RESEARCH

Our research focuses on the question why parasites harm their hosts. Intuitively, parasites that depend on their hosts for their survival should be benign to their hosts, yet many parasites cause harm (i.e. they cause virulence). Our research is aimed at understanding the ecological forces that shape the evolution of virulence, and currently focuses on two major lines of research.

1. Ecology and evolution of monarch butterfly parasites

Monarch butterflies are best known for the population in eastern North America that undergoes a spectacular and long distance (thousands of km) annual migration to overwinter in the oyamel forests of central Mexico. However, monarchs also occur in western North America, where they migrate to overwinter along the Californian coast, and in many populations where they do not migrate, such as Hawaii, South Florida and Central America. Monarchs are less known for the diseases that they carry, yet monarch diseases can be widespread. A common parasite of monarchs is the protozoan Ophryocystis elektroscirrha. This parasite is relatively rare in eastern North America (infecting between 5 and 10% of monarchs), but can reach very high prevalence (over 80%) in other populations. On the right: monarchs take to the sky in Mexico.

O. elektroscirrha forms spores on the outside of the monarch butterfly. When female monarchs lay eggs, they drop some of these spores onto their eggs and onto the milkweed plants on which they lay their eggs. When the caterpillar hatches, it eats the egg shell and the milkweed leaf, and thereby ingests the spores. These will break open in the monarch gut, to release sporozoites that travel through the midgut wall to infect the monarch tissues. During larval and pupal development, the parasites replicate to produce millions of spores on the adult butterfly when it emerges from its pupal case. We have shown that a single parasite spore is enough to cause an infection, and that the parasites from this single spore can produce over a million new spores on the adult butterfly.

This picture shows the parasite spores (14 μm long) on top of the monarch butterfly scales.

We study monarchs and their parasites to understand: (1) why parasites evolve to cause disease to their hosts; (2) how ecological factors can select for more or less dangerous parasites; and (3) how different parasites can affect each other’s prevalence in nature. We are currently addressing the following questions.

1.1 What are the effects of the parasite on the monarch butterfly?

We have shown that the parasite can cause severe disease to the monarch. In the worst case, the high numbers of parasite spores disrupt the monarch abdomen so badly that the monarch gets stuck in its pupal case and is unable to emerge (see picture on the right). Monarchs that do emerge successfully are smaller, live shorter, and lay fewer eggs over their lifetime. The monarch on the right is stuck to its pupal case and, defenseless as it is, gets attacked by a wasp.

1.2 Why do parasites harm their hosts?
Obviously, the disease caused by the parasite is bad for the host, but it may also be bad for the parasite. This is because the parasite needs the host to survive at least long enough to mate and start laying eggs, during which it is transmitted to the next generation of monarchs. By killing its host, reducing the chance that monarchs will start laying eggs, and reducing the numbers of eggs laid by the monarch, the parasite appears to shoot itself in the foot. Much of our work has gone into understanding why the parasite causes this disease despite the detrimental effects on its own fitness. Our studies suggest that the answer is that the parasite needs to produce these high numbers of spores to be able to transmit from infected monarchs to their offspring. Monarchs with higher numbers of spores do transfer more spores to their eggs and milkweed. Mathematical analyses on our data suggest that parasites that produce intermediate levels of spores obtain greatest lifetime transmission by balancing the costs of spore production (detrimental effects on the host) with its benefits (greater transmission). This suggests that the parasite has evolved its virulence as a consequence of natural selection on transmission, and supports a popular evolutionary theory for the evolution of parasite virulence (known as the trade-off model).

Increasing parasite spore loads result in lower proportions of monarchs surviving to the adult stage (a) and lower mating probabilities (b). Higher parasite loads also led to increased transmission through higher proportions of monarch eggs that acquired spores (c), and higher numbers of parasites per egg and milkweed leaf (d, f); these higher parasite numbers increase the probability of infection (e, g).

1.3 Do monarch populations differ in parasite incidence?
Although the parasite occurs in all monarch populations studied to date, its incidence varies among populations. For example, fewer monarchs are infected in the eastern North American migratory population (~5-10%) than in the western North American population (~15-30%); and in South Florida, almost all monarchs are infected (>80%). By monitoring these populations over the years, we have shown that these differences seem consistent over time (although the year-to-year incidence varies). One hypothesis that we are testing to explain these patterns is that monarch migration can select against high parasite prevalence.
	Parasite incidence in eastern monarchs that migrate to Mexico and western monarchs that migrate to California.

Because the parasite shortens the monarch’s lifespan the idea is that the cost of infection is higher for monarchs in populations that migrate long distances than for those that do not migrate, simply because migrating monarchs need to survive longer to complete the migration and to survive the winter. As such, migration may “weed out” infected monarchs and thereby reduce parasite incidence. We are testing this hypothesis by studying whether a greater proportion of monarchs are infected before than after their migration, and whether parasite incidence declines during the overwintering season. We do this work in collaboration with Sonia Altizer’s lab at the University of Georgia.

1.4 Do parasites in different populations vary in disease severity?
Apart from studying differences in parasite incidence, we also study whether parasites from different populations cause different levels of disease, and whether monarchs differ in their resistance to infection. So far, we have shown that parasite genotypes vary in the disease they cause, and that parasites from the western population are on average more virulent than those from the eastern population. Hosts from the two populations did not differ in their resistance.

Parasites obtained from the western and eastern North-American monarch populations varied in virulence, as indicated by the lifespan of infected monarchs; western parasites were on average more virulent than eastern parasites in both western (a) and eastern (b) hosts.

1.5 Do larval host plants affect the parasitic disease?
Monarch larvae feed on milkweed plants, which contain toxic chemicals (cardenolides). Monarchs sequester these chemicals into their own bodies and thereby make themselves toxic to predators such as birds. We are studying whether milkweeds also affect the monarch’s parasites. So far, we have done studies in which we reared monarchs on two different milkweed species: Asclepias incarnata (swamp milkweed) and A. curassavica (tropical milkweed). Our results show that the parasite produces fewer spores in monarchs reared on A. incarnata than in monarchs reared on A. curassavica. Consequently, monarchs reared on A. curassavica did not get as sick and lived longer. We are currently testing whether cardenolides affect the parasite directly, or whether monarchs reared on different milkweeds become more or less immune to infection. We also study whether monarchs are able to self-medicate by choosing host plants that make them less sick, and whether other milkweed herbivores, such as milkweed aphids, can affect monarch parasites through their effects on the host plant. We do this work in collaboration with Mark Hunter’s lab at the University of Michigan.
	Monarchs were reared on two different host plant species and infected with four different O. elektroscirrha strains. Monarchs lived longer when reared on A. incarnata than A. curassavica, a milkweed species with more toxic cardenolides.

1.6 Are parasites locally adapted to their hosts and larval host plants?   
Many evolutionary studies have shown that parasites become locally adapted to their hosts, such that they obtain greater fitness from the hosts they encounter in nature than from hosts that occur in other populations. We collect hosts and parasites from many different populations (including North America, Hawaii, Australia and Costa Rica), and use laboratory tests to study whether monarch parasites are locally adapted to their monarch hosts. We are also developing molecular tools to study the co-phylogeny of hosts and parasites. Because milkweeds importantly affect the fitness of host and parasite, one of our major hypotheses is that parasites are locally adapted to monarch larval host plants. In order to test this, we will determine which species of milkweed are used by monarchs in different populations and test whether the parasites in each population obtain highest fitness on monarchs reared on the species from that particular population.

1.7 How do parasitoid flies affect host and parasite fitness?
Monarchs have many more enemies than O. elektroscirrha. One of these is the fly Lespesia archippivora, which lays eggs on the monarch caterpillar. The fly maggots then hatch from their eggs and eat their way into the monarch caterpillar, in which they will grow. During the late monarch larval instar or early pupal stage, the maggots eat their way out, killing the monarch in the process. We study how co-infection of monarchs by both parasites affects the fitness of the host and each of the parasites. As part of this we are trying to find out what immune responses monarchs use against both parasites. We will also study whether the parasites affect each other’s prevalence in wild populations.

An adult parasitoid fly	Three fly maggots ate their way out of this monarch caterpillar.

2. Within-host ecology of malaria parasites and its consequences for the evolution of parasite virulence

Although not generally known to most people, infections often consist of multiple genotypes of the same parasite species. Such multiple infections could result in competition between parasites, and could have important implications for the evolution of virulence. We use the rodent malaria parasite Plasmodium chabaudi in laboratory mice to understand how such competition shapes the evolution of virulence.

In collaboration with Andrew Read's lab (University of Edinurgh) we have shown that more virulent parasites outcompete less virulent parasites in mixed-strain malaria infections and obtain higher transmission to the mosquito host, thereby implying that high virulence can evolve as a consequence of within-host competition. We have also shown that the extent of competition depends on the genetic background of the host, the order in which competing parasite strains infect their host, and the presence or absence of antimalarial drugs.

Currently we are collaborating with Rustom Antia (here at Emory) to combine mathematical models with experimental studies to study the ecological mechanisms that cause within-host competition and hence malaria virulence evolution. In particular we are trying to find out whether competition between malaria parasites is caused by a lack of resources (such as red lood cells) or determined by the host's immune system. We expect that this work will not only shed light on competition between parasites in mixed infections, but also on the more fundamental question of how and why malaria parasites cause disease. Such insights are not only important to understand malaria virulence, but also to predict how human interventions such as vaccination and blood transfusion could affect subsequent virulence evolution.