Access to this full-text is provided by American Society for Microbiology.
Content available from mBio®
This content is subject to copyright. Terms and conditions apply.
Mechanisms behind the Madness: How Do Zombie-Making
Fungal Entomopathogens Affect Host Behavior To Increase
Transmission?
Charissa de Bekker,
a
,
b
William C. Beckerson,
a
,
b
Carolyn Elya
c
a
Department of Biology, College of Sciences, University of Central Florida, Orlando, Florida, USA
b
Genomics and Bioinformatics Cluster, University of Central Florida, Orlando, Florida, USA
c
Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, Massachusetts, USA
ABSTRACT Transmission is a crucial step in all pathogen life cycles. As such, certain
species have evolved complex traits that increase their chances to find and invade
new hosts. Fungal species that hijack insect behaviors are evident examples. Many
of these “zombie-making”entomopathogens cause their hosts to exhibit heightened
activity, seek out elevated positions, and display body postures that promote spore
dispersal, all with specific circadian timing. Answering how fungal entomopathogens
manipulate their hosts will increase our understanding of molecular aspects underly-
ing fungus-insect interactions, pathogen-host coevolution, and the regulation of ani-
mal behavior. It may also lead to the discovery of novel bioactive compounds, given
that the fungi involved have traditionally been understudied. This minireview sum-
marizes and discusses recent work on zombie-making fungi of the orders
Hypocreales and Entomophthorales that has resulted in hypotheses regarding the
mechanisms that drive fungal manipulation of insect behavior. We discuss mechani-
cal processes, host chemical signaling pathways, and fungal secreted effectors pro-
posed to be involved in establishing pathogen-adaptive behaviors. Additionally, we
touch on effectors’possible modes of action and how the convergent evolution of
host manipulation could have given rise to the many parallels in observed behaviors
across fungus-insect systems and beyond. However, the hypothesized mechanisms
of behavior manipulation have yet to be proven. We, therefore, also suggest avenues
of research that would move the field toward a more quantitative future.
KEYWORDS animal behavioral change, coevolution, host specialization, effectors,
Hypocreales, Entomophthorales
Understanding a pathogen’s chain of transmission—how it enters a susceptible
host, causes infection, and transmits afterward—is fundamental to the study of in-
fectious disease (1). While adaptive evolution in any of these links leads to more fit
pathogen populations, it is the selective pressures on transmission that have given rise
to the intricate spore dispersal strategies of so-called “zombie-making”entomopatho-
genic fungi. Tiny insect hosts create a dual problem for transmission as they provide
small amounts of carbon to sustain spore production and little surface area for spore
structure presentation. The sheer biomass of epizootics, such as the large, seasonal
outbreaks in house flies caused by Entomophthora muscae (2–7), could compensate for
this. Moreover, certain entomopathogen species have optimized their dispersal by
instigating host behaviors that promote direct contact with conspecifics or abiotic
modes of spread. Indeed, E. muscae, like many other behavior-manipulating fungi,
induces its fly hosts to summit and attach to positional vantage points that positively
affect wind-mediated spore dispersal and to spread their wings to provide more
Citation de Bekker C, Beckerson WC, Elya C.
2021. Mechanisms behind the madness: how
do zombie-making fungal entomopathogens
affect host behavior to increase transmission?
mBio 12:e01872-21. https://doi.org/10.1128/
mBio.01872-21.
Editor Danielle A. Garsin, University of Texas
Health Science Center at Houston
Copyright © 2021 de Bekker et al. This is an
open-access article distributed under the terms
of the Creative Commons Attribution 4.0
International license.
Address correspondence to Charissa de Bekker,
charissa.debekker@ucf.edu.
Published
September/October 2021 Volume 12 Issue 5 e01872-21 ®mbio.asm.org 1
MINIREVIEW
5 October 2021
surface for spore production (8). As such, both infection quantity (e.g., number of indi-
viduals infected) and quality (e.g., elevation and body posture of infected individuals)
are seemingly important for effective transmission.
An estimated 1.5 million fungal species are entomopathogenic (9), demonstrating
their substantial roles across ecosystems. Nevertheless, fungal entomopathogens have
historically received less attention compared to their plant pathogen counterparts (10).
This leaves much to be discovered about their modes of infection and transmission,
natural products, and effector secretion during pathogen-host interaction. Beauveria
(also Cordyceps)bassiana and several species of Metarhizium are among the relatively
few fungal species that have been studied from a mechanistic perspective (11–13).
These few characterized species represent generalist fungi that kill and consume their
hosts in a matter of days upon infection without eliciting obvious pathogen-adaptive
behavioral alterations. As such, one could consider these fungi to have a necrotrophic
lifestyle. This is in contrast with behavior-manipulating species, which have a limited to
highly specific host range and can spend more time in a symbiotic relationship with
their insect host prior to killing it: a lifestyle that could be considered more hemibiotro-
phic, as a premature death of the host disrupts the fungal pathogen’s life cycle (14).
Therefore, Beauveria and Metarhizium might not be the most appropriate species to
draw parallels from when forming hypotheses about the molecular workings of manip-
ulating pathogens.
Moreover, there are many other examples of insect-manipulating fungi that
are not part of the order Hypocreales (Phylum: Ascomycota) like Beauveria and
Metarhizium. Another large, diverse group of zombie-making fungi reside in the
order Entomophthorales (Phylum: Zoopagomycota). Major fundamental differences
between the life cycles and structures of Ascomycota and Zoopagomycota likely impact
the molecular underpinnings of host manipulation. As such, our knowledge of the mech-
anisms that drive fungus-adaptive behavioral phenotypes is limited, especially with
regard to the less lab-amenable Entomophthorales species (15). Despite these chal-
lenges, mycologists have begun to expand fungal biology, ecology, and evolution stud-
ies to include more obscure entomopathogenic species and explore their genomes, tran-
scriptomes, proteomes, and metabolomes in recent years (16–22). This vastly growing
body of work has led to various hypotheses about the mechanistic underpinnings that
these fungi might employ to manipulate their insect hosts. Since host manipulation is a
trait that has evolved multiple times independently (23, 24), the question arises whether
zombie-making fungi could have evolved comparable mechanisms to infect and manip-
ulate their hosts. In this review, we discuss the mechanistic hypotheses that have been
posed based on manipulating-fungus research across the spectrum (Table 1). As such,
we review previously reported studies of behavior-manipulating Hypocreales and
Entomophthorales species, compare data to investigate if there are parallels that would
suggest the evolution of comparable host manipulation mechanisms, and propose next
steps to test those hypotheses.
MANIPULATED INSECT BEHAVIORS: PARALLELS ACROSS FUNGI
Manipulated insect behaviors increase fungal fitness by making spore dispersal
more effective, either prior to or after host death. Massospora, for example, produces
infective spores while keeping its cicada hosts alive, coercing them to actively transmit
the infection to conspecifics (25). Success of Massospora spp. depends on their ability
to consume the host and rupture its abdomen to produce spores, while keeping the
insect intact enough to remain active and facilitate dispersal. Behavior-manipulating
fungi that kill their insect hosts prior to spore production are not so different in their
approach to increase transmission. Many entomophthoralean and hypocrealean fungi
rely on host movement to transport them toward conditions that favor spore produc-
tion and dissemination. Fungal manipulators that infect eusocial insects (i.e., Pandora
and Ophiocordyceps species) appear to particularly benefit from increased locomotion
activities in their hosts (21, 26–28). Ophiocordyceps-infected ants display a directionless,
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 2
TABLE 1 Observed behavior manipulations by entomopathogenic fungi with their proposed fungal benefit, hypothesized underlying fungal mechanisms, and potential host pathways of
action
Induced behavior Fungal benefit Fungal mechanism Host pathway(s) Example fungi (hosts) Example references
Time-specific behaviors Aligns fungal emergence with
favorable abiotic factors
Effector secretion, disruption of
sensory periphery
Biological clock Ophiocordyceps spp. (Camponotus);
Entomophthora muscae (Musca domestica,
Drosophila)
17, 28, 36, 47, 48, 77
Light seeking Positions host in favorable
microenvironment
Effector secretion Biological clock,
phototaxis
Ophiocordyceps spp. (Camponotus,
Colobopsis)
32,33
Hyperactivity Avoidance of social immunity,
facilitates summiting
Effector secretion, host nutrient
depletion
Locomotion, arousal,
hunger
Ophiocordyceps spp. (Camponotus)21,28
Summit disease Increases spore dispersal Effector secretion Thermotaxis, phototaxis,
gravitaxis
Entomophthora muscae (Musca domestica,
Drosophila); Entomophaga grylli (Melanoplus
bivittatus); Ophiocordyceps spp.
(Camponotus,Colobopsis); Eryniopsis
lampyridarum (Chauliognathus
pensylvanicus)
8, 35, 36, 39, 49
Surface adherence Prevents falling from vantage
points that increase spore
dispersal
Hydrophobic protein secretion,
growth in/around mandibular
muscle, hyphal anchoring
Proboscis; mandibles
and legs
Entomophthora muscae (Musca domestica,
Drosophila); Pandora (Formica),
Ophiocordyceps spp. (Camponotus)
8, 17, 30, 36, 58, 121
Splayed wings Removes barriers for spore
dispersal
Growth patterns in/around
thoracic muscle
Intrathoracic pressure Entomophthora muscae (Musca domestica,
Drosophila); Eryniopsis lampyridarum
(Chauliognathus pensylvanicus)
8, 17, 36, 49
Increased sexual behavior Increases transmission via
direct contact
Effector secretion Sexual arousal,
locomotion
Massospora (Cicada)19,25
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 3
constant locomotion activity that hampers effective foraging efforts (28). Such wander-
ing behavior could cause infected individuals to stray from the ant colony before the
infection is noticed by nest mates. This is likely essential for fungal survival as healthy
ants attack infected individuals as part of their social immunity strategy, which interferes
with eventual spore formation and transmission (21, 28). Therefore, host activity can also
be viewed as a spore dispersal strategy used by fungi that transmit after host death.
Hyperlocomotion and wandering likely also aid in establishing the next manipula-
tion step that is often induced by fungi that kill their hosts prior to transmission: sum-
miting. Many manipulating Entomophthorales and Hypocreales species coerce their
insect hosts to ascend vegetation toward the end of the infection (8, 29). Ejecting
spores from an elevated position is thought to increase transmission through more
effective wind dispersal (30, 31). The upward locomotory movement may be accompa-
nied by “light-seeking behavior,”as suggested for Ophiocordyceps camponoti-atricipis
and Ophiocordyceps pseudolloydii (32, 33). Additionally, cooler-temperature seeking by
E. muscae has been observed during late stages of infection, though it is not yet known
if this behavior is a fungal manipulation or a general response to sickness (34). Overall,
inducing a preference for specific abiotic factors could increase the likelihood that
insect hosts die at elevated positions, serving to distance them from aggressive con-
specifics as well as facilitate the most optimal microhabitat (i.e., temperatures, humid-
ity, and light levels) for fungal growth (30, 32). Indeed, experiments that manipulated
incipient light levels or removed Ophiocordyceps-infected cadavers from their original
positions largely disrupted fungal fruiting body development (14, 32, 35).
Once the hijacked host has arrived at its final destination, fungal pathogens that
cause summiting often induce one final behavior to ensure substrate adherence.
Entomophthora species that infect dipterans cause them to attach to vegetation by
extending their proboscis (17, 36), while Pandora and Ophiocordyceps species induce
biting behavior in their hymenopteran hosts (30, 37). Depending on the substrate type
and fungal species involved, this behavior can be accompanied by the insect folding
its legs around the vegetation for additional support with the fungus sometimes fur-
ther securing the position by “fixing”the cadaver with hyphae (16, 23, 33, 36, 38, 39).
Additionally, Entomophthora-infected insects often extend their wings out of the line
of fire for ejected spores, providing a clear path to new hosts (17, 36).
Crucially, this final cascade of manipulated behaviors does not take place at just
any time of day. Studies noting the timing of manipulated behaviors for particular fun-
gus-insect interactions, as well as other infectious agents beyond the fungal kingdom
(40–46), indicate that they occur with a certain daily timing. Ophiocordyceps-infected
ants displayed biting behavior around solar noon (35), likely to facilitate light-seeking
behavior (32, 47, 48), Eryniopsis lampyridarum-infected soldier beetles die in the morn-
ing (49), while Entomophthora-infected flies die at dusk (17, 36). As such, the manipula-
tion of daily timing in the host appears to be an additional strategy that can be found
across the board.
Fungi that independently evolved behavior-hijacking strategies to enhance transmis-
sion cause manipulated behaviors with several parallels across fungus-insect interactions
(46, 50–52) (Fig. 1). Notably, ascension behaviors are also induced by nonfungal patho-
gens and parasites, such as baculoviruses and trematodes (38–40, 53). For one baculovi-
rus strain, two virus genes (i.e., ptp and egt) are seemingly involved in causing enhanced
locomotion and climbing behavior in its caterpillar host (54, 55). Additionally, histology
work on trematode species that infect ants showed that the parasite’s physical position-
ing relative to host nervous tissues determines if climbing is accompanied by biting
behavior (53, 56, 57). Behavioral parallels between these zombie-makers and fungi make
it plausible that comparable strategies to hijack host behaviors have evolved.
POSSIBLE MECHANISMS
The endoparasitic life cycle of entomopathogenic fungi is intertwined with host tis-
sue infiltration and consumption. Since zombie-making fungi rely on able-bodied hosts
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 4
to execute fungus-adaptive, manipulated behaviors, they are likely to interact with
host tissues in a more careful approach than a simple consume-all strategy. The deli-
cate relationship between fungal cells and host tissues might have led to the evolution
of successful host infiltration strategies that ultimately give rise to altered behaviors.
Such strategies can be lumped into two major categories: chemical and mechanical
processes. While chemical means of host manipulation include a wide breadth of
secreted biomolecules that disrupt and otherwise manipulate the host cellular machin-
ery at the molecular level, mechanical methods of manipulation involve broader physi-
cal changes to the host, e.g., tissue-specificinfiltration and destruction, changes in in-
ternal pressures, and spore production in particular compartments.
Mechanical processes. Entomopathogenic fungi vary in which host tissues they
occupy, nervous tissue being a prime example. There is mounting evidence that
entomophthoralean species (E. muscae and Entomophthora aphidis,Entomophaga
grylli,Strongwellsea castrans,Conidiobolus coronatus,Pandora formicae)infiltrate the
brain tissue of their hosts (fruit flies, aphids, grasshoppers, anthomyiid flies, mosqui-
toes, and ants, respectively) while the hosts are still alive (17, 26, 58–62). In contrast,
Ophiocordyceps species do not infiltrate the nervous tissue of their carpenter ant
hosts until after death, despite their unabated expansion into the muscle tissue (35,
63, 64). Additionally, these two groups differ in the rate of disease progression.
Ophiocordyceps infections take place slowly, over the course of weeks, providing
more time for manipulation of behaviors that drive the ant away from the nest and
upward (18, 21, 27). Pandora,ontheotherhand,infiltrates its ant hosts in a matter of
days, resulting in mound biting or gripping onto blades of grass near the ant mound
(26). The rate of pathogenicity and the rate of fungal expansion into various host tis-
sues are, therefore, likely tied to selective pressures requiring differing degrees of
host manipulation.
Another notable divergence between host-manipulating fungi is observed in the
physical manipulation of the host immediately perimortem or postmortem. Summiting
behavior, for example, favors a method to anchor the host to its substrate to provide
FIG 1 Differences between generalist and specialist, zombie-making fungal entomopathogens and the overlap between manipulated
behaviors observed across specialist pathogen-host systems.
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 5
enough time to produce spore-bearing structures without falling from its optimal
spore dispersal vantage point. Adherence of the host is in many species accomplished
via induction of biting behaviors collectively called the “death grip.”The mechanisms
behind the death grip are still being explored and likely involve a combination of me-
chanical and chemical means. The death grip behavior in Ophiocordyceps-infected
hosts may be, in part, caused by overgrowth of fungal tissue in the legs and mandibles
(64). The fungus forms interconnected cells around individual mandibular muscle cells,
occupying ;40% of the biomass and causing muscle fibers to become widely sepa-
rated. Separation of the muscle tissues causes Z-lines to become swollen, sarcomeres
to shorten, and muscle fiber sarcolemma to degrade. In addition, hyphal bodies and
extracellular vesicle-like particles were in direct contact with mandibular muscles, indi-
cating that compound secretion may also play a role in muscle contraction (64).
Mechanical means of adherence are likely facilitated by secretory compounds in other
species as well. In hosts without mandibles, such as fruit flies infected by E. muscae,the
proboscis is extended toward a substrate in a mechanical manner unlike the extension
observed during eating. This proboscis extension is paired with the production of sticky
secretions that are yet uncharacterized but hypothesized to be of fungal origin (17, 58).
Once the host is fixed to a surface, the fungus can emerge and disseminate spores.
To make way for emerging conidiophores that carry the primary conidia in winged
hosts, mechanical manipulation of the wings lifts them up and away from the dorsal
abdomen (17). The raising of the wings observed in E. muscae-infected Drosophila hap-
pens over approximately 15 min and is thought to be caused by an increase in pres-
sure against thoracic muscle (17). By raising the wings off the back, conidia can be for-
cibly ejected without being blocked (17). This splayed wing phenotype has also been
studied in the soldier beetle Chauliognathus pensylvanicus infected with E. lampyrida-
rum; however, Eryniopsis-induced wing raising takes much longer, occurring over the
course of several hours (49). In addition to applying physical pressure to host tissue,
entomopathogenic fungi consume and, thus, destroy it through the secretion of pro-
teases, lipases, and chitinases (16, 22, 27, 65). This results in physical stress by impairing
tissue integrity and in chemical stress by removing cells that produce signaling mole-
cules involved in physiological homeostasis, either of which could trigger behavioral
responses in the host. However, it seems unlikely that tissue destruction itself accounts
for modified host behaviors that are pathogen adaptive, given that generalist entomo-
pathogens like Beauveria and Metarhizium also destroy tissues in their living hosts and
do not evoke the same, precise, behavioral changes.
While many mechanical host manipulations seem to be conserved across behavior-
modifying entomopathogens, unique examples also exist. Massospora cicadina, a fun-
gus that infects cicadas, causes destruction of the terminal abdomen where the genita-
lia of the host are located and replaces it with a fungal spore mass (25). In addition to
removing the reproductive organs of the host to make way for infective conidiospores,
Massospora species seem to hijack host reproductive behavior to spread more rapidly
via direct contact. This strategy is likely facilitated by the secretion of compounds that
increase the host’s desire to mate (25). Fungal overgrowth has also been implicated in
the dysmotility observed in the antennae of Ophiocordyceps-infected ants. The anten-
nae of Ophiocordyceps camponoti-floridani-infected Camponotus floridanus are com-
monly observed to be locked in a bent L-shaped position late in infection (28). While
antennal movements are important for communication with nest mates and naviga-
tion, their obstruction likely inhibits these behaviors and may therefore contribute to
the wandering behavior that drives the ant away from the nest.
While mechanical manipulations are seen across the spectrum of zombie-making
fungi, their degree of influence throughout the infection process is still being explored.
Studies involving scanning and transmission electron microscopy on the mandibular
muscles of Ophiocordyceps-infected hosts are broadening our understanding of how
these fungi position themselves within the host at the point of behavior manipulation
(64). Further studies into the physical manipulation of hosts by host-specific behavioral
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 6
manipulators and more generalist nonmanipulators may provide further insight into
the elaborate colonization strategies that have evolved in zombie-making fungi. While
it is clear that physical manipulation of the host is vital for completion of the fungal life
cycle, many, if not all, of these forced behaviors are facilitated in part via chemical sig-
naling between these pathogens and the neuromuscular systems of their hosts.
Understanding the collaboration between physical and chemical means of host manip-
ulation is necessary to understand how these pathogen-driven modifications have
evolved.
Chemical signaling. Chemical signaling could be either directly or indirectly neuro-
modulatory: secreted fungal factors could act directly on the neural circuits underlying
a given behavior or they could alter upstream inputs to these circuits, thereby trigger-
ing pathways that ultimately result in changed behavioral output. As mentioned, zom-
bie-making fungi differ with respect to invasion of nervous tissue: hypocrealean fungi
do not invade while entomophthoraleans do. This suggests that the two major clades
of zombie-making fungi may employ complementary strategies to achieve broader be-
havioral changes. For Entomophthorales species, direct access to host neuropil may
indicate a more direct approach to altering circuits underlying behavioral changes (i.e.,
secreted factors could act directly on neurons to change behavioral output). Lack of
such direct access in hypocrealean fungi may indicate that behavior is altered through
indirect routes (i.e., modulating internal state or integrity of tissue). Alternatively, the
position of fungal cells relative to host nervous tissue may be dispensable for behav-
ior alteration (i.e., both direct and indirect chemical signaling are possible mecha-
nisms for behavior-manipulating fungi, regardless of their clade and mode of me-
chanical interaction).
Fungi could employ chemical signaling to alter host behavior via depletion of host
nutritional reserves or via modulating host physiology/neurobiology through the
release of chemical effectors. Depletion of host nutritional reserves resulting from con-
sumption by the fungal pathogen can induce a starvation state that ultimately leads to
host behavior changes as it seeks to replenish its stores. Transcriptomic data provide
ample evidence of host starvation: in Ophiocordyceps-manipulated ants, the expression
levels of genes implicated in response to starvation were found to be altered (e.g.,
lipase, amylase, insulin, juvenile-hormone-responsive cytochrome p450), while in E.
muscae-infected fruit flies, metabolism Gene Ontology (GO) terms were significantly
enriched among downregulated genes during late infection (17, 21). Moreover,
increased locomotor activity as a result of acute starvation has been reported in several
insects (66–69) and is reminiscent of the activity that insects demonstrate before they
are killed by fungal pathogens (21, 28). However, nutrient depletion alone seems
unlikely to account for the moribund behaviors of zombie fungus-infected insects
since starving insects do not manifest summiting behavior or distinct postural changes.
Given the tight temporal coupling of host resource depletion and manipulated behav-
iors, it will be important to tease out the role that starvation plays in driving observed
end-of-life behaviors (i.e., how much does shifting internal state contribute to behavior
manipulation). Studies using genetic tools or pharmaceuticals to prevent the host from
sensing starvation or studies ectopically inducing satiety in nutrient-depleted hosts
could be a reasonable starting point for this work.
Entomopathogenic fungi could also indirectly trigger behavioral responses through
effectors, which we define as any molecule (e.g., metabolite, protein, or other biopoly-
mer) that impacts host physiology. There is mounting evidence for the importance of
protein effectors in zombie ants: transcriptomic analyses of two distinct Ophiocordyceps-
ant interactions (i.e., Ophiocordyceps kimflemingiae-infected Camponotus castaneus and
O. camponoti-floridani-infected C. floridanus) have revealed a putative enterotoxin to be
highly expressed during manipulated summiting and biting behavior (27). While ento-
mopathogenic fungi contain several enterotoxin-encoding genes in their genomes (with
Ophiocordyceps species so far having been found to have the most), comparative
genomic analysis found that orthologs of this specific enterotoxin are seemingly
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 7
exclusively conserved among ant-manipulating Ophiocordyceps species (18). This sug-
gests that this compound plays an important role in altering host ant behavior (21).
In addition, genomics and transcriptomics analyses have identified a large reper-
toire of small secreted proteins (SSPs) encoded by Ophiocordyceps spp. of which a sig-
nificant number are upregulated during manipulated biting behavior (18, 21, 27). This
suggests that SSPs may be promising candidates for behavior modifications. Many of
these predicted SSPs have unknown functions since they cannot be classified based on
known Pfam domains or GO terms. Additionally, they are often highly species specific,
complicating homology-based annotations (21, 27). As such, SSPs comprise an interest-
ing class of bioactive molecules that warrant functional investigations, not only
because they could be key to the mechanisms underlying behavioral manipulation but
also because they may result in the discovery of novel compounds that have medicinal
or pest control applications.
Metabolite effectors have also been implicated in zombie behaviors. A recent study
identified the presence of the alkaloids psilocybin (the active ingredient in hallucino-
genic magic mushrooms) and amphetamine cathinone in Massospora levispora- (syn-
onymous with Massospora platypediae [70]) and M. cicadina-infected cicadas (19).
Analyses of genome sequences for these fungi revealed homologues for some of the
genes known to be involved in synthesizing these alkaloids, while others were conspic-
uously absent. This led to the speculation that Massospora might possess novel means
of biosynthesizing these compounds. Psilocybin and cathinone, like many other alka-
loids, have well-known behavioral effects. This makes it plausible that these com-
pounds are involved in the increased activity and hypersexuality behaviors observed in
Massospora-infected cicadas (25). Additionally, O. kimflemingiae and O. camponoti-flori-
dani contain at least one alkaloid-producing metabolite cluster that is predicted to be
an aflatrem-like compound. The cluster was highly expressed during manipulation (21,
27), further suggesting that metabolite effectors may play important roles during
behavior manipulation. However, several Metarhizium species have been shown to pro-
duce ergot alkaloids in live insect hosts but not in dead insects or on artificial media.
This suggests a role for ergot alkaloids in insect colonization (71). Thus, though these
findings are compelling, we cannot exclude the possibility that the role of these ergot
alkaloids is restricted to killing or colonizing the ant host rather than driving behavior
manipulation. Studies using fungal mutant strains for the genes that give rise to these
compounds, as well as experiments testing the behavioral effects of these compounds
in uninfected animals, will be helpful to determine the role they play in behavior
manipulation (19). Beyond alkaloids, other metabolite effectors have also been impli-
cated in Ophiocordyceps-ant brain interactions. In one study, brains of four different
ant species, of which one was the naturally infected and manipulated host, were cocul-
tured ex vivo with O. kimflemingiae and liquid chromatography-tandem mass spec-
trometry (LC-MS/MS) analysis was performed to detect the presence of secreted com-
pounds (72). Resultant analysis found that O. kimflemingiae secreted a specific set of
metabolites depending on the species of ant brain it encountered. Additionally, two
compounds with potential neurobiological function, guanidinobutyric acid (GBA) and
sphingosine, were found to be uniquely present in cocultures with the brain of the nat-
ural host, C. castaneus (72). A subsequent study looking at the metabolites in brains of
O. kimflemingiae-infected C. castaneus confirmed that infection has a significant impact
on brain metabolism even though the fungus does not physically contact the brain
during manipulation (20). The study also detected a dramatic increase of ergothioneine
in the brains of O. kimflemingiae-manipulated hosts compared to those of healthy
hosts and hosts dying from infection by the non-behavior-modulating generalist B.
bassiana. The authors of this study suggest that fungus-derived ergothioneine could
be preserving nervous tissue during fungal infection (20).
It is not yet clear which putative effector genes can be implicated in behavior
manipulation by entomophthoralean fungi. Both a paucity of genomic data (owing to
extremely large, repeat-rich, and assembly-averse genomes) and their much greater
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 8
phylogenetic distance from well-studied Ascomycota (often impeding inference of
function by homology) mean we currently know much less about the genes expressed
by these fungi over the course of infection and during manipulation. The predominant
hypothesis in the field is that entomophthoralean fungi refrain from producing effec-
tors to elude immune detection from their host, as well as to avoid premature host
death (73, 74), but this should be reevaluated as more information becomes available.
Possible sites of action. Based on altered behavioral phenotypes, several pathways
stand out as potential targets for fungal effectors (Table 1). Given the consistent circa-
dian timing of “zombified”insect behaviors, one system likely targeted by fungal effec-
tors is the host circadian clock. Work in E. muscae has found that house flies infected
and housed in complete darkness (i.e., free-running conditions) die without consistent
timing (36), even though healthy house flies maintain rhythmicity under free-running
conditions (75). In contrast, flies infected and housed under a light-dark cycle for 72 h
(i.e., entrainment conditions) prior to housing in darkness do demonstrate synchron-
ized timing of death (36). The most parsimonious explanation for this observation is
that the host clock does not drive timing of moribund behaviors and death. These
events follow another clock, which requires Zeitgeber cues during early infection.
Transcriptomic data have revealed that E. muscae expresses a homologue of white-col-
lar 1 (17), a photoreceptor and core component of the molecular clock in Neurospora
crassa (76). This raises the possibility that the alternative clock that drives moribund
behaviors could belong to E. muscae (17). In a similar vein, the O. kimflemingiae ge-
nome contains several homologues of N. crassa clock components, which are
expressed in a circadian manner and drive the daily expression of genes annotated to
be involved in pathogen-host interactions (77). Moreover, some of the host clock
genes appear to be dysregulated in ants that display manipulated biting behavior (21,
27), suggesting that the fungal clock is also at play in the zombie ant system. In addi-
tion, or alternative to clock-regulated fungal effectors, changes in host timing may
result from diminished sensing of Zeitgeber cues due to fungal cell growth and tissue
integrity loss, which could lead to phase shifts and amplitude changes of host daily
rhythms. Future research that bridges chronobiology and pathogen-host interactions
(78) should address this possibility.
Pathways that control locomotion also seem likely targets of manipulation by fun-
gal effectors. For example, zombie ants show increased locomotion prior to death (21,
28). Such enhanced locomotion could arise through modulation of the host clock since
the circadian network affects locomotor and sleep output (79). Given that summit dis-
ease is closely linked with locomotion and that directionality is indicated by environ-
mental cues such as light and gravity, perhaps fungal effectors target phototaxis or
gravitaxis pathways to drive hosts to climb nearby substrates. Work in O. camponoti-
atricipis has found that infected ants are more likely to be found in well-lit than shaded
areas and that the height that infected ants climb before their death varies with the
amount of light in their immediate environment (32) This suggests that phototaxis
pathways are altered in zombie ants to make them “light seekers.”
So far, our best understanding of pathogen-induced enhanced locomotory activity
(ELA) and summiting comes, not from behavior-modulating fungi, but from baculovi-
ruses. In Spodoptera exigua and Bombyx mori, it has been found that the viral protein
tyrosine phosphatase (ptp) is essential for baculovirus-elicited ELA (54, 80); in
Lymantria dispar, the viral gene ecdysteroid UDP-glucosyltransferase (egt) is essential
to elicit climbing behavior (81). While the mechanism by which ptp acts is unknown,
egt inhibits the molting hormone 20-hydroxyecdysone (20E) (82), which disrupts molt-
ing behavior and is hypothesized to cause the host to continue feeding at elevated
locations (83). However, follow-up work has found that viral ptp and egt are not neces-
sary for manipulated locomotion and climbing in other baculovirus-caterpillar systems
(83, 84). This suggests that a variety of mechanisms are used to achieve these behav-
iors. Given this diversity of mechanisms seen in the realm of other behavior-modifying
pathogens, it seems reasonable to expect that, while parallels might exist, zombifying
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 9
fungal entomopathogens may also use a variety of mechanisms to elicit host
behaviors.
THE EVOLUTION OF COMMONLY ENCOUNTERED MANIPULATED BEHAVIORS
The ability to manipulate host behavior has convergently evolved in an array of fun-
gal entomopathogens and beyond. This complex trait is thought to be the result of a
long, intimate coevolution between pathogens and hosts in which both organisms are
in a constant arms race. Despite the independent evolutionary trajectories of each of
these pathogens with their respective insect hosts, they have elicited similar types of
behaviors. As discussed, fungi across two distinct phyla manipulate host locomotion
activity, induce climbing behavior, trigger preference for or attraction to certain abiotic
factors, activate mouthparts to adhere to vegetation, and do so with timed precision
(Fig. 1). This begs the question: why are these manipulated behaviors so frequently
encountered?
One potential answer is that the mechanisms and pathways that lead to these
behaviors are simply the easiest to alter. Animals, including insects, can change their
behavior as their environment changes. Neuroactive chemicals like hormones and neu-
romodulators can activate and deactivate behavioral pathways in the central nervous
system to allow for quick responses to unpredictable alterations in an individual’s
direct surroundings (85). Additionally, behavioral outputs of the biological clock regu-
late activity patterns to anticipate predictable biotic and abiotic daily changes in the
environment, (86, 87) but do so flexibly enough to give rise to interindividual variation
within a population (i.e., chronotypes) (88, 89) and adjust to sudden changes in envi-
ronmental and social cues (90–92). While such behavioral plasticity is of utmost impor-
tance for survival, it also provides opportunities for parasites to hijack and coopt (93,
94). Indeed, there is evidence that titers of neuromodulatory biogenic amines, such as
serotonin and dopamine, are affected by parasites and pathogens across vertebrate
and invertebrate hosts that adaptively manipulate behavior (21, 27, 95–98). Moreover,
the clock of C. floridanus, the ant host of O. camponoti-floridani, appears to be rather
plastic and has been suggested to underlie at least some of the plasticity that gives
rise to the behavioral division of labor in colonies of this species (99). Reports on the
loss of daily activity patterns in Ophiocordyceps-infected C. floridanus (28), establish-
ment of synchronized manipulated summiting (21), and differential expression of clock
and clock-controlled genes in infected individuals compared to healthy conspecifics
(21) suggest that behavioral plasticity resulting from biological clocks is corruptible.
Another explanation for the convergent evolution of similar behavioral manipula-
tions could be that these particular behaviors provide such a big advantage in trans-
mission to new hosts that they conferred very strong selection over other altered
behaviors. Parasites and pathogens that can alter host behavior are more likely to do
so by casting a wide net, instead of selectively attacking discrete areas of nervous tis-
sue (93). As such, a suite of insect host behaviors is likely affected by fungal cells and
effectors. Combinations of tissue occupancy and chemical secretions that lead to host
phenotypes resulting in higher pathogen fitness will be selected for over time.
In addition, the host’s immune and nervous system are undeniably connected
(100–102). Immune responses against invading pathogens result in the release of fac-
tors that affect neural function. This, in turn, leads to altered host behaviors that poten-
tially aid in its recovery: so-called sickness behaviors (103–105). Sickness behaviors of-
ten involve altered locomotion and feeding activity and diminished interactions with
conspecifics. In addition, house flies infected with the generalist B. bassiana demon-
strate a behavioral thermoregulation response by seeking out hotter and colder tem-
peratures at different times of day (106). Moreover, the nonmanipulating fungus
Metarhizium brunneum causes ant hosts to become phototropic and less attracted to
social cues, which would eventually result in them straying from the nest (107). As
such, infections with nonmanipulating fungi also appear to involve host behavioral
changes. However, the interpretation of the function of these behaviors is host
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 10
adaptive for some (106, 108–110) or pathogen adaptive for others (111), or they are
deemed nonfunctional/mere by-products of disease. Nevertheless, generalist entomo-
pathogens may have the tools needed to manipulate, and either we have not observed
the behavioral effects in their preferred hosts or their manipulations are much less con-
spicuous compared to the classic examples of active host transmission and summiting.
Regardless of their function, there seems to be a degree of overlap between the com-
mon behaviors that behavior-manipulating fungi induce and those that arise from
infections with nonmanipulating generalists. This suggests that there could be a slid-
ing scale along which entomopathogenic fungi can evolve with less precise infection-
related behaviors on one end and more fine-tuned transmission-benefitting manipu-
lated behaviors on the other.
Considering that behavior is a complex phenotype with multiple underlying mecha-
nisms, the ability to induce a stereotypical set of behaviors at the right time such that
it consistently benefits the pathogen’s life cycle is unlikely to evolve quickly. The ability
to produce a complex cocktail of compounds that changes behavioral outputs of the
brain with exquisite precision requires a long, tight coevolutionary history with the
host’s nervous system. Logically, this would come at the cost of promiscuity (Fig. 1),
which would explain why only select pathogens have evolved this trait and why those
that have are fairly to highly host specific.
The question of why the same manipulated behaviors are so frequently encoun-
tered across fungus-insect interactions is currently left unanswered. However, research
that would increase our understanding of this topic would be a worthwhile endeavor
and is also likely to lead to general insights into the mechanisms underlying behavior
plasticity and fungus-insect coevolution.
CONCLUSIONS
The study of behavior-modifying fungal pathogens is a rapidly growing field, as
exemplified by the discovery and (re-)classification of many new species of zombie-
making fungi in just the last decade (e.g., references 70 and 112 to 120). During this
time, various -omics and molecular biology techniques have also become more acces-
sible and applicable to nontraditional fungal models. This has allowed for deeper ex-
ploration into the complex relationship between zombie-making fungal pathogens
and their hosts, resulting in many hypotheses about the underlying mechanisms of
host behavior manipulation (Table 1). These hypotheses currently place research on
behavior-modifying fungal pathogens at an exciting crossroads toward linking the be-
havioral phenomes of manipulated insects with the underlying genomes of the fungi
that infect them. However, the application of cutting-edge -omics and molecular tech-
niques does not come without challenges. Many zombie-making fungi, especially
Entomophthorales, are extremely fastidious, making them difficult to isolate and cul-
ture in the lab environment. This complicates the implementation of molecular meth-
ods, such as CRISPR-Cas9, and transformation techniques necessary to determine the
functional roles of secreted fungal molecules in behavioral manipulation. Work that
makes some of these more recalcitrant species lab amenable is, therefore, paramount
for continued mechanistic studies. We recognize the difficulties that come along with
these endeavors and offer encouragement to researchers who have already or will
soon be venturing into this field. With persistence, creativity, and luck, we are optimis-
tic that new zombie-making fungus systems will continue to be harnessed and investi-
gated, as we have seen in recent years.
Future research into the molecular underpinnings of host manipulation is also
dependent on a complementary expansion in genomics, transcriptomics, and
metabolomics work. For instance, the current relatively low availability of high-qual-
ity, annotated genomes, scattered across a few distantly related fungal species, com-
plicates meaningful comparative analyses between these groups. Increasing the
number of sequenced genomes within genera and families that harbor at least sev-
eral zombie-making fungal species would allow for the identification of evolutionary
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 11
patterns, genes, and gene clusters that give rise to behavior manipulation traits.
This, too, however, poses various challenges: where the genomes of hypocrealean
Ophiocordycipitaceae and Cordycipitaceae are relatively compact (21.91 to
32.31 Mb for Ophiocordyceps species [18, 21]) and straightforward to assemble, ento-
mophthoralean fungi have much larger and repeat-rich genomes (;1GbforE. mus-
cae [NCBI accession no. PRJNA479887]), which are assembly averse. The advent of
long-read sequencing and ever-improving future technological developments are
promising options for solving this issue.
As with any new field of study that ventures forward from descriptive to quantita-
tive research on wild, nontraditional model systems, there are many hurdles to over-
come before the mystery of mechanisms underlying fungal manipulation of insect
behavior can be fully unraveled. Understanding these pathways will lead to valuable
insight into the mechanisms of pathogen-host coevolution and genes underlying host
specificity in animal models. Additionally, these efforts will shed new light on how and
why animals behave in certain ways and may open the road toward the discovery of
novel fungal biomolecules with potential pharmaceutical and industrial applications.
Furthermore, characterization of biomolecules secreted by zombie-making fungi and
their role in host manipulation at the molecular level will finally allow for a more holis-
tic comparison between these animal pathogens and well-studied host-pathogen
interactions in fungus-plant models.
ACKNOWLEDGMENTS
We thank Roel Fleuren from Science Transmitter for helping us with the visual
aspects of our paper.
Charissa de Bekker is supported by NSF Career Award 1941546. Carolyn Elya is
supported by a Hanna H. Gray fellowship from HHMI.
REFERENCES
1. van Seventer JM, Hochberg NS. 2017. Principles of infectious diseases:
transmission, diagnosis, prevention, and control, p 22–39. In Interna-
tional encyclopedia of public health, 2nd ed. Elsevier, Amsterdam,
Netherlands.
2. Mullens BA, Rodriguez JL, Meyer JA. 1987. An epizootiological study of
Entomophthora muscae in muscoid fly populations on Southern Califor-
nia poultry facilities, with emphasis on Musca domestica. Hilgardia 55(3):
1–41. https://doi.org/10.3733/hilg.v55n03p041.
3. Watson DW, Petersen JJ. 1993. Seasonal activity of Entomophthora mus-
cae (Zygomycetes: Entomophthorales) in Musca domestica L. (Diptera:
Muscidae) with reference to temperature and relative humidity. Biol
Control 3:182–190. https://doi.org/10.1006/bcon.1993.1026.
4. Carruthers RI, Haynes DL, MacLeod DM. 1985. Entomophthora muscae
(Entomophthorales: Entomophthoracae) mycosis in the onion fly, Delia
antiqua (Diptera: Anthomyiidae). J Invertebr Pathol 45:81–93. https://doi
.org/10.1016/0022-2011(85)90053-9.
5. Mullens BA. 1990. Entomophthora muscae (Entomophthorales: Ento-
mophthoraceae) as a pathogen of filth flies, p 231–245. In Rutz DA,
Patterson RS (ed), Biocontrol of arthropods affecting livestock and poul-
try. Westview Press, Inc, Boulder, CO.
6. Eilenberg J, Philipsen H. 1988. The occurrence of Entomophthorales on
the carrot fly [Psila rosae F.] in the field during two successive seasons.
Entomophaga 33:135–144. https://doi.org/10.1007/BF02372648.
7. Steinkraus DC, Geden CJ, Rutz DA. 1993. Prevalence of Entomophthora
muscae (Cohn) Fresenius (Zygomycetes: Entomophthoraceae) in house
flies (Diptera: Muscidae) on dairy farms in New York and induction of epi-
zootics. Biol Control 3:93–100. https://doi.org/10.1006/bcon.1993.1015.
8. Gryganskyi AP, Mullens BA, Gajdeczka MT, Rehner SA, Vilgalys R, Hajek
AE. 2017. Hijacked: co-option of host behavior by entomophthoralean
fungi. PLoS Pathog 13:e1006274. https://doi.org/10.1371/journal.ppat
.1006274.
9. Hywel-Jones NL. 1993. A systematic survey of insect fungi from natural,
tropical forest in Thailand, p 300–301. In Isaac S, Frankland JC, Watling R,
Whalley AJS (ed), Aspects of tropical mycology. Cambridge University
Press, Cambridge, United Kingdom.
10. Vega FE, Blackwell M. 2005. Insect-fungal associations: ecology and evo-
lution. Oxford University Press, Oxford, United Kingdom.
11. Wang J, Lovett B, St Leger RJ. 2019. The secretome and chemistry of
Metarhizium; a genus of entomopathogenic fungi. Fungal Ecol 38:7–11.
https://doi.org/10.1016/j.funeco.2018.04.001.
12. Iwanicki NSA, Pereira AA, Botelho ABRZ, Rezende JM, de Moral RA,
Zucchi MI, Delalibera Júnior I. 2019. Monitoring of the field application
of Metarhizium anisopliae in Brazil revealed high molecular diversity of
Metarhizium spp in insects, soil and sugarcane roots. Sci Rep 9:4443.
https://doi.org/10.1038/s41598-019-38594-8.
13. Bilgo E, Lovett B, St Leger RJ, Sanon A, Dabiré RK, Diabaté A. 2018. Native
entomopathogenic Metarhizium spp. from Burkina Faso and their viru-
lence against the malaria vector Anopheles coluzzii and non-target insects.
Parasit Vectors 11:209. https://doi.org/10.1186/s13071-018-2796-6.
14. Loreto RG, Elliot SL, Freitas MLR, Pereira TM, Hughes DP. 2014. Long-term
disease dynamics for a specialized parasite of ant societies: a field study.
PLoS One 9:e103516. https://doi.org/10.1371/journal.pone.0103516.
15. Lovett B, Macias A, Stajich JE, Cooley J, Eilenberg J, de Fine Licht HH,
Kasson MT. 2020. Behavioral betrayal: how select fungal parasites enlist
living insects to do their bidding. PLoS Pathog 16:e1008598. https://doi
.org/10.1371/journal.ppat.1008598.
16. Małagocka J, Grell MN, Lange L, Eilenberg J, Jensen AB. 2015. Transcrip-
tome of an entomophthoralean fungus (Pandora formicae) shows molecu-
lar machinery adjusted for successful host exploitation and transmission. J
Invertebr Pathol 128:47–56. https://doi.org/10.1016/j.jip.2015.05.001.
17. Elya C, Lok TC, Spencer QE, McCausland H, Martinez CC, Eisen M. 2018.
Robust manipulation of the behavior of Drosophila melanogaster by a
fungal pathogen in the laboratory. Elife 7:e34414. https://doi.org/10
.7554/eLife.34414.
18. de Bekker C, Ohm RA, Evans HC, Brachmann A, Hughes DP. 2017. Ant-
infecting Ophiocordyceps genomes reveal a high diversity of potential
behavioral manipulation genes and a possible major role for enterotox-
ins. Sci Rep 7:12508. https://doi.org/10.1038/s41598-017-12863-w.
19. Boyce GR, Gluck-Thaler E, Slot JC, Stajich JE, Davis WJ, James TY, Cooley
JR, Panaccione DG, Eilenberg J, De Fine Licht HH, Macias AM, Berger MC,
Wickert KL, Stauder CM, Spahr EJ, Maust MD, Metheny AM, Simon C,
Kritsky G, Hodge KT, Humber RA, Gullion T, Short DPG, Kijimoto T,
Mozgai D, Arguedas N, Kasson MT. 2019. Psychoactive plant- and mush-
room-associated alkaloids from two behavior modifying cicada
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 12
pathogens. Fungal Ecol 41:147–164. https://doi.org/10.1016/j.funeco
.2019.06.002.
20. Loreto RG, Hughes DP. 2019. The metabolic alteration and apparent
preservation of the zombie ant brain. J Insect Physiol 118:103918.
https://doi.org/10.1016/j.jinsphys.2019.103918.
21. Will I, Das B, Trinh T, Brachmann A, Ohm RA, de Bekker C. 2020. Genetic
underpinnings of host manipulation by Ophiocordyceps as revealed by
comparative transcriptomics. G3 (Bethesda) 10:2275–2296. https://doi
.org/10.1534/g3.120.401290.
22. De Fine Licht HH, Jensen AB, Eilenberg J. 2017. Comparative transcrip-
tomics reveal host-specific nucleotide variation in entomophthoralean
fungi. Mol Ecol 26:2092–2110. https://doi.org/10.1111/mec.13863.
23. Loreto RG, Araújo JPM, Kepler RM, Fleming KR, Moreau CS, Hughes DP.
2018. Evidence for convergent evolution of host parasitic manipulation
in response to environmental conditions. Evolution 72:2144–2155.
https://doi.org/10.1111/evo.13489.
24. Araújo JPM, Hughes DP. 2019. Zombie-ant fungi emerged from non-
manipulating, beetle-infecting ancestors. Curr Biol 29:3735–3738.e2.
https://doi.org/10.1016/j.cub.2019.09.004.
25. Cooley JR, Marshall DC, Hill KBR. 2018. A specialized fungal parasite (Mas-
sospora cicadina) hijacks the sexual signals of periodical cicadas (Hemi-
ptera: Cicadidae: Magicicada). Sci Rep 8:1432. https://doi.org/10.1038/
s41598-018-19813-0.
26. Csata E, Billen J, Barbu-Tudoran L, Markó B. 2021. Inside Pandora’s box:
development of the lethal myrmecopathogenic fungus Pandora formi-
cae within its ant host. Fungal Ecol 50:101022. https://doi.org/10.1016/j
.funeco.2020.101022.
27. de Bekker C, Ohm RA, Loreto RG, Sebastian A, Albert I, Merrow M,
Brachmann A, Hughes DP. 2015. Gene expression during zombie ant bit-
ing behavior reflects the complexity underlying fungal parasitic behav-
ioral manipulation. BMC Genomics 16:620. https://doi.org/10.1186/
s12864-015-1812-x.
28. Trinh T, Ouellete R, de Bekker C. Getting lost: the fungal hijacking of ant
foraging behaviour in space and time. Anim Behav, in press.
29. Araújo JPM, Hughes DP. 2016. Diversity of entomopathogenic fungi:
which groups conquered the insect body? Adv Genet 94:1–39. https://
doi.org/10.1016/bs.adgen.2016.01.001.
30. Andersen SB, Gerritsma S, Yusah KM, Mayntz D, Hywel-Jones NL, Billen J,
Boomsma JJ, Hughes DP. 2009. The life of a dead ant: the expression of
an adaptive extended phenotype. Am Nat 174:424–433. https://doi.org/
10.1086/603640.
31. Maitland DP. 1994. A parasitic fungus infecting yellow dungflies manipu-
lates host perching behaviour. Proc R Soc Lond B Biol Sci 258:187–193.
https://doi.org/10.1098/rspb.1994.0161.
32. Andriolli FS, Ishikawa NK, Vargas-Isla R, Cabral TS, de Bekker C, Baccaro
FB. 2019. Do zombie ant fungi turn their hosts into light seekers? Behav
Ecol 30:609–616. https://doi.org/10.1093/beheco/ary198.
33. Chung T-Y, Sun P-F, Kuo J-I, Lee Y-I, Lin C-C, Chou J-Y. 2017. Zombie ant
heads are oriented relative to solar cues. Fungal Ecol 25:22–28. https://
doi.org/10.1016/j.funeco.2016.10.003.
34. Watson DW, Mullens BA, Petersen JJ. 1993. Behavioral fever response of
Musca domestica (Diptera: Muscidae) to infection by Entomophthora
muscae (Zygomycetes: Entomophthorales). J Invertebr Pathol 61:10–16.
https://doi.org/10.1006/jipa.1993.1003.
35. Hughes DP, Andersen SB, Hywel-Jones NL, Himaman W, Billen J, Boomsma
JJ. 2011. Behavioral mechanisms and morphological symptoms of zombie
ants dying from fungal infection. BMC Ecol 11:13. https://doi.org/10.1186/
1472-6785-11-13.
36. Krasnoff SB, Watson DW, Gibson DM, Kwan EC. 1995. Behavioral effects
of the entomopathogenic fungus, Entomophthora muscae on its host
Musca domestica: postural changes in dying hosts and gated pattern of
mortality. J Insect Physiol 41:895–903. https://doi.org/10.1016/0022
-1910(95)00026-Q.
37. Boer P. 2008. Observations of summit disease in Formica rufa Linnaeus,
1761 (Hymenoptera: Formicidae). Myrmecol News 11:63–66.
38. Roy HE, Steinkraus DC, Eilenberg J, Hajek AE, Pell JK. 2006. Bizarre inter-
actions and endgames: entomopathogenic fungi and their arthropod
hosts. Annu Rev Entomol 51:331–357. https://doi.org/10.1146/annurev
.ento.51.110104.150941.
39. Pickford R, Riegert PW. 1964. The fungous disease caused by Entomoph-
thora grylli Fres., and its effects on grasshopper populations in Saskatch-
ewan in 1963. Can Entomol 96:1158–1166. https://doi.org/10.4039/
Ent961158-9.
40. Rijo-Ferreira F, Carvalho T, Afonso C, Sanches-Vaz M, Costa RM,
Figueiredo LM, Takahashi JS. 2018. Sleeping sickness is a circadian disor-
der. Nat Commun 9:62. https://doi.org/10.1038/s41467-017-02484-2.
41. Thomas F, Schmidt-Rhaesa A, Martin G, Manu C, Durand P, Renaud F.
2002. Do hairworms (Nematomorpha) manipulate the water seeking
behaviour of their terrestrial hosts? J Evol Biol 15:356–361. https://doi
.org/10.1046/j.1420-9101.2002.00410.x.
42. Biron DG, Ponton F, Marché L, Galeotti N, Renault L, Demey-Thomas E,
Poncet J, Brown SP, Jouin P, Thomas F. 2006. ‘Suicide’of crickets har-
bouring hairworms: a proteomics investigation. Insect Mol Biol 15:
731–742. https://doi.org/10.1111/j.1365-2583.2006.00671.x.
43. Lively CM. 1987. Evidence from a New Zealand snail for the mainte-
nance of sex by parasitism. Nature 328:519–521. https://doi.org/10
.1038/328519a0.
44. Levri EP. 1999. Parasite-induced change in host behavior of a freshwater
snail: parasitic manipulation or byproduct of infection? Behav Ecol 10:
234–241. https://doi.org/10.1093/beheco/10.3.234.
45. Badie A, Vincent M, Morel-Vareille C, Rondelaud D. 1973. Cycle of Dicro-
coelium dendriticum (Rudolphi, 1819) in Limousin. Ethology of ants par-
asited by the metacercaria. C R Seances Soc Biol Fil 167:725–727.
46. Manga-González MY, González-Lanza C, Cabanas E, Campo R. 2001. Con-
tributions to and review of dicrocoeliosis, with special reference to the
intermediate hosts of Dicrocoelium dendriticum. Parasitology 123
(Suppl):S91–S114. https://doi.org/10.1017/s0031182001008204.
47. de Bekker C. 2019. Ophiocordyceps–ant interactions as an integrative
model to understand the molecular basis of parasitic behavioral manipu-
lation. Curr Opin Insect Sci 33:19–24. https://doi.org/10.1016/j.cois.2019
.01.005.
48. de Bekker C, Merrow M, Hughes DP. 2014. From behavior to mecha-
nisms: an integrative approach to the manipulation by a parasitic fungus
(Ophiocordyceps unilateralis s.l.) of its host ants (Camponotus spp.).
Integr Comp Biol 54:166–176. https://doi.org/10.1093/icb/icu063.
49. Steinkraus DC, Hajek AE, Liebherr JK. 2017. Zombie soldier beetles: epi-
zootics in the goldenrod soldier beetle, Chauliognathus pensylvanicus
(Coleoptera: Cantharidae) caused by Eryniopsis lampyridarum (Ento-
mophthoromycotina: Entomophthoraceae). J Invertebr Pathol 148:
51–59. https://doi.org/10.1016/j.jip.2017.05.002.
50. Helluy S, Holmes JC. 1990. Serotonin, octopamine, and the clinging
behavior induced by the parasite Polymorphus paradoxus (Acanthoce-
phala) in Gammarus lacustris (Crustacea). Can J Zool 68:1214–1220.
https://doi.org/10.1139/z90-181.
51. Wesołowska W, Wesołowski T. 2014. Do L eucochloridium sporocysts
manipulate the behaviour of their snail hosts? J Zool 292:151–155.
https://doi.org/10.1111/jzo.12094.
52. Goulson D. 1997. Wipfelkrankheit: modification of host behaviour during
baculoviral infection. Oecologia 109:219–228. https://doi.org/10.1007/
s004420050076.
53. Mehlhorn H. 2015. The brain worm story, p 101–108. In Mehlhorn H (ed),
Host manipulations by parasites and viruses. Springer International Pub-
lishing, Cham, Switzerland.
54. Katsuma S, Koyano Y, Kang W, Kokusho R, Kamita SG, Shimada T. 2012.
The baculovirus uses a captured host phosphatase to induce enhanced
locomotory activity in host caterpillars. PLoS Pathog 8:e1002644. https://
doi.org/10.1371/journal.ppat.1002644.
55. Kamita SG, Nagasaka K, Chua JW, Shimada T, Mita K, Kobayashi M,
Maeda S, Hammock BD. 2005. A baculovirus-encoded protein tyrosine
phosphatase gene induces enhanced locomotory activity in a lepidop-
teran host. Proc Natl Acad Sci U S A 102:2584–2589. https://doi.org/10
.1073/pnas.0409457102.
56. Carney WP. 1969. Behavioral and morphological changes in carpenter
ants harboring Dicrocoeliid Metacercariae. Am Midl Nat 82:605–611.
https://doi.org/10.2307/2423801.
57. Romig T, Lucius R, Frank W. 1980. Cerebral larvae in the second interme-
diate host of Dicrocoelium dendriticum (Rudolphi, 1819) and Dicrocoe-
lium hospes Looss, 1907 (Trematodes, Dicrocoeliidae). Z Parasitenkd 63:
277–286. https://doi.org/10.1007/BF00931990.
58. Brobyn PJ, Wilding N. 1983. Invasive and developmental processes of
Entomophthora muscae infecting houseflies (Musca domestica). Trans
Br Mycol Soc 80:1–8. https://doi.org/10.1016/S0007-1536(83)80157-0.
59. Lowe RE, Kennel EW. 1972. Pathogenicity of the fungus Entomophthora
coronata in Culex pipiens quinquefasciatus and Aedes taeniorhynchus.
Mosquito News 32:614–620.
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 13
60. Humber RA. 1976. The systematics of the genus Strongwellsea (Zygomy-
cetes: Entomophthorales). Mycologia 68:1042–1060. https://doi.org/10
.2307/3758721.
61. Funk CJ, Ramoska WA, Bechtel DB. 1993. Histopathology of Entomo-
phaga grylli pathotype 2 infections in Melanoplus differentialis. J Inver-
tebr Pathol 61:196–202. https://doi.org/10.1006/jipa.1993.1035.
62. Brobyn PJ, Wilding N. 1977. Invasive and developmental processes of
Entomophthora species infecting aphids. Trans Br Mycol Soc 69:349–366.
https://doi.org/10.1016/S0007-1536(77)80072-7.
63. Fredericksen MA, Zhang Y, Hazen ML, Loreto RG, Mangold CA, Chen DZ,
Hughes DP. 2017. Three-dimensional visualization and a deep-learning
model reveal complex fungal parasite networks in behaviorally manipu-
lated ants. Proc Natl Acad Sci U S A 114:12590–12595. https://doi.org/10
.1073/pnas.1711673114.
64. Mangold CA, Ishler MJ, Loreto RG, Hazen ML, Hughes DP. 2019. Zombie
ant death grip due to hypercontracted mandibular muscles. J Exp Biol
222(Pt 14):jeb200683. https://doi.org/10.1242/jeb.200683.
65. Grell MN, Jensen AB, Olsen PB, Eilenberg J, Lange L. 2011. Secretome of
fungus-infected aphids documents high pathogen activity and weak
host response. Fungal Genet Biol 48:343–352. https://doi.org/10.1016/j
.fgb.2010.12.003.
66. Browne LB, Evans DR. 1960. Locomotor activity of the blowfly as a func-
tion of feeding and starvation. J Insect Physiol 4:27–37. https://doi.org/
10.1016/0022-1910(60)90065-2.
67. Wheater CP. 1991. Effect of starvation on locomotor activity in some pre-
dacious Coleoptera (Carabidae, Staphylinidae). Coleopt Bull 45:371–378.
68. Leonard DE. 1970. Effects of starvation on behaviour, number of larval
instars, and developmental rate of Porthetria dispar. J Insect Physiol 16:
25–31. https://doi.org/10.1016/0022-1910(70)90109-5.
69. Yang Z, Yu Y, Zhang V, Tian Y, Qi W, Wang L. 2015. Octopamine mediates
starvation-induced hyperactivity in adult Drosophila. Proc Natl Acad Sci
U S A 112:5219–5224. https://doi.org/10.1073/pnas.1417838112.
70. Macias AM, Geiser DM, Stajich JE, Łukasik P, Veloso C, Bublitz DC, Berger
MC, Boyce GR, Hodge K, Kasson MT. 2020. Evolutionary relationships
among Massospora spp. (Entomophthorales), obligate pathogens of cica-
das. Mycologia 112:1060–1074. https://doi.org/10.1080/00275514.2020
.1742033.
71. Leadmon CE, Sampson JK, Maust MD, Macias AM, Rehner SA, Kasson MT,
Panaccione DG. 2020. Several Metarhizium species produce ergot alka-
loids in a condition-specific manner. Appl Environ Microbiol 86:e00373-
20. https://doi.org/10.1128/AEM.00373-20.
72. de Bekker C, Quevillon LE, Smith PB, Fleming KR, Ghosh D, Patterson AD,
Hughes DP. 2014. Species-specific ant brain manipulation by a special-
ized fungal parasite. BMC Evol Biol 14:166. https://doi.org/10.1186/
s12862-014-0166-3.
73. Boomsma JJ, Jensen AB, Meyling NV, Eilenberg J. 2014. Evolutionary
interaction networks of insect pathogenic fungi. Annu Rev Entomol 59:
467–485. https://doi.org/10.1146/annurev-ento-011613-162054.
74. Humber RA. 1984. Foundations for an evolutionary classification of the
Entomophthorales (Zygomycetes), p 166–183. In Wheeler Q, Blackwell M
(ed), Fungus-insect relationships: perspectives in ecology and evolution.
Columbia University Press, New York, NY.
75. Codd V, Dolezel D, Stehlik J, Piccin A, Garner KJ, Racey SN, Straatman KR,
Louis EJ, Costa R, Sauman I, Kyriacou CP, Rosato E. 2007. Circadian
rhythm gene regulation in the housefly Musca domestica. Genetics 177:
1539–1551. https://doi.org/10.1534/genetics.107.079160.
76. Cheng P, Yang Y, Wang L, He Q, Liu Y. 2003. WHITE COLLAR-1, a multi-
functional neurospora protein involved in the circadian feedback loops,
light sensing, and transcription repression of wc-2. J Biol Chem 278:
3801–3808. https://doi.org/10.1074/jbc.M209592200.
77. de Bekker C, Will I, Hughes DP, Brachmann A, Merrow M. 2017. Daily
rhythms and enrichment patterns in the transcriptome of the behavior-
manipulating parasite Ophiocordyceps kimflemingiae. PLoS One 12:
e0187170. https://doi.org/10.1371/journal.pone.0187170.
78. Westwood ML, O’Donnell AJ, de Bekker C, Lively CM, Zuk M, Reece SE.
2019. The evolutionary ecology of circadian rhythms in infection. Nat
Ecol Evol 3:552–560. https://doi.org/10.1038/s41559-019-0831-4.
79. Beer K, Helfrich-Förster C. 2020. Model and non-model insects in chrono-
biology. Front Behav Neurosci 14:601676. https://doi.org/10.3389/fnbeh
.2020.601676.
80. van Houte S, Ros VID, Mastenbroek TG, Vendrig NJ, Hoover K, Spitzen J,
van Oers MM. 2012. Protein tyrosine phosphatase-induced hyperactivity
is a conserved strategy of a subset of baculoviruses to manipulate
lepidopteran host behavior. PLoS One 7:e46933. https://doi.org/10
.1371/journal.pone.0046933.
81. Hoover K, Grove M, Gardner M, Hughes DP, McNeil J, Slavicek J. 2011. A
gene for an extended phenotype. Science 333:1401. https://doi.org/10
.1126/science.1209199.
82. O’Reilly DR, Miller LK. 1989. A baculovirus blocks insect molting by pro-
ducing ecdysteroid UDP-glucosyl transferase. Science 245:1110–1112.
https://doi.org/10.1126/science.2505387.
83. Ros VID, van Houte S, Hemerik L, van Oers MM. 2015. Baculovirus-
induced tree-top disease: how extended is the role of egt as a gene for
the extended phenotype? Mol Ecol 24:249–258. https://doi.org/10.1111/
mec.13019.
84. Kokusho R, Katsuma S. 2021. Bombyx mori nucleopolyhedrovirus ptp
and egt genes are dispensable for triggering enhanced locomotory ac-
tivity and climbing behavior in Bombyx mandarina larvae. J Invertebr
Pathol 183:107604. https://doi.org/10.1016/j.jip.2021.107604.
85. Huber R. 2005. Amines and motivated behaviors: a simpler systems
approach to complex behavioral phenomena. J Comp Physiol A Neuroe-
thol Sens Neural Behav Physiol 191:231–239. https://doi.org/10.1007/
s00359-004-0585-5.
86. Yerushalmi S, Green RM. 2009. Evidence for the adaptive significance of
circadian rhythms. Ecol Lett 12:970–981. https://doi.org/10.1111/j.1461
-0248.2009.01343.x.
87. Paranjpe DA, Sharma VK. 2005. Evolution of temporal order in living organ-
isms. J Circadian Rhythms 3:7. https://doi.org/10.1186/1740-3391-3-7.
88. Randler C. 2014. Sleep, sleep timing and chronotype in animal behaviour.
Anim Behav94:161–166. https://doi.org/10.1016/j.anbehav.2014.05.001.
89. Roenneberg T, Wirz-Justice A, Merrow M. 2003. Life between clocks: daily
temporal patterns of human chronotypes. J Biol Rhythms 18:80–90.
https://doi.org/10.1177/0748730402239679.
90. Maury C, Serota MW, Williams TD. 2020. Plasticity in diurnal activity and
temporal phenotype during parental care in European starlings, Sturnus
vulgaris. Anim Behav 159:37–45. https://doi.org/10.1016/j.anbehav.2019
.11.004.
91. van der Vinne V, Riede SJ, Gorter JA, Eijer WG, Sellix MT, Menaker M,
Daan S, Pilorz V, Hut RA. 2014. Cold and hunger induce diurnality in a
nocturnal mammal. Proc Natl Acad Sci U S A 111:15256–15260. https://
doi.org/10.1073/pnas.1413135111.
92. Graham JL, Cook NJ, Needham KB, Hau M, Greives TJ. 2017. Early to rise,
early to breed: a role for daily rhythms in seasonal reproduction. Behav
Ecol 28:1266–1271. https://doi.org/10.1093/beheco/arx088.
93. Adamo SA. 2013. Parasites: evolution’s neurobiologists. J Exp Biol 216:
3–10. https://doi.org/10.1242/jeb.073601.
94. Adamo SA. 2002. Modulating the modulators: parasites, neuromodula-
tors and host behavioral change. Brain Behav Evol 60:370–377. https://
doi.org/10.1159/000067790.
95. Adamo SA. 2012. The strings of the puppet master: how parasites change
host behavior. In Hughes DP, Brodeur J, Thomas F (ed), Host manipulation
by parasites. Oxford University Press, Oxford, United Kingdom.
96. Helluy S. 2013. Parasite-induced alterations of sensorimotor pathways in
gammarids: collateral damage of neuroinflammation? J Exp Biol 216:
67–77. https://doi.org/10.1242/jeb.073213.
97. Libersat F, Gal R. 2013. What can parasitoid wasps teach us about deci-
sion-making in insects? J Exp Biol 216:47–55. https://doi.org/10.1242/jeb
.073999.
98. Webster JP, Kaushik M, Bristow GC, McConkey GA. 2013. Toxoplasma
gondii infection, from predation to schizophrenia: can animal behaviour
help us understand human behaviour? J Exp Biol 216:99–112. https://doi
.org/10.1242/jeb.074716.
99. Das B, de Bekker C. 2021. Time-course RNASeq of Camponotus florida-
nus forager and nurse ant brains indicate links between plasticity in the
biological clock and behavioral division of labor. bioRxiv https://doi.org/
10.1101/2021.03.27.433505.
100. Adamo SA. 2006. Comparative psychoneuroimmunology: evidence
from the insects. Behav Cogn Neurosci Rev 5:128–140. https://doi.org/
10.1177/1534582306289580.
101. Adamo SA. 2008. Bidirectional connections between the immune sys-
tem and the nervous system in insects, p 129–149. In Beckage NE (ed),
Insect immunology. Academic Press, San Diego, CA.
102. Mangold CA, Hughes DP. 2021. Insect behavioral change and the poten-
tial contributions of neuroinflammation-a call for future research. Genes
12:465. https://doi.org/10.3390/genes12040465.
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 14
103. Hart BL. 2009. Beyond fever: comparative perspectives on sickness
behavior, p 205–210. In Encyclopedia of animal behavior. Elsevier, Am-
sterdam, The Netherlands.
104. Hart BL, Hart LA. 2019. Sickness behavior in animals: implications for
health and wellness, p 171–175. In Encyclopedia of animal behavior, 2nd
ed. Elsevier, Amsterdam, The Netherlands.
105. Dantzer R. 2001. Cytokine-induced sickness behavior: mechanisms and
implications. Ann N Y Acad Sci 933:222–234. https://doi.org/10.1111/j
.1749-6632.2001.tb05827.x.
106. Anderson RD, Blanford S, Jenkins NE, Thomas MB. 2013. Discriminating
fever behavior in house flies. PLoS One 8:e62269. https://doi.org/10
.1371/journal.pone.0062269.
107. Leclerc J-B, Detrain C. 2017. Loss of attraction for social cues leads to
fungal-infected Myrmica rubra ants withdrawing from the nest. Anim
Behav 129:133–141. https://doi.org/10.1016/j.anbehav.2017.05.002.
108. Blanford S, Thomas MB. 2001. Adult survival, maturation, and reproduc-
tion of the desert locust Schistocerca gregaria infected with the fungus
Metarhizium anisopliae var acridum. J Invertebr Pathol 78:1–8. https://
doi.org/10.1006/jipa.2001.5031.
109. Braga GUL, Rangel DEN, Fernandes
EKK, Flint SD, Roberts DW. 2015. Mo-
lecular and physiological effects of environmental UV radiation on fun-
gal conidia. Curr Genet 61:405–425. https://doi.org/10.1007/s00294-015
-0483-0.
110. Cremer S, Armitage SAO, Schmid-Hempel P. 2007. Social immunity. Curr
Biol 17:R693–R702. https://doi.org/10.1016/j.cub.2007.06.008.
111. Blanford S, Thomas MB. 1999. Role of thermal biology in disease dynam-
ics. Aspects Appl Biol 53:73–82.
112. Araújo JPM, Evans HC, Geiser DM, Mackay WP, Hughes DP. 2014. Unravel-
ling the diversity behind the Ophiocordyceps unilateralis complex: three
new species of zombie-ant fungi from the Brazilian Amazon. bioRxiv
https://doi.org/10.1101/003806.
113. Araújo JPM, Evans HC, Kepler R, Hughes DP. 2018. Zombie-ant fungi across
continents: 15 new species and new combinations within
Ophiocordyceps. I. Myrmecophilous hirsutelloid species. Stud Mycol 90:
119–160. https://doi.org/10.1016/j.simyco.2017.12.002.
114. Evans HC, Araújo JPM, Halfeld VR, Hughes DP. 2018. Epitypification and
re-description of the zombie-ant fungus, Ophiocordyceps unilateralis
(Ophiocordycipitaceae). Fungal Syst Evol 1:13–22. https://doi.org/10
.3114/fuse.2018.01.02.
115. Saltamachia SJ, Araújo JPM. 2020. Ophiocordyceps desmidiospora, a ba-
sal lineage within the “Zombie-Ant Fungi”clade. Mycologia 112:
1171–1183. https://doi.org/10.1080/00275514.2020.1732147.
116. Araújo JPM, Evans HC, Fernandes IO, Ishler MJ, Hughes DP. 2020. Zom-
bie-ant fungi cross continents: II. Myrmecophilous hymenostilboid spe-
cies and a novel zombie lineage. Mycologia 112:1138–1170. https://doi
.org/10.1080/00275514.2020.1822093.
117. Hodge KT, Hajek AE, Gryganskyi A. 2017. The first entomophthoralean
killing millipedes, Arthrophaga myriapodina n. gen. n. sp., causes climb-
ing before host death. J Invertebr Pathol 149:135–140. https://doi.org/
10.1016/j.jip.2017.08.011.
118. Eilenberg J, Michelsen V, Humber RA. 2020. Strongwellsea tigrinae and
Strongwellsea acerosa (Entomophthorales: Entomophthoraceae), two
new species infecting dipteran hosts from the genus Coenosia (Musci-
dae). J Invertebr Pathol 175:107444. https://doi.org/10.1016/j.jip.2020
.107444.
119. Görg LM, Eilenberg J, Jensen AB, Jensen AH, Gross J. 2021. Pathogenicity
against hemipteran vector insects of a novel insect pathogenic fungus
from Entomophthorales (Pandora sp. nov. inedit.) with potential for bio-
logical control. J Invertebr Pathol 183:107621. https://doi.org/10.1016/j
.jip.2021.107621.
120. Olsen NL, Herren P, Markussen B, Jensen AB, Eilenberg J. 2019. Statistical
modelling of conidial discharge of entomophthoralean fungi using a
newly discovered Pandora species. PLoS One 14:e0215914. https://doi
.org/10.1371/journal.pone.0215914.
121. Balazy S. 1984. On rhizoids of Entomophthora-muscae (Cohn) Fresenius
(Entomophthorales, Entomophthoraceae). Mycotaxon 19:397–407.
Minireview ®
September/October 2021 Volume 12 Issue 5 e01872-21 mbio.asm.org 15
Content uploaded by Carolyn Elya
Author content
All content in this area was uploaded by Carolyn Elya on Oct 07, 2021
Content may be subject to copyright.
