Title: Do predators keep prey healthy or make them sicker? A meta-analysis
Short Title: Testing the healthy herds hypothesis
Article Type: Letter
Authors: Robert L. Richards1,2*, John M. Drake1,2, Vanessa O. Ezenwa1,2,3
1Odum School of Ecology, University of Georgia, 140 E. Green Street, Athens, GA, 30602
2Center for the Ecology of Infectious Diseases, University of Georgia, 203 D. W. Brooks Drive, Athens, GA, 30602
3Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, 501 D.W. Brooks Drive, Athens, GA, 30602
* Corresponding author: Odum School of Ecology, 140 E. Green Street, Athens, GA, 30602; Phone: 617-312-4802; e-mail:robert.richards@uga.edu
Keywords: predator, parasite, disease ecology, healthy herds hypothesis, meta-analysis, trophic interaction
Statement of authorship: RLR, JMD, and VOE conceived of the study. RLR performed literature search, collected and analyzed data, and wrote the first draft of publication. All authors contributed substantially to revisions.
Data accessibility statement: Data and code used in these analyses will be published on figshare upon acceptance of the manuscript and a do will be included in the article. Prior to acceptance data and code can be accessed at this private figshare link:https://figshare.com/s/ae13262817c42e4e82d9
Abstract: 149 Words
Main Text: 3986 Words
Number of figures: 3
Number of Tables: 2
Number of References: 53
ABSTRACT
Ecological theory suggests that predators should keep prey populations healthy by reducing parasite burdens. However, empirical studies show that predators often have minimal effects on, or even increase, parasitism in prey. To quantify the overall magnitude and direction of the effect of predation on parasitism in prey, we conducted a meta-analysis of 50 empirical studies. We also examined how key attributes of these studies, including parasite type, study design, and predator interaction type (consumptive vs. non-consumptive) contributed to variation in the predator-prey-parasite interaction. We found that the overall effect of predation on parasitism differed between parasites and parasitoids and that predator interaction type, and whether a predator was a non-host spreader of parasites were the most important traits predicting the parasite response. Our results suggest that the mechanistic basis of predator-prey interactions strongly influences the effects of predators on parasites and that these effects, while context dependent, are predictable.
INTRODUCTION
Organisms navigate a complex set of interspecific interactions, among the most important of these being victimization by natural enemies. Both predators (Krebset al. 1995, 2018) and parasites (Hudson et al.1992b; Tompkins & Begon 1999) can affect the population demography and dynamics of the species they attack. However, few organisms are victim to only a single natural enemy. Competition between predators of a single prey population (Holt & Lawton 1994; Holt & Polis 1997; Tallian et al. 2017) and between parasites within a single host organism (Pedersen & Fenton 2007; Jolles et al. 2008; Ezenwa & Jolles 2011) have both been studied for the effects that these interactions have on natural enemy and victim populations. But predators and parasites of a single victim population also interact in a variety of potentially important ways. Parasites may weaken their hosts, making them easier to catch and consume (Hudsonet al. 1992a; Moore 2002), while the killing and consuming of prey by predators also kills parasites (Hatcher et al.2006; Borer et al. 2007), except when the predator itself becomes the next host (Lafferty 1999; Kuris 2003; Logiudice 2003). Therefore, like other natural enemy interactions, interactions between predators and parasites are likely critical to understanding the dynamics of natural populations.
Ecologists have long recognized the importance of predator-prey-parasite interactions (Hudsonet al. 1992a). Among the most influential hypotheses about the consequences of predator-prey-parasite interactions is Packer et al.(2003)’s prediction, based on a mathematical model, that predators reduce parasitism in their prey. This Healthy Herds Hypothesis (HHH) phenomenon might be produced by multiple mechanisms. First, predators directly, and often preferentially, kill infected individuals, decreasing the number of infected individuals in the population. Second, predators often reduce prey population sizes, which can decrease the spread of parasites with density dependent transmission. Empirical studies have tested the underlying predictions of the HHH in a variety of systems, but results are conflicting. Some studies show a strong negative effect of predators on parasites, while others show strong positive effects. For example, experimentally increased bird predation on lizard hatchlings (Acanthodactylus beershebensis ) decreased parasitic trombiculid mite loads in the lizards (Hawlena et al.2010), while sunfish (Lepomis gibbosus ) predators introduced into tanks with infected tadpoles (Lithobates spp. ), increased trematode cercarial load in tadpole prey (Szuroczki & Richardson 2012). Interestingly, these empirical studies differ along multiple axes, including the transmission traits of the parasite (Holt & Roy 2007; Roy & Holt 2008) and the type of predator or predatory interaction manipulated (Cácereset al. 2009; Strauss et al. 2016; Duffy et al.2019)which may help explain the variation in outcomes.
In this study, we used a meta-analytical approach to quantify the overall magnitude and direction of the effect of predation on parasitism, providing a synthesis of the empirical work on this topic. We also tested the prediction that differences among studies along two key axes, (i) parasite type, and (ii) type of predatory interaction, explain variation in observed parasite responses. Specifically, we predicted that effects of predators on macroparasites and parasitoids would be more negative than effects on microparasites, because macroparasites and parasitoids tend to be highly aggregated among hosts and spatial locations (Hassell 1982; Chesson & Murdoch 1986; Shaw & Dobson 1995) allowing small amounts of selective predation pressure to nearly eliminate their populations. Parasitoids in particular have free-living adult stages which may fall prey to or avoid predators of their hosts (Heimpel et al.1997; Brodeur & Rosenheim 2000). We also predicted that consumptive predatory interactions would have more negative effects on parasites than non-consumptive interactions, except when consumptive effects facilitate parasite spread. In this case, consumptive interactions should actually increase parasitism. The HHH predicts that, on average, consumptive interactions decrease parasitism as infected individuals are removed from populations (Packer et al.2003). However, this average effect of consumption on parasites should not apply in all circumstances (Duffy et al.2019). In particular, “predator-spreaders” may facilitate the spread of parasites from their prey items by dispersing infectious agents more widely (Cácereset al. 2009). On the other hand, non-consumptive interactions can alter prey movement and space use behavior (Brown et al.1988; Spieler 2003; Jones & Dornhaus 2011; Creel et al. 2014) in ways that can either increase or decrease parasite transmission (Ezenwa 2004; Patterson & Ruckstuhl 2013), meaning that the effects of non-consumptive interactions on parasites should be less consistently negative than those of consumptive interactions. While multiple syntheses of predator-prey-parasite interactions have been published over the past 20 years (Ostfeld & Holt 2004; Hatcher et al. 2006; Duffy et al. 2019), these studies take a qualitative approach while here we use an approach that explicitly quantifies the the typical effect of predators on parasites in their prey and the most important drivers of variation in this response. Here we ask: (i) what is the average overall effect of predators on parasites in their prey and (ii) does this effect vary by parasite or interaction type? We expect to find a negative overall effect of predation on parasitism but this effect should be more negative for macroparasites and parasitoids than microparasites and for interactions involving consumption than non-consumptive interactions and that consumptive interactions including identified “predator-spreaders” should have more positive effects than those with non-spreaders.
MATERIALS AND METHODS