Differential host mortality explains the effect of high temperature on the prevalence of a marine pathogen
October
Differential host mortality explains the effect of high temperature on the prevalence of a marine pathogen
Timothy J. Sullivan¤ 0 1 2 3
Joseph E. Neigel 0 1 2 3
0 Department of Biology, the University of Louisiana at Lafayette , Lafayette, Louisiana , United States of America
1 Funding: This research was funded by grants from The Gulf of Mexico Research Initiative (gulfresearchinitiative.org) (GRI-013 and GoMRI2012-II-523) awarded to JEN and by a University of Louisiana Doctoral Fellowship to TJS. The funders had no role in the study design , data , USA
2 Editor: Torsten Thomas, University of New South Wales , AUSTRALIA
3 Current address: Department of Biological Sciences, University of Arkansas , Fayetteville, Arkansas , United States of America
Infectious diseases threaten marine populations, and the extent of their impacts is often assessed by prevalence of infection (the proportion of infected individuals). Changes in prevalence are often attributed to altered rates of transmission, although the rates of birth, recovery, and mortality also determine prevalence. The parasitic dinoflagellate Hematodinium perezi causes a severe, often fatal disease in blue crabs. It has been speculated that decreases in prevalence associated with high temperatures result from lower rates of infection. We used field collections, environmental sensor data, and high-temperature exposure experiments to investigate the factors that change prevalence of infections in blue crab megalopae (post-larvae). These megalopae migrate from offshore waters, where temperatures are moderate, to marshes where temperatures may be extremely high. Within a few days of arriving in the marsh, the megalopae metamorphose into juvenile crabs. We found a strong negative association between prevalence of Hematodinium infection in megalopae and the cumulative time water temperatures in the marsh exceeded 34ÊC over the preceding two days. Temperatures this high are known to be lethal for blue crabs, suggesting that higher mortality of infected megalopae could be the cause of reduced prevalence. Experimental exposure of megalopae from the marsh to a temperature of 34ÊC resulted in higher mortality for infected than uninfected individuals, and decreased the prevalence of infection among survivors from 18% to 3%.
Introduction
Outbreaks of infectious diseases in marine animals are of growing concern. Pollution,
nonnative pathogens, and climate change are generally expected to increase the frequency and
severity of marine disease outbreaks and their effects on ecosystems, fisheries, and threatened
species [1±4], although some research has suggested these effects will be taxon-specific [
5, 6
].
Models that accurately predict outbreaks are needed to develop strategies for limiting their
impacts [7±9]. However, well-developed epidemiological models for human and wildlife
collection and analysis, decision to publish, or
preparation of the manuscript.
populations are not easily applied to marine systems [
1, 3, 10
]. The parameters of these models
represent processes that are difficult to observe or quantify in the marine realm, such as disease
transmission, disease-induced mortality, and recovery from infection [
2, 3, 10
]. In some
cases, basic etiological aspects of a disease, such as the identity of the pathogen or its mode of
transmission, are unknown [3, 11±13]. Consequently, efforts to identify the causes of marine
outbreaks and predict their occurrences have focused on finding associations between
environmental factors, such as temperature and salinity, and disease prevalence, the proportion of
infected or diseased individuals in a population [14±20]. However, the use of prevalence, as an
indicator of the progress or severity of an outbreak is potentially misleading. The rates of
multiple processes, including infection, recovery from infection, birth of new individuals, and the
deaths of both infected and uninfected individuals jointly determine disease prevalence [Fig 1,
21, 22]. Environmental factors could influence any of these processes, so that correlations
between environmental variables and disease prevalence can arise in different ways with
different implications for host populations.
The parasitic dinoflagellate Hematodinium perezi is a widespread and virulent pathogen of
the blue crab, Callinectes sapidus. In laboratory experiments, mortality of infected adult blue
crabs has ranged from 87 to 100% [
23
]. Prevalence of infection along the US Atlantic coast is
often as high as 30% and may approach 100% in severe outbreaks [
24
]. Prevalence of infection
by H. perezi in blue crabs is correlated with both water temperature and salinity [
14, 24
].
Infected blue crabs are rarely found at salinities below 18 ppt and have not been reported at
salinities below 11 ppt [
15, 24
]. There is evidence that low salinity reduces prevalence by
Fig 1. SI model of prevalence. Boxes represent groups that differ in disease status, with n (...truncated)