S36.2: The coevolution of virulence and immune defence in birds

Laboratoire d'Ecologie, CNRS URA 258, Université Pierre et Marie Curie, Bât. A, 7ème étage, 7 quai St. Bernard, Case 237, F-75252 Paris Cedex 05, France, fax 33 1 44 27 35 16, e-mail smerino@snv.jussieu.fr

Parasites are organisms that live on or within other organisms for some or all of their time, by exploiting resources provided by these organisms (Price 1980). A higher rate of host exploitation by a parasite resulting in a higher reproductive rate of the parasite is thus equated with a higher level of virulence. This damage or cost of parasitism is what we understand by the word virulence. Virulence is simply the increased host mortality resulting from parasite infection (May & Anderson 1983). Hence a common denominator of parasite damage of hosts is measured in terms of fitness, i.e. the reduction in contribution to the next generation of parasitised hosts as compared to individuals not suffering from a parasite. Virulence can be considered a phenotypic trait like any other character, and it can thus be subject to selection pressure, and provided a genetic basis of virulence as found in some fruit-fly and human parasites, virulence can change during evolutionary time. Since parasites generally have short generation times compared to their hosts, such evolutionary change in virulence may take place relatively rapidly.

Some years ago parasitologists believed that once a host-parasite relationship had become established, this would eventually equilibrate at a low level of virulence. One reason for this common belief is the study of myxomatosis in Australian rabbits (Fenner 1983). Although the virulence was initially very high, killings lots of rabbits, and virulence subsequently became reduced, this is not the same as the parasite eventually becoming avirulent. A parasite that initially kills more than 95% of all hosts, but subsequently only kills 50%, has become relatively avirulent, but is by all means still very virulent.

A number of different hypotheses and mathematical models have been proposed to account for the evolution of virulence (reviews in Bull 1994; Frank 1996). These can be classified in a number of different ways which depend on whether natural selection affects (1) the dynamics of parasite transmission between hosts; (2) the dynamics of parasite genotypes within hosts; or (3) a combination (Bull 1994). Between-host transmission dynamics may affect the evolution of virulence by affecting the relative importance of horizontal and vertical transmission (e. g. Ewald 1983). This idea is based on the argument that a parasite that is horizontally transmitted can afford in an evolutionary context to damage its host more than a parasite that is vertically transmitted. Horizontal transmission refers to transmission of a parasite from one unrelated host to another, while vertical transmission reflects transmission from a host to a close relative like an offspring.

Within-host dynamics of a parasite may also dramatically affect the evolution of virulence because multiple infections tend to increase virulence (e. g. Bremermann & Pickering 1983). This hypothesis is based on the idea that the resources of a single host are limiting. If the damage of a host by a parasite is directly related to its rate of multiplication, we should expect that a parasite strain may multiply at a relatively low rate if alone in a host. However, if more than a single strain is present, restraint on host exploitation is no longer important. The parasite strain that can exploit the host most efficiently is likely to be at a selective advantage because it will competitively exclude other strains. More efficient exploitation is equated with a higher reproductive rate by the parasite and hence a higher rate of host resource use.

Empirical examples of the evolution of virulence are still relatively rare (Bull et al. 1991; Herre 1993; Agnew & Koella 1997). A comparative study by Clayton & Tompkins (1994) of the damage to bird hosts by a range of parasites suggested that parasites with vertical transmission showed higher benevolence than those transmitted horizontally. An alternative interpretation is horizontal transmission is associated with multiple infection and competition among parasite strains of different genetic origin.

If hosts rarely recover from parasitism, investment in resistance should be maximised. However, if hosts are able to recover, we can have a situation where parasite virulence and host resistance evolve towards an evolutionary equilibrium. How much should hosts invest in immune function in order to optimise resistance to parasites (van Baalen 1998)? If the risk of infection is low, natural selection may favour hosts with less effective immune systems. Since the optimal allocation to defence depends on the force of infection, and the force of infection depends on the level of defence in the rest of the population, a game theoretical approach is useful.

The model by van Baalen (1998) is based on a trade-off between transmissibility, which implies that parasites can always increase transmissibility, but at an accelerating cost in terms of disease-induced mortality rate. The faster hosts are able to expel the parasite and recover, the greater virulence is favoured. This makes sense because recovery robs the parasite of the benefit associated with reduced virulence, namely prolonged infectiousness. A more efficient system therefore compels the parasite to switch to increased transmissibility, which amounts to increased virulence. The force of infection may rise sharply when recovery ability decreases. The parasites' evolutionary response is to decrease their virulence, and if virulence decreases, so does the parasites' ability to control the host population. The host population then grows to infinity, and with the host population the parasite population too. Two different stable evolutionary equilibria occur with an unstable intermediate equilibrium that leads to switches from one stable equilibrium to another. One of the coevolutionary stable strategy pairs is characterised by heavily defended hosts and virulent parasites, the other by undefended hosts and abundant, relatively avirulent parasites. Even when density-dependent regulation of host populations is introduced in the model, there is still this coevolutionary bistability with many, avirulent parasites or relatively few, virulent parasites. Hosts match these levels of virulence by matching levels of immune defence.

In conclusion, if parasites are not allowed to evolve, the outcome is a single stable strategy. If parasites coevolve, multiple outcomes are possible, one where parasites are relatively avirulent and common and hosts invest little in immune function, the other where parasites are rare but virulent and the hosts invest heavily in defence.

Empirical tests of host defence in relation to virulence are few and scattered. The main problem is that it is still very difficult to predict virulence without direct measurement. However, there might be exceptions to this rule. A number of bird species frequently re-use their nest sites because of nest site limitation. This is the case for two broad categories of birds; hole nesters and colonially breeding birds. It is well-known that hole nesters can be attracted to nest boxes simply because there are no other nest sites available. Similarly, colonial birds often re-use the same site for decades or even centuries. This implies that parasites are readily transmitted horizontally from one host to another because the limiting sites are visited by many individuals, and in the case of colonial breeders because of close proximity of nests. Furthermore, these frequent visits by different hosts result in multiple infections by different strains of parasites. Hence, we should expect that hole nesters and colonially breeding birds are exploited by particularly virulent parasites and that they during evolutionary time have invested differentially in immune defence (Møller & Erritzøe 1996). A comparative study of the size of the bursa of Fabricius and the spleen, which are both crucial for antibody production, revealed consistently larger immune defence organs in species of birds that re-use nests (Møller & Erritzøe 1996). The assumption of higher virulence of parasites in hosts that re-use nest sites has not been directly tested. However, it is known that the prevalence and the intensity of some parasite infections are similar in colonial and solitarily nesting birds (Rózsa 1997). We are currently testing by the use of standard immune defence challenge tests whether there is a relationship between parasite virulence and host immunocompetence associated with the evolution of coloniality in the family Hirundinidae.

A potential second example of high levels of virulence and anti-parasite defence concerns tropical birds (Møller 1998). Tropical regions are characterised by not having resource levels and populations of animals and plants reduced during winter, which should give rise to continuously high population densities of hosts and parasites. Such high densities and intense competition for resources should facilitate multiple infections and horizontal transmission of parasites. Hence, we should expect that tropical bird species invest differentially in immune function as compared to closely related non-tropical species. This is indeed the case since the concentration of leukocytes and the size of the spleen is consistently higher in the tropics as compared to non-tropical regions (Møller 1998). The supposedly higher virulence in the tropics and the associated investment in immune function may be linked to other peculiarities of tropical birds (and other organisms) such as an extremely high species diversity, relatively low population densities compared to species in the temperate zone, small clutch sizes and high adult survival rates (Møller 1998).

An aspect of bird migration that is not well studied is the fact that migratory species have to cope physiologically with at least two different parasite faunas; that of the breeding grounds and that of the wintering range (Møller & Erritzøe 1998). Given that defence against multiple parasites costs more than defence against fewer parasites, investment in immune function should be increased in migratory birds. Elevated investment in immune function by migratory birds could arise as a consequence of more frequent horizontal transmission of parasites or more cases of multiple infections with parasite strains of different genetic origin. The size of both bursa of Fabricius and spleen was larger in migratory bird species compared to closely related resident species (Møller & Erritzøe 1998). High levels of site fidelity in the breeding grounds, on migration and in the wintering areas may also be explained from the perspective of host-parasite interactions. Once a host has acquired immunity to a particular parasite, it pays to stay rather than become exposed to a different strain of the same parasite. Hence, change of sites should only occur when the costs of multiple infections are balanced by the benefits of movement to a novel site.

The study of parasite virulence and host anti-parasite defence is still very much in its infancy. A number of different questions are in need of investigation: (1) Currently, we know very little about the relative role of within and between-host population dynamics of parasites as a determinant of virulence. Experiments should be able to address this question. (2) We know very little about the immune response of individual hosts in relation to the virulence of a parasite. Independent assessment of parasite virulence and host immune response could resolve this question. (3) We know very little about the ecological conditions that promote parasite virulence and host investment in immune function. Careful comparative analyses of tropical and temperate species of birds or colonial and solitary birds may help resolve this question.

Our research was supported by grants from the Spanish Ministry of Education (S.M.) and an ATIPE BLANCHE from CNRS.

Bremermann, H.J. & Pickering, J. 1983. A game-theoretical model of parasite virulence. Journal of theoretical Biology 100: 411-426.

Bull, J.J., Molineux, I.J. & Rice, W.R. 1991. Selection of benevolence in a host-parasite system. Evolution 45: 875-882.

Bull, J.J. 1994. Perspective: Virulence. Evolution 48: 1423-1437.

Clayton, D.H. & Tompkins, D.M. 1994. Ectoparasite virulence is linked to mode of transmission. Proceedings of the Royal Society of London B 256: 211-217.

Ewald, P.W. 1983. Host-parasite relations, vectors, and the evolution of disease severity. Annual Review of Ecology and Systematics 14: 465-485.

Fenner, F. 1983. Biological control as exemplified by smallpox eradication and myxomatosis. Proceedings of the Royal Society of London B 218: 259-285.

Frank, S.A. 1996. Models of parasite virulence. Quarterly Review of Biology 71: 37-78.

Herre, E.A. 1993. Population structure and the evolution of virulence in nematode parasites of fig wasps. Science 259: 1442-1445.

May, R.M. & Anderson, R.M. 1983. Parasite-host coevolution. In: Futuyma, D.J. & Slatkin, M, (eds) Coevolution. Sunderland; Sinauer: 186-206.

Møller, A.P. 1998. Evidence of larger impact of parasites on hosts in the tropics: Investment in immune function within and outside the tropics. Oikos (in press).

Møller, A.P. & Erritzøe, J. 1996. Parasite virulence and host immune defense: Host immune response is related to nest reuse in birds. Evolution 50: 2066-2072.

Møller, A. P. & Erritzøe, J. 1998. Host immune defence and migration in birds. Evolutionary Ecology (in press).

Price, P.V. 1980. Evolutionary biology of parasites. Princeton; Princeton University Press: 237 pp.

Rózsa, L. 1997. Patterns in the abundance of avian lice (Phthiraptera: Amblycera, Ischnocera). Journal of Avian Biology 28: 249-254.

van Baalen, M. 1998. Coevolution of recovery ability and virulence. Proceedings of the Royal Society of London B 265: 317-325.