The role of parasites in regulating host populations

Because parasites are rarely seen and frequently cause only mild or sublethal effects, it is commonly thought that their impacts on animal abundance are minor, perhaps accounting for occasional mortality among very weak, young, or old animals. Counter to this view, epidemiological models and a growing number of wildlife-parasite examples point to a number of conditions under which parasites can regulate host population size (Scott and Dobson 1989). In some extreme cases, infectious agents have caused precipitous losses of 50% or more of existing host populations, as occurred with morbillivirus epidemics in black-footed ferrets and harbor seals and more recently with transmissible facial tumors affecting Tasmanian devils (Thorne and Williams 1988; Harding et al. 2002; Bostanci 2005). In other cases, removal of endemic parasites revealed that parasites were a significant factor depressing host population size or were responsible for generating dramatic population cycles in host abundance (McCallum and Dobson 1995; Hudson et al. 1998a). It is important to note that host regulation can arise from both within-species processes, such as competition for limited resources, and from species interactions like predation, interspecific competition, and parasitism. Confusion in identifying regulatory mechanisms often arises because most species are affected by a combination of biotic density-dependent factors and extrinsic environmental variation (May 1983; Bj0rnstad and Grenfell 2001), making it difficult to tease apart processes that contribute to population dynamics in non-experimental systems.

4.3.1 Theoretical predictions

Parasites can impact total host population size (N) through their effects on individual host fitness, including parasite-induced host mortality and reductions in host fertility. To illustrate this mathematically for the microparasite model shown in Box 4.1, the change in total host population size can be written as, dt=(r~ay)N (4.6)

where the intrinsic growth rate of uninfected hosts is r = a-b, the prevalence of disease is y = I/N, and a is disease-induced host mortality. Equation (4.6) implies that one mechanism by which parasites regulate their hosts is through disease-induced mortality (a) that offsets the host's intrinsic growth rate. Somewhat surprisingly, if pathogens affect host mortality alone, those with intermediate virulence will depress host density to a greater degree (the upper-right face of Fig. 4.7). This is because more lethal parasites are more likely to also kill their hosts before transmission to other hosts occurs (Anderson and May 1979; Anderson 1982a; McCallum 1994), so that the more virulent a parasite, the lower its expected prevalence in the population. Indeed, highly virulent parasites do not appear to reduce equilibrium host density, although they could induce short-term population declines.

A striking element of Fig. 4.7 is that equilibrium host abundance is lowest when parasites completely sterilize infected animals with no additional host mortality (as shown in the lower corner of this graph). Such negative effects on host reproduction have been demonstrated across a range of animal-parasite systems including helminths infecting red grouse, hares, and reindeer (Hudson et al. 1985, 1992; Stien et al. 2002; Newey and Thirgood 2004). Further investigation of these issues in wild primates would undoubtedly produce similar examples (e.g. Milton 1996). In the case of helminth infections, parasite-induced reductions in host fecundity can also trigger oscillations in host abundance (May and Anderson 1978; Hudson et al. 1998a), especially when parasites have long-lived infectious stages that persist in the environment (Dobson and Hudson 1992). Collectively, these points suggest that infectious diseases with low or moderate effects on host survival or those that sterilize their hosts may cause far greater conservation concerns and should not be overlooked when assessing potential causes of wildlife declines.

Fig. 4.7 Parasite-mediated reduction of host population size in relation to disease-induced host mortality and the relative fecundity of infected hosts. Higher values of host mortality (towards the left) and lower levels of fecundity (to the bottom) indicate greater negative effects on the host (i.e. virulence). Results are based on a modified version of the model in Box 4.1, with frequency-dependent transmission and density-dependent births (i.e. additional host regulation in the absence of disease). Shown on the vertical axis is host population size relative to the disease free carrying capacity (i.e. N*/K) in the presence of the pathogen.

Other parameters used were: a = 0.5, b = 0.35, and p = 3.

Fig. 4.7 Parasite-mediated reduction of host population size in relation to disease-induced host mortality and the relative fecundity of infected hosts. Higher values of host mortality (towards the left) and lower levels of fecundity (to the bottom) indicate greater negative effects on the host (i.e. virulence). Results are based on a modified version of the model in Box 4.1, with frequency-dependent transmission and density-dependent births (i.e. additional host regulation in the absence of disease). Shown on the vertical axis is host population size relative to the disease free carrying capacity (i.e. N*/K) in the presence of the pathogen.

Other parameters used were: a = 0.5, b = 0.35, and p = 3.

Simple inferences about host regulation assume that parasites have a narrow host range and cannot rely on a reservoir host for persistence. As discussed below, pathogens with a wide host range that are relatively benign in reservoir hosts can have severe consequences for endangered or rare species (McCallum and Dobson 1995). Furthermore, parasites whose transmission is density-dependent should have stronger effects on high-density host populations (Anderson 1978; Getz and Pickering 1983), and can induce striking host population cycles. Because density-dependent diseases in theory require a threshold host density for establishment and persistence, they should be unlikely to cause host extinction when acting alone (Anderson and May 1979). Pathogens with frequency-dependent transmission, on the other hand, can persist and continue to spread even in low-density host populations.

Relative to microparasites, host regulation by macroparasites further depends on the degree to which parasites are aggregated among hosts (Anderson and May 1978; May and Anderson 1978; Tompkins et al. 2001). This effect arises because hosts that harbor high numbers of parasites are most likely to be removed from the population, whereas host with few parasites might experience little or no reductions in fitness. When a large proportion of a macroparasite population is aggregated in a small proportion of the hosts, stable regulation is more likely, although at the other extreme, parasites can be so aggregated that the host escapes regulation entirely. As most macroparasites show aggregated distributions (Shaw and Dobson 1995; Shaw et al. 1998; Wilson et al. 2001), it seems probable that these parasites play some role in regulating wild populations. A related point is that wildlife managers might expect regulating parasites to be abundant in a high proportion of the host population, including a large number of dead animals. Counter to this expectation, mathematical models suggest that regulation by endemic macroparasites is probably more likely when high parasite burdens are seen in only a few infected animals (McCallum and Dobson 1995).

4.3.2 Regulation in experimental and natural populations

A common misconception is that parasite effects on host abundance can be inferred using information on prevalence alone, or observations of parasite-induced host mortality (McCallum 1994; McCallum and Dobson 1995). Unfortunately, modeling approaches suggest that counter to common wisdom, the most frequently observed causes of mortality are not necessarily the most important regulatory factors (Anderson and Gordon 1982). In natural systems, therefore, observing host population abundance and demographic rates in both the presence and absence of parasites is probably the best way to examine the population-level impact of infectious disease (Scott and Dobson 1989; Tompkins and Begon 1999; Hochachka and Dhondt 2000).

Only a handful of studies have been conducted to examine the population level effects of disease in wild populations. In extreme cases, the effects of disease on host abundance are obvious, as when populations of European rabbits (Oryctolagus cuniculus) in Australia and Europe collapsed following the intentional introduction of myxoma virus during the 1950s, and later, calicivirus during the late 1990s

(Fenner and Fantini 1999). Similar dramatic declines in host abundance were observed when populations of harbor seals in the North Sea crashed during outbreaks of phocine distemper in 1988 and 2002 (Heide-Jorgensen et al. 1992; Jensen et al. 2002). One thorough and groundbreaking analysis quantified the impacts of a bacterial eye disease (caused by Mycoplasma gallisepticum) that emerged in wild house finches (Carpodacus mexicanus) in North America starting in 1993. Researchers used observed prevalence and host abundance data at a continent-wide scale to show that as this disease spread across the house finches' eastern range, host populations dropped sharply to around 40% of their expected disease-free abundance (Hochachka and Dhondt 2000). This analysis also showed that higher density populations suffered more severe declines, relative to areas with lower host density. Evidence indicates that the eye disease probably caused host population declines through effects on individual survival rather than fecundity, as the timing of outbreaks generally occurred during the fall and winter (outside of the breeding season, Altizer et al. 2004) and birds with severe infections, where one or both eyes swelled shut, probably died of exposure, starvation, or predation (Dhondt et al. 2005).

Although a few studies have the advantage of comparing host abundance before and after pathogen introduction, experiments are essential to document parasite effects to the exclusion of other regulatory factors, in part because it is difficult to establish regulation when populations harboring endemic parasite infections are in equilibrium (Tompkins et al. 2002). In manipulative experiments, researchers treat a fraction of animals or a subset of populations—either by experimentally adding parasites or by using anti-parasitic drugs or vaccination to lower infections—and treat other animals (controls) with placebos. Survival and fecundity at the individual level, together with population size and growth rates, can be compared among treatment and control groups. A classic experimental study of population regulation by parasites was conducted in a freely breeding colony of mice. In large arenas housing up to 1000 individual mice, Scott (1987b) introduced a helminth (Heligmosomoides polygyrus) that parasitizes mice in the wild. Whereas the unexposed control population increased and maintained a high population size, the parasite-treated populations crashed rapidly to very low abundance (Fig. 4.8), only recovering after antihelminthic treatment was given. Because mice in these enclosures reproduced freely and had access to abundant resources, this study underscored the potential importance of parasites relative to competition for food or space. In the field, experimental approaches have demonstrated impacts on host survival or population size induced by nematode parasites on feral Soay sheep (see fig. 1.5, Gulland 1992; Gulland et al. 1993b), botfly parasites in wild mice (Munger and Krasnov 1991), and caecal nematodes on population cycles of red grouse (Hudson et al. 1998a).

We currently have limited knowledge of parasite-induced population regulation in primate hosts, but this should not discourage experimental work on suitable primate subjects (Janson 2000). Although some researchers have proposed that parasites can regulate primate populations (Freeland 1976; Smith 1977; Milton 1996), no experimental studies of population regulation in wild primate have been conducted to date. Records of severe population-level mortality have been recorded in a number of primate populations (summarized in Chapters 1 and 7). In the absence of experimental

Was this article helpful?

0 0

Post a comment