AN ABSTRACT OF THE THESIS OF
Hugh G. Lefcort for the degree of Doctor of Philosophy
Zoology
in
presented on
March 10, 1993
Title: The Effects of Parasites on Host Behavior:
Whp Benefits?
Abstract approved:
Redacted for Privacy
Andrew R. Blaustein
Some parasites may modify the behavior of their
hosts.
Altered behaviors may: 1) benefit the host in
that they defend against the pathogen, 2) benefit the
pathogen and represent manipulations of the host
response, and 3) benefit neither the host or the
pathogen and simply be a product of the host response to
infection.
In this thesis I examine four host/parasite systems.
For each system, I explore host/parasite behavioral
interactions, and examine them with regard to selective
pressures that may be acting on both the host and the
parasite.
I test the Hamilton and Zuk hypothese in 26 species
of lizards.
I find an inverse relationship between a
lizard species' brightness and parasite prevalence.
My
result lend credence to criticisms of the Hamilton and
Zuk Hypothesis.
If infection does occur, animals may alter their
behavior to impair the growth and reproduction of the
parasite.
To test this prediction, I examine behavioral
thermoregulation in two strains of the snail
Biomphalaria glabrata, one resistant to, and one
susceptible to, the parasite Schistosoma mansoni.
The
preferred temperature of infected snails drops five
weeks after exposure to the parasite.
I propose the hypothesis that pathogen-induced host
defense responses result in altered host behaviors and
enhanced predation.
In particular, I examine the
effects of the acute phase response (a physiological
response whose symptoms include fever, reduced activity
and malaise) on antipredatory behavior in bullfrog (Rana
catesbeiana) tadpoles.
This host response is associated
with the preliminary stages of infection with many
pathogens yet its behavioral effects have received
little attention.
I find that the stereotypical effects
of the acute phase response can lead to increased
predation.
I suggest that altered behaviors may afford
some parasites a potential pathway to their next host.
I examine the behavioral effects of a yeast, Candida
spp., a single-host parasite species in its natural
host, the red-legged frog (Rana aurora).
Infected
tadpoles exhibit the same behavioral modifications that
are noted in bacteria injected bullfrog tadpoles.
These
results suggest that some altered behaviors may occur
due to a host response to infection and not due to
parasitic manipulation.
The Effects of Parasites on Host Behavior:
Who Benefits?
by
Hugh G. Lefcort
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirement for the
the degree of
Doctor of Philosophy
Completed March 10, 1993
Commencement June 1993
APPROVED:
Redacted for Privacy
Professor of Zoology
charge of Iajor
Redacted for Privacy
air of Depart/ment of Zoology
Redacted for Privacy
an of Gr
uate School
Date thesis is presented
March 10, 1993
Typed by
Huah G. Lefcort
ACKNOWLEDGEMENTS
This thesis would honestly not be possible without the
help of a number of people.
First, I want to
acknowledge the help and support of my major professor
Andy Blaustein.
Andy not only suggested that I explore
the effects of parasites on host behavior but he has
guided me along every step of my scientific development.
My graduate committee - George Bailey, Mark Hixon,
Frank Moore and Phil Rossignol - have also been pivotal
in helping me to design and analyze my experiments.
I
thank George for being diligent to procedural
requirements.
Mark's thorough reading of proposals and
manuscript drafts tightened my output and ensured speedy
publication.
Frank encouraged me to think
scientifically and critically.
Phil, a true
intellectual, allowed me to see the forest through the
trees.
Chris Bayne and Charles King have also greatly
influenced me.
Joe Beatty and Doc Storm helped me to
understand the Natural History of frogs and lizards.
Susan Walls answered numerous questions and, while her
name does not appear on my graduate committee, it
should.
Eric Berlow has been a constant sounding board for
ideas, both scientific and otherwise.
he has been my friend.
More importantly,
Zhang Lei has helped and
collaborated on a number of projects. Sergio Navarrete
has suffered though my attempts at statistical analysis.
Clyde Hakelev has been a constant field companion and
persuaded me to do more fieldwork than I might normally
have done.
I also appreciate the scientific help I have
received from people outside the department.
Ray Huey,
Matt Kluger, Frances Lefcort and Malcolm Lefcort have
all helped in innumerable ways.
Specifically, Steve
Eiger deserves credit for deriving the core concept of
my thesis.
Finally I owe special appreciation to Dorothy Lefcort
for constant support and encouragement and to Diane
Wilson for being both a scientific collaborator and for
focussing and enriching my life.
TABLE OF CONTENTS
CHAPTER ONE:
INTRODUCTION
CHAPTER TWO:
PARASITE LOAD AND BRIGHTNESS IN LIZARDS:
A TEST OF THE HAMILTON AND ZUK HYPOTHESIS
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
CHAPTER THREE:
TEMPERATURE DISTRIBUTIONS OF RESISTANT AND
SUSCEPTIBLE STRAINS OF BIOMPHALARIA GLABRATA
(GASTROPODA) EXPOSED TO SCHISTOSOMA MANSONI
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
CHAPTER FOUR:
ANTIPREDATORY BEHAVIOR OF FEVERISH TADPOLES:
IMPLICATIONS FOR PATHOGEN TRANSMISSION
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
1
21
22
23
27
32
34
40
46
47
48
52
55
60
63
70
71
72
76
85
90
93
CHAPTER FIVE:
FEVER AND PREDATOR AVOIDANCE OF PARASITIZED
TADPOLES
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
100
101
102
105
111
114
117
SUMMARY
120
BIBLIOGRAPHY
125
LIST OF FIGURES
Figure
III.1
Page
Mean temperatures measured at the
65
centers of zones within the thermal
gradient
111.2
Mean temperatures ( ±S.E.) of
67
uninfected susceptible snails and five
week infected susceptible snails.
111.3
Mean temperatures (+S.E.) at which
69
snails were located one hour after
placement in a thermal gradient.
IV.1
Mean change in body temperature of
94
tadpoles.
IV.2
Mean temperature +S.E. of tadpoles in
either the presence or absence of
salamanders.
96
LIST OF TABLES
Table
Page
11.1 Species brightness scores
41
11.2 Relationship between species
42
brightness, number of parasite genera
and number of parasite species
11.3 Relationships between ecological
43
variables and parasite prevalence,
number of parasite genera and lizard
species brightness
11.4 Relationship between the percent
44
of a parasite taxon infecting a lizard
species and lizard species brightness
11.5 Variables and outcomes of interspecific
45
tests of the Hamilton and Zuk (1982)
hypothesis
IV.1 Mean temperatures of tadpoles 60 and 120
94
minutes post-injection
IV.2 Results of predator-cue experiment
95
V.1
Mean number of tadpoles in shallow zones
118
V.2
Results of predator-cue experiment
119
THE EFFECTS OF PARASITES ON HOST BEHAVIOR: WHO BENEFITS?
CHAPTER ONE
INTRODUCTION
In the broad sense, parasites compose the majority of
species on earth and strongly affect the evolutionary
biology, ecology and sexual interactions of the
organisms they infect.
Parasites, which can include
viruses, bacteria, protists, helminths and arthropods
(Margolis et al., 1982), can have wide ranging
physiological and behavioral effects on their hosts.
Physiological changes induced by pathogens have
concerned physicians for thousands of years and
physiologists for hundreds.
Recently, accounts of
behavior alterations in parasitized hosts have attracted
much attention in behavioral ecology.
Yet behavioral
and physiological changes cannot be separated.
Behavior
stems from physiology, and behaviors can alter
physiology.
An examination of the interplay of
physiology and behavior is key to understanding how
parasites may be associated with alterations of host
behavior.
Ewald (1980; 1983; 1987) has addressed the problem of
how to interpret the altered physiological responses of
some parasitized hosts.
Although he did not directly
consider them, behavioral changes associated with
2
parasites can be included in his system and tied-in with
physiological changes.
He states that alterations can
be explained in three ways: I) alterations that benefit
the host in that they defend against the pathogen, II)
alterations that benefit the pathogen and represent
manipulations of the host response, and III) alterations
that benefit neither the host or the pathogen and are
only "side effects" of infection (Ewald, 1980).
In this thesis I examine four host/parasite systems.
For each system, I explore the interrelationships
between parasites and host behaviors.
I also speculate
on the selective pressures that may be acting on both
the host and the parasite.
BEHAVIORAL ALTERATIONS THAT BENEFIT HOSTS - PARASITE
MEDIATED SEXUAL SELECTION
Darwin (1871) proposed that elaborate and apparently
deleterious secondary sex characteristics of male
animals exist due to sexual selection.
The existence of
exhaustive bellowing in mammals such as elk, elaborate
plumage and complex song in birds, and vivid coloration
in many reptiles, amphibians and fishes would therefore
be due to female choice for these traits.
Energetically
expensive and showy traits may evolve due to sexual
selection, but this begs the question of why females are
3
not attracted to dull males (Kirkpatrick, 1987; Clayton,
1991) .
Hamilton and Zuk (1982) suggested that, if parasite
resistance is a heritable trait, animals may reduce the
ravages of parasites on reproductive success by choosing
parasite-free (and therefore parasite-resistant) mates.
Not only would females avoid contact with directlytransmitted parasites but they also benefit indirectly
when their offspring inherit disease-resistant genes.
The Hamilton and Zuk hypothesis states that females
"discern" parasite resistance by using secondary sexual
traits that are only fully expressed when an individual
is healthy and parasite-free.
Hamilton and Zuk (1982) examined surveys of blood
parasites in 109 species of North American passerine
birds.
They found weak, but statistically significant,
associations between prevalence of infection (five
genera of protozoa and one nematode) and showy traits
(male "brightness", female "brightness" and male song).
To determine a species plumage-brightness and song
showiness Hamilton and Zuk used photographs and
recordings.
predictions.
Hamilton and Zuk (1982) made two major
First, that within a species, bright males
being naturally disease resistant would have fewer
parasites than dull males (tested by Borgia, 1986; Houde
4
and Torio, 1992; Kennedy et al., 1987;
Pomiankowski
1989; Milinski and Bakker, 1990; and Zuk, 1987; 1988;
1990).
Second, that between species, those species
with high parasite burdens would have experienced
selective pressure for bright characteristics as an
honest indicator of disease resistance
The intraspecific prediction that parasites alter
male showiness has been supported in some insects (Zuk,
1987; 1988), fishes (Houde and Torio; Kennedy, 1987;
Milinski and Bakker, 1990) and birds (Pruett-Jones et
al., 1990; Clayton, 1990; Moller, 1988; 1990; 1991;
Hillgarth, 1990; Weatherhead, 1990; Zuk, 1990).
The
prediction has also not been supported in some
amphibians (Hausfater et al., 1990), reptiles (Schall,
1986; Ressel and Schall, 1989), and birds (Borgia, 1986;
Borgia and Collis 1989; 1990; Gibson, 1990).
The interspecific prediction has been supported in
some birds (Hamilton and Zuk, 1982; Read, 1987), fish
(Ward, 1988; 1989).
The hypothesis has also not been
supported in some birds (Read and Harvey, 1989a; 1989b;
Read and Weary, 1990).
The inconclusive nature of tests of parasite-mediated
sexual selection may be due to inherent problems
associated with tests of the Hamilton and Zuk
hypothesis.
These include misidentification of
5
parasites, inappropriate choice of parasite taxa,
inaccurate measurement of parasite load, not controlling
for unequal exposure of hosts to parasites, measurements
of host showiness by human standards, and the role of
parasite manipulation of hosts to augment parasite
transmission (Clayton, 1991).
Tests of the Hamilton and Zuk hypothesis in lizards
are particularly interesting.
Hamilton and Zuk
predicted an inverse correlation between male showiness
and parasite load.
Indeed, as a general trend, sick
vertebrates do tend to be more drab than healthy
conspecifics (Hamilton and Zuk, 1982).
Yet Schall
(1986) found that bright blue male Cnemidophorus
arubensis lizard were more likely to be infected with a
haemogregarine parasite that were dull males.
Similarly, Ressel and Schall (1989) found that malaria
infected male fence lizards (Sceloporus occidentalis)
were significantly more showy than conspecifics of the
same age because sub-adult infected males display
markings normally restricted to adult males.
In this
species, adult males develop vivid black coloration on
their ventral surface.
This black coloration contrasts
sharply with parallel blue stripes on their ventral
surface.
Males engage in "push-up" displays that reveal
their venters and are important in intersexual
6
interactions (Noble, 1934).
Evidence from Sceloporus occidentalis contradicts the
intraspecific prediction of the Hamilton and Zuk
hypothesis since infected males are brightly colored yet
physiologically debilitated and suffer lowered
reproductive success (Schall, 1983; Schall and Dearing,
1987).
Nor does the malarial parasite Plasmodium
mexicanum benefit from making its host more attractive
to females since the parasite is not directly
transmitted (such as lice) but is vector borne.
Ressel
and Schall (1989) suggested that the striking color of
infected males may support Zahavi's (1976) handicap
principle, i.e., showiness of infected males may be used
by females as a mark of quality in that only truly
exceptional males could survive with such handicaps (see
also Dawkins, 1992).
BEHAVIORAL ALTERATIONS THAT BENEFIT HOSTS
BEHAVIORS
THAT MINIMIZE CONTACT WITH, OR SLOW THE GROWTH OF
PARASITES
Besides choosing parasite-free mates, animals often
change their behavior in other ways both to avoid being
parasitized and to enhance their ability to fight off
infection (reviewed by Hart, 1988; 1990).
Many animals
preen to remove ectoparasites, avoid reusing nesting
7
areas, avoid areas where ticks are abundant, and flee
from aerial parasites such as botflies that are seeking
hosts in which to lay eggs (Hart, 1990; 1992).
Endoparasites are avoided by altering foraging
strategies.
Many ungulates avoid feeding near recently
deposited feces (Taylor, 1954; Michel, 1955).
Other
strategies are to defecate only in specific areas of a
territory, to avoid large congregations of possibly
disease-carrying conspecifics, to avoid cannibalism, and
to change their diet when infected (Hart, 1990; see also
Freeland, 1983).
Even if potential hosts are unable to avoid being
infected, many are able to use behavioral means to slow
the growth of, or to eliminate, a parasite.
Infected
vertebrates exhibit reductions in food intake which may
serve to reduce iron levels (a necessary nutrient for
gram negative bacteria, Hart, 1990).
Indeed, mice
starved for three days have increased survival when
challenged with the bacterium Listeria monocytogenes
(Wing and Young, 1980).
Sick mammals often are inactive
and lethargic (Hart, 1990).
Hart (1990) interprets
these behaviors as mechanisms to husband resources to
fight off infection. Chernin (1967) reported that snails
(Biomphalaria glabrata) infected with trematodes
(Schistosoma glabrata) were sluggish.
Infected animals
8
often seek-out and dwell in warm microhabitats.
This
raises the temperature of the animal (behavioral fever)
and results in increased survivorship during infection
for both ectotherms (Kluger, 1975) and endotherms
(Dimock et al., 1991).
Fever increases survival rate
during bacterial infections by lowering blood iron
concentrations, mobilizing the immune response,
enhancing cytolysis, and exposing some bacteria to
lethal temperatures (Kluger, 1990).
To examine behavioral alterations that benefit hosts,
I test Hamilton and Zuk's interspecific hypothesis using
lizards in chapter two.
Lizards are often brightly
colored and species vary in parasite loads.
Has
selective pressure from female choice in heavily
infested species resulted in the evolution of showy
males as a mark of disease resistance?
In chapter
three, I examine if trematode infected snails alter
thermoregulatory behaviors.
BEHAVIORAL ALTERATIONS THAT BENEFIT PARASITES - PARASITE
MANIPULATIONS OF HOST BEHAVIOR AND GROSS PHYSIOLOGY
Host behavior may also be altered due to parasitic
manipulations that enhance a parasite's rate of
transmission.
Bethel and Holmes (1972) proposed three
possible strategies that parasites may use to augment
9
the capture of intermediate hosts by definitive hosts:
1) decreasing host stamina, 2) increasing host
conspicuousness, 3) disorienting the host and altering
the ability of the host to elude predators.
Decreased host stamina is a common symptom of
infection (Hart, 1988).
As early as 1906, it was
noticed that butterfish (Peprilus triacanthus) with
cestodes in their musculature had reduced activity
(Linton, 1906).
Shinner fish infected with cestodes
were also found to be sluggish and inhabited warmer,
shallow water (Dence, 1958; Bethel and Holmes, 1972).
Hydatid cestode cysts in ungulates have been correlated
with severe decreases in stamina and ability to elude
capture by predators (Fenstermacher, 1937; Cowan, 1951;
Ritcey and Edwards, 1958 and Mech, 1966).
Reduced
ambulatory activity has been observed in CD-1 mice (Mus
musculus) with Trichinosis-encysted muscles (Zohar and
Rau, 1986).
Increased conspicuousness is another strategy that
parasites use to attract visually stimulated predators
(Holmes and Bethel, 1972).
Altered pigment patterns in
fish are associated with trematode infections
(Rothschild, 1962).
Hares exhibit inappropriate
seasonal color changes when infected with cestodes
(Leiby and Dyer, 1971).
Coral polyps infected with
10
trematode metacercaria appear as bright pink swollen
nodules that the corals are unable to retract into
protective calices (Aeby, 1991).
The trematode's rate
of transmission was enhanced as colored corals were
consumed by fish-definitive hosts at a higher rate than
uninfected corals.
The most dramatic example of
parasite-induced conspicuousness is that of a snail that
when infected by a trematode exhibits pulsating
tentacles which attract avian definitive hosts (Hecker
and Thomas, 1965).
Disoriented hosts often display inappropriate
behaviors, have a reduced fright-response, and may
separate from the herd due to neurological damage.
Mice
artificially infected with nematodes have increased
activity levels due to brain lesions (Rau, 1984; Hay et
al., 1985; Dolinsky et al., 1985).
Lemmings also
exhibit less of a fright response when infected with a
coccidian parasite (Quinn et al., 1987).
Blindness is a
common result of trematode infections in the eyes of
fish (reviewed by Bethel and Holmes, 1972; Crowden and
Broom, 1980).
Ants infected with trematode metacercaria
exhibit a suite of behaviors that increase their chance
of being accidentally ingested by sheep - the
trematode's definitive host (Carney, 1969; see also
Curtis, 1990).
Some of the above examples illustrate
11
the exhibition by hosts of behaviors that have no
relation to normal activities, while others are normal
behaviors exhibited at inappropriate times.
Gammarid crustaceans infected with acanthocephalans
have reversed phototaxis and an altered fright response
to predacious birds (Bethel and Holmes, 1972; 1977).
Isopods parasitized by acanthocephalans are
disproportionately found on light colored substrate,
unsheltered areas and in dry areas of their habitat.
These altered behaviors make the isopods more
susceptible to starlings which are the definitive host
of the parasite (Moore, 1983).
Indeed, behavioral
alterations are so common among acanthocephalan
intermediate hosts, that Moore (1984) has suggested that
all acanthocephalans may have evolved from an ancestor
that altered the behavior of its intermediate host.
Cestode infected beetles fail to exhibit normal
photophobic behavior (Hurd and Fogo, 1991).
Plasmodium
infected rodents exhibit reduced ability to fend off
biting mosquito vectors during the period when the
rodents are most infective (Day and Edman, 1983).
Mosquitoes carrying Plasmodium have reduced ability to
locate blood vessels.
This results in both increased
probing times and an increase in the number of hosts
contacted during a feeding session (Rossignol et al.,
12
1984).
Sand flies infected with Leishmania exhibit a
response similar to Plasmodium infected mosquitoes
(Beach, et al., 1985).
BEHAVIORAL ALTERATIONS THAT BENEFIT PARASITES - PARASITE
MANIPULATIONS OF HOST IMMUNE AND HORMONAL FUNCTIONS
Observations of altered behaviors are common.
However, studies that explore humoral and hormonal
mechanisms associated with altered behaviors are rare.
Changes in the clinging behaviors of gammarids, noticed
by Bethel and Holmes (1972; 1977), are due to
acanthocephalan-induced changes in serotonin-sensitive
pathways (Helluy and Holmes, 1990).
Stickleback fish
infected with cestodes spent more time at the surface
than uninfected controls.
Lester (1971) found that this
was due to infected fish having higher respiration
rates.
To understand how the physiology of an animal is
liable to change during infection, it is necessary to
examine the changes that occur during infection.
During
the initial stages of infection tissue macrophages and
blood monocytes are activated and they release cytokines
such as Interleukin-1 (IL-1), Interleukin-6 (IL-6) and
Tumor Necrosis Factor (TNF).
These cytokines serve as
an early warning system to nearby cells which begins a
13
cascading response of fever, the acute phase response,
and the differentiation and activation of T-cells, Bcells and macrophages. (Titus et al., 1991).
Interestingly, mosquito (Aedes aeaypti) salivary glands
contain a substance that inhibits TNF release
(Bissonnette et al., 1992).
IL-1 causes the release of IL-6 and TNF.
IL-1
reduces appetite (McCarthy et al., 1985), induces
general malaise (Bluthe et al., 1992) and slow-wave
sleep (Kluger, 1990).
11-6 also induces fever and is
released during psychological stress (LeMay et al.,
1990) .
The acute phase response also occurs during the
initial stages of infection and has two components.
The
first local phase involves hemostasis, inflammation, and
kinin generation.
The second systemic phase results in
pain, leukocytosis, the manufacture of acute phase
proteins such as C-reactive protein, increased protein
catabolism, increased gluconeogenesis, changes in blood
concentrations of heavy metals, increases in cortisol,
glucagon, catecholamines, and thyroxin (reviewed by
Stadnyk and Gauldie, 1991).
Febrile (i.e. elevated) temperatures are detrimental
to the growth of some bacteria and some stages of
Plasmodium.
Yet overproduction of cytokines can also be
14
detrimental.
Brain damaging and fatal temperatures for
hosts are usually just a few degrees above febrile
temperatures.
Furthermore, overproduction of TNF causes
the wasting and gauntness of cancer and AIDS patients.
TNF's side effects in humans are muscle stiffness,
muscle pain, nausea, and vomiting (Clark, 1987). Indeed,
the pathology of cerebral malaria may largely be due to
TNF (Clark, 1987).
During the acute phase response the
liver of humans has a reduced oxidase system which
results in a decreased ability to detoxify pathogenderived toxins (Stadnyk and Gauldie, 1991). Nematode
infected mice show a similar response (Tekwani et al.,
1987) .
Paradoxically, the immune system of vertebrates may
enhance the transmission of certain parasites.
Doenhoff
et al. (1978) examined mice infected with the human
blood fluke Schistosoma manson
.
In this system, an
adult parasite-pair inhabit mesenteric arteries of their
mammalian definitive host.
Eggs are released and, to be
infective, must travel from mesenteric arteries through
a thick intestinal wall to the intestinal lumen from
where they are eventually excreted.
How the nonmotile
eggs are able to reach the lumen is unclear.
Interestingly, although immunocompetent mice release
parasite eggs, immunodeficient mice do not (Doenhoff et
15
al, 1978).
Yet both groups of mice have identical worm
burdens with identical fecundity rates.
When infected
immunodeficient mice were given T-cells, they began to
excrete Schistosome eggs (Doenhoff et al., 1985).
Baboons with Schistosoma mansoni infections possess
intestinal granuloma that contain Schistosome eggs and
host derived neutrophils, macrophages, and lymphocytes
(Damian et al., 1984).
Damian (1987) suggests that
although the eggs are nonmotile, immune cells such as Tlymphocytes, B-lymphocytes, plasma cells, eosinophils,
neutrophils, fibroblasts and macrophages are motile.
He
proposes, and provides evidence, that the egg/immune
cell granuloma travels as a unit from the mesenteric
arteries to the lumen of the intestine. Once in the
lumen the granuloma breaks down and the egg is released.
Some parasites may actively cause a host defense
response and then benefit from the resulting behavioral
changes.
parasites.
The inflammatory response can be utilized by
Mycobacterium and Salmonella utilize
traveling monocytes to spread from their initial point
of invasion (Parker and Schneider, 1981).
Schistosomes
and trypanosomes are able to rapidly shed host antigens
(Parker and Schneider, 1981), and mammals infected with
bacteria often have high levels of bacterial proteins in
their blood (Rake, 1933).
This may act as "chaff", to
16
confuse the immune system as to the actual location of
the pathogen.
Bacterial releases of this nature would
result in massive production of cytokines which would
result in a cascade of physiological and behavioral
effects.
Some of these changes in behavior, i.e.
drowsiness and altered thermoregulation, may result in
the death of the host through predation.
Some multi-
host parasites may be under selective pressure to induce
such changes at a parasite life-history stage that is
conducive for transfer to a subsequent host.
In chapter four, I explore the behavioral effects of
the host defense response.
This response may be
available to parasites as a mechanism to alter
intermediate-host behavior in ways that lead to the
definitive host consuming the intermediate host.
III) BEHAVIORAL ALTERATIONS THAT BENEFIT NEITHER THE
HOSTS NOR THE PARASITES
Evolutionary biologists rarely examine host changes
that benefit neither the host or the pathogen.
Although
Ewald (1980) brings attention to these changes, he uses
it as a "lump-all" category for alterations of hosts
that do not fit into his other two categories.
An
examination of these "side-effects of infection" is
overdue because alterations of hosts are almost always
17
viewed as adaptive to either the host or the parasite
(see also Dobzhansky, 1956; Gould and Lewontin, 1979;
Mayr, 1983 and Wright, 1949).
Although unexplainable from the adaptationist
viewpoint, altered host behaviors are also associated
with parasites that do not gain fitness-benefits from
these altered behaviors.
For example, mice (Peromyscus
maniculatus) are more susceptible to predation by
weasels if the mice are infected with two or more bot
fly maggots (Smith, 1978).
active.
consumed.
Infected mice are less
Bot fly larvae die when their mouse host is
Thus, altered behaviors are harmful to both
the host and the parasite.
Mite infestations of an
homopteran insect result in increased predation rates yet the mite is not transmitted to most predators
(LaMunyon and Eisner, 1990).
Ectoparasitic copepods are
attracted to moving trout (Poulin et al., 1991), yet
once a trout is infected its activity levels increase
which results in further infestations.
The net result
is a weak host and fewer resources for each individual
parasite.
The trout dies and the parasites are thereby
deprived of a source of nutrients.
Unless the copepods
are kin (untested), behavioral alterations that result
from infection benefit neither the host nor the
parasite.
Some of these studies may have examined a
18
predator that is not the definitive host but other
unobserved predators may be (or have been, Connell,
1980) the definitive host.
Yet many of the parasites in
this category have only a single definitive host and are
directly transmitted.
The unexplainable nature of these
studies deserves closer attention.
In this thesis, I will examine the hypothesis that
pathogens cause selective pressure for: 1) hosts to
modify their behavior to avoid becoming infected 2)
altered host behaviors that, along with physiological
changes, reduce the growth and detrimental effect of
parasites; 3) altered behaviors that may enhance the
growth and reproduction of the pathogen; and 4) altered
behaviors that do not benefit the host or the pathogen
and are only a byproduct of substances hosts release
during infection.
In chapter two, I explore how potential hosts may
alter their mating systems to avoid becoming infected.
Specifically I examine the predictions of the Hamilton
and Zuk hypothesis (1982) using lizards as a model
system.
The hypothesis predicts that species with high
parasite loads should have evolved showy characteristics
as a true indicator of parasite resistance. Contrary to
the predictions of the hypotheses, I found a significant
inverse correlation between a species' parasite load and
19
its showiness.
In chapter three, I explore how a snail,
Biomphalaria glabrata, alters its normal
thermoregulatory behavior when infected with the
trematode Schistosoma mansoni.
I found that infected
snails thermoregulate at lower temperatures than healthy
controls.
This may be advantageous to the snail because
the parasite has impaired development at lower
temperatures.
In chapter four, I examine behavioral alterations
that occur in tadpoles during the acute phase response.
Compared to invertebrates, anurans have sophisticated
defense systems, possessing many of the immune system
components of mammals, including cell-mediated and
humoral immunity.
They posses IgM, IgY (IgG-like), and
IgX (related to IgA) immunoglobulin.
Anurans have T and
B cells, lymphoid tissue, histocompatibility complexes,
complement system, and phagocytosing cells, and
magainins (peptides with antibiotic and antifungal
abilities, Tinsley, 1989).
Furthermore, they are
capable of antibody responses that include immunological
memory.
I found that tadpoles with activated defense systems
have impaired refuge seeking behavior and suffer
increased predation by newts (Taricha qranulosa).
20
Manipulations of this response may provide multi-host
parasites with a method to augment transition to a
definitive host.
In chapter five, I build on my results from chapter
four and investigate how infection with a pathogenic
yeast alters the behavior of tadpoles.
Infection
results in many symptoms of the acute phase response,
e.g. fever, and results in a similar suite of behaviors
that I discovered in chapter four.
However, in this
system neither the host nor the parasite benefits from
behavioral alterations.
Finally, in the summary, I tie together these results
and propose a framework to interpret behavioral
alterations of parasitized hosts.
21
CHAPTER TWO
PARASITE LOAD AND BRIGHTNESS IN LIZARDS:
AN INTERSPECIFIC TEST OF THE HAMILTON AND ZUK HYPOTHESIS
by
Hugh Lefcort
and
Andrew R. Blaustein
This paper appeared in
JOURNAL OF ZOOLOGY (LONDON), 1991. 224:491-499
22
ABSTRACT
Hamilton and Zuk (1982) hypothesized a positive
correlation between a species sexual showiness and its
level of parasitic infection.
We tested the hypothesis
in 26 species of lizards, members of a class of
vertebrates never before used to test the model.
The
prevalence of parasites was determined using published
lists of parasites found in wild lizard populations.
An
index of showiness (brightness) was derived by scoring
photographs of lizards in natural settings.
Contrary to
expectations of Hamilton and Zuk, we found an inverse
correlation between a lizard species' brightness and
parasite prevalence.
No correlation was found between a
species' brightness and the number of parasite genera,
species, or percentage of individual infecting parasite
taxa.
These results are discussed in relation to other
interspecific tests of the hypothesis.
23
INTRODUCTION
Hamilton and Zuk (1982) proposed that there should be a
positive association between a species' sexual showiness
and its level of parasitic infection.
They predicted
that interspecifically, highly parasitized species are
more showy than those with less of a parasite load, and
intraspecifically, that males with more conspicuous
secondary sexual characters have a lower parasite load
than those with less conspicuous characters.
In species
where female choice of mates is prevalent, it is assumed
that females will choose the showiest males because
amplified secondary sexual characters should reflect a
relatively low parasite burden.
Individuals that choose
mates with a lower parasite (or disease) load, would be
favored by natural selection if parasite (or disease)
resistance is at least partially heritable.
The Hamilton and Zuk (1982) hypothesis has been
tested in several species from varying taxonomic groups.
The intraspecific aspect of the hypothesis has been
tested by Borgia (1986), Kennedy, Endler, Poynton et al.
(1987), Milinski and Bakker (1990), Pomiankowski (1989),
and Zuk (1987; 1988; 1990).
Many studies have also
focused on the interspecific element of the hypothesis.
Positive correlations of parasite prevalence and
showiness have been suggested for birds (Hamilton and
24
Zuk, 1982; Read, 1987; Read and Harvey, 1989a and Read
and Weary, 1990) and fish (Ward, 1988; 1989).
However,
support for the hypothesis is not conclusive because
some analyses of particular data sets seem to support it
(Hamilton and Zuk, 1982; Read, 1987; Ward, 1988) whereas
others do not (Read and Harvey, 1989a; Read and Weary,
1990).
Discrepancies in how data sets are analyzed has
added to the confusion.
For example, using
independently derived plumage scores Read and Harvey
(1989a) reexamined Hamilton and Zuk's original data and
found no evidence to support the hypothesis.
An Interspecific Test
To further test the interspecific aspect of the
Hamilton and Zuk hypothesis, we examined the
relationship between sexual showiness (brightness) and
parasitic infection among lizard species.
Lizards seem
to be an ideal group in which to test the Hamilton and
Zuk hypothesis.
The presence of bright male colors are
important in lizard sexual displays (Noble, 1934; Schall
and Sarni, 1987; Schall and Dearing, 1987) and lizard
color varies with parasite prevalence (Ressel and
Schall, 1989).
However, Ressel and Schall (1989) found
that contrary to expectations of the Hamilton and Zuk
hypothesis, male Sceloporus occidentalis lizards
25
possessed an increased degree of bold black ventral
pigment when infected with the malarial pathogen
Plasmodium mexicanum.
This is particularly interesting
considering the drastic decrease in male reproductive
success when suffering from malaria due to an inability
to engage in energetic male/male interactions (Schall,
1983; Schall and Dearing, 1987).
The dark venters of
sexually mature but young infected males, contrasts
sharply with the dull colored venters of young
noninfected males.
A heavily pigmented ventral surface
is normally the sign of an older mature male.
Ressel
and Schall (1989) thought that to their eyes "... a male
Sceloporus occidentalis venter accented with copious
black is the most impressive color pattern".
Similarly,
in the lizard Cnemidophorus arubensis, bright blue males
had higher levels of infection with a haemogregarine
parasite (Schall, 1986).
Previous tests (e.g. Hamilton and Zuk, 1982; Read,
1987; Read and Weary, 1990; and Ward, 1988) encompassed
potentially confounding variation by examining extremely
broad taxonomic groups.
Read and Weary (1990) have
suggested that interspecific comparisons may suffer from
statistically non-independent species points.
This
could arise if the species examined were not
phylogenetically independent.
Clusters of closely
26
related species within a larger data set would be
treated as independent points.
This could skew the
results when these species clusters are compared to
other clusters (see also Pagel and Harvey, 1988).
We investigated predictions of the hypothesis in 26
species of lizards from 14 genera and 8 families (Table
11.1).
Furthermore, we attempted to reduce and equalize
the phylogenetic variation between species by
concentrating on two families of iguanian lizards, the
Phrynosomatidae (eight species examined) and the
Polychridae (six species examined, see Frost and
Etheridge, 1989 for recent phylogenies).
The taxonomic
level of family was chosen since it is the most
restrictive classification for which an adequate sample
size of lizard species and their parasites was
available.
Key ecological characteristics of the
species were also examined to determine their
relationship with parasite prevalence and sexual
showiness.
Since this is the first interspecific study
using reptiles to specifically test Hamilton and Zuk's
hypothesis, results from our analysis will add to the
generalization between parasite prevalence and sexual
showiness.
27
METHODS
To locate studies numerically listing the
endoparasites of lizard families containing four or more
species of lizards, we examined all studies of iguanian
parasites listed in Zoological Record from 1960 - 1989.
To our knowledge only the six studies examined meet this
criterion.
Only families from the Americas were
examined since only for these groups were we able to
obtain adequate photographs.
We determined the mean number of parasite species and
genera in 26 lizard species using lists of endoparasites
found in 1580 free ranging lizards that were compiled by
Waitz (1961), Telford (1970), Pearce and Tanner (1973),
Benes (1985), Lyon (1986) and Bundy et al (1987).
The
investigators made no attempts to associate parasite
prevalence with color.
We obtained mean parasite
prevalence by summing the individuals infected by each
parasite species and dividing by the number of
individuals of that lizard species in the study (for
explanation of terms see Margolis et al., 1982).
Furthermore, to examine if differences in mode of
transmission and life cycle stage of the parasites
affected the color/parasite prevalence relationships, we
derived and examined the percentage of the total number
28
of parasite species and parasite genera made up of
flagellates, ciliates, amoebas, sporozoans,
platyhelminths, nematodes, or acanthocephala separately
for each lizard species.
Because some lizards were
infected with more than one parasite species, the total
number of infections exceeded the number of lizards (see
Hamilton and Zuk, 1982).
In the examination of parasite
prevalence, all parasites were considered equivalent to
each other in pathogenic effects whether they were in
their intermediate or definitive host.
This may be
justified in that the theory predicts only that males
must demonstrate overall parasite resistance and not
resistance to particular types (Ward, 1988).
However,
certain parasites such as blood-borne hematozoa can have
far greater deleterious effects on reproductive fitness
than, for example, intestinal parasites (Schall, 1983;
Ewald, 1983).
For this reason we investigated the
relationship between the parasite taxa examined and a
species' brightness.
As in Read (1987) and Ward (1988), brightness scores
for each lizard species were obtained by asking naive
associates to score brightness.
Six colleagues judged
brightness on a ten point scale (1 = dull, 10 = bright
and conspicuous) from either Kodachrome slides
(collection of R. M. Storm) or pictures of male lizards
29
from field guides (Behler and King, 1988; Schwartz and
Henderson, 1985) for head, ventral body, dorsal body and
tail brightness.
derived.
An average value for each species was
All photographs included areas of the body
used during displays.
Two judges scored the Kodachrome
slides and four scored the book pictures.
For species
exhibiting large morphological variation over their
range, more than one photograph was used and the results
were averaged.
The statistical methods used to
determine a species' color score were similar to those
of Read (1987) and Ward (1988).
Read and Harvey (1989a; 1989b) used two-tailed
statistical tests because they believed that the null
hypothesis would be rejected had a significant negative
relationship been found.
Zuk (1989) has questioned this
assumption claiming that a negative correlation would
not support the Hamilton and Zuk hypothesis, stating
that only one-tailed tests should be used.
Given the
unresolved nature of this disagreement and the increased
tendency of a type I statistical error if one-tailed
tests are used, we believe that two-tailed tests should
initially be used, followed by one-tailed tests if Pvalues lie between 0.05 and 0.1.
All values reported
are two-tailed P-values unless otherwise specified.
As in Read (1987) and Read and Harvey (1989a), we
30
examined the association between parasite prevalence and
a lizard species' brightness scores using Spearman's
Rank correlation.
We also examined if brightness is
correlated with the number of parasite genera or
species.
Most tests (Hamilton and Zuk, 1982; Read,
1987; Read and Harvey, 1989a; Read and Weary, 1990;
Ward, 1988, 1989) of the Hamilton and Zuk hypothesis
pooled data from more than one study to determine the
parasites load of a vertebrate species.
We believe that
the literature on the parasites of lizards is inadequate
to indicate the parasite prevalence of individual lizard
species.
For this reason, and to control for
differences in methods of data collection between
parasite studies (see for example Kuris and Blaustein,
1977), we not only pooled data from separate studies but
we also examined data from each study separately, with
regard to brightness scores.
Furthermore, in an attempt
to control for differences between lizard families,
studies that included the Phrynosomatidae (Telford 1970)
and Polychridae (Bundy et al. 1987), the families with
the largest number of lizard species samples, were used
to determine if within these two families brightness was
correlated with any index of parasite load.
Ecological factors were analyzed using Spearman's
Rank correlation to determine if additional variables
31
might affect parasite prevalence or brightness score.
The variables were
:
(A) food gathering, whether a
species is a sit-and-wait predator or an active forager;
(B) strata, whether a species is most often found (1) on
the ground,
(2) between ground level and one meter,
or
(3) above one meter [for the Anoles of Bundy et al.,
(1987) the criteria were (1) trunk/ground,
(2) twig,
(3)
trunk/crown]; (C) month of initial breeding; (D) total
length of breeding season (weeks); (E) size (total
length in mm); and (F) altitude of capture (m).
These
data were obtained from Behler and King (1988) and
Stebbins (1954).
32
RESULTS
The color scores among judges (Table II.1) were
normally distributed and they were correlated [mean
Spearman rank coefficient (rj = 0.64
0.81,
(P < 0.05)].
,
range 0.36 -
The mean scores of the Kodachrome
and the book values were highly correlated (r, = 0.89, P
< 0.05).
In all of the studies we examined we failed to find a
positive correlation between a lizard species'
brightness and any measure of parasite load (Table
11.2).
Results from the single large study used in our
analyses (Telford, 1970) indicated a significant (onetailed) inverse correlation between parasite prevalence
and brightness (r, = -0.448, P = 0.038).
Analyzed using
a two-tailed test the correlation would no longer be
significant with a P-value of 0.076.
Telford's study
encompassed a taxonomically broad group of seven
families, twelve genera and eighteen species, including
two subspecies.
When we restricted our analyses of the
Telford study to the monophyletic Phrynosomatidae (N = 8
species), no correlation was found between brightness
and any of the measures of parasite load.
The analyses of the pooled studies revealed no
correlation between species' brightness and any measure
of parasite load.
For all studies pooled the altitude
33
of capture was the only ecological variable that was
correlated with overall parasite prevalence
[r, = -0.87 P < 0.01 (Table I1.3)].
When altitude-of-
capture was controlled for (A. Read, pers. comm.), by
dividing the lizard species into three separate altitude
groups (0 - 600m, 601 - 1000m and 1001+m), none of the
studies individually, or pooled, showed any correlation
between species' brightness and parasite prevalence, the
number of parasite genera, parasite species,
or any of
the ecological variables.
Using data from each of the smaller studies and the
Polychridae study individually, no correlation was found
between a species brightness score and either overall
parasite prevalence, number of parasite genera, or
parasite species.
The presence of any particular
parasite taxon was not correlated with a species
brightness (Table 11.4).
34
DISCUSSION
Our analysis does not support the Hamilton and Zuk
hypothesis that highly parasitized species are more
showy than those species with reduced parasite
prevalences.
Among 951 lizards from 18 lizard species
in our one large study (Telford, 1970) we actually found
an inverse correlation, using one-tailed tests, between
species' brightness and parasite prevalence, thus,
brighter lizard species had fewer parasites.
Our
finding supports the contention of Read and Harvey
(1989b) that continued female choice of bright parasite
resistent males may lead to a negative correlation
between parasite load and species brightness.
We report
this one-tailed value because previous studies of lizard
parasites (Schall, 1986; Ressel and Schall, 1989)
suggest a negative correlation between species color and
parasite load.
Reporting the results of one-tailed
tests is important because this may indicate a
significant trend in the correlation between parasite
load and lizard color.
The number of parasite species, genera, or
percentages of individual parasite taxa were not
correlated with a lizard species' brightness.
Nonsignificant results were also found in the five
smaller studies, both individually and when we pooled
35
them with Telford's (1970) study.
Of six ecological
attributes examined in relation to parasite prevalence
and brightness, only the altitude of capture was
correlated, being inversely related to parasite
prevalence.
When we controlled for altitude-of-capture
in Telford's (1970) study, by dividing the lizards into
three separate altitudes groups, no correlation was
noted between a lizard species' brightness and parasite
prevalence.
We do not feel that this detracts from the
significance of the inverse correlation we found in
Telford's study (1970) between species' brightness and
parasite prevalence since the lack of significance could
be due to the small number of lizard species (4, 6 and
8)
in each of these groups.
The fact that only
Telford's study (1970), and not the pooled analyses,
yielded a significant correlation between a lizard
species' brightness and parasite prevalence could be due
to a number of factors.
Telford's data set is the
largest (18 species and 954 animals) and most
comprehensive of the studies we examined.
Furthermore,
widely different methods between studies in examining
the lizards for parasites, may render a pooled analysis
of limited use and could lead to spurious correlations
(see discussion in Kuris and Blaustein, 1977).
Hamilton and Zuk (1982) may be correct that a male's
36
brightness advertises health and vigor to females.
Yet
brightness traits are also under many other selective
pressures, largely from male-male interaction,
camouflage from predators, and thermoregulatory factors.
For example, Norris (1967) showed that desert lizards
were darker dorsally in the morning, thereby increasing
UV absorption, and lighter at mid-day to facilitate
reflection of incident solar radiation.
Bright secondary sexual characteristics are important
in lizard social communication (Schall and Sarni, 1987;
Schall and Dearing, 1987; Cooper, 1988), yet non-visual
cues such as vocalizations or odors, might also be
involved.
Although the lizards censused in this study
are not known to vocalize, the degree and form of
chemical signals could be important variables in
determining a species' "showiness", as has been
suggested for mammals (Blaustein, 1981).
Moreover,
color may significantly change with behavioral context.
For example, an individual may show intense colors when
aggressive and be rather drab when not aroused.
An interesting form of showiness has been described
in the lizard Dipsosaurus dorsalis (Alberts, 1989).
While the color score we obtained for this species was a
drab 2.3 (out of 10), Alberts (1989) has shown that D.
dorsalis is actually conspicuous compared to its desert
37
background when viewed in the near ultraviolet
(wavelengths the animal is capable of perceiving).
If
D. dorsalis females are using the near ultraviolet to
determine brightness then the color score we obtained
using the visible spectrum could be incorrect.
Thus, it
is possible that human investigators may not accurately
be assessing color patterns as they are perceived by
some animals.
Difficulties are inherent when testing the Hamilton
and Zuk hypothesis.
If the analyses are conducted among
too broad a taxonomic group, then morphological,
ecological and historical differences between species
can reduce the significance of any parasite/color
relation (Read and Harvey, 1989a).
Hamilton and Zuk
(1982) examined a sampling of species from an entire
order (Passeriformes).
Read (1987) considered a
similarly broad group although he attempted to control
for taxonomic and behavioral variables.
Ward (1988)
limited his analyses to comparisons between families,
yet his study averaged fewer than three species per
family (Table 11.5).
Using Telford's (1970) broad based
study encompassing seven families we found a negative
correlation between brightness and parasite prevalence the reverse of what Hamilton and Zuk (1982) would
predict.
We also limited part of our study to a narrow
38
monophyletic group yet we failed to support the
hypothesis.
One potential problem with correlational studies
testing the Hamilton and Zuk hypothesis (Hamilton and
Zuk, 1982; Read, 1987; Ward, 1988; 1989) is that all
parasite species are given equal weight regardless of
the host's position in the parasite's life cycle,
pathogenicity, or mode of transmission.
However, Ewald
(1987) has shown that modes of transmission are
important determinants in parasite pathogenicity.
Perhaps it is prudent to differentiate among types of
parasites rather than to lump entire taxa under headings
such as "blood-borne".
Instead of approaching parasites
as homogenous units with constant effects, tests of the
hypothesis should explore how ecological variables among
parasite groups affect host reproductive success.
For
this reason we investigated not only overall parasite
prevalence, but also if any particular parasite taxa was
correlated with a species brightness.
Similarly, the accuracy of measuring parasite
prevalence differ among parasite taxa.
While counting
helminths infesting the gut may be an accurate
indication of parasite prevalence, data on hematozoa
numbers from stained blood smears are less reliable
(Cox, 1989).
39
A further potentially more serious problem with
previous studies, and one only partially addressed by
our study, is that of measuring parasite prevalence at
the species level rather than on a population level.
Variation in pathogen resistance among populations is
not addressed, nor is it clear if the geographic range
of all parasites is coincident with all host
populations.
Furthermore, brightness may vary among
populations.
To properly address these issues parasites
and brightness should be measured between closely
located ecologically similar populations.
By limiting part of our analyses to the family level,
historical and ecological variance between species was
minimized yet our results failed to find evidence of a
positive association between a species' sexual showiness
and its degree of parasitic infection as proposed by
Hamilton and Zuk (1982).
We found that a negative
correlation exists in lizards which lends support to
Read and Harvey's (1989b) interpretation of the
hypothesis.
Additional studies, testing other
vertebrates hosts and their parasites, and addressing
the problems listed above, will be necessary before the
overall significance of the hypothesis is ascertained.
40
ACKNOWLEDGEMENTS
We wish to thank A. Kuris and M. Zuk for reviewing the
manuscript and for valuable suggestions on ways to
improve it. A. Read provided extremely helpful additions
on more than one occasion.
Various drafts were greatly
improved by the critical readings of D. Olson, C. Fuller
and K. Chang. We thank K. Chang, C. Fuller, J. Harding,
D. Meadows, D. Olson, J. Peterson, P. Van Tamelen, L.
Walsh and L. Zhang for scoring lizard colors.
thank R. Storm for providing lizard slides.
We also
We are
especially grateful to the large contributions of Ann
Darrow, Jack Driscoll and Carl Denham. C. Englehorn
provided logistic support.
Partial financial support
was provided by NSF grant BNS-8709047 to ARB.
TABLE II . 1
141
Species Brightness Scores
No.
Lizard species
Individuals/
Study
Anolis grahami
14
12
Anolis sari:Tel
Eumeces skiltonianus
Sceloporus magister
Coleonyx varieqatus
Sceloporus orcutti
Anolis qarmani
Xantusia henshawi
Anolis opalinus
Callisaurus draconoides
Eumeces ailberti
Anolis valencienni
Uta steineaeri steinegeri
Gambelia wislizenii
Uta stansburiana hesperis
Cnemidophorus tiaris
Xantusia riversiana
Anolis lineatopus
Xantusia viqilis
Sceloporus occidentalis
Sceloporus undulatus
Sceloporus qraciosus
Uma notata
Phrynosoma solare
Dipsosaurus dorsalis
Sauromalus obesus
Urosaurus qraciosus
1
2
3
4
5
6
=
=
=
=
=
=
21
115
71
23
4
51
18
19
6
7
418
24
22
131
78
41
18
151
11
220
Benes (1985)
Bundy et al., (1987)
Lyon (1986)
Pearce and Tanner (1973)
Telford (1970)
Waitz (1961)
24
14
21
12
34
Derived
Mean
Brightness
score
Study
6.8
5.9
5.8
5.8
5.4
5.2
5.0
5.0
4.9
4.6
4.4
4.3
4.2
4.1
4.1
3.8
3.6
3.2
3.2
3.1
2.9
2.8
2.6
2.5
2.3
2.3
1.7
2
2
5,
1,
1,
6
4,
5
5
5
2
5
2
1,
5
2
1,
3,
5,
1,
5
5
5
5,
6
3,
6
5
2
5
3,
4,
5,
6
4,
5,
6
4
3,
5
1
5
5
5
TABLE 11.2
Relationships (r1) Between Species Brightness, Parasite Prevalence,
Number of Parasite Genera and Number of Parasite Species
Study
#Spp
#Liz
Benes (1985)
Bundy et al.(1987) X
Pearce and Tanner (1973)
Lyon (1986)
Telford (1970)
Waitz (1961)
Telford (1970) XX
All Studies Pooled
6
6
4
4
271
Parasite
Prey.
-0.700
0.257
-0.200
0.800
-0.488*
0.800
-0.200
1580
-0.316
96
55
183
954
21
679
18
4
8
26
> 0.05 in all tests.
P = 0.0383 (one-tailed). Two-tailed P = 0.076.
X
Polychridae
XX Phrynosomatidae
P
*
Brightness vs.
No. Par.
No. Par.
Genera
Species
0.738
-0.406
-0.949
0.447
0.050
0.258
0.400
-0.055
0.738
-0.406
0.633
0.447
0.040
0.258
0.400
-0.085
TABLE 11.3
Relationships (r0 between Ecological Variables and Parasite
Prevalence, Number of Parasite Genera and Lizard Species Brightness
Variable
correlation
with
parasite
prevalence
correlation
with
number of
par. gen. r,
Food gathering
Strata
Month of breeding
Length of breeding
Size
Altitude
0.358
0.073
-0.078
0.144
0.340
-0.871*
-0.184
-0.755
-0.099
-0.045
0.054
0.381
* = P < 0.05
P > 0.05 for all other tests.
correlation
with
brightness r,_
-0.103
0.368
0.178
0.245
-0.386
0.357
TABLE 11.4
Relationship (r,) between the Percent of a Parasite Taxon
Infecting a Lizard Species and Lizard Species Brightness
Parasite Taxa
Flagellates
Ciliates
Amoebas
Sporozoans
Platyhelminths
Nematodes
Acanthocephalans
P > 0.05 in all tests.
r
-0.136
0.000
0.014
-0.236
-0.312
0.037
0.314
TABLE II . 5
Variables and Outcomes of Interspecific Tests
of the Hamilton and Zuk (1982) Hypothesis
Study
Host
Taxa
Present study sauria
Present study sauria
Present Study iguanids
Present study
"
Taxonomic
Level
#Spp.
Parasite
Taxa
suborder
26
1 - 7
1
family
6
& 8
It
n.s.
Read &
Weary, 1990
Read &
Harvey, 1989a
Ward, 1989
Watd, 1988
Read, 1987
Hamilton &
Zuk, 1982
passerines
order
Measure of
Parasitic
Infection
Sig.
prevalence
n.s.*
I gen. & spp.
n.s.
prevalence
n.s.
I gen. & spp.
131
hematozoa
prevalence
n.s.
passerines order
114
agnatha/
superclass/ 25
osteichthyes class
agnatha/
superclass/ 24
osteichthyes class
passerines order
227
hematozoa
1 - 7
prevalence
/ gen.
n.s.
sig.
gen.
sig.
passerines
order
1 - 7
hematozoa
prevalence
109
1=amoebas, 2=ciliates, 3=flagellates, 4=sporozoans, 5=platyhelminthes,
6=nematodes, 7=acanthocephala.
* Significant as interpreted by Read & Harvey (1989b).
sig.
sig.
46
CHAPTER THREE
THERMAL PREFERENCES OF RESISTANT AND SUSCEPTIBLE
STRAINS OF BIOMPHALARIA GLABRATA (GASTROPODA) EXPOSED
TO SCHISTOSOMA MANSONI (TREMATODA)
by
Hugh Lefcort
and
Christopher J. Bayne
This paper appeared in
PARASITOLOGY, 1991. 103:357-362
47
ABSTRACT
The thermal preferences of two strains of the snail
Biomphalaria glabrata, one resistant to, and one
susceptible to, the parasite Schistosoma mansoni were
determined in an aquatic thermal gradient.
Snails were
tested without exposure to the parasite, and both two
hours and five weeks after exposure to trematode
miracidia.
The mean temperature selected by susceptible
strain snails two hours post exposure was 0.5 +0.3°C.
lower than that of unexposed controls.
In this strain,
at five weeks post exposure, the preferred temperature
dropped a further 1.9 ±0.5 °C.
The resistant strain
displayed a significant drop of 1.8 +0.6°C two hours
post exposure.
These results are consistent with the
hypothesis that a drop in mean temperatures selected by
snails is due to altered levels of endogenous cytokines
such as IL-1 or TNF in association with parasite
activation of the snail internal defense system.
48
INTRODUCTION
Diseased states, including parasitism, have long been
associated with alterations in animal body temperatures.
In almost all cases, invasion by pathogens result in an
elevation of body temperature (Kluger, 1979).
Fevers
(defined as an elevation of the thermoregulatory set
point above the normal set point), and pyrogenic agents
that elicit fevers, have been correlated with infectious
diseases for over 2000 years (Atkins, 1982).
Fevers in
endothermic birds or mammals generally are achieved
either metabolically or behaviorally.
In ectotherms,
fevers are achieved only behaviorally, e.g. individuals
can seek out and dwell in microhabitats where heat
energy can be obtained.
A lowering of temperature set-points has been less
commonly associated with disease.
However, starvation
often leads to a lowering of set-points (Regal, 1966;
Lillywhite, Licht and Chelgren, 1973) and impaired renal
function is correlated with lowered set-points in
rabbits (Kluger et al., 1981; Eiger and Kluger, 1985).
Kluger et al. (1981) postulated the existence of an
endogenous cryogen that when released into the
bloodstream of vertebrates acts to lower set-points.
this time no such substance has been identified (see
review by Kluger, 1991).
At
49
TEMPERATURE AND HOST/PARASITE GROWTH
The natural aquatic habitats of the snail
Biomphalaria glabrata are thermally heterogeneous,
providing for possible temperature selection.
Small
alterations in ambient temperature have strong effects
on the biology of snails that are host to trematode
parasites of the genus Schistosoma (Shiff, 1964; Shiff
and Husting, 1966; Sturrock and Sturrock, 1972; Sodeman
and Dowda, 1974).
For example, growth rates were
progressively higher for snails reared at 20°C, 25°C and
30°C, and lower for those reared at 35°C (Sturrock and
Sturrock, 1972).
30°C.
Fecundity was highest between 25 and
Indeed, temperatures around 28°C have been found
to be the mean selected temperature of B. olabrata
(Chernin, 1967).
While higher temperatures favor the growth of B.
glabrata, the growth of Schistosoma mansoni in its
molluscan host is also greatest at temperatures above
26°C (Stirewalt, 1954).
It has also been reported that
some snails kept at low temperatures lost their
infections (Stirewalt, 1954; see also Webbe and James,
1972) .
In a test of the thermoregulatory behavior of B.
glabrata, Chernin (1967) found that snails infected with
S. mansoni behaved more erratically and had more
50
lethargic movements than uninfected individuals, but he
found no difference in their thermal preference.
With the knowledge that appropriate behavioral
changes in thermally heterogeneous habitats could affect
the host-parasite balance, and intrigued by Chernin's
(1967) mention of erratic behavior, we have further
explored the effects of S. mansoni on the thermal
behavior of B. glabrata.
Our interest was further
piqued by the report of interleukin-1 (IL-1) and tumor
necrosis factor (TNF) in molluscs (Hughes et al., 1990);
in vertebrates, both of these cytokines are secreted
into the blood during immune responses, and influence
temperature set-points (Dinarello, 1984; Dinarello et
al., 1986; see also Kluger et al., 1990).
IL-1 is
pyrogenic (Dinarello 1984), as is TNF at pharmacological
doses (Dinarello, 1986); TNF is also cryogenic at
physiological doses (Long et al., 1990a; 1990b).
We examined the thermal behavior of snails during the
two periods when their internal defense systems would be
maximally challenged by schistosomes.
For resistant
snails this occurs immediately following miracidial
encounter as the host's encapsulation response leading
to parasite death is initiated (Bayne, 1983).
In
susceptible snails, entering miracidia, and the mother
sporocysts into which they quickly transform, do not
51
elicit an immune response by the host internal defense
system (Bayne, 1983).
However, when the infection is
patent (peak level is approximately at five weeks) and
mature cercariae are leaving this host as their first
step in locating a final (mammalian) host, the
emigration of cercariae can be quite disruptive to the
snail, at least as seen in histological examinations
(Pan, 1963; 1965; Loker, 1979; van der Knaap and Loker,
1990; see also Sturrock and Sturrock 1970; 1971 for
parasite-induced shell abnormalities).
In measuring the
snail's thermal preference, Chernin (1967) either
randomly assigned snails along a thermal gradient, or
placed groups of snails at temperatures well above or
well below their thermal preference.
In contrast, in
our experiments to determine thermal preferences post
exposure, we first determined the thermal preference of
unstressed unexposed snails of the susceptible strain.
Subsequently we placed snails at that temperature, thus
forcing them to choose between warmer or cooler water.
With this protocol we investigated if differences
existed between the thermoregulatory behavior of
a) unexposed snail strains known to be either
susceptible to, or resistant to, S. mansoni PR1 strain
and b) snails of each strain exposed previously to
miracidia of PR1 S. mansoni.
52
METHODS
The PR1 strain of S. mansoni was propagated using
male CF-1 Mus musculus mice and susceptible M-line B.
glabrata snails.
Miracidia were procured from the
livers of mice 6 - 8 weeks after they were exposed to
250 to 300 cercariae (Stibbs et al., 1979).
The two snail strains used in this study were M-line
animals (80% of which are susceptible in our laboratory
to PR1 S. mansoni) and 13-16-R1 animals [90% of which
are resistant to PR1 miracidia (Fryer and Bayne, 1990)].
Histological examination of resistant snails three days
post-exposure reveal no persistent effects of S. mansoni
(Bayne, 1983).
Six treatments were used: 1) two hour unexposed
susceptible strain snails, 2) two hour exposed
susceptibles, 3) five week unexposed susceptibles,
4) five week exposed susceptibles, 5) two hour unexposed
resistants and 6) two hour exposed resistants.
Snails were reared in a 600 L tank containing about
500 individuals.
All snails used had a shell diameter
of 4 - 5 mm and were four weeks old when first used.
Snails were fed ad libitum with romaine lettuce and were
kept at a constant 26°C.
The laboratory was illuminated
between 6:00 and 19:00 PDT.
For exposure to the parasite, individual snails were
53
confined for two hours in 3 ml plastic wells containing
15 miracidia in 2 ml of artificial spring water at 26 °C.
To control for the potential stress of such confinement
(see Fryer, Dykes-Hoberg and Bayne, 1989) all unexposed
animals were placed for two hours, prior to testing, in
identical wells containing only artificial spring water.
After two hours in these small wells both exposed and
sham-exposed snails were either directly tested, or
transferred to 8 L tanks for five weeks at which time
they too were tested.
Prior to testing five-week-
exposed and sham-exposed snails were again placed in 3
ml wells containing only artificial spring water.
Experiments were run in a gray epoxy-painted
container measuring 273 cm (1) x 27 cm (w) x 46 cm (d).
Aged tap water was added to a depth of 1.5 cm.
The warm
end of a thermal gradient was created by placing three
200 Watt immersion coils in a 2.8 liter beaker of water
at one end of the container.
At the opposite end a 15
liter glass jar was filled with ice.
Snails were
isolated from the cold and the hot end of the gradient
by fine mesh screen placed 19 cm from the container
ends.
The resulting length available to the snails was
234 cm. This was divided by faint lines into 24 zones of
9.7 cm length.
Temperatures to the nearest tenth of a
degree were recorded at the center of each zone with a
54
K-type thermocouple connected to an Omega HH82
electronic thermometer.
No vertical stratification of
water temperatures was present.
Air stones at either
end of the container (behind the screens) served to make
an almost linear thermal gradient (Figure III.1).
Temperatures in each zone varied by less than two
degrees Celsius between trials.
All animals within a
given zone were assigned the same temperature.
Experiments were conducted daily between 13:00 and 16:00
hours.
Individual animals were tested only once.
Preliminary experiments confirmed Chernin's (1967)
finding of no difference in thermal preference in
illuminated or dark thermal gradients.
Consequently for
the experiments reported here, the container was evenly
illuminated by fluorescent lighting to an intensity of
525 lux.
Data for the movement experiment were analyzed using
two two-way ANOVAs with significance at P < 0.05.
experimental design was not orthogonal.
The
In order to
avoid a type I error, all ANOVA generated P values were
corrected using the Bonferroni multiple comparison
procedure (Johnson and Wichern, 1988).
The test of
exposure x time involved two hour unexposed susceptible
strain snails, two hour exposed susceptibles, five week
unexposed susceptibles and five week exposed
55
susceptibles.
The test of exposure x resistance
involved two hour unexposed susceptibles, two hour
exposed susceptibles, two hour unexposed resistants and
two hour exposed resistants.
MOVEMENT EXPERIMENT USING SUSCEPTIBLE SNAILS
To identify individual snails, shells were uniquely
marked with three drops of enamel nail polish.
Twenty-
four snails were placed along a line separating zones 12
and 13 at an average temperature of 21.7 +0.4°C [a
temperature below that reported by Chernin (1967) to be
the mean selected temperature].
After 30 minutes the
zonal position of each snail was recorded.
The color
markings were not always apparent so all snails were
individually lifted from the container, their identity
and zonal position recorded, the temperature of the zone
noted, and then the snails were replaced in their
original position.
This process lasted fifteen minutes.
Therefore, the following four readings were taken at 45
minute intervals.
210 minutes.
The total time of the experiment was
Five trials each with 24 unexposed snails
and five trials each with 24 snails five weeks past
exposure to miracidia were conducted.
Snails were not
placed in 3 ml infection wells prior to testing.
56
THERMAL PREFERENCE EXPERIMENT
For each trial ten unmarked snails of a single
experimental group were placed at the previously
determined thermal preference of the unexposed
susceptible strain (zone 20 at an average temperature of
28.7 +0.4°C).
One hour later their positions and the
temperature of each zone were recorded and a mean
temperature of the group was derived.
57
RESULTS
CONTROLS
Lengthwise, in the absence of a thermal gradient, no
preference was noted for one half of the tank over the
other half (X2 = 0.78, d.f. = 1, P = 0.37).
The stress of confinement in the 3 ml infection wells
for two hours could have influenced thermoregulatory
behavior.
Uninfected snails confined for two hours in
the infection wells tended to select temperatures
slightly, but not statistically, lower than unexposed
snails taken from a 600 liter tank (31.0 +0.5 and
31.5 +0.6°C respectively, P = 0.103).
As a result of
this, confinement regimens were carefully controlled for
in all further experiments.
In contrast, the mean temperature selected by snails
confined for five weeks to 8 L rearing tanks, followed
by two hours in the 3 ml infection wells, was not
significantly different from that of snails from the 600
L tank (Tukey test, P > 0.050).
MOVEMENT EXPERIMENT
Of the 192 individuals tested, all but 11 snails
moved from zone 12/13 (21.7°C) towards zones containing
warmer water.
At the readings taken at 30, 75, 120,
165, and 210 minutes, the snails harboring five week
58
infections were consistently located at lower (but not
statistically so) temperatures than those of unexposed
snails (Figure 111.2).
The total movement of infected
snails (13.2 ±0.6 zones) was less than that of unexposed
snails [18.3 ±0.8 zones (P = 0.008)].
Since the mean
temperatures at which unexposed and infected snails were
located were not significantly different (P > 0.5 at all
readings), this indicates that infected snails proceeded
more directly to preferred temperature regions while
unexposed animals moved back and forth from warmer and
cooler waters.
THERMAL PREFERENCE EXPERIMENT
Five weeks after exposure it was determined that
greater than 80% of the exposed susceptible strain
snails developed patent infections. The mean temperature
of each experimental group at the end of one hour is
illustrated in Figure 111.3.
The mean temperature selected by susceptible strain
snails two hours post exposure was 0.5 +0.3°C. lower
than that of unexposed controls.
In this strain, at
five weeks post exposure, the preferred temperature
dropped a further 1.9 +0.5°C.
The resistant strain
displayed a significant drop of 1.8 +0.6°C two hours
post exposure.
59
Exposure to the parasite for two hours resulted in a
lowering of selected temperatures for both the
susceptible and the resistant strain (two-way ANOVA,
F = 7.012, d.f. = 1, P = 0.032).
Resistant snails
consistently selected lower temperatures than
susceptible snails; both in the presence and the absence
of the parasite (two-way ANOVA, F = 6.941, d.f. = 1,
P = 0.036).
The temperature, that two hour exposed
susceptibles selected, was lower than that of the
baseline temperature of unexposed resistant snails.
When comparing two hour exposure versus five week
exposure, we found that exposure to the parasite lead to
lower temperatures in susceptible snails at both two
hours and five weeks (two-way ANOVA, F = 6.316,
d.f. = 1, P = 0.034).
The magnitude of the changes in
temperature was no different at two hours post-exposure
versus five weeks post-exposure (two-way ANOVA,
F = 0.072, d.f. = 1, P = 0.793)
60
DISCUSSION
Our results imply that molluscs regulate body
temperature as part of their response to parasitic
infection.
The selection of lower temperatures at times
of immunological response was unexpected, and the
significance of this direction of change warrants
evaluation.
The snail Lymnaea stagnalis appressa has an increased
metabolic rate when infected by trematodes (Hurst and
Walker, 1935) and B. glabrata infected with S. mansoni
have higher heart rates than noninfected snails (Lee and
Cheng, 1971).
The possibility exists that infected
snails move to lower temperatures to shed excess heat.
However, the high thermal conductivity of water reduces
the likelihood that changes of the observed magnitude
would be required to achieve this.
We therefore propose
that the movement of snails to lower temperatures is an
evolutionary strategy of snails to lessen the impact of
parasitism.
There are data from other snail/parasite systems
suggesting that lower temperatures aid the survival of
the host relative to the parasite (Webbe and James,
1972; Blankespoor et al., 1989).
Using the same snail
and parasite as we used, Stirewalt (1954) found that
lower temperatures (23 - 25°C) favored the survival of
61
infected snails, partly because they lost their
infections.
However, the optimal temperature range for
B. glabrata infected with S. mansoni appears to be
relatively narrow and above 25°C.
Mortality of the
snail increases below 26°C and above 30°C (Standen,
1952).
Therefore, while an infected B. glabrata may
lessen the impact of cercarial shedding by seeking
waters cooler than 26°C, this would decrease the
survival of the snail.
Much work has been done towards finding mechanisms
that raise thermoregulatory set-point (Kluger, 1979),
yet little has been done to find mechanisms that result
in a lowering of set-points.
Two endogenous cryogens
have been proposed: Glucocorticoids (Besedovsky et al.,
1986) and TNF (Long et al., 1990a; but see also LeMay et
al., 1990).
It is possible that the lower thermal
preference of the resistant strain is due to chronic low
level releases of endogenous cryogens.
The marked drop
in temperature preference (1.8°C) in resistant snails
after exposure to miracidia may be caused by larger
releases at times of immunological stress (see Kluger et
al., 1990). Selection by unexposed resistant snails of
a temperature that is lower by 1.5°C than that of
unexposed susceptible snails, could, in thermally
heterogeneous habitats, contribute to protection against
62
S. mansoni.
Our results are consistent with the hypothesis that
behavioral changes which take certain snails into cooler
water are mediated by factors released into the
hemolymph, and with the postulate that such factors may
be cytokines or stress hormones of host origin.
This
surprising drop in temperature preference following
infection may be beneficial to the host (Stirewalt,
1954).
Since most animals exhibit an increase in
thermoregulatory set-point following infection (Kluger,
1979), this snail/trematode model may prove to be an
excellent system in which to search for the hypothetical
(Kluger et al., 1981), yet still undiscovered,
endogenous cryogen.
63
ACKNOWLEDGEMENTS
We are grateful to Steven Eiger and Matthew Kluger
for extensive help in both formulation of concepts and
interpretation of results.
The manuscript was greatly
improved by the critical readings of Andrew Blaustein,
Steven Eiger, Sarah Fryer, Matthew Kluger, Frances
Lefcort, R. F. Sturrock and an anonymous reviewer.
We
thank Kattarina Matteus for help with preliminary
experiments, and Maureen Larson and Jane Menino for care
of experimental animals.
This study is dedicated to the
memory of Dr. Marshall Hurlich who introduced H. L. to
the wondrous world of environmental physiology.
Financial support from NIH grant AI-16137 to C.J.B. and
a Sigma Xi Grant-In-Aid-Of-Research to H.L. are
gratefully acknowledged.
64
Figure II1.1
Mean temperatures measured at the centers of
zones within the thermal gradient.
40
CD
35
CD
0
Q)
Ld
30
I-CC
LLI
0
I
I
LU
25
CC
L LI
._,.<
20
15
2
4
6
8
10
12
14
ZONES
FIGURE III . 1
16
18
20
22
24
66
Figure 111.2
Mean temperatures ( ±S.E.) of uninfected
susceptible snails (n = 96) and five week infected
susceptible snails (n = 96) over 210 minutes.
were placed at 21.7°C.
The snails
30 -
26
5 Week Infected Susceptible
Unexposed Susceptible
o
24
22
0
30
60
90
120
TIME (minutes)
FIGURE III . 2
150
180
210
68
Figure 111.3
Mean temperatures ( ±S.E.) at which snails
were located one hour after placement at 28.7°C in a
thermal gradient.
All snails were placed individually in
identical infection wells two hours prior to testing.
Susceptible strain snails were either not exposed to S.
mansoni miracidia (unexp), unexposed and then reared for
five weeks in 8 L tanks (5wk unex), exposed for two hours
(2hr exp) or exposed for two hours and then reared for
five weeks in 8 L tanks (5wk inf).
Resistant strain
snails were either not exposed (unexp) or exposed for two
hours (2hr exp).
32
SUSCEPTIBLE
RESISTANT
31
30
29
28
27
26
UNEXP
5WK UNEX
2HR EXP
5WK INF
EXPERIMENTAL GROUPS
FIGURE III . 3
UNEXP
2HR EXP
70
CHAPTER FOUR
ANTIPREDATORY BEHAVIOR OF FEVERISH TADPOLES:
IMPLICATIONS FOR PATHOGEN TRANSMISSION
by
Hugh Lefcort
and
Steven M. Eiger
71
ABSTRACT
In this paper we propose the hypothesis that
pathogen-induced host defense responses result in
altered host behaviors and enhanced predation.
In
particular we examine the effects of the acute phase
response (whose effects include fever, reduced activity
and malaise) on antipredatory behavior in bullfrog
catesbeiana) tadpoles.
(Rana
This host response is associated
with the preliminary stages of infection with many
pathogens yet its behavioral effects have received
little attention.
Bullfrog tadpoles were injected with
alcohol-killed bacteria to induce a response to
infection and their ability to detect and avoid capture
by predatory salamanders (Taricha granulosa) was
explored.
We predicted that acute phase responses
increase tadpole vulnerability to predation by
influencing thermoregulatory behavior and their ability
to detect, and avoid capture by, salamanders.
We found that the stereotypical effects of the acute
phase response can lead to increased predation.
Malaise
affected the refuge seeking behavior of the tadpoles in
the presence of salamanders.
We suggest that for
tadpoles provided with refuges, altered behaviors are a
liability.
This endogenous response may afford some
parasites a potential pathway to their next host.
72
INTRODUCTION
Pathogens, including viruses, bacteria and parasites,
have long been known to affect the behavior of their
hosts (Bethel and Holmes, 1972; 1977; Camp and Huizanga,
1979; Brassard et al., 1982; Hay and Hutchison, 1983;
Rau, 1983; Moore, 1984; Quinn et al., 1987; Lefcort and
Bayne, 1991).
Altered behavior due to damaged host
physiology and neurology have been documented (Livesey,
1980; Minchella, 1985; Helluy and Holmes, 1990; Holmes
and Zohar, 1990; Thompson, 1990; Kavaliers and Colwell,
1992).
However, few studies have examined the
behavioral effects of endogenous substances (e.g.
proteins such as the interleukins) released into the
bloodstream by infected, but otherwise unimpaired,
animals.
Because this response occurs in most
vertebrates when infected with pathogens, some pathogens
may benefit from behavioral changes associated with host
sickness, such as behavioral fever and "malaise".
Traditionally, nonmemory host defense responses
(e.g. the acute phase response, which typically results
in fever, perturbations of trace metal concentrations in
blood, and the release of cytokines such as the
interleukins), have been viewed as acting as protection
against pathogens and not as mechanisms that enhance
pathogen success.
For example, Kluger et al.
(1975)
73
showed that the lizard Dipsosaurus dorsalis, placed in a
thermal gradient, chose a higher ambient temperature
(a behavioral fever) after being injected with the
bacterium Aeromonas hydrophila.
Infected lizards
prevented from raising their body temperature had
decreased survival relative to lizards which maintained
high temperatures post-injection.
Thus, increased body
temperature was deleterious to the pathogen.
However, some host defense responses facilitate
parasite transmission (Hayunga, 1979; Shinn, 1985a;
1985b; Soulsby, 1987; Kluger, 1990).
Damian's (1987)
prediction that manipulations of host defenses would
either aid pathogen reproduction within the host, or
transmission of the pathogen to a subsequent host (see
also Ewald, 1986; 1991; Williams and Nesse, 1990), has
had some support.
For example, some tapeworms employ
the intestinal granulomatous (i.e. encapsulation)
response of their fish host to ensure a protected
environment and a means of attachment (Hayunga, 1979).
Similarly, turbellarian flatworms occupying the coelom
of sea cucumbers may rely on physiological defense
mechanisms of the host to ensure transmission (Shinn,
1985a; 1985b).
Furthermore, schistosome egg expulsion
from the mouse gut is facilitated by the immune response
of the host (Mus musculus, Doenhoff et al., 1985).
In
74
these examples, pathogens receive fitness benefits from
the physiological effects of an active host defense
response.
Additionally, altered host behavior facilitate the
transmission of multi-host parasites to secondary hosts
by either enhancing predation of primary hosts, or
increasing accessibility of hosts to vector (Rossignol
et al., 1984; Rossignol and Rossignol, 1988;
Kwiatkowski, 1989; Kwiatkowski and Greenwood, 1989).
After infection with acanthocephalan parasites, isopods,
(Armadillidium vulgare) are consistently more
susceptible to predation by starlings (Moore, 1983).
Lemmings infected with the protozoan Sarcocystis
rauschorum have increased exploratory activity which can
increase their susceptibility to avian predators (Quinn
et al., 1987; see also Dolinsky et al., 1985).
Predation of infected hosts could be increased if
malaise, a symptom of the host defense response,
resulted in a reduced fright response.
For example,
minnows observed to be "sick" show little or no reaction
to alarm substances from conspecifics (von Frisch, 1941)
and stickleback fish parasitized by a pseudophyllid
cestode recover more quickly from a frightening overhead
stimulus than do non-parasitized fish (Giles, 1983;
Godin and Sproul, 1988).
Malaise may also alter refuge
75
seeking behavior.
Many uninfected amphibians and fishes
respond to predators by increasing refuge use and
decreasing activity (Sih, 1987; Sih and Kats, 1991;
Wilson and Lefcort, In Press; reviewed by Lima and Dill,
1990)
.
In this paper we propose the hypothesis that
pathogen-induced host defense responses lead to altered
host behavior that enhance predation.
In particular we
examine the effects on antipredator behavior of acute
phase responses in bullfrog (Rana catesbeiana) tadpoles.
We predict that acute phase responses increase tadpole
vulnerability to predation by influencing their ability
to detect, and avoid capture by, predatory salamanders
(the roughskin newt, Taricha granulosa).
76
METHODS
Sympatric bullfrogs and rough-skinned newts were
collected from two field sites: bullfrog tadpoles were
collected at Pisgah Pond (Linn County, Oregon), and
newts were collected at Soap Creek Pond (Benton County,
Oregon).
Bullfrogs are preyed upon by newts at these
sites (pers. obs.).
Tadpoles (Gosner stages
23 - 25,
i.e. without leg buds, [Gosner, 1960] mass 0.8 - 1.2 g)
were housed at Oregon State University, fed alfalfa
pellets ad libitum, and maintained between 18 and 22°C.
The laboratory was illuminated between 06:00 - 19:00
hours.
Newts were houses under similar conditions and
fed midge fly larvae (Chirononmus spp.).
We conducted laboratory experiments between June and
October, 1991, and field experiments in September and
October, 1991.
Experiments were conducted between 13:00
- 17:00 hours.
We tested individual tadpoles only once.
Aeromonas hydrophila, a naturally occurring gram
negative bacterium, was chosen to experimentally induce
the acute phase response in bullfrog tadpoles (sensu
Casterlin and Reynolds, 1977).
This bacterium causes
the often fatal "red-leg" disease in anurans (Gibbs,
1963).
A. hydrophila is not transmitted through
predation from bullfrog tadpoles to T. granulosa
(Lefcort, Unpub.).
To test our hypothesis, the acute
77
phase response's effect on behavior must be isolated
from physical or neurological damage caused by a
pathogen or by the products and toxins a pathogen
releases.
We used alcohol-killed bacteria to induce
fever in hosts, rather than a live pathogen, to separate
the behavioral effects of the acute phase response from
the myriad of physiological changes and tissue damage
that may occur during infections with living pathogens.
Lyophilized bacteria (American Type Culture
Collection) were grown in nutrient broth, then killed by
suspension in ethanol, washed in sterile pyrogen-free
saline (0.9% sodium chloride), centrifuged, and
resuspended in saline.
agar produced no growth.
Plating of bacteria on blood
A concentration of
1 x 1010
dead bacteria/ml was assessed by its turbidity (Difco
Laboratories; Kluger, 1977).
In all experiments,
experimental tadpoles were injected intraparitonealy
with 0.1 ml of this solution to induce the acute phase
response.
Control tadpoles were injected with 0.1 ml of
pyrogen-free saline.
To test for significant treatment effects we used
Wilcoxin rank sum tests (a nonparametric test), whenever
a violation of the assumption of normality prevented the
application of Student's t-test and ANOVA parametric
78
tests.
Statistical significance of all tests was
P < 0.05.
I. LABORATORY THERMAL PREFERENCE EXPERIMENT
One indication of the acute phase response in
ectotherms is behavioral fever - i.e. altered
thermoregulatory behavior (Kluger, 1990).
We determined
the thermal preferences of control and bacteria-injected
tadpoles in the laboratory.
We conducted experiments in
a gray epoxy-painted container measuring 273 cm (1) x 27
cm (w) x 46 cm (d).
We divided the container lengthwise
into two lanes by a dark plexiglass partition.
dechlorinated tap water to a depth of 1.5 cm.
We added
We
created the warm end of a thermal gradient by placing
two 200 W immersion coils in a 2.8 liter beaker of water
at one end of the container.
A compressor driven
cooling coil was submerged at the opposite end in a 5.0
liter beaker of water.
Test-lanes were isolated from
the extreme cold and the hot ends of the gradient by
fine mesh screen placed 19 cm from the container ends.
The resulting lane length of 234 cm was divided
widthwise by faint pencil lines into 24 zones of 9.7 cm
length.
We recorded temperatures to the nearest tenth
of a degree at the center of each zone with a K-type
thermocouple connected to an Omega HH82 electronic
79
thermometer.
By keeping the water depth low, no
vertical stratification of water temperatures was
detected.
Air stones at each end of the container
(behind the screens) provided continuous water flow and
an essentially linear thermal gradient.
each zone varied by < 2°C between trials.
Temperature in
To determine
the instantaneous temperature of a tadpole, a tadpole
was assigned the temperature of the zone it occupied.
The container was evenly illuminated by diffuse
incandescent lighting to an intensity of 320 lux.
We introduced a single tadpole into the central zone
(zone 12, mean temperature = 19.9 +0.6°C) of each lane,
and then left them undisturbed for 24 hours.
During
hour 25, we recorded the positions of the tadpoles every
five minutes.
We then injected the tadpoles with either
dead A. hvdrophila (to induce fever) or sterile pyrogenfree saline (control treatment), and we returned the
tadpoles to the zones from which they were found.
the next two hours we recorded the positions of the
tadpoles every five minutes.
Lane assignments of
tadpoles (experimental vs. control treatment) were
alternated between runs.
II. FIELD THERMAL PREFERENCE EXPERIMENT
To determine if bacteria injections also induced
For
80
altered thermal preferences in the field, we conducted
the following experiment.
Soap Creek Pond, Oregon.
Experiments were conducted at
We placed 12 tadpoles in each
of 12 enclosures (the experimental design was similar to
Walls, 1991) constructed of 5 cm wide plywood strips,
composing a frame covered on three sides with fiberglass
mesh insect screening (mesh size = 1 mm).
Enclosures
measured 137 x 30 x 30 cm and were placed lengthwise,
along the natural slope of the pond bottom.
The
uncovered top of the enclosure extended above the
surface of the water.
Enclosures were placed 30 cm
apart and were parallel to each other and perpendicular
to the pond's edge.
Due to the slope of the pond, the
end farthest from the pond's edge was situated in "deep"
water (20 cm) and the opposite end of the enclosure was
situated in "shallow" water (2 cm).
Mean temperatures in the enclosures were
:
23.7
+0.12°C at the shallow end, 21.8 +0.12°C at the deep end,
and 23.4 +0.14°C 1 cm below the surface of the deep end
(air temperature = 22.0°C).
Temperatures were measured
only at the end of the experiment because preliminary
results indicated that wading into the pond to record
temperatures stirred-up pond silt, agitated the tadpoles
and disturbed any existing thermal gradient.
Temperatures at other areas of the pond increased by
81
less than 0.3 °C over the 120 minutes of the experiment.
We placed tadpoles to be used in field experiments in a
"holding" enclosure identical to the 12 experimental
enclosures and allowed them to habituate to the pond for
24 h.
At the start of the experiment, we injected 12
tadpoles with either alcohol-killed A. hydrophila
bacteria or saline (control treatment) solution and
added them to the center of each enclosure.
Treatments
were interspersed to control for a gradient of
temperatures along the pond margin.
After 60 minutes
(the time at which maximum fevers were observed in the
laboratory), the number of animals in both the shallow
half of the enclosure and top 1 cm of the deep half were
recorded by observers using binoculars at a distance of
4 m.
Sixty minutes later (i.e. 120 minutes post-
injection, after laboratory fevers subsided), the
positions of the tadpoles were again recorded and
temperatures were recorded at the geometric center of
the shallow end, the geometric center of the deep end,
and the water 1 cm below the surface of the deep end.
We then terminated the experiment and counted the number
of tadpoles in each container to determine if any had
escaped.
The mean temperature of the tadpoles within a
given enclosure was estimated by assigning to the
82
animals within one of the three zones the temperature of
that zone.
III. FIELD EXPERIMENT TO TEST THE EFFECT OF SALAMANDERS
ON THERMOREGULATION
This experiment tested how the presence of a
predatory salamander affected behavioral
thermoregulation of bacteria-injected tadpoles.
Prior
to the experiment ten tadpoles were allowed to acclimate
in a holding enclosure for 24 hours.
Each of the 12
enclosures contained 10 tadpoles and received one of the
following four treatments:
a two by two design had all
combinations of tadpoles, injected with either bacteria
or saline, crossed with salamanders either present or
absent.
The salamander was placed in a container
constructed of a plastic frame with mesh sides (15 x 15
x 8 cm).
This container was attached to the geometric
center of each enclosure.
Empty containers were placed
in enclosures whose treatment did not include a
salamander.
We carried out observations as in the
previous experiment.
Mean temperatures in the
enclosures were: 17.15 +0.13°C at the shallow end, 13.92
+0.15°C at the deep end, and 17.09 +0.14 °C 1 cm below the
surface of the deep end (air temperature = 19.6°C).
83
IV. LABORATORY EXPERIMENT TO TEST THE EFFECT OF REFUGES
ON PREDATION BY SALAMANDERS
To determine if bacteria injections affect the use of
refuges, we conducted the following experiment.
Twenty-
four hours before testing we dyed the tadpoles either
blue (methylene blue, EM Science Co.) or red (neutral
red, Matheson, Coleman & Bell Co).
Dyes were used for
identification and colors were reversed on alternate
trials.
We placed tadpoles in seven-liter opaque bowls,
filled with five liters of water.
Two treatments were
used: half of the bowls contained plastic plant refugia
and half did not.
We added four bacteria-injected and
four saline-injected tadpoles to each bowl.
We added a
starved (24 hours) salamander after 40 minutes.
After a
further 60 minutes we recorded the number and color of
the remaining tadpoles.
We tested a total of 320
tadpoles.
V. PREDATOR-CUE EXPERIMENT
This experiment tested the behavioral response of
tadpoles to chemical cues present in water that had been
in contact with predatory salamanders.
The experimental
apparatus was a modification of that used by Petranka et
al. (1987).
We tested tadpoles in a gravitational flow-
though system composed of three 25 liter plastic tubs
84
(51 x 37 x 21 cm deep).
The tubs were arrayed at
different heights so that water flowed from one to
another at 0.5 1/min.
23 liters of water.
The uppermost tub was filled with
In half the trials, the middle tub
contained a salamander, from which water flowed to the
lower-most tub, which contained eight tadpoles and was
divided lengthwise such that half the tub contained
plastic aquarium plants and half did not.
The two
lower-most tubs had output openings and never contained
more than 10 liters of water.
Ten minutes after flow
was initiated, we recorded the tadpoles' positions every
minute for ten minutes.
We recorded the number of
tadpoles in each half of the enclosure, along with the
number of times a tadpole crossed between the two halves
of the enclosure (i.e. moves per trial).
Four
experimental treatments were used: 1) saline-injected
tadpoles, salamander absent, 2) saline-injected
tadpoles, salamander present, 3) bacteria-injected
tadpoles, salamander absent, 4) bacteria-injected
tadpoles, salamander present.
Preliminary results indicated that handling and
injecting animals had no significant affect on the
distribution of tadpoles.
85
RESULTS
I. LABORATORY THERMAL PREFERENCE EXPERIMENT
From 0 to 60 minutes prior to injection, the mean
water temperature ( ±S.E.) of both bacteria
(experimental) and saline-injected (control) animals was
21.6 +0.71°C (n=31 tadpoles).
Control animals (n=13)
did not exhibit a rise in temperature after injection
(Wilcoxin rank sum test, P >0.05).
Experimental animals
(n=18) exhibited a mean rise in temperature (in the
zones they occupied) of 2.8 +0.90 °C between 30 and 90
minutes after injection (mean rise was calculated from
the difference of each animal from its own preinjection
mean, Wilcoxin rank sum test, P < 0.05; Figure IV.1).
This compares closely to a similar rise of 2.6 - 2.7°C
for bullfrog tadpoles found by Casterlin and Reynolds
(1977) using a similar dose of the same bacterium.
We
found that both experimental and control tadpoles
selected lower temperatures than those reported by
Casterlin and Reynolds (1977).
This difference may have
been due to either the acclimation of tadpoles to our
laboratory conditions affecting their thermoregulatory
set-point, or to Casterlin and Reynolds' use of a
"shuttle box".
86
II. FIELD THERMAL PREFERENCE EXPERIMENT
Sixty minutes after injection, experimental animals
(n=6 enclosures) were present in water with a mean
temperature of 23.2 +0.23°C and control animals (n=6
enclosures) were in deeper water at a cooler mean
temperature of 22.4 +0.14°C.
After 120 minutes post-
injection, experimental and control animals had similar
temperatures, 22.6 +0.17°C vs. 22.4 +0.17°C.
There was a
significant interaction term between type of injection
and the amount of time post-injection (repeated-measure
ANOVA, F = 7.184, d.f.= 1, P = 0.014).
This suggests
that experimental animals were able to "clear" the
bacteria from their systems after 120 minutes and had
returned to control temperatures.
III. FIELD EXPERIMENT TO TEST THE EFFECT OF SALAMANDERS
ON THERMOREGULATION
Sixty minutes after injection, experimental tadpoles
were found at higher temperatures than control tadpoles,
both in the presence and absence of salamanders (ANOVA,
F = 29.98, d.f.= 1, P = 0.006, Table IV.1).
The
presence of salamanders resulted in lower temperatures
for both experimental and control tadpoles (ANOVA, F =
12.63, d.f.= 1, P = 0.008).
Control and experimental
tadpoles responded to the presence of salamanders in
87
different ways.
In the presence of the salamander,
experimental animals were found at temperatures 1.9°C
higher than control animals (Student's t-test,
P = 0.002).
Control tadpoles, in the absence of the
salamander, were found at temperatures 1.4 °C higher than
control tadpoles in the presence of a salamander
(Student's t-test, P = 0.006).
Experimental tadpoles
were also found at slightly higher temperatures in the
absence, than in the presence, of salamanders (0.6°C),
but this difference was not statistically significant
(Student's t-test, P = 0.656).
One-hundred and twenty minutes post-injection the
effect of injection was no longer significant (ANOVA,
F = 2.05, d.f.= 1, P = 0.190) but tadpoles in the
presence of a salamander (both treatment groups examined
together) were still found at lower temperatures (ANOVA,
F = 20.85, d.f.= 1, P = 0.018).
When the readings at 60 and 120 minutes are analyzed
together, bacterial injections resulted in higher
selected temperatures (repeated-measure ANOVA,
F = 20.47, d.f.= 1, P = 0.002) and the presence of
salamanders resulted in lower selected temperatures
(repeated-measure ANOVA, F = 23.02, d.f.= 1, P = 0.001,
Figure IV.2).
There were no significant interaction
88
terms (repeated-measure ANOVA,
F = 1.38, d.f.= 1,
P = 0.275).
IV. LABORATORY EXPERIMENT TO TEST THE EFFECT OF REFUGES
ON PREDATION BY SALAMANDERS
The presence of plant refuges reduced the number of
tadpoles eaten by salamanders, regardless of whether the
tadpoles had been injected with bacteria or saline.
When plant refuges were absent, salamanders consumed a
mean of 1.56 +0.22 total (bacteria plus saline) tadpoles
per trial.
When plants were present salamanders
consumed significantly fewer tadpoles per trail (0.83
+0.18, Student's t-test, P = 0.032).
When no plants were provided, the mean number of
experimental tadpoles consumed (0.55 +0.16) tended to be
lower, but was not significantly so, than the mean
number of control tadpoles consumed (1.00 +0.21;
Student's t-test, P = 0.22).
In contrast, when refuges
were provided, significantly more experimental than
control tadpoles were consumed (0.67 +0.16 and
0.07 +0.17 respectively; Student's t-test, P = 0.022).
V. PREDATOR CUE EXPERIMENT
A summary of the differences in the distributions of
animals in the lower-most tub is given in Table IV.2.
89
No significant treatment effects were found for refuge
use (% of tadpoles in open vs % among plastic plants,
ANOVA, F = 2.64, d.f. = 3, P = 0.073).
Control animals
used refuges in greater frequency than all other
treatments but this difference was not significant.
Significant treatment effects were found for mobility
effects (total number of times tadpoles crossed from one
half of the tub to the other half, ANOVA, F = 17.36,
d.f. = 3, P < 0.001).
All experimental treatment groups
exhibited less movement than control groups.
In
contrast to experimental animals, control tadpoles
exhibited significantly less mobility in the presence of
salamander exposed water.
Mobility of experimental
animals was not significantly altered by the presence of
salamander exposed water.
90
DISCUSSION
We found that the stereotypical effects of a
tadpole's physiological responses to infection
(indicated by the presence of behavioral fever, expts. I
& II) can affect predation.
This response reduced
activity and the refuge seeking behavior of tadpoles in
the presence of a predator (expts. III, IV and V).
In
both the laboratory and the field, bacteria-injected
experimental tadpoles did not seek shelter as actively
as saline-injected controls.
Control tadpoles reduced their activity when exposed
to water that had been in contact with predatory
salamanders, but experimental tadpoles did not.
While
some behavioral alterations may be beneficial to hosts,
in that lethargic hosts conserve resources (Hart, 1988;
1990), we suggest that behavioral alterations can be
costly to hosts because they may lead to higher
predation.
When plant refuges were available,
salamanders consumed a greater number of experimental
than control tadpoles.
Behavioral fevers in an
ectotherm can have wide reaching effects.
Thermoregulation may take precedence over predator
avoidance.
While control tadpoles avoided warm water
when a predator was present, experimental tadpoles did
91
not exhibit this response.
Physiological fevers have
been documented in a wide range of mammals and birds
(D'Alecy and Kluger, 1975; Atkins, 1982) and behavioral
fevers have also been found in many species of reptiles,
amphibians, fishes and arthropods (Kluger, 1990).
Therefore, maladaptive behaviors associated with a host
defense response may be widespread among diverse taxa.
Our findings with bacteria are consistent with results
we have obtained using red-legged frog tadpoles (Rana
aurora infected with live yeast Candida spp., a singlehost pathogen, Lefcort, unpub.).
Given the dramatic effects of a host defense response
on behavior we suggest that some multi-host parasites
may capitalize on normal host defense responses by
eliciting acute phase responses at key periods of the
parasite's lifecycle and then benefit from the resulting
behavioral changes of the host.
A large number of
acute phase responses are triggered by pathways
dependent on a common mediator, such as interleukin-1
(IL-1).
A pathogen may benefit from some of these
responses, whereas other responses may remain
deleterious to the pathogen.
Recently it was discovered
that IL-1 enhances the growth of virulent strains of the
bacterium Escherichia coli (Porat et al., 1991).
If the
net effect of the host defense response is harmful to
92
the invader, then a strategy of avoidance of the host
defense system (Damian, 1964; Soulsby, 1987) would be
expected.
However, if certain elements of the defense
response are helpful to the reproductive efficiency of
the pathogen, then we would expect detection to occur.
In fact, as pathogens undergo changes within their
hosts, and their needs change, we expect that the
various life stages might alternate between enhanced and
suppressed recognition by the host's defense system.
Currently we are exploring if altered behaviors also
occur in a system where tadpoles are the intermediate
host for a parasite that is transmitted through
predation.
93
ACKNOWLEDGEMENTS
Susan Walls provided extensive help in the formation
and analyses of the experiments.
The critical readings
of E. Berlow, A. Blaustein, M. Hixon, R. Huey, M.
Kluger, A. Kuris, F. Lefcort, D. Olson, P. Rossignol,
S.
Walls and H. Wilbur considerably improved earlier drafts
of the manuscript.
D. Wilson assisted both in the field
and laboratory. D. Ingle and K. Hoff provided advice on
anuran anti-predator behavior.
Financial assistance to
H. L. was provided by an Oregon State Univ. Zoology
Dept. grant.
TABLE IV.1
MEAN TEMPERATURE
( °C ±S.E.)
OF TADPOLES 60 and 120 MINUTES POST-INJECTION
All adjacent pairs of means are significantly different
except those labelled "ns" (see text).
60 MINUTES
120 MINUTES
SALAMANDERS
SALAMANDERS
ABSENT
SALINE
INJECTION
BACTERIA
16.23 +0.01
ns
17.30 +0.48
*
PRESENT
ABSENT
14.86 +0.19
16.36 +0.08
*
ns 16.75 +0.14
ns
16.60 +0.35
PRESENT
*
15.40 +0.08
ns
ns 15.73 +0.15
TABLE IV.2
RESULTS OF PREDATOR-CUE EXPERIMENT
Means that are not significantly different at P < 0.05 are identified with
similar letters (Tukey test)
TREATMENT
% OUTSIDE
REFUGES(MEAN ±S.E.)
Bacteria-injected,
salamander exposed water
n = 8 trials
51.5 +4.9 a
Bacteria-injected,
control water
n = 6 trials
46.5
Saline-injected,
salamander exposed water
n = 8 trials
35.4 +3.3 a
25.3 ±3.1 c
Saline-injected,
control water
n = 6 trials
47.5 ±2.3 a
37.7 ±3.5 d
a
MOVES/TRIAL
9.0 ±1.8 b
13.3
b
96
Figure IV.1
Mean change in body temperature +S.E. minus the mean
body temperature during of the period 60 minutes preinjection of alcohol-killed bacteria-injected tadpoles
(n-18) and saline-injected tadpoles (n=13) from 60
minutes before injection (at time 0) and 120 minutes
post-injection.
Saline
o
TIME
(minutes)
FIGURE IV.1
Bacteria
98
Figure IV.2
Mean temperature +S.E. of tadpoles 60 minutes after
injection of alcohol-killed bacteria or saline in either
the presence or absence of salamanders.
represents 3 enclosures of 10 tadpoles.
Each treatment
SALII\ E
BACTERIA
18
17
16
15
14
0
ABSENT
PRESENT
ABSENT
PRESENT
EXPERIMENTAL GROUPS
FIGURE IV.2
100
CHAPTER FIVE
FEVER AND PREDATOR AVOIDANCE OF PARASITIZED TADPOLES
by
Hugh Lefcort
and
Andrew R. Blaustein
101
ABSTRACT
Many parasites alter the behaviors of their hosts.
In most cases that have been studied, predation of the
intermediate host by the definitive host is necessary
for parasite transmission.
Single-host parasites have
been studied less frequently and the fitness benefits
from altering host behaviors are not obvious.
Parasitic
infection in vertebrates usually results in a
physiological response to infection.
The behavioral
effects of this response have similarly received little
attention.
If altered behaviors are a generalized host
response to infection, irrespective of the infecting
parasite's life history, i.e. single-host or multiplehost species, then alterations of host behavior may be a
general phenomenon of infection.
In this paper, we
examine the behavioral effects of a yeast, Candida spp.,
a single-host parasite species in its natural host, the
red-legged frog (Rana aurora).
Infected tadpoles were
found more frequently in warm shallow water, exhibited
reduced movement, failed to detect chemical cues from
predatory newts (Taricha aranulosa), and suffered
increased predation by newts, compared to uninfected
controls.
102
INTRODUCTION
Some parasites alter the behavior of their animal
hosts in ways that increase the host's susceptibility to
predation (e.g. Bethel and Holmes, 1972; Holmes and
Zohar, 1990; Thompson, 1990).
Many of these parasites
exhibit multiple host life-cycles that require the
definitive host to consume the intermediate host for
transmission to occur ( Bethel and Holmes 1977; Brassard
et al., 1982; Camp and Huizanga 1979; Giles, 1983;
Lester, 1971; Moore, 1984; Quinn et al., 1987).
However, the behavioral effects of single-host parasites
has received less attention (Baer, 1973; LaMunyon and
Eisner, 1990).
Single-host parasites often die when their host dies.
Thus, single-host parasites may gain no obvious
selective advantage from inducing host behaviors that
result in the death of their host.
One mechanism by
which pathogens alter host behavior is by causing their
host to release substances into the bloodstream as part
of the infection response (Kavaliers and Colwell, 1992;
Lefcort and Eiger, unpub.).
When a vertebrate host
becomes infected, an initial host defense response
results that involves inflammation, increased levels of
stress hormones, cytokines (reviewed by Stadnyk and
Gauldie, 1991), and a subsequent immune response.
Some
103
cytokines have behavioral effects.
For example,
interleukin-1 (IL-1) and tumor necrosis factor (TNF) can
cause ectothermic vertebrates to behaviorally
thermoregulate (i.e. seek out and dwell in warm
microhabitats) at higher (feverish) temperatures
(Kluger, 1991).
Other cytokines causes fatigue and
induce sleep in mammals (Shohan et al., 1987; Spiggs et
al., 1988).
Lefcort and Eiger (unpub. data) elicited a
host-defense response in bullfrog tadpoles (Rana
catesbeiana).
They found that the host-defense response
induces behavioral fevers, alters the tadpole's predator
avoidance behavior, and increases the tadpoles
susceptibility to predation by newts (Caudata: Taricha
granulosa).
Infection of vertebrates with viruses, bacteria and
most eukaryotic parasites results in a host defense
response (Titus et al., 1991).
This response occurs
regardless of the parasite's single or multi-host lifehistory.
If changes in behavior are associated with the
host defense response, then alterations of host behavior
may be a more widespread phenomenon than is currently
believed.
To determine if behavioral alterations
associated with infection occur in systems where
alterations are detrimental to the parasite's genetic
fitness, we examined the behavioral effects of a one-
104
host parasitic yeast (Candida spp., Richards, 1958;
1962; Steinwascher, 1979) in tadpoles of the red-legged
frog, Rana aurora.
Specifically we tested: 1)
if
infected tadpoles have altered thermal preferences, 2)
how the presence of a predator affects thermoregulation,
and measure 3) the ability of infected tadpoles to
detect and respond to chemical cues from a predator, and
4) the susceptibility of infected tadpoles to predation.
105
METHODS
We gathered all tadpoles and newts at a pond located
eighteen km south of Waldport, Oregon (elev. 20m).
Tadpoles were gathered as eggs, housed at Oregon State
University, and fed alfalfa pellets ad libitum.
maintained the animals between 18 and 22 °C.
tests when tadpoles reached Gosner stages
We
We began
22 - 25 (i.e.
without leg buds, Gosner, 1960, mass 0.8 - 1.2 g).
The
laboratory was illuminated between 06:00 - 18:00 hours.
We conducted experiments in March and April of 1992
between 13:00 - 17:00 hours.
The yeast, Candida spp. has a one-host life-cycle and
is naturally transmitted among individual tadpoles
through water and feces (Richards, 1958).
We infected
the tadpoles by exposing them for seven days to water
and feces from high density (15/liter) groups of R.
aurora that were already infected when captured.
These
crowded tadpoles were collected from the same pond as
the experimental tadpoles at Gosner (1960) stages 22-25.
After the seven day exposure period, experimental
tadpoles exhibited signs of wasting and microscopic
examination of gut contents revealed colorless, round,
vacuolated cells, as described by Richards (1958).
These cells were not recovered from control tadpoles
kept at a lower density (2/liter).
Although Candida
106
sap. was present in all experimental tadpoles examined,
it is possible that other pathogens were also present.
Microscopic examination of newts that had fed on
infected tadpoles (three days post-exposure) indicated
that the yeast was probably not transmitted to newts,
although transmission at undetectable low levels could
have occurred.
Although Candida yeasts, when parasitic,
are generally single-host parasites, (Lodder, 1970) i.e.
no intermediate and definitive hosts, it is possible
that predators other than newts can act as secondary
hosts.
Tadpoles, seven to nine days post initial
exposure, were tested only once.
Field experiments were
conducted at Soap Creek Pond, located 15 kilometers
north of Corvallis, Oregon.
To test for significant treatment effects, paired
t-tests, Student's t-test and two-way analyses of
variance (ANOVA) tests were used.
Statistical
significance of all tests was P < 0.05.
EXPERIMENT 1: FIELD THERMAL PREFERENCE
One indication of an initial host defense response in
ectotherms is behavioral fever - i.e. altered
thermoregulatory behavior (Kluger, 1991).
To determine
the thermal preferences of control and exposed tadpoles
in the field, we used an experimental design similar to
107
that of Walls (1991).
We used 12 enclosures constructed
of 5 cm wide plywood strips, composing a frame covered
on three sides with fiberglass mesh insect screening
(mesh size = 1 mm).
Enclosures measured 137 cm (1) x 30
cm (w) x 30 cm (d) and were placed lengthwise, along the
natural slope of the pond bottom.
The uncovered top of
the enclosure extended above the surface of the water.
Enclosures were placed 30 cm apart and were parallel to
each other and perpendicular to the pond's edge.
Due to
the slope of the pond, the end farthest from the pond's
edge was situated in "deep" water (20 cm).
The opposite
end of the enclosure was situated in "shallow" water (2
cm).
We placed aquatic plants in the enclosures.
We placed 15 tadpoles in the center of each enclosure
and allowed them to acclimate to the pond.
Six
enclosures contained uninfected tadpoles and six
contained infected tadpoles.
After 24 hours, we
recorded the number of animals in both the shallow half
of the enclosure and the top 1 cm of the deep half by
viewing with binoculars at a distance of 4 m.
We also
recorded water temperatures at the geometric center of
the shallow end, the geometric center of the deep end,
and 1 cm below the surface of the pond's deep end.
To
control for a gradient of temperatures along the pond
margin the treatments were alternately interspersed.
108
The positions of the tadpoles were again recorded 30,
60, and 90 minutes later.
At the last reading the
temperatures of the three zones were again recorded.
We
then terminated the experiment and counted the number of
tadpoles in each container to determine if any had
escaped.
Temperatures were not recorded more frequently
because preliminary results indicated that wading into
the pond to record temperatures stirred-up pond silt,
agitated the tadpoles and disturbed any existing thermal
gradient.
EXPERIMENT 2: EFFECT OF PREDATORY NEWTS ON
THERMOREGULATION OF TADPOLES
This experiment tested how the presence of a
predatory newt (Taricha qranulosa) affected behavioral
thermoregulation of infected tadpoles.
We used the same
plywood and mesh enclosures as experiment 1.
We allowed
15 tadpoles to acclimate for 24 hours in each of the
enclosures, and we then added a newt satiated on
tadpoles to half of the enclosures.
Thus, each of three
enclosures received one of the following four treatments
in a randomized block design:
1) newt present, tadpoles uninfected, 2) newt absent,
tadpoles uninfected, 3) newt present, tadpoles infected,
or 4) newt present, tadpoles infected.
109
Observations were carried out as in experiment 1.
EXPERIMENT 3: SUSCEPTIBILITY OF TADPOLES TO PREDATION BY
NEWTS
To determine if yeast infections affect the
susceptibility of tadpoles to predation we used the same
design as the previous two experiments, but with the
following treatments: six enclosures contained 10
uninfected tadpoles and six contained 10 infected
tadpoles.
After a 24 hour acclimation period we added a
24 hour-starved newt to each enclosure.
Sixty minutes
later, we counted the number of surviving tadpoles.
EXPERIMENT 4: PREDATOR-CUE
This experiment tested the behavioral response of
tadpoles to chemical cues present in water that had been
in contact with newts.
The experimental apparatus was
similar to that of Petranka et al. (1987).
Tadpoles
were tested in a gravitational flow though system
composed of three 25-1 plastic tubs (51 cm (1) X 37 cm
(w) X 21 cm (d).
The tubs were arrayed at different
heights so that water flowed from one to another at 0.5liter /min.
The two lower tubs had output openings and
never contained more than 10 liters of water.
The
uppermost tub was filled with 23 liters of water.
In
110
some of the trials (see below) the middle tub was either
unoccupied or contained two newts.
From the middle tub,
water flowed to the lower-most tub, which contained
eight tadpoles and was divided across its width such
that half the tub contained the input opening and half
contained the output opening.
Ten minutes after flow
was initiated, we recorded the number of tadpoles in
each half of the enclosure along with the number of
times a tadpole crossed between the two halves of the
enclosure. These data were recorded for 10 minutes at 1
minute intervals.
Three treatments were used: 1, infected tadpoles,
newt absent; 2, infected tadpoles, newt previously fed
insect larvae (midge flies, Chirononmus spp.); and 3,
infected tadpoles, newt previously fed R. aurora
tadpoles.
Since R. aurora alter their behavior in
response to predators who have fed on conspecifics
(Wilson and Lefcort, in press) the newts were maintained
on, and fed, either R. aurora tadpoles or insect larvae
ad libitum 24 hours prior to testing.
111
RESULTS
EXPERIMENT 1: FIELD THERMAL PREFERENCE
Mean temperatures in the enclosures at 13:30 were as
follows: 17.57 +0.19°C at the shallow end, 16.06 +0.24°C
at the deep end, and 17.28 +0.16°C 1 cm below the
surface of the deep end (air temperature = 13.4
14.5°C)
.
Significantly more infected than uninfected
animals were found in the two combined shallow zones
(Paired t-test P < 0.001).
EXPERIMENT 2: EFFECT OF PREDATORY NEWTS ON
THERMOREGULATION OF TADPOLES
Mean temperatures in the enclosures at 14:30 were as
follows: 21.57 +0.14°C at the shallow end, 20.18 +0.18°C
at the deep end, and 21.47 +0.16°C 1 cm below the
surface of the deep end (air temperature = 21.4 22.4°C).
When the readings at 30, 60 and 90 minutes are
analyzed together, infection with the parasite resulted
in a slight, but not significant, increase in the number
of tadpoles in warm shallow zones (Table V.1, repeated
measure ANOVA, F = 4.04, d.f.= 1, P = 0.079).
The
presence of newts had no affect on the distribution of
tadpoles (repeated measure ANOVA, F = 0.96, d.f.= 1, P =
0.355, Table V.1).
There was no significant interaction
112
between newt presence and infection (repeated measure
ANOVA, F = 0.02, d.f.= 1, P = 0.892).
EXPERIMENT 3: SUSCEPTIBILITY OF TADPOLES TO PREDATION BY
NEWTS
Significantly more infected (3.17 + 0.48) than
uninfected (1.33 +0.42) tadpoles were consumed
(Student's t-test, P = 0.016)
EXPERIMENT 4: PREDATOR-CUE
When only the three treatments of infected tadpoles
were examined, no significant treatments effects were
found for the mean number of animals on the side of the
tub closest to incoming treated water (ANOVA, F = 0.48,
d.f. = 2, P = 0.62, Table V.2).
Significant treatment
effects were found for the number of moves (total number
of times tadpoles crossed from one half of the tub to
the other half) made by the tadpoles (F = 13.34, d.f. =
2, P < 0.0001).
When data (Wilson and Lefcort, In Press; identical
design and tadpoles) from uninfected tadpoles are
included in an examination with infected tadpoles,
no
differences were found among the six treatments based on
the mean number of animals occupying the tub-side near
the influx of treated water (P > 0.380 for all
113
treatments).
Significant differences in the mean number
of moves were found for the presence of the parasite
(F = 68.57, d.f. = 1, P < 0.001) and the diet of the
predators (F = 34.44, d.f. = 1, P < 0.001).
However,
there was also a significant effect for the interaction
between parasite and predator-diet (F = 12.42, d.f. = 1,
P < 0.001).
114
DISCUSSION
Physiological responses of R. aurora are associated
with altered behavior and increased predation.
tadpoles were found in warm shallow water.
Infected
Behavioral
fevers have been associated with the release of
cytokines such as IL-1 and TNF in anurans (Kluger, 1977;
1991).
The behavioral effects of these endogenously
produced substances may be related to alterations in the
behavior of infected R. aurora; i.e., infected tadpoles
suffered increased predation by newts.
Furthermore,
infected tadpoles exhibited less movement than did
uninfected tadpoles when exposed to chemical cues from
newts fed insect larvae or conspecific tadpoles.
However, there was a significant interaction between the
presence of the parasite and the predator's diet,
suggesting that the parasite alters the tadpole's normal
response to newts.
Interestingly, thermoregulatory
behavior of infected and uninfected tadpoles was not
altered by the presence of a caged newt.
Theoretical models indicate that, under certain
circumstances, two-host parasites have higher genetic
fitness if their hosts' exhibit altered behavior
(Dobson, 1988; Freedman, 1990; see also Lafferty, 1992).
However, the benefits to a one-host parasite may be more
variable.
For example, parasites such as the rabies
115
virus have increased transmission if the infected host
becomes more aggressive (i.e., in the form of biting).
However, parasites such as Candida may have decreased
fitness if their host is eaten since we were unable to
recover the yeast from newts that had consumed infected
tadpoles.
For hosts, physiological responses to infection have
behavioral ramifications.
Behavioral effects of the
host defense response may provide a mechanism by which
behavioral alterations occur in some one and two-host
systems.
These behavioral alterations may have adaptive
value to hosts [e.g. febrile ectothermic lizards have
higher survival than nonfebrile lizards (Kluger et al.,
1975) or may simply be by-products of the host defense
response that lead to increased predation.
The host
defense response was selected for, and is maintained
because it provides a valuable tool for fighting
infection.
But aspects of the response may have
negative effects on host survival.
cytokines can be detrimental.
Overproduction of
Brain damaging and fatal
temperatures are usually just a few degrees above
feverish temperatures.
Furthermore, overproduction of
TNF in mammals causes the wasting and gauntness of
cancer associated with many chronic diseases (Titus et
al., 1991).
TNF's side effects in humans are muscle
116
stiffness, muscle pain, nausea, and vomiting (Clark,
1987).
Indeed, the pathology of cerebral malaria may
largely be due to TNF (Clark, 1987).
Cytokines, released during almost all vertebrate
infections are extremely powerful and have a wide array
of physiological (Titus et al., 1991) and,
suggests, behavioral effects.
as this study
Altered host behavior due
to the host defense response may occur in many systems
(see Damian, 1987).
This may result in augmented
transmission in some systems (Ewald, 1991)
or, as in
this system, result in decreased transmission.
Adaptive
arguments to explain altered host behavior should
therefore be made with caution since altered hosts
behaviors may be a generalized response to infection
that is independent of parasite life-history variables.
117
ACKNOWLEDGMENTS
We wish to thank Mark Hixon, Phil Rossignol and Susan
Walls for critically reading and improving earlier
drafts of this manuscript.
We are especially indebted
to Robert M. S. Hightower, Pedro R. Fuerte and William
Kearney for their guidance.
This work was partially
supported by NSF grants BNS-9020957 and BNS-9107172.
118
TABLE V.1
MEAN NUMBER OF TADPOLES (+S.E.)IN SHALLOW ZONES
ABSENT
INFECTION
NEWTS
PRESENT
ABSENT
5.00 +1.87
4.67 +1.99
PRESENT
7.22 +2.28
6.22 +1.61
TABLE V.2
RESULTS OF PREDATOR-CUE EXPERIMENT
TREATMENT
# TRIALS
MOVES/TRIAL
% TOWARD INCOMING
SCENT
(MEAN +S.E.)
(MEAN +S.E.)
Infected
Untreated Water
21
28.57 +3.47
39.82 +2.46
Insect-fed Newts
23
15.96 +4.04
43.15 +3.17
Tadpole-fed Newts
19
3.89 +1.03
39.14 +3.51
RESULTS FROM WILSON AND LEFCORT (IN PRESS)
Uninfected
Untreated water
18
46.55 +5.77
43.13 +2.18
Insect-fed newts
16
64.75 +6.17
40.48 +2.54
Tadpole-fed newts
14
23.00 +4.35
43.03 +3.29
120
SUMMARY
Explanations and interpretations of the adaptive
significance of parasitic effects on hosts must consider
both selective pressures on hosts and on parasites.
Host responses involve a change in physiology, and
often, behavior.
The responses may be due to many
factors: host modifications to avoid or to fight off
infection, manipulations of hosts by parasites,
or
alterations that are not adaptive for hosts or parasites
and are simply the consequences of the host response to
infection.
Hamilton and Zuk (1982) hypothesized that animals
alter their mate choosing behavior to avoid a
parasitized mate.
Support of the hypothesis has been
mixed, although it has found general support in fishes
and birds.
In the first test of the hypothesis in
reptiles, I failed to find supportive evidence (chapter
two).
In fact, contrary to the hypothesis, my results
indicate that lizard species with high parasite loads
are showy.
Therefore, I did not find evidence that
lizards have modified their mating behavior to avoid
becoming parasitized.
Snails (Biomphalaria glabrata) do seem to modify
their behavior in ways that lessen the impact of
parasites.
The timing of these modifications appears to
121
be dependent on whether the snail is susceptible to the
parasite, or has innate resistance.
Naturally
resistant-strain snails seek out cooler temperatures
(two degrees Celsius) - temperatures that have been
correlated with impaired parasite fecundity - hours
after contact with the invading parasite (chapter
three).
The parasite is unable to establish itself and
hence the snail is able to quickly return to normal
preferred temperatures.
Susceptible-strain snails lower
their temperature only slightly (0.5 degrees Celsius).
This minor change in temperature of the susceptible
strain allows the parasite to successfully establish
itself.
Later, at five weeks post exposure, the
parasite begins to release progeny that damage snail
tissues.
At this point the susceptible snail responds
by seeking out cool water temperatures.
In chapter four I explored fever as a mechanism that
may be available for parasites to modify host behavior.
In chapter five I found that many behavioral effects of
parasitism do not aid in the elimination of the parasite
(i.e., are not for the host's benefit) or the increased
transmission of the parasite (i.e., for the parasite's
benefit).
Tadpoles parasitized with a yeast exhibited
much of the same alterations of behavior that tadpoles
exhibit when injected with killed bacteria that are not
122
growing.
In both cases, the behavioral alterations
resulted in increased rates of predation.
The host does
not benefit from being killed (although tenuous
inclusive fitness arguments could be made), and since
the bacterium and yeast die when its host is consumed,
the behavioral alterations of the host do not benefit
the pathogen.
My findings in chapters four and five extend our
understanding of host-parasite interactions.
While it
is generally accepted that parasites are under selective
pressure to maximize growth, development, fecundity
and/or their own transition to a subsequent host; the
relative success of these "goals" is largely dependent
on the physiological and behavioral responses of their
host.
The host is far more than just a source of
nutrients.
The host provides a habitat, a controlled
environment and a mode of transportation.
Host defense
systems (i.e., the acute phase response), due to its
effects on host behavior, represent an opportunity for
parasitic manipulation.
The acute phase response involves the release of
cytokines such as IL-1, IL-6 and TNF whose effects are
far reaching and result in a cascade of responses that
are often detrimental to the elimination of some
parasites.
Fever mobilizes the immune system and
123
exposes some bacteria to lethal temperatures but it also
taxes nutrient stores and can lead ectothermic animals
to engage in thermoregulatory behavior that result in
exposure to predators (chapters four and five).
Casterlin and Reynolds (1977) showed that when bullfrogs
are injected with killed bacteria they exhibit
behavioral fevers.
This response was later shown to be
due to the release of cytokines (Kluger 1990).
Cytokines induce sleep that help animals to conserve
resources and possibly avoid activities that expose them
to predators (Hart, 1988), but that same lethargic
behavior can make a foraging or behaviorally
thermoregulating animal susceptible to predation.
Part of the host defense system is specific, i.e.,
antibody production, but part is extremely generalized.
Host defenses must be broad enough to counter bacteria,
viruses, fungi, and both single and multi-celled
animals.
By its very nature, a generalized response
must include aspects that are not always beneficial in
dealing with all parasites.
For example, the release of
histamine by mast cells causes inflammation which limits
the movements of microorganism through epithelial
layers.
However, the same response, when misdirected
against harmless plant pollen, can be incapacitating and
lead to asthmatic attacks.
Because the acute phase
124
response is so easily induced - the exposure macrophages
to any foreign antigen - parasites have the potential to
spark the acute phase response and then benefit from the
resulting response.
Each of the systems studied in this thesis involves
associations of parasites and behaviorally modified
hosts.
In every chapter adaptive arguments are
presented that may explain why (potential) hosts do or
do not alter their behavior.
Yet the effects of
parasites on host behavior cannot be simplified into a
"who gains" question (see Gould and Lewontin, 1979).
Each host/parasite system is unique.
While the
behavioral alterations that animals exhibit may seem to
be either for the host or the parasite, many of these
alterations benefit neither, therefore the driving
forces that cause and conserve these alterations must be
examined on a case-by-case basis.
125
BIBLIOGRAPHY
Aeby, G. 5.444 1991. Behavioral and ecological
relationships of a parasite and its hosts within
a coral reef system. Pac. Sci. 45:263-269.
Alberts, A. C. 1989. Ultraviolet visual sensitivity in
desert iguanas: implications for pheromone detection.
Anim. Behv. 38:129-137.
Atkins, E. 1982. Fever: its history, cause and function.
Yale J. Biol. Med. 55:283-289.
Baer, G.M. 1973. The Natural History of Rabies, Vols I
and II. Academic Press, N.Y.
Bayne, C. J. 1983. Molluscan immunobiology. In: Biology
of the Mollusca. Vol. 5, Physiology Pt. 2. Eds.:
Saleuddin, A. S. M. and Wilbur, K. M. Academic Press.
San Diego. pp.407-486.
Beach, R., Kiilu, G., and Leeuwenburg, J. 1985.
Modification of sand fly biting behavior by
Leishmania leads to increased parasite transmission.
Am. J. Trop. Med. Hyg. 34:278-282.
126
Field guide to North American reptiles and amphibians.
1988. Eds.: Behler, J. L., and King, F. W. Audubon
Society. Alfred A. Knopf. New York.
Benes, E. S. 1985. Helminth parasitism in some central
Arizona lizards. Southwest. Nat. 30:467-473.
Besedovsky, H., Del Rey, A., Sorkin, E. and Dinarello,
C. H. 1986. Immunoregulatory feedback between
interleukin-1 and glucocorticoid hormones. Science
233:652-654.
Bethel, W. M. and Holmes, J. C. 1972. Modifications of
intermediate host behaviour by parasites.
In:
Behavioural aspects of parasite transmission. Eds.:
Canning, E. V. and Wright, C. A. Linnean Society of
London. London. pp.123 - 149.
Bethel, W. M. and Holmes, J. C. 1977. Increased
vulnerability of amphipods to predation owing to
altered behavior induced by larval acanthocephalans.
Can. J. Zool. 55:110-115.
127
Bissonnette, E. Y., Rossignol, P. A. and Befus, A. D.
Extracts of mosquito salivary gland inhibit tumor
necrosis factor alpha release from mast cells.
Parasit. Immunol. In Press.
Blankespoor, H. D., Babiker, S. M. and Blankespoor, C.
L. 1989. Influence of temperature on the development
of Schistosoma haematobium in Bulinus truncatus. J.
Med. App. Mal. 1:123-131.
Blaustein, A. R. 1981. Sexual selection and mammalian
olfaction. Am. Nat. 117:1006-1010.
Bluthe, R. M., Dantzer,R., and Kelley, K. W. 1992.
Effects of interleukin-1 receptor antagonist on the
behavioral effects of lipopolysaccharide in rat.
Brain Res. 573:318-323
Borgia, G. 1986. Satin bowerbird parasites: a test of
the bright male hypothesis. Behay. Ecol. Sociobiol.
19:355-358.
128
Borgia, G. and Collis, K., 1989. Female choice for
parasite-free male satin bowerbirds and the evolution
of bright male plumage. Behay. Ecol. Sociobiol.
25:445-454.
Borgia, G. and Collis, K. 1990. Parasites and bright
male plumage in the satin bowerbird (Ptilonorhynchus
violaceus). Am. Zool. 30:279-286.
Brassard, P., M. E. Rau, and Curtis, M. A. 1982.
Parasite - induced susceptibility to predation by
larval acanthocephalans. Parasit. 85:495-501.
Bundy, D. A. P., Vogel, P. & Harris, E. A. 1987.
Helminth parasites of Jamaican anoles (Reptilia:
Iguanidae): a comparison of the helminth fauna of 6
Anolis species. J. Helminthol. 61:77-83.
Camp, J. W. and Huizanga, H. W. 1979. Altered color,
behavior and predation susceptibility of the isopod
Asellus intermedius infected with Acanthocephalus
dirus. J. Parasit.
65:667-669.
129
Carney, W. P. 1969. Behavioral and morphological changes
in carpenter ants harboring dicrocoelid
metacercaria. Am. Midl. Nat. 82:605-611.
Casterlin, M. E. and Reynolds, W. W. 1977. Behavioral
fever in anuran amphibian larvae. Life Sci.
20:593-596.
Chernin, E. 1967. Behavior of Biomphalaria glabrata and
of other snails in a thermal gradient. J. Parasit.
53:1233-1240.
Clark, I. A. 1987. Cell mediated immunity in protection
and pathology of malaria. Paristol. Today.
3:300-305.
Clayton, D. H. 1990. Mate choice in experimentally
parasitized rock doves: lousy males lose. Am. Zool.
30:251-262.
Clayton, D. H. 1991. The influence of parasites on host
sexual selection. Parasit. Today 7:329-334.
Connell, J. H. 1980. Diversity and the coevolution of
competitors, or the ghost of competition past. Oikus
35:131-138.
130
Cooper, W. E. 1988. Aggressive behavior and courtship
rejection in brightly and plainly colored female
keeled earless lizards (Holbrookia propinqua).
Ethology. 77:265-278.
Cox, F. E. G. Parasites and sexual selection. 1989.
Nature 341:289-290.
Cowan, I. McT., 1951. The diseases and parasites of big
game mammals of western Canada. Proc. Fifth Ann. Game
Conv. pp. 37-64.
Crowden, A. E. and Broom, D. M. 1980. Effects of the
eyefluke, Diplostomum spathaceum, on the behaviour of
dace (Leuciscus leuciscus). Anim. Behv. 28:287-294.
Curtis, L. A. 1990. Parasitism and the movements of
intertidal gastropod individuals. Biol. Bull. (Woods
Hole). 179:105-112.
D'Alecy, L. G. & Kluger, M. J. 1975. Avian febrile
response. J. Physiol. 253:223-232.
131
Damian, R. T. 1964. Molecular mimicry: antigen sharing
by parasite and host and its consequences. Am. Nat.
98:128-149.
Damian, R. T. 1987. Presidential address. The
exploitation of host immune responses by parasites.
J. Parasit. 73:3-13.
Damian, R. T., Roberts, M. L., Powell, M. R., Clark, J.
D., Lewis, F. A., and Stirewalt, M. A. 1984.
Schistosoma mansoni egg granuloma size reduction in
diated cryopreserved schitosomula. Proc. Nat. Acad.
Sci. USA. 81:3552-3556.
Darwin, C. 1871. The descent of man and selection in
relation to sex. J. Murray, London.
Dawkins, R. The selfish gene. Oxford Univ. Press, Second
Edition. Oxford.
Day, J. F. and Edman, J. D. 1983. Malaria renders mice
susceptible to mosquito feeding when gametocytes are
most infective. J. Parasit. 69:163-170.
132
Dence, W. A. 1958. Studies on Lingula-infected common
shinners (Notropis cormutus Agassiz) in the
Adirondacks. J. Parasit. 44:334-338.
Dimock, K. A., Davis, C. D. and Kuen, R. D. 1991.
Effects of elevated environmental temperature on the
antibody response of mice to Trypanosoma cruzi during
the acute phase of infection. Infec. Immun.
59:4377-4382.
Dinarello, C. A. 1984. Interleukin-1 and the
pathogenesis of the acute-phase response. New England
Journal of Medicine 311:1413-1418.
Dinarello, C. A., Cannon, J. G. Wolff, S. M., Bernheim,
H. A., and Beutler, B. 1986. Tumor necrosis factor
(cachectin) is an endogenous pyrogen and induces
production of interleukin 1. J. Exp. Med.
163:1433-1450.
Dobson, A. P. 1988. The population biology of parasiteinduced changes in host behavior. Q. Rev. Biol.
63:139-165.
133
Dobzhansky, T. 1956. What is an adaptive trait? Am. Nat.
90:337-347
Doenhoff, M. J., Hassounah, 0. A. and Lucas, S. B. 1985.
Does the immunopathology induced by schistosome eggs
potentiate parasite survival?
Immun. Today
6:203-206.
Doenhoff, M. J., Musallam, R., Bain, J., and McGregor,
A. 1978. Studies on host-parasite relationship in
Schistosoma mansoni-infected mice: The immunological
dependance of parasite egg excretion. Immunol.
35:771-778.
Dolinsky, Z. S., Hardy, C. A. Burright, R. G. and
Donovick, P. J. 1985. The progression of behavioral
and pathological effects of the parasite Toxocara
canis in the mouse. Physiol. Behay. 35:33-42.
Eiger, S. M. and Kluger, M. J. 1985. Cyanate and
temperature regulation in anephric rabbits. Am. J.
Physiol. 249:F69-F73.
134
Ewald, P. W. 1980. Evolutionary biology and the
treatment of signs and symptoms of infectious
disease. J. Theor. Biol. 86:169-176.
Ewald, P. W. 1983. Host-parasite relations, vectors, and
the evolution of disease severity. Ann. N. Y. Acad.
Sci. 503:295-305.
Ewald, P. W. 1986. Evolutionary biology and the
treatment of signs and symptoms of infectious
disease. J. Theor. Biol. 86:169-176.
Ewald, P. W. 1987. Transmission modes and evolution of
the parasitism-mutualism continuum. Ann. N.Y. Acad.
Sci. 503:295-305.
Ewald, P. W. 1991. Waterborne transmission and the
evolution of virulence among gastrointestinal
bacteria. Epidem. Infec. 106:83-119.
Fenstermacher, R. 1937. Further studies of diseases
affecting moose. II. Cornell Vet. 27:25-37.
135
Freedman, H. I. 1990. A model of predator-prey dynamics
as modified by the action of a parasite. Math.
Biosci. 99:143-155.
Freeland, W. J. 1983. Parasites and the coexistence of
animal host species. Am. Nat. 121:223-236.
Frisch, K. Von. 1941. Uber einen Schreckstoff der
Fischhaut and seine biologische Bedeutung.
Z. Vergl. Phys. 29:46-145.
Frost D. R. and Etheridge, R. 1989.
A phylogenetic
analysis and taxonomy of iguanian lizards (Reptilia:
Squamata). University of Kansas Museum of Natural
History. Mis. Pub. /81.
Fryer, S. E., Dykes-Hoberg, M. & Bayne, C. J. 1989.
Changes in plasma opsonization of yeast after
isolation of Biomphalaria glabrata in small volumes
of water. J. Invert. Path. 54:275-276.
Fryer, S. E. and Bayne, C. J. 1990. Schistosoma mansoni
modulation of phagocytosis in Biomphalaria glabrata.
J. Parasit. 76:42-52.
136
Gibbs, E. L. 1963.
An effective treatment for red-frog
disease in Rana pipiens. Lab. Animal Care 13:781-783.
Gibson, R. M. 1990. Relationships between blood
parasites, mating success and phenotypic cues in male
sage grouse Centrocercus urophasianus. Am. Zool.
30:271-278.
Giles, N. 1983. Behavioural effects of the parasite
Schistocephalus solidus (Cestoda) on an intermediate
host, the three-spined stickleback, Gasterosteus
aculeatus L. Anim. Behv. 31:1192-1194.
Godin, J. & Sproul, C. D. 1988. Risk taking in
parasitized sticklebacks under threat of predation:
effects of energetic need and food availability. Can.
J. Zool. 66:2360-2367.
Gosner, K. L. 1960. A simplified table for staging
anuran embryos and larvae with notes on
identification. Herpetolog. 16:183-190.
137
Gould, S. J. and Lewontin, R. 1979. The spandrels of San
Marco and the Panglossian paradigm: a critique of the
adaptationists programme. Proc. R. Soc. Lond. B.
205:581-598.
Haahr, S. and Mogensen, S. 1978. Function of fever in
infectious disease. Biomed. 28:305-307.
Hamilton, W. D. and Zuk, M. 1982. Heritable true fitness
and bright birds: a role for parasites? Science
218:384-387.
Hart, B. L. 1988. The biological basis of the behavior
of sick animals. Neurosci. Behay. Rev. 12:123-137.
Hart, B. L. 1990. Behavioral adaptations to pathogens
and parasites: five strategies. Neurosci. Behay. Rev.
14:273-294.
Hart, B. L., Hart, L. A., Mooring, M. S. and Olubayo, R.
1992. Biological basis of grooming behaviour in
antelope: the body-size, vigilance and habitat
principles. Anim. Behv. 44:615-631.
138
Hausfater, G., Gerhardt, H. C. and Klump, G. M. 1990.
Parasites and mate choice in grey treefrogs, Hyla
versicolor. Am. Zool. 30:299-312.
Hay, J. & Hutchison, W. M. 1983. Congenital Toxoplasma
infection and the response to novelty in mice. Ann.
Trop. Med. Par. 77:437-439.
Hay, J., Aitken, P. P., Arnott, M. A. 1985. The effect
of Toxocara canis on the spontaneous running activity
of mice. Ann. Trop. Med. Par. 79:221-222.
Hayunga, E. G. 1979. Observations on the intestinal
pathology caused by three caryophyllid tapeworms of
the white sucker Catostomus commersoni Laecpede
(sic).
J. Fish Dis. 2:239-248.
Hecker, V. and Thomas, E. 1965. Uber
sporocystenschlauche von Leucochloridium macrostomum
Rud. Verh Dtsch Zool Ges Kiel, Zool. Anz. Suppl.
28:444-456.
139
Helluy, S. and Holmes, J. C. 1990. Serotonin,
octopamine, and the clinging behavior induced by the
parasite Polymorphus marodoxus (Acanthocephala) in
Gammarus lacustris (Crustacea). Can. J. Zool.
68:1214-1220.
Hillgarth, N. 1990. Parasites and female choice in the
ring necked pheasant. Am. Zool. 30:227-234.
Holmes, J. C. and Zohar, S. 1990. Pathology and host
behaviour. In: Parasitism and host behaviour Eds.:
Barnard, C. J. and Behnke, J. M. Taylor and Francis.
London. pp. 34-63.
Houde, A. E. and Torio, A. J. 1992. Effect of parasitic
infection on male color pattern and female choice in
guppies. Behay. Ecol. 3:346-351.
Hughes, K. Jr., Smith, E. M., Chin, R., Cadet, P.,
Sinisterra, J.,
Leung, M. K., Shipp, M. A.,
Scharrer, B. & Stefano, G. B. 1990. Interaction of
immunoactive monokines (interleukin 1 and tumor
necrosis factor) in the bivalve mollusc Mytilus
edulis. Proc. Nat. Acad. Sci.
(USA) 87:4426-4429.
140
Hurd, H. and Fogo, S. 1991. Changes induced by
Hvmenolepis dimunuta (Cestoda) in the behaviour of
the intermediate host Tenebrio molitor (Coleoptera).
Can. J. Zool. 69:2291-2294.
Hurst, C. T. and Walker, C. R. 1935. Increased heat
production in a poikilothermous animal in parasitism.
Am. Nat. 69:461-466.
Johnson, R. A. and Wichern, D. W. 1988. Applied
multivariate statistical analysis. Prentice Hall.
Englewood Cliffs, New Jersey.
Kavaliers, M. and Colwell, D. D. 1992. Parasitism,
opioid systems and host behavior. Adv. in
Neuroimmunol. 2, In Press.
Kennedy, C. E. J., Endler, J. A., Poynton, S. L.
&
McMinn., H. 1987. Parasite load predicts mate choice
in guppies. Behay. Ecol. Sociobiol. 21:291-295.
King, M. K. and Wood, W. B. 1958. Studies on the
pathogenesis of fever. J. Exp. Med. 107:291-303.
141
Kirkpatrick, M. 1987. Sexual selection by female choice
in polygynous animals. Ann. Rev. Ecol. Syst.
18:43-70.
Kiuger, M. J. 1977. Fever in the frog Hvla cinerea. J.
Therm. Biol. 2:79-81.
Kiuger, M. J. 1979. Phylogeny of fever. Fed. Proc.
38:30-34.
Kluger, M. J. 1990. The febrile response. In: Stress
proteins in biology and medicine. Cold spring harbor
laboratory press. pp. 61-78.
Kluger, M. J. 1991. The adaptive value of fever. In:
Fever: basic mechanisms and management. Ed.:
Mackowiak, P. A. Raven Press, New York. pp. 105-122.
Kiuger, M. J., Conn, C. A., Franklin, B., Freter, R. and
Abrams, G. D. 1990. Effects of gastrointestinal flora
on body temperature of rats and mice. Am. J. Physiol.
258:R552-R557.
Kiuger, M. J., Ringler, D. H. and Anver, M. R. 1975.
Fever and survival. Science 118:166-168.
142
Kluger, M. J., Turnbull, A. J., Cranston, W. I., Wing,
A. J., Gross, M. P.& Rothenburg, B. A. 1981.
Endogenous cryogen secreted by the kidneys. Am. J.
Physiol. 281:R271-R276.
Kuris, A. M. and Blaustein, A. R. 1977. Ectoparasitic
mites on rodents: application of the island
biogeography theory? Science 195:596-598.
Kwiatkowski, D. 1989. Febrile temperatures can
synchronize the growth of Plasmodium falciparum in
vitro. J. Exp. Med. 169:357-361.
Kwiatkowski, D. & Greenwood, B. M. 1989. Why is malaria
fever periodic? A hypothesis. Parasit. Today
5:264-266.
Lafferty, K. D. 1992. Foraging on prey that are modified
by parasites. Am. Nat. 140:854-867.
LaMunyon, C. and T. Eisner. 1990. Effect of mite
infestation on the anti-predator defenses of an
insect. Psyche (Cambridge) 97:31-41.
143
Lee, 0. F. and Cheng, T. C. 1971. Schistosoma mansoni
infection in Biomphalaria alabrata: alterations in
heart rate and thermal tolerance of the host. J.
Invert. Path. 18:412-418.
Lefcort, H. and Bayne, C. J. 1991. Thermal preference of
resistant and susceptible strains of Biomphalaria
glabrata (Gastropoda) exposed to Schistosoma mansoni
(Trematoda). Parasit. 103:357-362.
Lefcort, H. and Blaustein, A. R. 1991. Parasite load and
brightness in lizards: an interspecific test of the
Hamilton and Zuk hypothesis. J. Zool.
(London)
224:491-499.
Leiby, P. D. and Dyer, W. G. 1971. Cyclophillidean
tapeworms of wild Carnivora. In: Parasitic diseases
of wild mammals. Eds: Davis and Anderson.
Iowa St.
Univ. Press. Ames. pp. 174-234.
Lemay, L. G., Vander, A. J., and Kluger, M. J. 1990. The
effects of psychological stress on plasma
interleukin-6 activity in rats. Phys. Behay.
47:957-961.
144
Lester, R. J. G. 1971. The influence of Schistocephalus
plerocercoids on the respiration of Gasterosteus and
a possible resulting effect on the behavior of fish.
Can. J. Zool. 49:361-366.
Lillywhite, H. B.,
Licht, P., and Chelgren, P. 1973.
The role of behavioral thermoregulation in the growth
energetics of the toad Bufo boreas. Ecology
52:375-383.
Lima, S. L. and Dill, L. M. 1990. Behavioral decisions
made under the risk of predation: a review and
prospectus. Can. J. Zool. 68:619-640.
Linton, E. 1906. A cestode parasite in the flesh of the
butterfish. Bull. Bur. Fish. Wash. 26:111-132.
Livesey, J. L., Molyneux, D. H. and Jenni, L. 1980.
Mechanoreceptor-trypanosome interaction in the labrum
of Glossina: fluid mechanics. Acta Tropica
37:151-161.
Lodder, J. 1970. The yeasts: a taxonomic study.
Amsterdam: North-Holland.
145
Loker, E. S. 1979. Pathology and host responses induced
by Schistosomatium douthitti in the freshwater snail
Lymnaea catascopium. J. Invert. Path. 33:265-273.
Long, N. C., Vander, A. J., Kunkel, S. L., and Kluger,
M. J. 1990a. Antiserum against tumor necrosis factor
increases stress hyperthermia in rats. Am. J.
Physiol. 258:R591-R595.
Long, N. C., Otterness, I., Kunkel, S. L., Vander, A.
J., and Kluger, M. J. 1990b. Roles of interleukin 1B
and tumor necrosis factor in lipopolysaccharide fever
in rats. Am. J. Physiol. 259:R724-R728.
Lyon, R. E. 1986. Helminth parasites of six lizard
species from southern Idaho. Proc. Helm. Soc. Wash.
53:291-293.
Margolis, L., Esch, G. W., Holmes, J. C., Kuris, A. M.,
and Schad, G. A. 1982. The use of ecological terms in
parasitology (report of an ad hoc committee of the
American Society of Parasitologists). J. Parasit.
68:131-133.
146
Mayr, E. 1983.
How to carry out the adaptationist
program? Am. Nat. 121:324-334.
McCarthy, D. 0., Kluger, M. J. and Vander, A. J. 1985.
Suppression of food intake during infection: Is
interleukin-1 involved? Am J. Clin. Nutr.
42:1179-1182.
Mech, L. D. 1966. The wolves of Isle Royale. Fauna of
the national parks of the U. S., Fauna series, 7.
Washington, U. S. Govt. printing office
Michel, J. F. 1955. Parasitological significance of
bovine grazing behaviour. Nature 175:1088-1089.
Minchella, D. J. 1985. Host parasite variation in
response to parasitism. Parasit. 90:205-216.
Milinski, M. and Bakker, T. C. M. 1990. Female
sticklebacks use male coloration in mate choice and
hence avoid parasitized males. Nature 344:330-333.
Moller, A. P. 1991. Parasite load reduces song output in
a passerine bird. Anim. Behv. 41:723-730.
147
Moller, A. P. 1990. Effects of a haematophagous mite on
the barn swallow (Hirundo rustica): a test of the
Hamilton and Zuk hypothesis. Evolution 44:771-784.
Moller, A. P. 1988. Female choice selects for male
sexual tail ornaments in the monogamous swallow.
Nature 332:640-642.
Moore, J. 1983. Responses of an avian predator and its
isopod prey to an acanthocephalan parasite.
Ecology
64:1000-1015.
Moore, J. 1984. Altered behavioral responses in the
intermediate host - an acanthocephalan parasite
strategy. Am. Nat. 123:572-577.
Noble, G. K. 1934. Experimenting with the courtship of
lizards. Nat. Hist. 34:3-15.
Norris, K. S. 1967. Color adaptation in desert reptiles
and its thermal relationships. In: Lizard ecology: a
symposium. Ed.: Milstead, W. W. University of
Missouri Press. Columbia.
148
Parker, C. D. and Schneider, D. R. 1981. Microorganism
adaptation to host defenses. In Infection: The
physiologic and metabolic responses of the host. Ed.:
Powanda, M. C. and Canonico, P. G. Elsevier/NorthHolland Biomedical Press. London
Poulin, R., Rau, M. E. and Curtis, M. A. 1991. Infection
of brook trout fry, Salvelinus fontinalis, by
ectoparasitic copepods: the role of host behavior and
initial parasite load. Anim. Behv. 41:467-476.
Quinn, S. A., Brooks, R. J. and Cawthorn, R. J. 1987.
Effects of the protozoan parasite Sarcocystis
rauschorum on open-field behaviour of its
intermediate vertebrate host, Dicrostonyx
richardsoni. J. Parasit. 73:265-271.
Pagel, M. D. and Harvey, P. H. 1988. Recent developments
in the analysis of comparative data. Quart. Rev.
Biol. 63:413-440.
Pan, C. 1963. Generalized and focal tissue response in
the snail, Australorbus glabratus, infected with
Schistosoma mansoni. Ann. N. Y. Acad. Sci.
113:475-485.
149
Pan, C. 1965. Studies on the host-parasite relationship
between Schistosoma mansoni and the snail
Australorbus alabratus. Am. J. Trop. Med. Hyg.
14:931-976.
Pearce, R. C. and Tanner, W. W. 1973. Helminths of
Sceloporus lizards in the great basin and upper
colorado plateau of Utah. Great Basin Nat.
33:1-18.
Petranka, J. W., Kats, L. B. and Sih, A. 1987. Predatorprey interactions among fish and larval amphibians:
use of chemical cues to detect predatory fish.
Anim. Behv. 35:420-425.
Pomiankowski, A. 1989. Choosing parasite-free mates.
Nature 338:115-116.
Porat, R., Clark, B. D. Wolff, S. M. and Dinarello, C.
A. 1991. Enhancement of growth of virulent strains of
Escherichia coli by interleukin-1. Science
254:430-432.
150
Pruett-Jones, S. G., Pruett-Jones, M. A. and Jones, H.
I. 1990. Parasites and sexual selection in birds of
paradise. Am. Zool. 30:287-298.
Rake, G. 1933. The presence of meningococcus
precipitinogens in the cerebrospinal fluid. J. Exp.
Med. 58:375-383.
Rau, M. E. 1983. The open field behavior of mice
infected with Trichinella spiralis. Parasit.
86:311-318.
Rau, M. E. 1984.
The open field behavior of mice
infected with Trichinella pseudospiralis. Parasit.
88:415-419.
Read, A. F. 1987. Comparative evidence supports the
Hamilton and Zuk hypothesis on parasites and sexual
selection. Nature 327:68-70.
Read, A. F. and Harvey, P. H. 1989a. Reassessment of
comparative evidence for Hamilton and Zuk theory on
the evolution of secondary sexual characters. Nature
339:618-620.
151
Read, A. F. and Harvey, P. H. 1989b. Validity of sexual
selection in birds (A Reply). Nature 340:105.
Read, A. F. and Weary, D. M. 1990 Sexual selection and
the evolution of bird song: A test of the Hamilton
and Zuk hypothesis. Behay. Ecol. Sociobiol. 26:47-56.
Regal, P. M. 1966. Thermophilic responses following
feeding in certain reptiles. Copeia 1966:588-590.
Ressel, S. and Schall, J. J. 1989. Parasites and showy
males: malarial infection and color variation in
fence lizards. Oecologia 78:158-164.
Richards, C. M. 1958. The inhibition of growth in
crowded Rana pipiens tadpoles. Physiol. Zool.
31:138-151.
Richards, C. M. 1962. The control of tadpole growth by a
alga-like cells. Physiol. Zool. 35:285-296.
Ritcey, R. W. and Edwards, R. Y. 1958. Parasites and
diseases of the Wells Grey moose herd. J. Mammal.
39:139-145.
152
Rossignol, P. A., Ribeiro, J. M. C. and Spielman, A.
1984. Increased intradermal probing time in
sporozoite - infected mosquitoes. Amer. J. Trop. Med.
Hyg. 33:17-20.
Rossignol, P. A. and Rossignol, A. M. 1988. Simulations
of enhanced malaria transmission and host bias
induced by modified vector blood location behavior.
Parasit. 97:363-372.
Rothschild, M. 1962. Changes in behaviour in the
intermediate hosts of nematodes. Nature
193:1312-1313.
Schall, J. J. 1983. Lizard malaria: cost to a vertebrate
host's reproductive success. Parasit. 87:1-6.
Schall, J. J. 1986. Prevalence and virulence of a
haemogregarine parasite of the Aruban whiptail
lizard, Cnemidophorus arubensis. J. Herpetol.
20:318-324.
153
Schall, J. J. and Dearing, M. D. 1987. Malarial
parasitism and male competition for mates in the
western fence lizard, Sceloporus occidentalis. Copeia
1987:84-93.
Schall, J. J. and Sarni, G. A. 1987. Malarial parasitism
and the behavior of the lizard, Sceloporus
occidentalis. Copeia 1987:84-93.
Schwartz, A. and Henderson, R. W. 1985. A guide to the
identification of the amphibians and reptiles of the
West Indies exclusive of Hispaniola. Milwaukee Public
Museum.
Shiff, C. J. 1964. Studies on Bulinus (Physopsis)
globosus in Rhodesia. 1. The influence of temperature
on the intrinsic rate of natural increase. Ann.
Trop. Med. Parasit. 58:94-105.
Shiff, C. J. and Husting, E. L. 1966. An application of
the concept of intrinsic rate of natural increase to
studies on the ecology of freshwater snails of the
genera Biomphalaria and Bulinus (Physopsis).
Proc. Trans. Rhod. Sci. Assoc. 51:2-8.
154
Shinn, G. L. 1985a. Reproduction of Anoplodium hymanae,
a turbellarian flatworm (Neorhabdocoela, Umagillidae)
inhabiting the coelom of sea cucumbers; production of
egg capsules, and escape of infective stages without
evisceration of the host. Biol. Bull.
169:182-198.
Shinn, G. L. 1985b. Infection of new hosts by Anoplodium
hymanae, a turbellarian flatworm (Neorhabdocoela,
Umagillidae) inhabiting the coelum of the sea
cucumber Stichopus californicus. Biol. Bull.
169:199-214.
Shoham S., D. Davenne, A. B. Cady, C. A. Dinarello and
J. M. Krueger. 1987. Recombinant tumor necrosis
factor and interleukin-1 enhance slow wave sleep. Am.
J. Physiol. 253:R142-R149.
Sih, A. 1987. Predators and prey lifestyles: an
evolutionary and ecological overview. In: Predation:
direct and indirect effects on aquatic communities.
Eds.: Kerfoot, W. C. and Sih, A. University Press of
New England. Hanover. pp. 203-224.
155
Sih, A. and Kats, L. B. 1991. Effects of refuge
availability on the responses of salamander larvae to
chemical cues from predatory green sunfish. Anim.
Behv. 42:330-332.
Smith, D. H. 1978. Vulnerability of bot fly (Cuterebra)
infected Peromyscus maniculatis to shorttail weasel
predation in the laboratory. J. Wild. Dis. 14:40-51.
Sodeman, W. A. Jr. and Dowda, M. C. 1974. Behavioral
responses of Biomphalaria glabrata. Physiol. Zool.
47:198-206.
Soulsby, E. J. L., 1987. The evasion of the immune
response and immunological unresponsiveness:
parasitic helminth infections. J. Immun. Letters.
16:315-320.
Spiggs, D. R., M. L. Sherman, H. Michie, K. A. Arthur,
K. Imamura, K. Wilmore, D. Frei, and D. W. Keefe.
1988. Recombinant human tumor necrosis factor
administered as a 24-hour intravenous infusion. A
phase 1 and pharmacological study. J. Nat. Can. Inst.
80:1039-1044.
156
Stadnyk, A. W. and J. Gauldie. 1991. The acute phase
protein response during parasitic infection. Parasit.
Today 7:A7-Al2.
Standen, 0. D. 1952. Experimental infection of
Australorbis glabratus with Schistosoma mansoni. I.
Individual and mass infection of snails, and the
relationship of infection to temperature and season.
Ann. Trop. Med. Parasit. 46:48-53.
Stebbins, R. C. 1954. Amphibians and reptiles of Western
North America. McGraw-Hill. New York.
Steinwascher, K. 1979. Host-parasite interaction as a
potential population-regulating mechanism. Ecology
60:884-890.
Stibbs, H. H., Owczarzak, A., Bayne, C. J., and Dewan,
P. C. 1979. Schistosome sporocyst-killing amebae
isolated from Biomphalaria glabrata. J. Invert.
Path. 33:159-170.
Stirewalt, M. A. 1954. Effect of maintenance
temperatures on development of Schistosoma mansoni.
Exp. Parasit. 3:504-516.
157
Sturrock, B. M. and Sturrock, R. F. 1970. Laboratory
studies of the host-parasite relationship of
Schistosoma mansoni and Biomphalaria glabrata from
St. Lucia, West Indies. Ann. Trop. Med. Parasit.
64:357.
Sturrock, R. F. and Sturrock, B. M. 1971. Shell
abnormalities in Biomphalaria glabrata infected with
Schistosoma mansoni and their significance in field
transmission studies. J. Helm. 45:201-210.
Sturrock, R. F. and Sturrock, B. M. 1972. The influence
of temperature on the biology of Biomphalaria
glabrata (Say), intermediate host of Schistosoma
mansoni on St. Lucia, West Indies. Ann. Trop. Med.
Par. 66:385-390.
Taylor, R. J. 1954. Grazing behaviour and helmintic
disease. Br. J. Anim. Behay. 2:61-62.
Tekwani, B. L., Tripathi, L. M., Mukerjee, S., Misha,
A., Shukla, 0. P. and Ghatak, S. 1987. Impairment of
the hepatic-microsomal drug-metabolizing system in
rats parasitized with Nippostrongylus brasiliensis.
Biochem. Pharm. 36:1383-1386.
158
Telford, S. R. Jr. 1970. A comparative study of
endoparasitism among some Southern California lizard
populations. Amer. Midl. Nat. 83:516-533.
Thompson, S. N. 1990. Physiological alterations during
parasitism and their effects on host behavior.
In:
Parasitism and host behaviour Eds.: Barnard, C. J.
and Behnke, J. M. Taylor and Francis. London.
pp. 64-94.
Tinsley, R. C. 1989. The effect of host sex on
transmission success. Parasit. Today 5: 190-195.
Titus, R. C., B. Sherry, and A. Cerami. 1991. The
involvement of TNF, IL-1 and IL-6 in the immune
response to protozoan parasites. Parasit. Today
7:A13-A16.
Van Der Knaap, W. P. W. and Loker, E. S. 1990. Immune
mechanisms in trematode - snail interactions.
Parasit. Today 6:175-182.
159
Walls, S. 1991. Mechanisms of coexistence in a guild of
Ambystomatid salamanders: the importance of stagedependant regulation in species with a complex life
cycle. Ph.D. thesis. Univ. Southern Louisiana.
Lafayette Louisiana.
Waitz, J. A. 1961. Parasites of Idaho reptiles. J.
Parasit. 47:51
Ward, P. I. 1988. Sexual dichromatism and parasitism in
British and Irish freshwater fish. Anim. Behv.
36:1210-1215.
Ward, P. I. 1989. Sexual showiness and parasitism in
freshwater fish: combined data from several isolated
water systems. Oikos 55:428-429.
Weatherhead, P. J. 1990. Secondary sexual traits,
parasites and polygyny in red-winged blackbirds,
Actelaius phoeniceus. Behay. Ecol. 1:125-130.
Webbe, G. and James, C. 1972. Host-parasite
relationships of Bulinus globosus and B. truncatus
with strains of Schistosoma haematobium. J. Helm.
46:185-199.
160
Williams, G. C. and Nesse, R. M. 1991. The dawn of
Darwinian medicine. Quar. Rev. Biol. 66:1-22.
Wilson, D. J. and H. Lefcort. The effect of predator
diet on the alarm response of red-legged frog (Rana
aurora) tadpoles. Anim. Behv. In Press.
Wing, E. J. and Young, J. B. 1980. Acute starvation
protects mice against Listeria monocytogenes. Infec,
Immun. 28:771-776.
Wright, S. 1949. Adaptation and natural selection. In:
Genetics, paleontology, and evolution. Eds.: Jepsen,
G., Mayr, E. and Simpson, G. G. Princeton Univ.
Press. Princeton N. J. pp 365-389.
Zahavi, A. 1976.
Mate selection
a selection for a
handicap. J. Theor. Biol. 53:205-214.
Zohar, A. S. and Row, M. E. 1986. The role of muscle
larvae of Trichinella spiralis in the behavioral
alterations of the mouse host. J. Parasit.
72:464-466.
161
Zuk, M. 1987. The effects of gregarine parasites, body
size, and time of day on spermatophore production and
sexual selection in field crickets. Behay. Ecol.
Sociobiol. 21:65-72.
Zuk, M. 1988. Parasite load, body size, and age of wildcaught male field crickets (Orthoptera: Gryllidae):
effects on sexual selection. Evol. 42:969-976.
Zuk, M. 1989. Validity of sexual selection in birds.
Nature 340:104-105.
Zuk, M. 1990. Parasites and mate choice in red jungle
fowl. Am. Zool. 30:235-244.