|
In North America, three main populations of monarch
butterflies (Danaus plexippus) are infected by the neogregarine
protozoan parasite Ophryocystis elektroscirrha but show very different
prevalence. To test whether climatic variations could explain those
differences, we reared larvae exposed to cold, moderate and hot
temperatures and increasing inoculation dose of the parasite. We
observed that hot temperatures led to a significant decrease of
parasite load for every level of infection. At the same time, larvae
from the cold treatment were less harmed by the infection with higher
survival to eclosion and fewer deformities. These results suggest
that temperature is an important variable in dynamic of this host-pathogen
system but its effect alone is not sufficient to explain the differences
in prevalence among wild populations of monarchs.
Monarch butterflies are spread world-wide and all
populations examined to date harbor the parasite O. elektroscirrha.
The development of this parasite is highly related to the metamorphic
cycle of the host, which goes through a larval, pupal and adult
stage like all Lepidoptera (Figure 1).
This parasite is carried by the adults as dormant
spores on their abdomen and wings. Adults spread their spores to
other adults but only larvae can be infected when they ingest some.
Ingested spores are activated and replicate in the larvae hypoderm
during larval and pupal stage. At eclosion, the adult is covered
by hundreds of thousands spores on its abdomen and wings. Infection
by this parasite increases larval mortality and adult fitness, as
they can be deformed after eclosion and have decreased life expectancy
and wingspan (Figure 2).
Monarch butterflies can be found in three major areas
in North America (Figure 3).
Each of them present a different prevalence of the
parasite: - The eastern population breeds in a temperate climate
and migrates long distances to overwinter in central Mexico. Only
8% of the population is heavily infected - The western population
also migrates a shorter distance and 30% of adults are infected.
- The South Florida population does not migrate; monarchs live in
tropical conditions and 80% of them are heavily infected by this
parasite. Questions: What factors might explain these differences
in prevalence among wild populations? Three hypothesis to explore:
(1) Role of migratory behavior: as the butterflies who migrate the
farthest are less infected, selection during migration could reduce
parasite spread. (2) Role of genetic diversity: parasite strains
from South Florida could be more virulent and monarchs from migratory
populations more resistant to parasite infection. (3) Role of climate:
temperature and other local environmental factor could benefit to
the infection in South Florida. Objective of this study: To test
the effect of temperature on the monarch-parasite interaction.
Monarchs larvae were distributed into 3 temperature
treatments and 4 inoculation dose treatments (Table 1).
Larvae were obtained by crossing South-Florida freshly
collected males with eastern and South-Florida female monarchs,
after the spore checking showed they were uninfected. After inoculation
with calibrated doses of a South Florida strain of O. elektroscirrha,
larvae were reared in individual containers in growth chambers programmed
with a day-night cycle (light and temperature fluctuation) and fed
daily (Figure 3 and 4).
Data analyzed: we measured the following variables:
- Development time from inoculation to pupation and to eclosion.
- Infection level and spore load (using a standardize method to
remove scales and spore from monarch abdomens and count them by
image processing with Adobe Photoshop - Figure 5).
- Levels of deformity: Any sign of wing deformity
or failure to emerge by themselves was recorded among adults
- Adult lifespan and mass: after emergence, adults were removed
from the treatment. Their mass was recorded 24 hours after eclosion
and their lifespan (in days) was also reported.
All the data results were tested by Analysis of Variance (continuous
variable) or Logistic Regression (binary variables) tests to evaluate
the significance of the hypotheses.
Spore load (Figure 6):
- Spore load decreased with increasing inoculation doses. This effect
was due to difficulty in removing scales and spores from highly
infected adults, which had damaged abdomens.
- Spore load decreased with increasing temperature. This significant
result (P<0.001) was observed throughout all inoculation treatments.
Adult lifespan (Figure 7):
- Adult lifespan decreased with increasing inoculation doses. In
treatments 3 and 4, adults did not survive more than 2 or 3 days.
- In the low dose treatment, hot temperature favored survival of
monarchs after eclosion (P<0.001 - shown by the star on figure
7). But no global effect of temperature was observed in other treatments.
Deformity (Figure 8):
- Wing deformity increased with greater inoculation doses.
- For the medium dose treatment, low temperature limited levels
of deformity. But this moderating effect was not observed in the
other dose treatments.
Other results (Table 2):
Adult mass showed a decrease with increasing infection but no effects
of temperature. Results showed a very high proportion of infected
adults in comparison with previous experiments, suggesting the parasite
from South Florida may be more virulent. The effect of temperature
on development time of insects is well-known and a faster development
was observed in high temperatures, as expected. But infection also
slowed development times.
Effect of temperature: what we did not know before.
- Hot temperatures caused a 30% reduction on spore loads.
- Monarch deformities when inoculated with moderate doses were lower
in the coldest treatments.
- Monarchs reared in hot temperatures survived longer to a low dose
infection
This experiment showed that temperature has important effects
on the development of this parasite.
However,
- The results show that high temperature lead to a decrease of the
infection level: geographic differences in prevalence cannot be
explained by this single effect. Hypothesis (3) is not the right
explanation.
- Monarch fitness is strongly affected by the parasite, reinforcing
the relevancy of hypothesis (1).
- The high virulence of the parasite from South Florida supports
the idea that genetic differences may explain some differences in
the wild (hypothesis (2)). The mechanism of this system is more
complex than expected, with a probable trade-off between several
effects. Further investigations and modeling will be useful to be
able to explain the current and future dynamic of this host-parasite
system and foresee the exact consequences of global warming.
We thank Andrew Davis for his critical review of this work, advice,
suggestions and assistance. We also thank Nick Vitone and Zack Baumann
for their helpful laboratory assistance. We thank Les Real from
the Biology department for the loan of the greenhouse facilities.
We thank the SURE program of Emory University and the Howard Hughes
Medical Institute for their financial support. We thank the Ecole
Polytechnique for offering to Hugo Valin the opportunity to take
part in this project.
In North America, three main populations of monarch butterflies
(Danaus plexippus) are infected by a parasite but show very different
proportions of infected adults. To test whether climatic variations
could explain those differences, we reared larvae exposed to cold,
moderate and hot temperatures and increasing inoculation dose of
the parasite. We observed that hot temperatures led to a significant
decrease of parasite load for every level of infection. At the same
time, larvae from the cold treatment were less harmed by the infection
with higher survival to eclosion and fewer deformities. These results
suggest that temperature is an important variable in dynamic of
this host-pathogen system but its effect alone is not sufficient
to explain the differences in prevalence among wild populations
of monarchs.
Larvae rearing In environmental chambers, spore count standardized
method.
|
