfected cockroaches (Carmichael and Moore 1991).  Moore and 
Gotelli (1992) observed decreased travel velocity and distance, and increased use of horizontal 
substrates for two species of cockroach, Periplaneta americana and Blattella germanica, infected with 
the acanthocephalan Moniliformis moniliformis.   They attributed three possible factors to the shift in 
substrate use: cockroaches on horizontal surfaces may be more susceptible to predation (a behavior 
beneficial to the parasite), reduced sexual sensitivity because cockroaches usually stand on vertical 
surfaces to enhance the ability to contact females (altered sexual selection behavior), and using 
horizontal surfaces may require less energy than clinging on to vertical surfaces (a behavior reflecting 
the possible pathological constraints on energy level of the host).  

	Moore and Gotelli (1996) also looked at the phylogeny of behavior modifications in several 
cockroach species to further test the possibility that the behavioral changes were adaptive.  The fact 
that phylogenetic analysis of the components of behavioral alterations in many cockroach species to 
infection were not always shared by close relatives was encouraging, according to Moore and Gotelli 
(1996).  It supports the hypothesis that there exists a possible adaptive value at the species level to 
the altered behavior conferred on either the cockroach or the parasitic acanthocephalan.  They maintain 
that "if behavioral alterations result from physiological disruptions (neurological, hemolymph 
components, and so on) in one cockroach species, we would expect such alterations to be shared more 
readily by close relatives than distant ones, regardless of the adaptive nature of those alterations in 
every association" (Moore and Gotelli 1996).  This corresponds with Poulin's (1995) criteria that 
behaviors are more likely to be adaptations if they have arisen in several lineages independently.   

	Robb and Reid (1996) were interested in determining whether or not the flour beetle's, 
Tribolium confusum, behavioral modifications were due to pathology caused by the parasite, 
Hymenolepis diminuta, or if the parasite altered the intermediate host's behavior as an adaptive 
manipulation.  Their findings indicated that both mated status and infection affected the survivorship of 
the host, infected mated females surviving less than both infected virgin beetles and uninfected beetles.  
Behavior, however, was only altered significantly by infection of the parasite and not by mated status.  
They concluded that behavior did not appear to be a pathological response to the parasite but rather 
supported the hypothesis of  the host's behavior being modified by the parasite.  However, in Zuk's 
(1988) experiment on parasitized crickets, the findings suggest pathological consequences to 
parasitism.   Male parasitized crickets produced fewer spermatophores and had lower mating success 
than uninfected males did.  This is one example of how sexual selection was affected in an insect host due 
to the pathological consequences of the parasite load.

	Bumblebees, Bombus spp., on the other hand appear to have successfully mastered the use of 
altered behavior for their own advantage.  Mueller and Schmid-Hempel (1993) reported that the 
parasitized worker bumblebees stayed in the field overnight instead of returning to the nest.  These 
workers spent significantly more time in cold areas than did nonparasitized workers.  The cold 
temperatures experienced by the bumblebees retarded parasitoid development and decreased the 
parasitoid's survival chances.  The parasitized worker's colony benefited from the prolonged foraging in 
the cold night air, and the worker had a prolonged life span as a result of the reduced development rate 
of the parasitoid.  Poulin (1992) argues that these changes in behaviors of parasitized bumblebee 
workers are likely to be an adaptive response of the host resulting in greater inclusive fitness.  He 
notes that this may be one of the few examples of Smith Trail's (1980) kin selected-host suicide 
hypothesis in practice in nature.

	In certain instances, however, where it has been shown that a parasitized animal could benefit 
from behavioral modifications, there may be no adaptation.  For example, in nematode-parasitized 
Drosophila high temperatures had deleterious effects on the parasites.  But when given a choice of 
temperature in a thermal gradient, neither species of D. falleni nor D. neotestacea modified their 
behavior in favor of the higher temperature (behavioral fever) (Ballabeni et al. 1995).  

Modified behavior induced by endotoxins

	A host would benefit if, by altering its behavior, it could effectively harm the parasite.  Many 
ectotherms respond to a parasite by changing their microhabitat in such a way as to elevate their own 
body temperature (Horton and Moore 1993).   By doing so the animal exhibits behavioral fever.  
Recent literature on insects infected with endotoxins indicates that the host insect is able to use 
behavioral fever to its advantage against microsporidian protozoan, intracellular prokaryotes, 
bacterial endotoxins or prostaglandins, and fungus.  The  adaptive value of behavioral fever to pathogens 
can be expressed in three ways:  fever may be an adaptation of the host as a defense against the pathogen, 
a modification of the host by the pathogen to enhance growth, dispersal, and survival of the pathogen, or 
merely a side effect of infection and benefiting neither host not pathogen (Boorstein and Ewald 1987). 

	Boorstein and Ewald (1987) inoculated grasshoppers Melanoplus sanguinipes with the 
microsporidian protozoan Nosema acridophagus and showed that their preferred temperature increased.  
By maintaining the grasshoppers at both febrile and nonfebrile temperatures they were able to show 
that the febrile temperatures benefited the infected grasshoppers in survival and growth.  There is a 
cost to the fever, and febrile uninfected animals were negatively affected in growth.  However, for an 
infected animal these costs were outweighed by the benefits.  Infected insects maintained at nonfebrile 
temperatures had significantly lower fecundity, survival, and growth rates than controls.  Fever was 
beneficial to the infected animals in that they lacked significant differences from controls in fecundity, 
survival, and growth.

	Crickets Gryllus bimaculatus exhibit  behavioral fever in response to infection with 
Rickettsiella grylli, a chlamydia-like pathogen (Louis 1986).   When allowed to regulate their own 
temperature in a thermal gradient, infected insects chose higher temperatures than noninfected insects 
did.  This higher temperature caused the pathogen to degenerate.

	Two different insects display behavioral fever to endotoxins, specifically purified 
lipopolysaccharides isolated from E. Coli.  Lipopolysaccharides, components of the cell wall of Gram-
negative bacteria (Bronstein and Conner 1984), are potent pyrogens (McClain et al. 1988) that elicit 
a sequence of host-defense responses in animals.  Bronstein and Conner (1984) conducted a study of 
endotoxin-induced behavioral fever in the Madagascar cockroach Gromphadorhina portentosa.  
Lipopolysaccharide-W was injected into the insect and the mean temperature preference was 3.6 
degrees higher than that for control cockroaches.  Also, the tenebrionid beetle Onymacris plana was 
observed by McClain et al. in 1988 to develop behavioral fever in response to lipopolysaccharide 
injection and sought higher preferred temperatures in a gradient.  To be certain that the body 
temperature was indeed affected by the position of the beetle on the thermal gradient, thermocouples 
were implanted in the thoracic musculature.  

	A final example of behavioral fever is the modification in temperature preference induced in 
house flies Musca domestica when infected with the fungi Entomophthora muscae (Watson et al. 1992).  
In experiments where flies were not given a choice of temperature, higher temperatures increased the 
survival period of the infected flies.  In thermal gradients the flies preferred the higher temperatures, 
exhibited behavioral fever, and were able to eliminate the pathogen using heat therapy. 

	With few exceptions, ectothermic animals, like endotherms will develop fevers in response to 
injections of endotoxins or other pyrogenic substances as a result of the animal "feeling" cold and 
selecting a warmer microclimate (Kluger et al. 1996).  Kluger et al. suggests that febrile responses 
are most likely adaptive because if they weren't, than "it would be unlikely that this energetically 
expensive phenomenon would have persisted for millions of years in so many groups of animals".  With 
the exception of the bumblebee worker, most other insects are harmed by pathogens when their body 
temperature is lowed and protected when they develop fevers (Kluger et al. 1996).  Kluger et al. also 
indicates that this observation and the fact that fevers are highly regulated, that they follow a 
relatively common sequence of events within the body, both support the argument that fever has 
evolved as a host defense response.  

Conclusion

	Insects alter their behavior when they are under the stress of infection by parasites or 
pathogens.  Such behavioral changes include seeking out warmer or cooler temperatures, changing their 
microhabitat, and acting more conspicuously.  Many times the behavior will increase the probability 
that the insect will be preyed upon.  This is often attributed as an adaptation of the parasite to alter its 
host's behavior in order to increase its transmission to a final host.  It is also often the case that the 
host insect will harm the parasite or pathogen by changing its behavior, such as increasing its body 
temperature, as seen in many of the insects infected with endotoxins.   Seeking out warmer 
environments in order to induce a physiological fever constitutes a behavioral fever and has been shown 
to effectively protect the insect from the pathogen.  Whether some adaptations favor the parasite's 
survival or the host's is still unclear in many of the parasite-host relationships.  Further studies are 
necessary, as Poulin (1995) suggests, in order to critically determine whether a behavior is adaptive.  
It is important to determine the fitness costs of behavioral modification, the phylogeny of the behavior 
among close and distant relatives,  and the pathological changes caused by the parasite and the effects 
that those changes may have on host behavior.

References

Ballabeni, P., H. Benway, and J. Jaenike. 1995. Lack of Behavioral Fever in Nematode-Parasitized 
Drosophila. Journal of Parasitology 81(5): 670-674.

Boorstein, S.M., and P.W. Ewald.  1987. Costs and Benefits of Behavioral Fever Melanoplus sanguinipes 
Infected By Nosema acridophagus. Physiological Zoology 60(5): 586-595.

Brodeur J., and J.N. McNeil. 1989. Seasonal Microhabitat Selection by an Endoparasite Through 
Adaptive Modification of Host Behavior. Science 244: 226-228.

Bronstein, S.M., and W.E. Conner. 1984. Endotoxin-Induced Behavioral Fever in Madagascar 
Cockroach, Gromphadorhina portentosa. Journal of Insect Physiology 30(4): 327-330. 

Carmichael, L.M., and J. Moore. 1991. A Comparison of Behavioral Alterations in the Brown Cockroach, 
Periplaneta brunnea, and the American Cockroach, Periplaneta americana, Infected With the 
Acanthocephalan, Moniliformis moniliformis. Journal of Parasitology 77(6): 931-936.

Gotelli, N.J., and J. Moore. 1992. Altered Host Behavior in a Cockroach- Acanthocephalan Association.  
Animal Behavior 43: 949-959.

Horton, D.R., and J. Moore. 1993. Behavioral Effects of Parasites and Pathogens in Insect Hosts. In: 
Parasites and Pathogens of Insects (Ed. by N.E. Beckage, S.N. Thompson, and B.A. Federici), pp. 107-
124. San Diego: Academic Press.

Kluger, M.J., W. Kozak, C.A. Conn, L.R. Leon, and D. Soszynski. 1996. The Adaptive Value of Fever. 
Infectious Disease Clinics of North America 10(1): 1-21.

Louis, C., M. Jourdan, and M. Cabanac. 1986. Behavioral Fever and Therapy in a Rickettsia-Infected 
Orthoptera. American Journal of Physiology 250: R991-R995. 

McClain, E., P. Magnuson, and S.J. Warner. 1988. Behavioral Fever in a Namib Desert Tenebrionid 
Beetle, Onymacris plana. Journal of Insect Physiology 34(4): 279-284. 

Moore, J., M. Freehling, and N.J. Gotelli. 1994. Altered Behavior in Two Species of Blattid Cockroaches 
Infected With Moniliformis moniliformis (Acanthocephala). Journal of Parasitology 80(2): 220-223.

Moore, J. and N.J. Gotelli. 1990. A Phylogenetic Perspective on the Evolution of Altered Behaviors: a 
Critical Look at the Manipulation Hypothesis. In: Parasitism and Host Behavior (Ed. by C.J. Barnes and 
J.M. Behnke), pp. 193-233. London: Taylor & Francis.

Moore, J. and N.J. Gotelli. 1996. Evolutionary Patterns of Altered Behavior and Susceptibility in 
Parasitized Hosts. Evolution 50(2): 807-819.

Mueller, C.B., and P. Schmid-Hempel. 1993. Exploitation of Cold Temperature as Defense Against 
Parasitoids in Bumblebees. Nature 363: 65-67.

Poulin, R. 1992. Altered Behavior in Parasitized Bumblebees: Parasite Manipulation or Adaptive 
Suicide? Animal Behavior 44: 174-176.

Poulin, R. 1995. "Adaptive" Changes in the Behavior of Parasitized Animals: A Critical Review. 
International Journal of Parasitology 25(12): 1371-1383.

Robb, T., and M.L. Reid. 1996. Parasite-Induced Changes in the Behavior of Cestode-Infected Beetles: 
Adaptation or Simple Pathology? Canadian Journal of Zoology 74: 1268-1274.

Smith Trail, D.R. 1980. Behavioral Interactions Between Parasites and Hosts: Host Suicide and the 
Evolution of Complex Life Cycles. The American Naturalist 116(1): 77-91.

Stamp, N.E. 1981. Behavior of Parasitized Aposematic Caterpillars: Advantageous to the Parasitoid or 
the Host? The American Naturalist 118(5): 715-725.

Watson, D.W., B.A. Mullens, and J.J. Petersen. 1992. Behavioral Fever Response of Musca domestica 
(Diptera: Muscidae) to Infection by Entomophthora muscae (Zygomycetes: Entomophthorales). Journal 
of Invertebrate Pathology 61: 10-16.

Zuk, M. 1988. Parasite Load, Body Size, and Age of Wild-Caught Male Field Crickets (Orthoptera: 
Gryllidae): Effects on Sexual Selection.  Evolution 42(5): 969-976.