Host–Pathogen Evolution, Biodiversity, and Disease Risks …pedersen.bio.ed.ac.uk/page3/assets/Altizer_Pedersen_2008.pdf · est birds as a result of avian malaria (Van Riper et

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 262 — #1

17

Host–Pathogen Evolution, Biodiversity, andDisease Risks for Natural Populations

SONIA ALTIZER

AMY B. PEDERSEN

P athogens can play important roles in naturalsystems, from influencing host genetic diversity

to altering the composition of ecological commu-nities. Infectious diseases can also threaten nat-ural populations (Harvell et al., 1999; Lafferty& Gerber, 2002). Virgin ground epidemics thatquickly spread throughout previously unexposedplant and animal populations can cause highmortality and reductions in host abundance (see,for example, Anagnostakis, 1987; Osterhaus &Vedder, 1988; Sherald et al., 1996). Althoughdocumented cases of pathogen-driven host extinc-tion are rare (Smith et al., 2006), several studiessuggest that pathogens can cause declines in previ-ously healthy populations and can be one of manythreats to already declining species (Anderson etal., 2004; Daszak et al., 2000; Lafferty & Gerber,2002; Pedersen et al., 2007). Recent examplesinclude population crashes in African apes resultingfrom Ebola virus (Walsh et al., 2003), amphib-ian declines caused by chytridiomycosis (Lips et al.,2006), the near extinction of several Hawaiian for-est birds as a result of avian malaria (Van Riper etal., 1986), recent declines in oaks in the westernUnited States because of sudden oak death syn-drome (Rizzo & Garbelotto, 2003), and widespreadmortality among flowering dogwoods in easternNorth America after the spread of anthracnoseblight (Sherald et al., 1996).

The overarching theme of this chapter is thatinfectious diseases provide a model system to

understand how evolutionary principles are rele-vant to biodiversity and conservation. Throughoutwe use the terms parasite, pathogen, and infec-tious disease interchangeably, and we consider bothmicroparasites (viruses and bacteria) and macropar-asites (protozoa, fungi, arthropods, and helminths)that can infect hosts and, in many cases, lower hostfitness and cause outward signs of disease. Becauseof their potential for rapid evolution and impactson host survival and reproduction, host–parasiteinteractions can generate a number of evolution-ary outcomes, ranging from the maintenance ofgenetic variation to significant shifts in the geneticcomposition of both host and parasite populations.Because parasites, at times, can both contribute toand threaten biological diversity, understanding theevolutionary dynamics of host–parasite interactionsis crucial for biological conservation.

In this chapter we begin by considering how hostresistance and genetic diversity can help buffer wildpopulations against epidemics of new and existingpathogens. We discuss the genetic basis of host resis-tance, how parasite infection may maintain geneticdiversity, and the importance of resistance variationfor the conservation of threatened species. Next wedemonstrate how evolutionary processes may influ-ence pathogen emergence and host shifts. Parasitescan evolve to capitalize on new transmission oppor-tunities, alter their virulence, and adapt to novelhost species or changing environments. From a con-servation perspective, these issues can be important

262

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 263 — #2

Host–Pathogen Evolution, Biodiversity, and Disease Risks for Natural Populations 263

for captive breeding programs, control strategiesfor existing pathogens, and landscape-level mana-gement approaches. Recent and historical exam-ples from wild systems are provided to illustratekey points, although we caution that they are notintended to represent an exhaustive list.

As a final point, parasites can influence hostdiversification and, more generally, are a majorcomponent of biodiversity themselves. Parasitesthat live uniquely on threatened host species couldgo extinct long before their hosts, and more accu-rate knowledge of parasite biodiversity will likelycompound estimates of future biodiversity loss. Weconclude by echoing recent assertions that inter-active networks of host and parasite populationsmight be necessary to protect biological diversityand evolutionary processes (Crandall et al., 2000;Thompson, 2005).

HOST GENETIC DIVERSITY ANDRESISTANCE TO INFECTION

Parasites represent powerful selective agents innatural populations, in part because they can spreadrapidly and cause significant negative effects on hostfitness. When exposure to pathogens is high in ahost population, traits conferring resistance are pre-dicted to increase in frequency. In animals, theseresistance strategies include behavioral defensesto avoid exposure or physically remove parasites,physiological and innate responses to infection,and humoral and cell-mediated immune defenses(Clayton & Moore, 1997). Resistance mechanismsin plants include biochemical defenses, receptor–protein interactions, and changes in phenology thatlower contact rates with infective stages (Fritz &Simms, 1992). These defenses can be innate, ormaintained in the absence of infection (and hencea “first line of defense” after infection) or adap-tive, or induced during the course of infection by apathogen (for example, antibody-mediated immu-nity in vertebrate animals). Moreover, defenses canbe highly specific, such that they recognize or defendagainst a particular type of pathogen or even a singlepathogen genotype, or general, in that they attack avariety of infectious organisms.

Given the strong selective pressures imposed byparasites and the benefits of host resistance traits,it is important to investigate factors that main-tain intra- and interpopulation variation in resis-tance to pathogens. In other words, why aren’t all

individuals resistant to infectious diseases? Modelsbased on simple host–parasite interactions showthat genetic variation can be maintained by at leastthree key mechanisms: frequency-dependent selec-tion, balancing selection, or negative correlationsbetween resistance and other fitness-conferringtraits (Box 17.1) (Schmid-Hempel & Ebert, 2002).In addition, specific defenses against different para-sites may act antagonistically, such that an immuneresponse against one agent might suppress resis-tance to other infectious diseases (Yazdanbakhshet al., 2002; but see also Pedersen & Fenton, 2007).

The Genetics and Maintenance ofHost Resistance

The genetics underlying variation in host immu-nity in wild populations has attracted much recentinterest (Frank, 2002; Schmid-Hempel, 2004),and many studies have demonstrated differentialsusceptibility among host genotypes in wild pop-ulations (reviewed in Altizer et al., 2003). Thefrequency of resistant genotypes in a populationcan be affected by and can feed back to localparasite dynamics. For example, long-term studiesof trematode parasites infecting the freshwatersnail Potamopyrgus antiopodarum (Fig. 17.1) haveshown that host genetic diversity can be maintainedthrough local adaptation of parasites to their hostsand through frequency-dependent selection (Lively,1992, 1999). Cross-infection experiments providedevidence of local adaptation by demonstrating thatcommon host clones were significantly more sus-ceptible to sympatric parasites than were rare hostclones (Dybdahl & Lively, 1998). Further researchrevealed that changes in the frequencies of com-mon and rare snail clones were driven by parasitetracking of susceptible genotypes. This evidenceof frequency-dependent selection suggests that highinfection rates can ultimately favor host sexualreproduction as a strategy for generating novel hostgenotypes that may resist infection (Dybdahl &Lively, 1998).

Plant–pathogen coevolution can similarly leadto a high diversity of host resistance and parasitevirulence alleles. In fact, natural plant populationshave been shown to harbor a staggering abun-dance of genetic polymorphisms for resistance tofungal diseases (Burdon & Thrall, 1999; Parker,1992), and wild plant populations have been citedas a source of resistance genes for pathogens and

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 264 — #3

box 17.1 Evolutionary Mechanisms That Can Maintain Variation in Host Resistance

Multiple processes can maintain genetic polymorphisms in host resistance in natural popu-lations. First, in the case of frequency-dependent selection, parasites can become locallyadapted to common host genotypes, and thus are better able to infect them. Hence, hostswith rare genotypes may escape parasite infection, conferring a selective advantage to rarealleles and a disadvantage to common alleles. This process can cause time-lagged cycles inboth host and parasite allelic frequencies, and may ultimately lead to the maintenance ofgenetic variation over longer timescales (Seger & Hamilton, 1988). The phenomenon ofparasites tracking common host genotypes has been demonstrated in several wild systemsand is important for arguments concerning the role of parasites in generating advantagesto host sexual reproduction (Dybdahl & Lively, 1998).

More generally, balancing selection refers to processes that favor the persistence of mul-tiple alleles, or genetic polymorphisms in a population, in contrast to directional selection,in which allelic frequencies tend to shift in a single direction. Balancing selection can berealized through frequency-dependent selection, as described earlier, or through heterozy-gote advantage, in which individuals with different alleles at any given locus tend to havegreater fitness than homozygous individuals. As one example addressed in the text, indi-viduals heterozygous for MHC alleles could experience an advantage in the face of diversepathogen strains by virtue of recognizing a greater variety of antigens for response by thevertebrate immune system (Penn et al., 2002). Spatial and temporal changes in the risk ofinfection by different pathogen genotypes or species, as generated in part by environmentalheterogeneity, could also favor the maintenance of multiple alleles over larger spatial andtemporal scales.

Third, resistance-conferring host traits may be costly in terms of reductions in otherfitness components (for example, fecundity, growth rates, resource competition) as a resultof pleiotropy or resource-based trade-offs. Modeling studies have indicated that even smallresistance costs (measured as differences in fitness between resistant and susceptible hostsin the absence of infection) should lead to genetic polymorphisms, such that both suscep-tible and resistant genotypes are maintained in the presence of parasites (Antonovics &Thrall, 1994). A growing number of field and experimental studies has identified measur-able costs of resistance to pathogens infecting many host species (see, for example, Sheldon& Verhulst, 1996), although other studies emphasize that the presence and size of costs willdepend on host and pathogen characteristics (see, for example, Carr et al., 2006; Mitchell-Olds & Bradley, 1996), and that the shape of the trade-off function will be important forthe longer term dynamics.

In light of potential costs of host resistance, inducible defenses that are activated onlyafter parasite infection may be beneficial when the risk of infection is rare or unpredictable(Harvell, 1990). Such defenses include antibody-mediated responses in vertebrates, anti-microbial proteins in invertebrates and plants, and behavioral avoidance in animals. Asone example, recent work in sea fan corals has shown that hosts can mount inducibledefenses against infection with emerging fungal pathogens, and that environmental factorsthat cause variation in this response may predict patterns of host susceptibility (Harvellet al., 2002; Ward et al., 2007). Understanding how inducible defenses can be mobilizedrapidly against novel pathogens, their rate of evolution, and the costs that they pose onhost fitness should help inform efforts to manage disease resistance for conservation.

264

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 265 — #4

A

B

figure 17.1 Field and experimental studies of the freshwater snailPotamopyrgus antipodarum and the trematode parasite Microphallus,for which the snails serve as intermediate hosts, have demonstratedlocal adaptation and time-lagged frequency-dependent selection in ahost–parasite interaction. (A) Pathology caused by infection withthe Microphallus (top) compared with the healthy snail (picturedbelow). Both snails are shown without shells; parasite eggs ingested bysnails develop into encysted intermediate stages and castrate the snail.(B) Experimental setup of snail clones after exposure to trematode eggsderived from specific parasite genotypes. (C) Lakes on the south islandof New Zealand, such as Lake Alexandrina (pictured here), serve askey sites for field studies of snails and parasites in their native environ-ments. (Reproduced with permission from C. Lively. Photos courtesy ofG. Harp and C. Lively.)

265

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 266 — #5

266 Conservation of the Coevolving Web of Life

C

figure 17.1 Continued

pests for crop plants (Jones, 2001). For exam-ple, long-term field studies of wild flax and flaxrust in natural populations in Australia (Fig. 17.2)indicate that a large number of resistance allelescan persist in plant metapopulations (Burdon &Jarosz, 1991). The distribution of genotypes canshift rapidly after local epidemics, and trade-offsarising from fitness costs have been linked withvariation in host resistance and pathogen virulence(Burdon & Thompson, 1995; Thrall & Burdon,2003). Studies of anther-smut infections in theirwildflower hosts also demonstrated variation inresistance among plant genotypes, impacts of hostresistance on pathogen prevalence, and costs ofresistance in terms of delayed flowering (see, forexample, Alexander et al., 1996; Carlsson-Granér,1997).

Among vertebrate animals, a variety of genesand gene complexes are important for mediatingdefenses against infectious diseases (see, for exam-ple, Acevedo-Whitehouse & Cunningham, 2007).The majority of scientific interest to date has focusedon MHC as playing a key role in acquired immu-nity. Major histocompatibility complex moleculesare immune proteins that recognize and bind topathogen proteins (antigens) inside infected hostcells and transport these antigens to cell outer mem-branes. Here they are presented to T cells to initi-ate antibody production and cell-mediated immuneresponses. Specific MHC molecules preferentially

bind to specific pathogen peptides, and hencedifferent MHC alleles confer resistance to differentpathogens. In natural populations, MHC class I andII genes show enormous variation and are impor-tant for recognizing a wide diversity of pathogens(Hedrick & Kim, 2000; Nei & Hughes, 1991). Indi-vidual hosts that are heterozygous across multipleMHC loci recognize a greater diversity of pathogensthan homozygous individuals (Doherty & Zinker-nagel, 1975), and at the population-level, highlevels of MHC allelic variation can increase thechance that at least some hosts’ immune systemswill recognize a single pathogen. Studies of humansand domesticated animals also highlight the impor-tance of non-MHC immune genes in protectingagainst infectious diseases (Acevedo-Whitehouse &Cunningham, 2006), and a broader understand-ing of resistance evolution in wild vertebrates willultimately require comprehensive genetic analysisof other immune regions that control aspects ofpathogen defense.

Costs, Trade-offs, andEnvironmental Variation

Research in the developing field of ecological immu-nity (McDade, 2003; Norris & Evans, 2000)examines the ecological causes and consequences ofvariation in immune function. One central issue isthat hosts that respond to parasite pressure through

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 267 — #6

A

B

figure 17.2 Long-term studies of the dynamics of wild flax (Linum marginale) and its fungal rust(Melampsora lini) illustrate coevolutionary dynamics in a naturally occurring plant–pathogen interac-tion. (A) The heavily infected leaves and stem of the L. marginale host plant are shown covered withuredial lesions after experimental inoculation. (B) Many populations used to investigate the ecological andevolutionary dynamics of this host–pathogen interaction occur in southeastern Australia in environmentssuch as this subalpine grassland surrounded by eucalypt forests (pictured here in the Kosciuzko NationalPark, N.S.W.). (Reproduced with permission from L. Barrett. Photos courtesy of C. Davies and L. Barrett.)

267

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 268 — #7


increased resistance could suffer lower competitiveability or reduced reproduction (and vice versa;Boxes 17.1 and 17.2) (reviewed in Lochmiller &Deerenberg, 2000). Thus, hosts might be expectedto invest in greater defenses only when the risk ofpathogen infection is high, leading to geographicvariation in genetically based host resistance traits.Also, it is well established that environmentalvariation, including temperature, resource availa-bility, and environmental stressors, can affect theexpression of host resistance in plants and animals(Rolff & Siva-Jothy, 2003). These environmentalprocesses could generate spatial and temporal vari-ation in host defenses, and might also representimportant constraints on host resistance evolution(Box 17.1).

Implications of Host Resistance forSpecies Conservation

Several studies have demonstrated that genetic vari-ation for host resistance is common in naturalpopulations and has significant consequences forparasite infection rates. For example, one compara-tive analysis showed that macroparasites were morelikely to colonize fish species that had low levels ofgenetic variation as indicated by mean heterozygo-sity (Poulin et al., 2000). Field studies of intestinalnematodes affecting Soay sheep demonstrated thathost allelic variation and levels of heterozygositywere associated with higher host survival and resis-tance to infection (Coltman et al., 1999; Patersonet al., 1998). But what are the implications of thesefindings for species conservation? One concern isthat small or endangered host populations mightsuffer disproportionate impacts from infectious dis-eases as a result of the loss of genetic variabilitythrough population bottlenecks, genetic drift, andinbreeding (Lyles & Dobson, 1993). Hosts bred incaptivity and treated to prevent and remove par-asitic organisms may further experience increasedsusceptibility caused by relaxed selection and costsassociated with resistance-conferring traits.

The relationship between the loss of geneticdiversity and increased disease susceptibility hasbeen found in many animal populations (Box 17.2)(Acevedo-Whitehouse et al., 2003). For example,recent studies of birds and amphibians have shownthat loss of heterozygosity (based on selectivelyneutral microsatellite markers) is associated withreduced immunocompetence, greater risk of infec-tion, and increased severity of infection (Hawley

et al., 2005; MacDougall-Shackelton et al., 2005;Pearman & Garner, 2005). Some evidence frompathogen resistance and inbreeding effects in thecontext of wild plant populations indicates thatinbreeding could increase susceptibility to fungaland viral disease (Carr et al., 2003; Ouborg et al.,2000).

Among vertebrate animals, allelic diversity atMHC loci can be lower than expected amongendangered species that have undergone dramaticdeclines in population size (see, for example,Aldridge et al., 2006). On the other hand, a highdiversity of MHC genotypes has been documentedacross small populations of red wolves, Arabianoryx, and some other endangered species (Hedricket al., 2000, 2002). In one extreme example, Aguilarand colleagues (2004) demonstrated homozygosityacross multiple selectively neutral loci in the SanNicholas Island fox (Urocyon littoralis dickeyi).This pattern is indicative of the loss of variationresulting from genetic drift after an extreme popula-tion bottleneck approximately 10 to 20 generationsbefore samples were collected. However, these ani-mals showed high levels of variation across fiveMHC loci, leading the authors to conclude thatintense balancing selection was necessary to preventthe loss of rare alleles and ultimately maintain MHCvariation in the face of past bottlenecks (circa <10individuals).

In light of these findings, one goal for conser-vation management is to design captive breeding,stocking regimes, and species management pro-grams that maintain levels of immunity or resistancevariation present in wild populations. One casestudy shows that the Hawaiian amakihi, a nativebird species whose population size and geographicrange were negatively affected by the introductionof avian malaria, has persisted and begun to repopu-late lowland forest habitats despite high infectionrates from Plasmodium relictum (Woodworth et al.,2005). One possible explanation for this recovery ishost evolution of genetically based resistance or tol-erance of infection. Findings such as these shouldmotivate future conservation efforts to focus notjust on protecting host populations in areas lessaffected by pathogens, but also to identify hostsinhabiting parasitized environments where signifi-cant evolutionary responses might have occurred.

A related point is that host and parasite move-ment among habitat patches may be crucial for bothhost persistence and the spread and maintenanceof resistance alleles. For example, Carlsson-Graner

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 269 — #8

box 17.2 Invasive Species as Model Systems for Studying Host and Parasite Evolution

Species introductions offer many opportunities to understand the strength of pathogensas agents of selection. First, most populations of introduced species are likely to originatefrom a few founders, thus limiting their genetic diversity and possibly increasing their sus-ceptibility to parasites (Sakai et al., 2001). Second, invasive species that escape infection bya broad range of parasites (Mitchell & Power, 2003; Torchin et al., 2003) might reallocateresources away from enemy defense and into growth and reproduction (Wolfe et al., 2004).Indeed, greater allocation to growth and reproduction after enemy release is widely thoughtto underlie the increased vigor or success of introduced species in their new range (Siemann& Rogers, 2001). Together, these ideas suggest that low genetic diversity combined withevolutionary reductions in parasite defense should make introduced species vulnerable tar-gets for future epidemics, and that detailed studies of pathogen interactions with invasivespecies in their native and introduced ranges are greatly needed.

Genetic variation tends to be lower among introduced populations, particularly ifcolonists come from a single source population or undergo an establishment phase duringwhich population sizes remain small (Sakai et al., 2001). Founder events resulting in geneticdrift and inbreeding could lower the fitness of introduced populations and limit their abilityto adapt to future challenges arising from parasites and infectious diseases. For example,reduced host genetic diversity has been suggested as the reason for the unusually high suscep-tibility of introduced house finches in eastern North America to mycoplasmal conjunctivitis(Hawley et al., 2005, 2006). In some cases, however, repeated introductions from multiplenative sites could lead to a mixing of alleles from different source populations (admixture),resulting in greater genetic variation—rather than less—in the introduced range (as hasbeen demonstrated with brown anole lizards [Kolbe et al., 2004]). Hosts with such novelgene combinations might be highly resistant to parasite infections (Sakai et al., 2001).

Many invasive species are larger bodied, more abundant, and more vigorous in theirintroduced range relative to their native range (Crawley, 1987; Grosholtz & Ruiz, 2003).A specific hypothesis termed the evolution of increased competitive ability proposes thatexotic species should adapt to the loss of natural enemies by allocating more energy togrowth and reproduction, rather than investing in costly defenses (Blossey & Nötzold,1995). Predictions of this hypothesis are that in the native range, growth and reproductionshould be lower, natural enemies should be more common, and investment in defenseshigher; whereas in the introduced range, natural enemies should be less common or absent,defenses lower, and growth and reproduction greater (Wolfe et al., 2004). Furthermore,these phenotypic differences should be genetically based and, given the choice, parasites andother natural enemies should better attack the invasive phenotypes—two predictions thatcan be tested using common-garden and reciprocal transplant experiments. Some recentstudies provide support for genetic divergence in enemy defense and reproductive strategiesbetween native and novel populations of introduced weeds and trees (Siemann & Rogers,2001; Wolfe et al., 2004), and suggest that introduced species might be vulnerable to futurepathogen introductions.

Another area for future research includes the role of evolutionary change in thosepathogens that persist or become established in populations of exotic host species. Forexample, do traits such as high mutation rates or more rapid generation times favorpathogen adaptation to hosts in their new range? How might pathogen virulence change inresponse to shifts in host genetic diversity or abundance in their new habitats? Currently,comprehensive studies of host resistance and pathogen evolution among invasive speciesin both their native and introduced ranges are rare, despite the potential insights that canbe gained from such comparisons. Future studies of exotic species and their parasites willlikely provide new evidence regarding the role of host genetic diversity in immunity andresistance, how costs and trade-offs affect host investment in immune defenses, and howpathogens adapt to populations of exotic organisms.

269

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 270 — #9


and Thrall (2002) demonstrated that isolated popu-lations of the wildflower Lychnis alpina were rarelyinfected with anther-smut disease, but when popu-lations were infected, the prevalence was high. Incomparison, highly connected populations showmore widespread pathogen occurrence across sites,with low prevalence within each location. This pat-tern was consistent with model simulations thatassumed higher rates of movement of both pathogenpropagules and host resistance alleles among con-nected populations (Carlsson-Graner & Thrall,2002). In bighorn sheep, larger population sizesand increased dispersal is associated with lowerextinction risk and rapid population recovery afterbronchopneumonial epidemics, indicating a poten-tial role for host dispersal in genetic variation andresistance evolution (Singer et al., 2001). Together,these studies suggest that habitat size and ecologicalcorridors can have significant effects on host disper-sal and movement of resistance alleles in ways thataffect host responses to future epidemics.

ROLE OF PATHOGEN EVOLUTIONIN DISEASE EMERGENCE

The evolutionary potential of pathogens sets themapart from other major threats to biodiversity. Mostpathogens have short generation times and largepopulation sizes; hence, strong selection pressuresfollowing ecological changes can increase the prob-ability of pathogen evolution and the likelihood ofdisease emergence in a novel host species. The rel-evance of pathogen evolution to human health isunderscored by recent threats from severe acute res-piratory syndrome, avian influenza, and antibiotic-resistant bacteria. However, empirical evidence ofhow evolutionary processes influence the likelihoodof emergence and pathogen spread in natural sys-tems is less clear and supported by only a fewexamples (Altizer et al., 2003; Schrag & Weiner,1995).

Two major questions are how do parasites estab-lish in new host populations, and does evolutionaffect the probability of long-term parasite success?Both epidemiology and evolutionary potential of thepathogen are important during and after pathogenentry into a novel host population. In terms of epi-demiology, a newly infected individual must trans-mit the pathogen to more than one other host forthe parasite to increase in frequency when initiallyrare (commonly referred to as R0 > 1 [Anderson

& May, 1991]). Novel infections are more likely toestablish if chains of transmission allow new muta-tions to arise that increase R0 above one, and henceevolution and mutations that influence transmis-sion and virulence can play a crucial role in theemergence process (Antia et al., 2003).

Despite general predictions, the direct role ofselection in the emergence of infectious diseases innatural systems remains relatively unknown. Recentstudies of host shifts among fungal pathogensinfecting plants indicate that geographic proxim-ity and opportunities for cross-species transmis-sion, rather than genetic changes in the parasitesthemselves, are primarily responsible for the ori-gin of new host–parasite combinations (Antonovicset al., 2002; Roy, 2001). Yet, after such intro-ductions, other studies demonstrate that geneticvariants of parasites from novel hosts can rapidlydiverge from ancestral genotypes (Ley & Yoder,1997; Oldroyd, 1999), suggesting that evolutionhas indeed occurred. Canine parvovirus (CPV),which first appeared in wolves, coyotes, and domes-ticated dogs during the mid 1970s, is an example ofa pathogen that likely arose as a new genetic variantof feline panleukopenia parvovirus, a closely relatedvirus from domesticated cats (Parrish, 1990).Genetic studies following CPV emergence suggestthat mutation and selection associated with a majorcapsid gene, rather than recombination of existingfeline variants, played a key role in viral emergence(Shackelton et al., 2005).

Evolutionary Potential, DiseaseEmergence, and Virulence

Several pathogen characteristics may predisposethem to high emergence potential. First, transmis-sion strategies that increase encounters with newhost species may increase the range of hosts that apathogen can infect (Woolhouse et al., 2005). Sec-ond, greater genetic variability, high mutation rates,and more rapid generation times might allow certainpathogens to exploit new host species more readily.Specifically, parasites with high antigenic variationshould have an increased ability to recognize hostproteins or evade host immune defenses (Bitter et al.,1998) and infect new host species (Cleaveland et al.,2001). Viruses dominate this group because theirhigh mutation rates, greater antigenic diversity, andshort generation times allow them to adapt rapidlyto a larger number of host species. In fact, emerging

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 271 — #10


infectious diseases in humans, domesticated ani-mals, plants, and wildlife are often dominated byviruses, especially RNA viruses that are character-ized by unusually high mutation rates and relativelylarge host ranges (Anderson et al., 2004; Holmes,2004; Pedersen et al., 2005; Woolhouse et al.,2005).

Evolution can also cause shifts in the virulenceof existing and emerging diseases. Theory on theevolution of parasite virulence suggests that para-sites should evolve toward an intermediate level ofvirulence. This is based on an a presumed trade-offbetween transmission and virulence, such that theproduction of infectious stages required for trans-mission to a new host is positively correlated withthe damage the parasites cause to the host. Inter-mediate virulence is favored because highly virulentpathogens may kill hosts before successful trans-mission occurs, whereas parasites with very lowvirulence will not produce enough infective stagesto be transmitted (Levin, 1996). Spatial structurechanges the evolution of virulence, however, suchthat if most transmission occurs at a local (ratherthan global) scale, then less virulent pathogens arefavored (Boots & Sasaki, 1999). Boots and col-leagues (2004) further demonstrated that acquiredimmunity can lead to the coexistence of both nonvir-ulent and highly virulent strains. These results helpexplain the sudden emergence of a virulent strainof rabbit hemorrhagic disease virus that spreadthroughout Europe, causing high mortality in free-living rabbit populations. Molecular evidence sug-gests that this highly virulent strain emerged from arecombination event involving less virulent strains.The sudden emergence of the highly virulent strainand coexistence with less virulent genotypes waslikely a product of the host’s social structure, asrabbits tend to have highly structured populations(Boots et al., 2004).

Anthropogenic Effects on PathogenEvolution and Risks for WildPopulations

Large-scale environmental changes caused byhumans could directly affect pathogen life cyclesand transmission, leading to evolutionary shiftsin other parasite traits. One example that high-lights the role of human activity and selectivepressure on sudden evolutionary changes is in theincreased transmission and host range of Toxo-plasma gondii, the protozoan parasite that causes

toxoplasmosis in humans and other mammalianhosts (Su et al., 2003). Toxoplasma gondii is main-tained by wild and domesticated cats that can shedinfectious stages in their feces and transmit the par-asite to humans and wildlife. Molecular geneticanalyses of this parasite indicate that the ability ofclonal lineages to transmit orally and without sex-ual recombination (as opposed to passing throughan intermediate host to complete the sexual phase ofthe life cycle) was associated with a selective sweepthat occurred around the time of human agriculturalexpansion several thousand years ago (Su et al.,2003). By creating high densities of multiple speciesof domesticated mammals, human activity mayhave selected for increased oral transmission. Thus,humans may be unknowingly selecting for newvariants of wildlife pathogens through global com-merce, changes in host density and habitat quality,and climate shifts over longer timescales (Harvellet al., 2002).

Spillover from domesticated plants and animalsis a particular threat to wild species because infectedreservoir hosts can facilitate epizootics in naturalpopulations with otherwise low population den-sity (Anderson et al., 2004; Daszak et al., 2000).As one example, African wild dogs (Lycaon pic-tus) became extinct in the Serengeti in 1991, inpart as a result of spillover by canine distemperfrom a domestic dog outbreak (Funk et al., 2001).Other pathogens of domesticated dogs have affectedwild populations—even in the ocean, where an out-break of Morbilivirus in harbor seals was geneticallysimilar to a strain from domestic dogs (Osterhaus& Vedder, 1988). Similarly, the anthropogenicintroduction of plant pathogens from imported tim-ber, agricultural crops, or nursery plant specieshas caused dramatic declines in wild plant popu-lations (with examples reviewed in Anderson et al.,2004).

Repeated pathogen introduction events, espe-cially from multiple sources, can influence geneticheterogeneity in the pathogen population. Forexample, the repeated introduction of the Dutch elmdisease fungus (Ophiostoma sps.) via infested tim-ber in the 20th century led to the death of billionsof elm trees in North America and Europe. Thiswas caused primarily by two different Ophiostomaspecies introduced several decades apart (Brasier,2001). Genetic and phenotypic evidence indicatesthat the more aggressive causal agent (introducedlater) gradually replaced the less aggressive species,and that hybridization between the two fungal

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 272 — #11


species produced novel genotypes that may havefurther increased pathogen virulence and environ-mental tolerance. This example emphasizes theneed for additional studies to examine the degreeto which evolutionary changes in pathogens haveaffected their spread and affects on novel hostspecies.

HOST–PARASITE COEVOLUTION,GEOGRAPHIC VARIATION, ANDDIVERSIFICATION

Most examples in natural systems suggest thatneither host nor parasite evolution operates inisolation, but that their interaction leads to coevo-lutionary dynamics. Over time this could resultin reciprocal adaptations of interacting lineages,possibly accompanied by cospeciation events andgenetic arms races (Box 17.3) (Page, 2003). Rela-tive to other types of species interactions, the inti-mate associations between hosts and parasites offermany opportunities for studying coevolution oncontemporary timescales. Collectively, studies ofhost–parasite interactions emphasize that coevolu-tion is common, causes heterogeneity in speciescharacteristics over space and time, and that geneticchanges linked with coevolution can drive signif-icant changes in species abundance (Thompson,2005).

An often-cited example of coevolution ofhost and parasite traits involves myxomatosisin Australian and European rabbit populations(Fenner & Fantini, 1999). After intentional releasesin Australia during the early 1950s, the myxomavirus rapidly shifted in virulence. Specifically, within5 years after the introduction of a highly virulentvirus, other strains with reduced severity (for exam-ple, longer times to host death and moderate mor-tality rates) became increasingly common (Marshall& Fenner, 1960). An evolutionary response of thehost followed, characterized by greater resistanceand lower viral-induced host death rates (Marshall& Fenner, 1958).

Selection operating on hosts and parasites canalso lead to coevolutionary arms races (Frank,2002; Hamilton, 1982), whereby parasites selectfor increased host immunity, which exerts recipro-cal selection for alternative transmission strategies,manipulation of host behavior, and changes invirulence of the parasite. This cycle of parasite adap-tation leads to further selection on host defenses. As

one example, studies of interactions between feralpigeons and feather lice demonstrated that selec-tion for improved defense systems (in other words,grooming for parasite removal) in the avian hostis accompanied by selection for parasite mecha-nisms to escape these host defenses (Clayton et al.,1999, 2003). More generally, studying coevolution-ary interactions in naturally occurring hosts andparasites can highlight selection pressures importantin the long-term maintenance of the association,including the degree to which limited evolutionaryresponses might lead to higher extinction risks foreither host or parasite lineages.

Phylogenetic Patterns andCospeciation

Over longer timescales, reciprocal adaptations ofhosts and parasites might lead to cospeciation andphylogenetic diversification (Clayton et al., 2003;Page, 2003) (Box 17.3). Cospeciation is definedas the parallel divergence of two or more inter-acting lineages (Page, 2003), and this process cangenerate congruent host and parasite phylogenies.Several studies have demonstrated patterns con-sistent with host–parasite cospeciation, yet otheroutcomes are possible, including host shifting, para-site extinction, and a process called missing theboat, during which the host splits into two ormore lineages but the parasite remains in only one(Page, 2003). Host shifting is likely to be commonamong parasites that can evolve rapidly relativeto host generation times, and where transmissiongenerates frequent opportunities for cross-speciestransfer.

Modern molecular tools allow researchers toexamine the origins and coevolution of host–parasite associations and the relative rates of hostand pathogen evolution (see, for example, Holmes,2004). Viruses and bacteria that have fast gene-ration times and large population sizes can evolvefaster than their hosts and might commonly showevidence for rapid genetic changes after shifts tonew host species or transmission among host popu-lations. Surprisingly, one study of cospeciationamong strains of simian foamy viruses (SFVs; inthe family Retroviridae) isolated from Old Worldmonkeys and apes provides evidence counter tothis general expectation (Switzer et al., 2005). Thisstudy showed strong evidence for congruent primateand virus molecular phylogenies (with evidence ofonly a few host jumps), and comparable divergence

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 273 — #12

box 17.3 Diversification and Coextinction of Host and Parasite Lineages

Interspecific interactions (in other words, predator–prey, plant–herbivore, and host–parasite) have been proposed to drive major diversification between assemblages ofcoevolving organisms (Farrell, 1998; Percy et al., 2004) and could ultimately be responsi-ble for the great species diversity found on earth. In terms of parasitism, one comparativestudy of primate–pathogen interactions demonstrated that parasite diversity (species rich-ness) was positively correlated with rates of primate host diversification (Nunn et al., 2004).In other words, primate host species from more diverse lineages harbored a greater num-ber of parasite species (including viruses, protozoa, and helminths). One mechanism thatcould give rise to this association is that parasites increase host evolutionary diversification,through, for example, their effects on sexual selection. Because sexually selected traits canalso correlate with parasite resistance (Hamilton & Zuk, 1982), high parasite pressuremight ultimately lead to greater potential for host speciation. A second explanation forthe patterns observed by Nunn and colleagues (2004) is that parasites infecting hosts frommore diverse lineages could experience greater opportunities for diversification. Specifically,related host species that overlap in geographic range might provide more opportunities forhost sharing by generalist parasites, and host shifting by specialist parasites—leading tohigher parasite species richness.

An arms race between hosts and parasites, involving an ongoing struggle to mountgreater host resistance against infection, and higher virulence and transmissibility of theparasites, could also account for positive correlations between host diversification andparasite species richness. As a final possibility, coextinctions of hosts and parasites mightalso drive the associations reported by Nunn and colleagues (2004). Specifically, parasitelineages might be lost as their hosts decline in population size and ultimately go extinct.In other words, higher extinction rates in declining host lineages could generally reduceparasite diversity, especially if parasites go extinct before their hosts (Koh et al., 2004).Another study using the same host–parasite data supported this idea by demonstrating thatmore threatened primate hosts harbored fewer parasite species (Altizer et al., 2007). Futurecomparative research on host–parasite diversification will likely benefit from investigatingthe geographic patterning of host–parasite interactions (Thompson, 1994), incorporatinginformation on parasite phylogeny into comparative tests (Hafner & Page, 1995), andexamining the degree to which parasites themselves have gone extinct along with theirhosts (Gompper & Williams, 1998).

Lastly, results of comparative studies, although provocative, do not directly examinemechanisms that underlie associations between host and parasite diversification. Theo-retical studies have shown that frequency-dependent selection between prey and naturalenemies can lead to evolutionary branching in both the host and enemy populations(Doebeli & Dieckmann, 2000). Small-scale experiments of coevolution between bacteriaand virulent phages in spatially structured environments have also demonstrated that para-sites can drive allopatric divergence among host populations, increasing host diversificationby selecting for antiparasite defenses genetically linked to different host traits in differentpopulations (Buckling & Rainey, 2002). Future studies aimed at developing and testinghypotheses for host and parasite evolutionary divergence will give insight to the degree towhich hosts and parasites show evidence for coadaptation, concordant phylogenies, andmechanisms that drive patterns of diversification.

273

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 274 — #13


times and rates of nucleotide substitution for hostsand parasites, suggesting that primates and SFVshave evolved at similar rates. Analysis of otherRNA viruses (for example, Morbilivirus, influenzaA viruses, and flaviviruses), however, shows evi-dence for rapid substitution rates and more frequentcross-species transmission events (see, for exam-ple, Chen & Holmes, 2006; Twiddy et al., 2003).Although there is variation in a parasite’s ability toestablish within a new host species, understandinghow host and parasite characteristics covary withthe risk of host shifts will be crucial for predictingdisease emergence events in declining or endangeredspecies, for which the impact of a novel diseasecould be devastating (Lafferty & Gerber, 2002;Pedersen et al., 2007).

PATHOGENS AS INDICATORS OFHOST CONTACT AND ISOLATION

Can genetic studies of infectious diseases revealinformation about host biology, behavior, anddisease transmission that enhance conservationefforts? In humans, it has been demonstratedthat host migration events can be revealed bypathogen genetic structure, especially for rapidlyevolving viruses like human immunodeficiency virus(Holmes, 2004). A handful of studies have appliedthis approach more recently to wildlife popula-tions. For example, low genetic diversity and theneed to examine patterns over relatively shorttime frames could pose difficulties in detectingpatterns of host evolutionary divergence, especiallyfor long-lived species. In such cases, evolution ofsome pathogens can be rapid enough to illumi-nate host geographic isolation and contact pat-terns. One analysis investigated sequence variationof two genes (pol and env) in the feline immun-odeficiency virus to assess patterns of isolationand physical contact among cougars in westernNorth America (Fig. 17.3) (Biek et al., 2006).Low cougar population densities during the pastcentury and restricted movements were expectedto result in pronounced genetic structure, butonly weak isolation-by-distance was revealed usingcougar microsatellite markers. By comparison, phy-logenetic analysis of viral strains isolated fromcougars in the northern Rocky Mountains revealedeight viral lineages and geographic population sub-structure, with different lineages dominating in

the central versus the periphery of the cougars’range (Biek et al., 2006). Genetic analysis furtherrevealed that the spatial occurrence of viral lineagesis expanding, most likely as a result in increases incougar population size or movement. This workillustrates the usefulness of pathogen molecularmarkers for understanding contemporary popula-tion movements and geographic structuring of theirhosts.

In another study, three species of whale lice(host-specific ectoparasitic crustaceans) revealedhistorical separation of populations of endangeredright whales (Kaliszewska et al., 2005). Geneticanalyses of parasite mtDNA sequences revealedthat three whale populations in the North Atlantic,North Pacific, and southern oceans diverged aroundfive to six million years ago, after the formationof the Isthmus of Panama. High genetic diver-sity among lice in the currently small populationsof North Atlantic right whales indicates that theirhost population sizes probably numbered in thetens of thousands before the modern era of com-mercial whaling, and that these populations hadnot experienced prolonged historical bottlenecksprior to the past few centuries (Kaliszewska et al.,2005).

Analysis of genetic divergence among pathogenlineages can also be used to predict patterns of dis-ease emergence and to develop control strategies forpathogens in endangered host populations. Statis-tical analysis of viral sequence changes over spaceand time has indicated dispersal patterns for thegeographic expansion of fox rabies virus in NorthAmerica (Real et al., 2005) and a wavelike spread ofEbola virus in Central Africa (Walsh et al., 2005).The case of Ebola in Africa is important for bothhuman health and the future of the remaining wildape populations, which are also susceptible to thevirus. The epidemic appears to have advanced as awavefront, and Walsh and colleagues (2005) sug-gest that the ladderlike phylogeny of viral isolatesand concordant spatial pattern support transmis-sion of the virus through populations of gorillas. Incontrast, Leroy and associates (2005) showed thatfruit bats harbor the virus, and proposed that Ebolaoutbreaks are driven less by direct ape-to-ape trans-mission. These recent analyses may help researcherspinpoint the origins of the epizootic and predict thefuture path of spread, with implications for concen-trating control efforts in wild ape populations at thehighest risk.

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 275 — #14

A

B

figure 17.3 Continued

275

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 276 — #15

C

figure 17.3 Evolutionary genetics of the feline immunodeficiency virus FIVPco have been used to elucidatethe population structure of its cougar hosts (Puma concolor) in North America. (A) Researchers draw ablood sample from a cougar to be used for FIVPco analysis. (Photo courtesy of T. Ruth.) (B) Spatialdistribution of FIVPco lineages in the natural habitat of cougars in the western United States and Canada.White circles represent negative samples (no FIVPco); filled circles represent samples from cougars thattested positive (with the several viral lineages indicated by various shadings). (C) Phylogeny of FIVPcoconstructed from the concatenated pol and env sequences. Eight viral lineages are designated based onmore than 5% divergence, and their colors match those shown in (B). Labeled nodes represent bootstrapsupport based on 1,000 neighbor-joining trees generated from maximum-likelihood distances estimatedfrom the original likelihood model and posterior proportions from the Bayesian analysis. (Figure providedby R. Biek; data are from Biek et al. [2006]. Reproduced with copyright permission from Biek, R., et al.2006. A virus reveals population structure and recent demographic history of its carnivore host. (Science311: 538–541.)

276

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 277 — #16


FUTURE DIRECTIONS

Parasites and Host Extinction Risk

Although managing wild host populations in lightof pathogen risks may seem trivial relative to numer-ous other threats to biodiversity, several specieshave been driven to extinction or to near extinctionas a result of infectious diseases (Lips et al., 2006;Smith et al., 2006; Thorne & Williams, 1988).On the one hand, models of directly transmit-ted infectious diseases suggest that highly virulentpathogens will disappear from small host popula-tions before their hosts go extinct (Anderson &May, 1991). However, there are several casesin which pathogens can cause host extinction (deCastro & Bolker, 2005). First, directly transmit-ted parasites could drive host population sizes solow that extinction by stochastic factors becomesa concern. Second, pathogens transmitted sexually,vertically, or by biting arthropods (in other words,with frequency-dependent transmission [Getz &Pickering, 1983]) should not suffer from reducedprevalence as host populations decline, and the sameis generally true for parasites that cause host steril-ity rather than mortality (O’Keefe & Antonovics,2002). Third, some generalist parasites maintainedin reservoir populations, particularly domesticatedspecies living at high density, can also affect threat-ened species (Fenton & Pedersen, 2005). In addi-tion, spatially explicit models show that even ifparasites do not cause global host extinction, theycan lead to local extinction and cause large-scaledeclines in host density as the infection spreads tonew patches (Boots & Sasaki, 2002; Sato et al.,1994).

One of the best-known examples of host nearextinction resulting from disease involves the black-footed ferret (Mustela nigripes), one of the mostendangered mammals in North America. In themid 1980s, outbreaks of canine distemper virus(affecting the ferrets) and sylvatic plague (affect-ing their prairie dog prey) effectively eliminatedblack-footed ferrets from the wild and severelythreatened a captive breeding program. At one pointthe entire species was reduced to fewer than 20captive animals (Dobson & Lyles, 2000). Despitecurrent success in breeding and release of captiveanimals, future threats from infectious diseases willprobably play a role in ferret recovery. Dramaticlosses of gorilla populations in Africa as a resultof Ebola virus outbreaks further underscore the

ecological relevance of infectious disease for wildlifeconservation. In one instance, an outbreak in theLossi Sanctuary eliminated an entire population of143 gorillas (Leroy et al., 2004). Disease-mediatedextinction risk is not limited to wildlife species, asseveral plants are also suffering dramatic declinesresulting from the introduction and spread of viru-lent pathogens (Anderson et al., 2004). In Australia,for example, the anthropogenic introduction of Jar-rah dieback disease (Phytophthora cinnamomi), aglobally occurring fungus with a host range of morethan 900 species, has caused a national threat toendemic flora, with significant risks for the healthof several ecological communities (Wills, 1993).

Host–Parasite Evolution andConservation of Biodiversity

Many studies have highlighted the importance ofgenetic variation in host resistance in causing dis-ease patterns in both field and experimental settings(reviewed in Altizer et al., 2003), but these almostalways involve single-host–single-pathogen interac-tions. In reality, host individuals, populations, andspecies are affected by large numbers of parasiticorganisms spanning divergent phyla from virusesto nematodes to parasitic insects, and for whichdifferent types of defenses vary in their effective-ness. Future studies of the types of pathogens thatpose the greatest threats to endangered host pop-ulations, together with broader genomic surveysthat examine a range of immune genes in ani-mals and plants are needed to shed further lighton the benefits and feasibility of selective main-tenance of resistance traits in captive breedingprograms (Acevedo-Whitehouse & Cunningham,2006). Another important area for future researchinvolves developing a better understanding of costsand trade-offs of resistance and how environmentalvariation affects host immunity. Such investigationswill require measuring ecologically relevant immunedefenses in the wild and understanding their corre-lations with other components of host survival andreproduction (Norris & Evans, 2000). In cases inwhich host defenses are costly, hosts that lose theirparasites during population bottlenecks or while incaptive breeding programs may also lose their abil-ity to respond to future disease threats after relaxedselection for immune defenses.

As a related point, when hosts decline towardextinction, their host-specific parasites will prob-ably also be lost (Gompper & Williams, 1998).

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 278 — #17

A

B

figure 17.4 (A, B) Parasite diversity in free-living mammals. Each point represents parasite species richnessbased on counts of viruses, bacteria, fungi, protozoa, helminths, and arthropods reported from a single hostspecies. These figures demonstrate positive correlations between study effort and the number of parasitesreported from free-living host populations, suggesting that scientists have uncovered only a fraction ofthe total diversity of parasites and pathogens in natural populations. More generally, this associationindicates that “parasites are like the stars—the more you look, the more you find” (J. Antonovics, personalcommunication, September 2000). Data are shown separately for nonhuman primates (n = 119) (A) andhoofed mammals (artio- and perissodactyls; n = 97) (B). In (A), the sampling effort is measured as thenumber of literature citations for each primate species using the online bibliographic resource PrimateLit(Wisconsin Primate Research Center and Washington National Primate Research Center). In (B), thesampling effort is based on the sum of the number of animals across all studies upon which parasite countdata were derived. Lines show least-squares regression based on a linear regression model. Data are asreported based on methods described by Nunn and Altizer (2005) and are available as part of the GlobalMammal Parasite Database (www.mammalparasites.org).

278

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 279 — #18


One question at the interface of conservation biol-ogy and disease ecology involves whether the lossof parasites is harmful or helpful to a host popu-lation. In other words, is the elimination of allnaturally occurring parasites beneficial to the host?Given that most of the examples highlighted in thischapter focus on parasites as a threat to wild pop-ulations, the answer might seem to be yes. Yet,many ecologists would argue that the best approachfor long-term conservation is to preserve geograph-ically structured populations of interacting species,including parasites, in part to maintain intact evo-lutionary processes. Keeping host–parasite relation-ships intact requires landscape management strate-gies that might include protecting corridors andnetworks of multiple habitat types important to abroad range of species.

Protecting parasitic organisms has not beenchampioned as a priority for current or futureconservation efforts. However, host declines andassociated coextinctions of parasites could dramat-ically compound estimates of future biodiversityloss (Fig. 17.4) (Koh et al., 2004). Parasites are anintegral part of life on earth, with their biodiver-sity projected to be significantly greater than thespecies richness of free-living hosts (Price, 1980);yet, biologists have uncovered only a minisculepercentage of the diversity of infectious organ-isms from natural host communities. Many micro-and macroparasites that live uniquely on threat-ened host species could go extinct long before theirhosts, and this poses a distinct threat for para-site extinction. As with most taxa, we do nothave accurate numbers of how many species ofparasitic organisms might be affected by futureextinctions. From a broader perspective, interac-tions between hosts and parasites may be a majorforce promoting both genetic and species diversity innatural communities. Conservation strategies thatresult in loss of parasites could ultimately reducehost populations’ behavioral, physical, and immunedefenses needed to respond to future ecologicalchanges.

Given the explosive growth in scientific under-standing of the ecology and evolution of infectiousdiseases, including noninvasive techniques for col-lecting information on host infection status, hor-mones, and host and parasite genetic identity (see,for example, Krief et al., 2005), scientists are nowwell positioned to address many exciting questions.How do changes in host movement patterns andpopulation structure influence the genetic structure

and evolution of their host-specific parasites? Cangenetic data from contemporary parasite popula-tions provide insights into the evolutionary historiesof host populations? To what degree are multi-ple host resistance traits genetically correlated witheach other and with fitness variables, and how doesthis affect host infection by a diversity of para-sites? Lastly, as humans disturb natural ecosystems,break transmission barriers among species, andreduce host population sizes, outbreaks of emerg-ing infectious diseases among rare or threatenedhost species may become more common. Under-standing the degree to which pathogen evolutionand genetic variation in host resistance traits playa role in disease emergence events in natural sys-tems will improve efforts to manage future dis-eases risks for human populations and in naturalsystems.

SUGGESTIONS FOR FURTHERREADING

Several well-known texts provide a comprehensiveoverview of the recent advances and status of popu-lation biology of infectious diseases in natural pop-ulations, including the edited volumes of Dobsonand Grenfell (1995) and Hudson and colleagues(2002). For a summary of host–pathogen evolution-ary dynamics, Frank (2002) uses a multidisciplinaryapproach to understand immunology within an evo-lutionary context, and Page (2003) provides anengaging review of phylogeny, coevolution, andcospeciation. Several reviews have been publishedrecently that aim to understand the risks posed byinfectious diseases in the context of animal and plantconservation; we specifically recommend Laffertyand Gerber (2002), de Castro and Bolker (2005),and Anderson and associates (2004). Conceptualissues and specific examples of the relevance of host–parasite evolutionary interactions for biodiversityand conservation are summarized in Altizer andcolleagues (2003).

Altizer, S., D. Harvell, & E. Friedle. 2003. Rapidevolutionary dynamics and disease threats tobiodiversity. Trends Ecol Evol. 18: 589–596.

Anderson, P. K., A. A. Cunningham, N. G. Patel,et al. 2004. Emerging infectious diseases ofplants: Pathogen pollution, climate change,and agrotechnology drivers. Trends Ecol Evol.19: 535–544.

CARRO: “CHAP17” — 2008/4/10 — 19:10 — PAGE 280 — #19


de Castro, F., & B. Bolker. 2005. Mechanisms ofdisease-induced extinction. Ecol Lett. 8:117–126.

Dobson, A., & B. Grenfell. 1995. Ecology ofinfectious disease in natural populations.Cambridge University Press,Cambridge, UK.

Frank, S. A. 2002. Immunology and evolution ofinfectious disease. Princeton University Press,Princeton, N.J.

Hudson, P. J., A. Rizzoli, B. T. Grenfell, et al.2002. The ecology of wildlife diseases.Oxford University Press, Oxford, UK.

Lafferty, K., & L. Gerber. 2002. Good medicinefor conservation biology: The intersection ofepidemiology and conservation theory.Conserv Biol. 16: 593–604.

Page, R. D. M. 2003. Tangled trees: Phylogenies,cospeciation and coevolution. University ofChicago Press, Chicago, Ill.

Documents

Host–Pathogen Evolution, Biodiversity, and Disease Risks …pedersen.bio.ed.ac.uk/page3/assets/Altizer_Pedersen_2008.pdf · est birds as a result of avian malaria (Van Riper et