Ecology, Epidemiology, and Evolution of Parasitism in Daphnia

  1. Introduction to the Ecology, Epidemiology, and Evolution of Parasitism in Daphnia
  2. Introduction to Daphnia Biology
  3. Some Parasites of Daphnia
  4. Parasitism in Natural Populations
  5. The Effects of Daphnia Parasites on Host Fitness
  6. Host Adaptations against the Costs of Parasitism
  7. Host Range of Daphnia Parasites
  8. Epidemiology
  9. Population Dynamics and Community Ecology
  10. Experiments with Daphnia and Parasites

5 The Effects of Daphnia Parasites on Host Fitness

Parasites use their hosts to foster their own needs, thus interfering with the hosts’ survival and reproduction needs and creating a conflict of interest. In this chapter, I describe what is known about the damage that parasites inflict on Daphnia. It has been shown that many parasite infections reduce host fecundity and survival. Parasites may also influence other host fitness components, such as predator escape, body size, and sex allocation. Some parasites show specialized modes of action, such as castration or the induction of enhanced body growth. The degree to which parasites damage their hosts varies greatly among parasite and host species, parasite and host genotypes, and also depends on the interaction between the two. Environmental factors, such as temperature and feeding conditions, also play a role in the expression of disease symptoms.

  1. Introduction
  2. Effects on Host Fecundity and Survival
    1. Environmental Effects
      1. Food Effects
      2. Temperature Effects
      3. Chemical Cues from Predators
      4. Dose Effects
    2. Genetic Effects
      1. Genetic Variation among Hosts and Parasites
      2. Genetic Variation across Populations and Local Adaptation
  3. Parasite Effects on Other Host Traits
  4. Parasites May Influence Predation on Their Hosts
  5. Conclusions and Open Questions

5.1 Introduction

Part of the standard definition of parasitism is that parasites harm their hosts. As mentioned above, a number of field studies have shown that parasitized females often have reduced fecundity as compared with healthy (i.e., not parasitized) females. However, field data for some parasites have not revealed significant effects. Large environmental noise in the data and rather small parasite effects may render tests insignificant. Furthermore, if the host population is already in poor health (low food levels may also reduce the female’s ability to carry eggs) or, alternatively, in very good health, the effect of the parasite may not easily be visible. Thus, it is not surprising that the apparent effect of parasites on host fitness varies if the same analysis is repeated in time or space (Yan and Larsson 1988; Bengtsson and Ebert 1998). Laboratory studies can reveal effects much more easily.

Because field studies usually cannot exclude the possibility that parasites infect hosts already weakened by other factors, such as poor nutrition, injuries, and inbreeding, their results must be considered with caution. Because laboratory experiments have demonstrated the clear fecundity costs of parasitism (see below), these confounding factors are unlikely to explain the bulk of the data. However, we need to be cautious when comparing field data across time, space, or species, because they are unlikely to reveal good quantitative data on parasite virulence.

The first attempts to demonstrate the effects of parasites under laboratory conditions used material from natural populations that had been brought to the laboratory for further observation (Green 1974; Brambilla 1983). Although these studies were able to observe differences between infected and uninfected females, they were not able to exclude various confounding factors. The infected and the (apparently) uninfected females may have differed in life history traits (e.g., age or size) or may have already been in different conditions when they became infected. By the time infected animals were collected, the ages of their infections were also different. Although I do not believe that these confounding factors are highly critical when demonstrating some negative effect of parasites on host fecundity, they certainly interfere with testing the effects of the parasites on survival (see Chapter 3). Furthermore, with field-caught animals, one cannot quantitatively determine the strength of the effects. Thus, such experiments are not suitable for comparing the effects of parasites across space, time, or species.

A number of studies have attempted to test and quantify the effect of parasitism using proper experimental procedures with random allocation of females to different treatment groups and controlled infections. To my knowledge, every experiment of this sort revealed some negative effect of the parasite on their Daphnia hosts. Unfortunately, not all Daphnia parasites can be easily used for experimentation.

5.2 Effects on Host Fecundity and Survival

The two fitness components that are typically considered with regard to parasitism are host fecundity and survival. For both variables, drastic effects have been observed, and the degree of harm done to the host varies greatly. The costs of parasitism differ not only across parasite species but also among isolates of the same parasite and across environmental conditions (Ebert 1994b; Ebert 1998a, 2000a; Bittner et al. 2002). Figure 5.1

Figure 5.1
shows to what degree parasites differ in the damage they inflict on their host. Currently, the most harmful parasite tested is the White Fat Cell Disease, a bacterial infection in D. magna that severely reduces both host fecundity and survival (Ebert et al. 2000a). On the other end of the spectrum are the microsporidian gut parasites, such as Glugoides intestinalis and Ordospora intestinalis. These common parasites reduce host fitness by only 15% to 20%.

Across the entire range of observed effects, most tested parasites reduced both host fecundity and survival to a similar degree. Thus, parasites that drastically reduce life span also considerably reduce fecundity (fecundity of the living host relative to uninfected hosts of the same age), whereas parasites benign in their effect on survival were also benign in their effect on fecundity. In a first approximation, the reduction of both fecundity and survival may be seen as a general sign of host morbidity. In contrast to this pattern, Pasteuria ramosa shows a different course of infection. This bacterium first castrates its host (around 10 days after infection) but then allows it to live for many more days (over 40 days after infection). It has been speculated that this specific pathology is adaptive for P. ramosa (Ebert et al. 2004). Castrating the host allows Pasteuria to monopolize resources that the host would otherwise invest into reproduction. Early castration results in more parasite transmission stages.

5.2.1 Environmental Effects

Although the harm caused by parasites may depend on the environmental conditions, few studies have tested for environmental effects. Thus, no clear generalizations have emerged thus far. However, environment-dependent or condition-dependent virulence is certainly rather the rule than the exception. Survival and fecundity of Daphnia depend strongly on the abiotic and biotic environment (e.g., food quality and quantity, temperature, host density, presence and density of competitors, kairomones, and toxins), and some of these factors also influence the parasites. Thus, it is likely that these factors also influence the interactions between host and parasite. Food Effects

The dependence of host fitness on the feeding conditions has been well documented for various Daphnia species. Lower food quantity or quality generally reduces fecundity but expands life span. The interaction between parasitic infections and the feeding conditions for the host has not yet been generally determined. Bittner et al. (2002) tested fecundity and survival of Caullerya mesnili-infected D. galeata in low and high food conditions. Although there was no significant difference in the survival of infected hosts, there was a strong effect on fecundity such that C. mesnili harms well-fed D. galeata more than poorly fed D. galeata. Infected D. galeata produced more eggs under low food conditions than under high food conditions. In contrast to the food study in D. galeata, a study on D. magna infected with P. ramosa found that well-fed infected hosts produced more eggs than poorly fed infected hosts (Ebert et al. 2004). Interestingly, the well-fed infected hosts also produced more P. ramosa transmission stages, indicating that good feeding conditions benefit both the host and the parasite. Both antagonists are possibly resource limited. Temperature Effects

Healthy Daphnia mature earlier and at a smaller size and have a shorter life span when growing under conditions of higher temperature. Surprisingly little is known about the influence of temperature for the expression of disease in Daphnia. Duffy et al. (2005) reported anecdotally that D. dentifera infected with Spirobacillus cienkowskii survive longer at lower temperatures. Because usually everything with invertebrates takes longer at lower temperature, this observation may simply be the result of the hosts' and parasites' lower metabolic rates. A more complex relationship between temperature and disease expression was reported by Mitchell et al. (2005). They found that the negative effect of P. ramosa on D. magna fecundity was more benign when the temperature was lower. At a lower temperature, the parasite gained later control over host fecundity. The authors emphasize that this effect weakens parasite-mediated selection during part of the season. Furthermore, this parasite effect interacted both with host genotype and temperature such that clonal ranks in host fitness differed under different temperature conditions. This effect cannot be explained by the temperature dependence of metabolic rates. Altered rank orders of host genotypes may have profound consequences for the evolution of host resistance. However, it is necessary to see these interactions in relation to the main effects and the seasonal dynamics of the disease to judge how evolution will be influenced. Chemical Cues from Predators

Daphnia have been a workhorse for the study of phenotypic plasticity. In particular, their reaction to chemical cues released by predators (i.e., kairomones) has received a lot of attention. Lass and Bittner (2002) tested for interactions between the effects of two antagonists on D. galeata, the protozoan gut parasiteC. mesnili and kairomones from planktivorous fish. They found no evidence for interactions between fish and parasite with regard to host fecundity and survival. Dose Effects

Another environmental effect that influences the harm caused by parasites is the dose of transmission stages to which a host is exposed. Typically, higher doses go hand-in-hand with a higher likelihood of infection and with more severe damage to the host (Ebert 1995; Ebert et al. 2000b; Regoes et al. 2003; Ebert et al. 2004). Very high doses may even harm the host so much that the parasite is not able to complete its development before the host dies (Ebert et al. 2000b).

5.2.2 Genetic Effects Genetic Variation among Hosts and Parasites

Parasite virulence varies across parasite isolates (strains, genotypes) and host clones. To my knowledge, every attempt to test for genetic variation within parasite-induced host damage in the Daphnia system has shown significant effects. Host clones originating from within or between populations differ in the degree with which they express disease symptoms, and parasite isolates vary greatly in the extent to which they cause damage to the same host clones (Ebert 1994a; Ebert 1998a; Little and Ebert 2000; Bittner 2001; Decaestecker et al. 2003). Furthermore, there are strong host clone x parasite isolate interactions: Within populations, the infectivity of P. ramosa depends strongly on the interaction between the Pasteuria and the D. magna genotypes (Carius et al. 2001) (Figure 5.2

Figure 5.2
). The same is true if fecundity reduction is considered among infected females only (Carius et al. 2001). What maintains these high rates of within-population variation is not fully understood, but it has been suggested that antagonistic arms races play a key role in maintaining genetic variation for virulence and resistance (Hamilton 1980; Ebert and Hamilton 1996; Carius et al. 2001). Genetic Variation across Populations and Local Adaptation

Genetic variation for parasite virulence is most pronounced across populations. This variation often follows a certain pattern, which is frequently discussed in the context of local adaptation (Kawecki and Ebert 2004). For four D. magna parasites, it has been shown that local parasite isolates cause more harm to their hosts than parasite isolates from other populations (Ebert 1994b; Ebert 1998a) (D. Refardt and D. Ebert, manuscript in preparation). These findings are consistent with the idea that parasites evolve local adaptation to the hosts they have encountered recently (Figures 5.3

Figure 5.3
and 5.4). Often (but not always) parasites that perform better in their local host than other foreign (or novel) parasites also perform better in their local hosts than in other hosts (Figures 5.3 and 5.5). Locally adapted parasites show not only higher levels of damage to their local hosts but also have higher levels of transmission-stage production (Ebert 1994b).

The finding of parasite local adaptation seems rather general in Daphnia systems but is not always found in other host–parasite systems. Some authors reported that hosts, rather than parasites, can be locally adapted (Morand et al. 1996; Kaltz and Shykoff 1998; Kaltz et al. 1999). It has been suggested that the key variable for the evolution of host or parasite local adaptation is the relative speed of evolution of the two antagonists (Gandon et al. 1996, 1997; Gandon 2002). Higher rates of mutation, recombination, and dispersal may facilitate local adaptation. Given these theoretical considerations and the finding that Daphnia parasites seem to be locally adapted, one may speculate that parasites of Daphnia usually have a higher evolutionary potential than their hosts.

A different approach to host–parasite interactions across populations is the question of how much a dispersing host suffers when it encounters a locally adapted parasite in a novel population. Note that this question is different from the question about parasite local adaptation. Kawecki and Ebert (2004) explain these differences in full detail. If parasites are locally adapted and thus cause more harm to their local hosts, a host that migrates into such a population should, one expects, suffer less on average from the local parasites than the local hosts. This observation has been reported in several experiments (Ebert 1994b; Ebert et al. 1998; Altermatt 2004). It is important to note that although this pattern is found when averaging across several host–parasite combinations, occasionally a host in a novel combination is much more affected by the new parasites than expected (Ebert 1994b). These instances are likely to be exceptions, but they may have profound consequences, because they may be the beginning of a devastating epidemic. Further information about the evolution of virulence can be found in a number of reviews (Bull 1994; Ebert 1998a, 1999; Ebert and Bull 2003).

5.3 Parasite Effects on Other Host Traits

Besides fecundity and survival, parasites may influence other aspects of host fitness, few of which have been studied. G. intestinalis (formerly Pleistophora intestinalis) reduces adult growth in its host D. magna (Ebert 1994b). The strength of this effect was shown to depend both on host clone and parasite isolate, with local parasite isolates having the strongest effect. Lass and Bittner (2002) showed that C. mesnili reduced the adult growth of its host D. galeata. In contrast, P. ramosa causes its host D. magna to grow to an unusually large size (Ebert et al. 1996, 2004). This form of parasite-induced host gigantism may be adaptive for the parasite, as larger hosts result in more parasite spores being produced (Ebert et al. 2004).

Parasites may also influence aspects of their hosts' sexual life cycle. For example, they may reduce the hosts' likelihood of finding mates or may increase or decrease the frequency with which a female produces ephippia and male offspring. Furthermore, vertically transmitted parasites may influence the survival of their host during resting (Lass and Ebert 2005).

5.4 Parasites May Influence Predation on Their Hosts

The potential effect that parasites have on host–predator interactions is also important. Parasites may lower the ability of their hosts to escape predators; infected hosts may swim and react more slowly than healthy hosts, for example. The sometimes dramatic visual effect that parasites have on Daphnia may even directly increase the hosts' attractiveness to visually hunting predators (Yan and Larsson 1988; Lee 1994; Duffy et al. 2005).

Lass and Bittner (2002) tested for more indirect effects of parasites on host–predator interactions. They tested whether hosts are less able to show adaptive phenotypic changes against predators when exposed to C. mesnili. Their experiments revealed no significant interactions between parasite and kairomon-induced life history changes. They concluded that this is because the host's adaptive response against fish predators changes life history traits expressed early during the host's life, whereas the parasite affects its host during later stages.

On the other hand, one can imagine that parasites alter their host's behavior so that hosts more effectively protect themselves from predators, e.g., by altering vertical migration. This may still be disadvantageous for the host because the parasite's interest is in host survival, while the host has to trade-off protection from predators against other fitness components, such as reproduction. Lee (1994) and Fels et al. (2004) showed that various parasite species influence the depth selection behavior of D. magna. Infected hosts stay deeper in the water than uninfected controls. It is not clear, however, whether this is adaptive for the host, the parasite, both, or none.

An extreme example of altered predator exposure would be a case in which the parasite manipulates its host's behavior to facilitate it own transmission to the next host. To my knowledge, none of the described unicellular parasites of Daphnia has a known second host, although this option has been speculated (Mangin et al. 1995). However, the macroparasites (helminth) parasites of Daphnia, which have not yet been extensively studied, have second hosts and may well manipulate their hosts to their own advantage (Stammer 1934; Green 1974; Schwartz and Cameron 1993).

5.5 Conclusions and Open Questions

There is little doubt that parasites of Daphnia and other Cladocerans are generally harmful. Occasional reports of "nonsignificant" effects of parasites have to be considered in the light of low statistical power or large environmental noise. Thus far, every species tested under controlled conditions proved harmful. What I find more interesting than the fact that the parasite harms its host are questions regarding the covariables of the degree of harm. There are a number of interesting questions about this:

  1. Why are some parasites more harmful than others? What role does the parasite's taxonomic position play for its virulence? What role does the mode of transmission play? What role does the specific tissue infected play?
  2. Are there further hidden costs of parasitism in Daphnia? For example, do parasites influence mate choice during sexual reproduction? Do parasites influence the survival of resting eggs?
  3. Does inter- and intra-specific competition of parasites influence virulence?