Host-Parasite Interactions of North American Guinea Worm (Nematoda: Dracunculus) and its Mammalian Hosts
By
Sarah Carissa Elsasser
Thesis submitted in partial fulfillment of the requirement for the degree of Doctor of Philosophy (PhD) in Boreal Ecology
School of Graduate Studies Laurentian University Sudbury, Ontario
© Sarah Carissa Elsasser, 2011 Library and Archives Bibliotheque et Canada Archives Canada
Published Heritage Direction du 1+1 Branch Patrimoine de I'edition 395 Wellington Street 395, rue Wellington Ottawa ON K1A0N4 Ottawa ON K1A 0N4 Canada Canada Your file Votre reference ISBN: 978-0-494-87730-2
Our file Notre reference ISBN: 978-0-494-87730-2
NOTICE: AVIS: The author has granted a non L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distrbute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.
The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.
In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.
While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis. Canada
Abstract
Parasites have the potential to negatively influence the fitness of host organisms.
Despite this, the ecology and effects of many wildlife parasites are largely unknown.
Dracunculus insignis is a nematode parasite that infects raccoon (Procyon lotor),
American mink (Neovison vison), and fisher (Martes pennant!). D. lutrae is a morphologically similar dracunculoid, but has only been recovered from river otter
(Lontra canadensis). Until now, species identification of these two North American guinea worms was only achievable by morphology of males and host identification. In this study, DNA barcoding was used to differentiate between D. insignis and D. lutrae, and validated the occurrence of the former in a newly discovered host: the river otter.
The occurrence of D. insignis in river otter highlights the need to supplement identification methods for certain nematodes using DNA techniques. The host specialization and host exploitation strategies of the generalist D. insignis and the specialist D. lutrae were compared. According to the trade-off hypothesis, specialist parasite species should be more 'successful' in terms of prevalence and intensity than generalist parasite species due to a trade-off between how many host species a parasite can exploit and its success. As predicted, the specialist D. lutrae infected their host at higher rates and intensities than the generalist D. insignis, which supports the trade-off hypothesis. This investigation of host exploitation strategies in the generalist D. insignis revealed that this species does not exploit all host species to the same degree. In general, mink and raccoon were more likely to be infected than fisher; however, guinea
iii worm grew to a larger size in fisher. This may reflect a particular strategy that D. insignis has developed in order to increase its chances of transmission in the environment.
Although no evidence of sex or age bias in guinea worm prevalence or intensity were found in this study, results indicate that condition of juvenile male fisher infected with guinea worm decreases with increasing guinea worm infection intensity. Because of the sexual size dimorphism in fisher and because males take longer to mature, males have increased energetic requirements compared to females, possibly making males more susceptible to negative effects of guinea worm parasitism because of their relatively already depleted resources. Decreased body condition due to guinea worm infection may have long-term implications for host fitness or host population health. Potential effects of guinea worm on fecundity of host female fisher were also investigated. No difference in fecundity between uninfected and infected hosts was found, and guinea worm parasitism did not appear to alter the proportion of females able to achieve pregnancy; however, a significant negative relationship was found between infection intensity and fecundity. In natural systems, guinea worm has the potential to play a regulating role in population dynamics of the host.
iv Acknowledgements
Thank you to my supervisor Dr. Albrecht Schulte-Hostedde, and advisory committee members; Drs. Mark Forbes, Mery Martinez-Garcia, Thomas Merritt and Jean-Francois Robitaille. Thanks also to Drs. Paul Hebert and Robin Floyd from the Biodiversity Institute of Ontario, University of Guelph. I am very thankful for the comments and suggestions provided by Drs. Eric Gauthier and Daniel McLaughlin.
Thanks so much to the many faculty and staff of the Biology Department and the Office of Graduate Studies throughout the years. I am very appreciative of my fellow lab mates and the waves of Biology and other LU graduate students with whom I have had the pleasure of interacting with and forming friendships with over the years, particularly Sophie, Anne, Vanessa B., Andreas, Ranji, Dean, Curtis, Mandy, the Spikin' Scientists, the Laurentian Voyeurs, and the Masters of the University.
I am very grateful to the fur trappers who assisted me with sample collections.
This work was supported financially by the Laurentian University Graduate Teaching and Research Assistantships, OGSST-Tembec Scholarship, Canadian Mink Breeders Association Arlen Kerr Memorial Scholarship, Ruffed Grouse Society Wildlife & Conservation Award, and National Science and Engineering Research Capacity Building grants.
I have had unlimited positive energy and encouragement from my family and friends (Sudbury, Yellowknife, and beyond) throughout this process. Very special thanks to my parents, my brother David and sister Mckenna, the Grattons, my Boivin and Elsasser clans, Sydney and the Cotts, and my little buddy Diego. My very caring and loyal family means the world to me. A big masicho to my employer, the Wek'eezhii Land and Water Board, and my coworkers for their support and positive energy in the final stretch.
I am eternally grateful for Pete's unwavering support and encouragement, his discussion and proofreading, his shoulder to cry on, his patience when I was going insane with this, and his optimism that the agony and suffering that persisted throughout the completion of this thesis would eventually end, and be worth it. Thank you Pete. 1 love you very much.
This work is dedicated to my parents. Dad and Mom: thank you for being there for me every step of the way. I am forever appreciative of your endless supply of love, encouragement, and support. I love you both.
v List of Original Papers
This thesis is based on four papers, which will be referred to by their Chapter title.
Chapter 2
Elsasser, S.C., R. Floyd, P.D.N. Hebert, and A.I. Schulte-Hostedde. 2009. Species identification of North American guinea worms (Nematoda: Dracunculus) with DNA barcoding. Molecular Ecology Resources 9: 707 - 712.
Chapter 3
Elsasser, S.C., and A.I. Schulte-Hostedde. Host specialization and exploitation strategies of Dracunculus spp. (manuscript)
Chapter 4
Elsasser, S.C., T. Kraus, J.-F. Robitaille, and G.H. Parker. The influence of guinea worm
(Dracunculus insignis) infection on host body condition and fecundity of fisher (Martes pennanti). (manuscript)
Chapter 5
Elsasser, S.C., and A.I. Schulte-Hostedde. The influence of guinea worm (Dracunculus spp.) on host body condition, (manuscript)
vi Table of Contents
List of Figures viii List of Tables x CHAPTER 1 - General Introduction 1 CHAPTER 2 - Species identification of North American guinea worms 18 (Nematoda: Dracunculus) with DNA barcoding 18 Introduction 18 Methods 21 DNA extraction, amplification and sequencing 23 Data analyses 24 Results 24 Discussion 30 CHAPTER 3 - Host Specialization and Exploitation Strategies of Dracunculus spp 33 Introduction 33 Methods 41 Results 43 Host exploitation strategies of D. insignis and D. lutrae 43 Host sex effect on Dracunculus spp. infection patterns 50 Host age effect on Dracunculus spp. infection patterns 52 Discussion 55 CHAPTER 4 - The influence of guinea worm (Dracunculus insignis) infection on body condition and fecundity of fisher [Martes pennanti) 60 Introduction 60 Methods 63 Prevalence and intensity of guinea worm infections 63 Body Condition 64 Fecundity 65 Statistical Analyses 67 Results 68 Prevalence and intensity of guinea worm infections 68 Body Condition 70 Fecundity 72 Discussion 75 CHAPTER 5 - The influence of guinea worm (Dracunculus sp.) on host body condition 79 Introduction 79 Methods 83 Results 85 Discussion 92 CHAPTER 6 - General Discussion 95 Literature Cited 101
vii List of Figures
Figure 1, Range of mink (Neovison vison) in North America (from Lariviere 1999). N. vison subspecies are represented by numbers 11 Figure 2. Range of fisher (Martes pennanti) in North America (from Powell et al. 2003). 11 Figure 3. Range of raccoon (Procyon lotor) in North America (from Lotze and Anderson 1979) 12 Figure 4. Range of otter (Lontra canadensis) in North America (from Lariviere and Walton 1998). Numbers refer to subspecies 14 Figure 5. Neighbour-joining tree of Kimura 2-Parameter (K2P) distances of COI Dracunculus sequences, with BOLD identifiers, from male specimens (n=15) 26 Figure 6. Neighbour-joining analysis of Kimura 2-Parameter (K2P) distances of COI Dracunculus sequences with BOLD identifiers (n=82). Asterisks indicate specimens from otter that yielded D. insignis sequences 27 Figure 7. Neighbour-joining analysis of Kimura 2-Parameter (K2P) distances of COI Dracunculus lutrae sequences (n=23) with BOLD identifiers. Asterisks indicate multiple nematodes from one individual otter 29 Figure 8. Prevalence (A) and intensity (B) of the generalist Dracunculus insignis and the specialist D. lutrae. Error bars represent 95% confidence interval 46 Figure 9. Prevalence (A) and intensity (B) of Dracunculus insignis between host species. Error bars represent 95% confidence interval. Lowercase letters denote statistical differences. Mink and raccoon have significantly higher prevalence than fisher. ...48 Figure 10. Guinea worm (Dracunculus spp.) length (cm) of infected hosts of each host species. Error bars represent standard deviation of the mean. Letters above error bar denote statistical differences. Guinea worm from fisher were significantly larger than those from mink, while those from raccoon and otter were smallest and not significantly different from each other 49 Figure 11. Prevalence (A) and intensity (B) of guinea worm (Dracunculus spp.) in each sex/age cohort of each host species. Error bars represent 95% confidence interval. 51 Figure 12. Guinea worm {Dracunculus spp.) length (cm) of infected hosts in each sex/age cohort of each host species. Error bars represent standard deviation of the mean. 54 Figure 13. Relationship between body condition (PFAT) and guinea worm intensity of infected individuals (n = 257). One data point of a highly infected (64 worms) fisher has been removed for graphical purposes 71 Figure 14. Relationship between corpora lutea counts and guinea worm intensity in infected female adult fisher 74 Figure 15. Semi-partial correlation between residual host body length (corrected for host body mass) and guinea worm abundance for adult male fisher (n = 39). Smaller adult male fisher tended to have more guinea worms than larger males.
VIII Repeating this analyses with the one outlying data point removed showed that this one case was driving the relationship, as it was no longer significant 89 Figure 16. Semi-partial correlation between residual body length (corrected for body mass) and guinea worm abundance for juvenile female raccoon (n = 19) indicating that small females tended to have more guinea worms than large females 90
ix List of Tables
Table 1. COI sequence divergence (K2P) within Dracunculus species 28 Table 2. Summary statistics of Dracunculus spp. prevalence and intensity with associated 95% confidence limits for each sex and age cohort of each host species examined. Guinea worm length (mean (cm) + standard deviation) in each sex/age cohort of each host species 45 Table 3. Differences in prevalence and intensity between male and female Dracunculus spp. hosts of each age cohort (see Table 2 for prevalence, intensities, and confidence intervals) 50 Table 4. Differences in guinea worm (Dracunculus spp.) length between male and female hosts of each age cohort (see Table 2 for mean lengths and standard deviations). Asterisks denote statistical differences 52 Table 5. Differences in prevalence and intensity between juvenile and adult Dracunculus spp. hosts of each sex cohort (see Table 2 for prevalence, intensities, and confidence intervals). Asterisks denote statistical differences 53 Table 6. Differences in guinea worm (Dracunculus spp.) length between juvenile and adult hosts of each sex cohort (see Table 2 for mean lengths and standard deviations). Asterisks denote statistical differences 53 Table 7. Prevalence, intensity, and associated confidence intervals (CI) of guinea worm (Dracunculus insignis) infection in fisher (Martes pennanti) from various regions of Ontario, Canada 68 Table 8. Prevalence, intensity, and associated confidence intervals (CI) of guinea worm (Dracunculus insignis) infection in fisher (Martes pennanti) from various regions of Ontario, Canada 69 Table 9. Numbers of adult, female fisher uninfected and infected with D. insignis that were able to achieve pregnancy (corpora lutea >1) and that were not able to achieve pregnancy (corpora lutea = 0) for each age class 72 Table 10. Corpora lutea counts of uninfected and D. insignis-infected adult, female fisher per age class 73 Table 11. Summary statistics of Dracunculus sp. prevalence and intensity with associated 95% confidence limits for each host species examined 86 Table 12. Results of logistic regression of host body condition on guinea worm infection status (infected or noninfected). Infection with guinea worm did not have an effect on body condition in mink, fisher, or raccoon. Analyses could not be completed for otter due to the lack of uninfected individuals 87 Table 13. Linear regressions of host body condition on guinea worm abundance 88 Table 14. Linear regressions of host body condition on guinea worm intensity 91
x CHAPTER 1 - General Introduction
Coevolution is the reciprocal evolution of two species - the response to selection imposed by each of the species on the other (Janzen 1980). Host-parasite coevolution, the reciprocal evolutionary change between interacting host and parasite species
(Webster et al. 2004), is one coevolutionary scenario. In a host-parasite relationship, the parasite receives benefits at the expense of the host. The mutual evolution of host resistance (the ability of the host to physiologically prevent establishment and survival of the parasite), and parasite infectivity (the ability of the parasite to infect the host) and/or virulence (the increased host mortality resulting from parasite infection) may drive host-parasite coevolution (Webster et al. 2004). It is believed that natural selection will favor host genotypes that minimize the negative effects of parasites, likely by favoring improved or altered defenses that work against the parasite (Hasu et al.
2009). Improved defenses in the host would then select for parasites with enhanced or altered mechanisms of infectivity and counter-defense. One significant assumption that underlies host-parasite coevolutionary theory is that parasites are detrimental to the fitness of an individual host organism, and animals better able to resist infection will experience higher fitness (Brown et al. 2006).
Due to the potential negative effect of a parasite on its hosts' energy budget, infected animals may have lower energy reserves available to them and thus be in poorer physiological condition (Hawlena et al. 2006). Experimental and observational work has shown that the gastrointestinal nematode Ostertagia gruehneri affects host condition in
1 Svalbard reindeer (Rangifer tarandus platyrhynchus) (Stein et al. 2002). In the wild, female caribou (Rangifer tarandus) show a significant decrease in body weight with increasing abomasal nematode burdens (Hughes et al. 2009). Variation in an individual's body condition - its energetic state - can have important fitness consequences (Seppala et al. 2008). For example, reproductive success is often dependent on body condition
(Hamilton and Bronson 1985, Atkinson and Ramsay 1995, Wauters and Dhondt 1995).
Hosts may also adaptively respond to parasitic infection by decreasing or increasing their investment into reproduction (Forbes 1993). Reduced host reproductive success due to parasitism has been revealed in many species. For example, Hudson (1986) demonstrated experimentally that red grouse (Lagopus lagopus) with reduced levels of intestinal nematode (Trichostrongylus tenuis) infection had larger clutches. Reduced probability of being pregnant due to high warble larvae (Hypoderma tarandi) burdens was found in caribou (Rangifer tarandus) (Hughes et al. 2009). A 'vicious circle' is often created when parasitic infection causes further deterioration in host condition and ultimately influences other factors such as reproductive success (Beldomenico and
Begon 2010). An experimental study of white-footed mice (Peromyscus leucopus) and deer mice (P. maniculatus) demonstrated, at the population level, that the impact of parasites on rodent abundance was exacerbated in the absence of host food supplementation (Pedersen and Greives 2008). In a snail-trematode interaction
(Potamopyrgus antipodarum-Microphallus sp.), hosts in poor condition had higher parasite-induced mortality than hosts in good condition (Krist et al. 2004). Many studies have failed however, to detect any measurable negative parasite- mediated affect on fitness, especially in wild populations (lason and Boag 1988).
Haemogregarine blood parasites were found to have no apparent effect on host fitness in the keelback snake Tropidoriophis mairii (Brown et al. 2006) even though detrimental effects of these parasites on fitness-related variables in the common lizard Lacerta vivipara were found (Oppliger et al. 1996). The benign appearance of some host- parasite associations suggests that these host-parasite interactions may have only trivial consequences for host fitness in natural populations (Brown et al. 2006). An explanation is that the coevolution of parasites and hosts may decrease or eliminate fitness costs of parasitism (Toft and Karter 1990, Toft 1991, Brown et al. 2006).
Hosts in poor condition are often assumed to be more susceptible to parasitic infection; however, there is a growing body of evidence that suggests that host susceptibility to infection is not at all affected by host condition. Krist et al. (2004) found that host condition did not affect susceptibility to infection in a snail-trematode interaction
(Potamopyrgus antipodarum-Microphallus sp.). In some cases, although hosts in poor condition have reduced immunocompetence, parasites may simply avoid them because individuals in poor condition do not provide adequate resources (Bize et al. 2008).
*
In addition to "choosing" hosts in poor condition, parasites have often evolved other host exploitation strategies in order to increase their reproductive success (Seppala et
3 al. 2008). Differences in parasitism rates between host sexes, for example, have been observed in a variety of mammalian taxa, with males generally carrying more parasites than females (Poulin 1996a,b, Schalk and Forbes 1997, Moore and Wilson 2002). Zuk and McKean (1996) report that, particularly with helminths in mammals, males generally show greater intensities (the number of parasites in an infected host) and prevalences (the percentage of host species infected). These differences have been attributed to behavioural, morphological, and physiological reasons such as greater male home range size, male-biased sexual size dimorphism, or immunosuppressive effects of testosterone and other androgens (Folstad and Karter 1992, Sheldon and
Verhulst 1996, Moore and Wilson 2002).
Studies have also found incongruities in parasite size depending on host sex (Poulin
1996b, Dare and Forbes 2008). For example, studies have found Rhabdias ranae lungworms to be larger in female Northern Leopard Frog (Rana pipiens) hosts compared to male hosts (Dare and Forbes 2009), whereas in male Wood Frog (Rana sylvatica) hosts, these same parasites were larger than in female hosts (Dare and Forbes 2008).
Variation in parasite size between male and female hosts may be due to differences in host size (Dare and Forbes 2009). Larger hosts may simply present more available resources to the parasite (Esch et al. 1990). However, differences in parasite size between the sexes may be due to other factors including sex-specific immune responses
(Dare and Forbes 2009).
4 Likewise, parasite prevalence, and intensity may be influenced by host age (Scott 1988).
For example, many parasites are common in young animals and remain prevalent through life, whereas others are very common in young animals, but decrease in prevalence and intensity with age (Scott 1988). This may be due to several factors including host behaviour and/or the adaptive immune response, a specialized system of nonself detection possessed by vertebrates which results in increased resistance to foreign intruders on repeated exposures (Roberts and Janovy 2009).
If direct or indirect effects on fitness due to parasites do occur, parasites may ultimately regulate host population dynamics (Anderson and May 1978, May and Anderson 1978).
This phenomenon has been observed in nature. Hudson et al. (1998), for example, demonstrated that parasitism by the gastrointestinal nematode Trichostrongylus tenuis is a significant factor in regulating populations of red grouse (Lagopus lagopus scoticus) in northern England. The evolutionary and ecological significances of host-parasite interactions, however, remains relatively poorly understood for many organisms, particularly in the wild (Deviche et al. 2001, Brown et al. 2006). This aspect is particularly important as it may have conservation and management implications for the host species.
In order to further address the ecological and evolutionary significances of host-parasite interactions in wild, multi-host populations, I have examined guinea worm Dracunculus spp. (Family Dracunculidae; Order Spirurida), a dioecious nematode endoparasite that
5 infects the tissues and serous cavities of several economically and ecologically important mammals (Chitwood 1933, Crichton 1972, Crichton and Beverley-Burton 1974). In many wild populations, the effects of parasitism on the host are entirely unknown. This is the case with North American guinea worm. This study focuses on guinea worm infection in
American mink (Neovison vison), fisher (Martes pennanti), raccoon (Procyon lotor), and the river otter (Lontra canadensis), and examines the potential effects of guinea worm on host body condition and fecundity to determine if these macroparasites play a potential role in the ecology of host populations. Relationships between guinea worm infection and host physiology could have significant implications for individual host or population fitness and could therefore be considered an important factor in the conservation biology of its host species.
The generalist macroparasite Dracunculus insignis (Leidy, 1858) parasitizes raccoon, fisher, and American mink, and has also been recovered from short-tailed weasel
(Mustela erminea), skunk (Mephitis spp.), opossum (Didelphis marsupialis), muskrat
(Ondatra zibethica) and occasionally domestic dogs in North America (Webster and
Casey 1970, Crichton 1971, Gibson and McKiel 1972, Beyer et al. 1999). D. lutrae
(Crichton and Beverley-Burton, 1973), on the other hand, has only been recovered from the river otter (Anderson 2000). D. insignis and D. lutrae are, therefore, two phylogenetically 'similar' species with different degrees of host specialization.
6 Generalist parasites such as D. insignis, capable of occupying multiple host species, might have a more serious effect on host population dynamics because this type of parasite would be able to exploit more hosts, thereby enhancing its own fitness.
Specialist parasites such as D. lutrae, on the other hand, inhabit only a single host. The advantages of these differing host exploitation strategies are poorly understood. It is acknowledged that generalist parasites can persist following host extinction by continuing to exploit alternative hosts whereas specialist parasites are highly host- specific and their fate is linked closely to that of the host (Poulin and Keeney 2008).
However, it is assumed that generalism must come with a cost given that it is a relatively uncommon lifestyle and, in keeping with the trade-off hypothesis, there would be costs of adaptations against multiple defense systems (Whitlock 1996, Woolhouse et al.
2001). It is thought that because of the costs of adaptations against multiple defense systems, parasites exploiting many host species would be less abundant than more specialist parasites (Woolhouse et al. 2001).
According to the trade-off hypothesis, specialist species, such as D. lutrae, should be more 'successful' in terms of parasitism than generalist species such as D. insignis. This study investigates the parasite ecology of the generalist nematode D. insignis in sympatric host species and compares it to that of the specialist parasite D. lutrae with a particular focus on whether guinea worm infection rates and intensities of infection differ between host sexes and ages. Evidence of a sex bias in parasite success could
7 have a range of evolutionary implications because it could suggest that higher levels of parasitism might be a relative cost associated with that sex.
Both D. insignis and D. lutrae are endemic to North America, however their distribution ranges are not precisely known. Crichton and Beverley-Burton (1973) found that the distributions of D. insignis and D. lutrae overlap in the southern part of Ontario. D. insignis prevalence values were also found to be greater than 50% in both raccoon and mink (Crichton and Beverley-Burton 1973). A higher prevalence of D. insignis in mink was found in southern Ontario, where raccoon are abundant, than in northern regions, where raccoon are seldom found and prevalence in mink is extremely low (Crichton and
Beverley-Burton 1974). There is also some experimental evidence that mink may be somewhat refractory to D. insignis infection compared with raccoon (Crichton 1972). It is possible then that raccoon may serve as a reservoir of D. insignis infection and that the range of D. insignis may be limited to that of raccoon (Crichton and Beverley-Burton
1974). In contrast, D. lutrae was found across the entire province of Ontario and, in most areas, the prevalence exceeded 75% (Crichton and Beverley-Burton 1974). D. lutrae is thought only to occur in river otter (Crichton and Beverley-Burton 1974), therefore its range could be limited to that of river otter.
American mink are found throughout Canada and most of the United States except
Arizona and parts of California, Nevada, Utah, New Mexico, and western Texas (Lariviere
1999; Figure 1). This mustelid was introduced in Russia and in other parts of Europe. In
8 addition, escapes from fur farms have also allowed this species to establish populations throughout Europe and South America. Wild mink in Canada may be declining (Bowman et al. 2007). It has been suggested that causes of declines in natural mink populations may include effects of organochlorine chemicals and mercury on mink reproduction, and/or the effects of interbreeding of escaped domestic mink with wild mink (Bowman et al. 2007, Kidd et al. 2009, Nituch et al. 2011). Mink are usually associated with water, although the species can be found in xeric habitats if food is abundant (Lariviere 1999).
Mink are strictly carnivorous, and their diet reflects the local prey-base, typically comprising offish, amphibians (mostly frogs), crustaceans (crayfish and crabs), muskrats, and small mammals (Lariviere 1999). Opportunistically, mink also consumes lagomorphs, birds and their eggs, reptiles and their eggs, aquatic insects, earthworms, and snails (Lariviere 1999).
The fisher is endemic to North America and restricted to coniferous forest of the boreal zone and its southern peninsular projections along the Appalachian and Pacific Coast mountain ranges (Powell et al. 2003; Figure 2). Between 1800 and 1940, fisher populations declined or were extirpated large portions of Canada and most of the
United States due to overtrapping and habitat destruction via logging and development
(Powell 1981). Conservation efforts (closed trapping seasons, habitat recovery programmes and re-introductions) have allowed fisher to return to much of their former range (Powell et al. 2003). Despite the efforts however, fisher are considered to be rare or even extirpated in many areas (Gibilisco 1994, Thompson 2000). Fisher prefer habitat
9 with extensive, continuous canopy and generally avoid forests with little overhead cover and open areas (Powell 1981). This member of Mustelidae is a generalized predator whose major prey are small to medium-sized mammals and birds, and carrion, however, when abundant, snowshoe hares, squirrels, mice, shrews, and porcupine are commonly consumed (Powell 1981).
10 Sylpl
PW \ r\ 0 »00 1000 ^\V
•" ^\ r I
Figure 1. Range of mink (Neovison vison) in North America (from La riviere 1999). N. vison subspecies are represented by numbers.
Kilometers 0 400 800
Figure 2. Range of fisher (Martes pennant!) in North America (from Powell et al. 2003).
11 The transcontinental range of raccoon, the only member of the Procyonidae in temperate regions of North America, extends from southern Canada to Panama (Lotze and Anderson 1979; Figure 3). At the turn of the nineteenth century, the northernmost boundary of its distribution was near the Canada-United States border, and raccoons occurred infrequently in Canada (Lariviere 2004). Since this time, raccoons have colonized most of southern Canada, and their range is currently expanding northward
(Lariviere 2004). Raccoons are now widely distributed throughout North America (Gehrt
2003). Raccoons eat a wide range of both animal and plant matter including arthropods, berries, nuts, grains, seeds, and anthropogenic food such as human refuse, gardens, and pet food (Lotze and Anderson 1979, Lariviere 2004).
Figure 3. Range of raccoon (Procyon lotor) in North America (from Lotze and Anderson 1979).
12 The North American river otter occurs in all Canadian provinces and territories, except for Prince Edward Island, and in the United States in New England, the states bordering the Great Lakes, Atlantic ocean and Gulf of Mexico, the forested regions of the Pacific coast, and Alaska, including the Aleutian Islands and the north slope of the Brooks
Range (Lariviere and Walton 1998; Figure 4). Historically, this mustelid was present throughout most of Canada and the United States; however, its range was greatly reduced due to pollution and urbanization (Lariviere and Walton 1998). Reintroductions have expanded the distribution of this species in recent years, especially in the mid- western United States (Lariviere and Walton 1998). Otter inhabit freshwater habitats ranging from alpine rivers, lakes, and streams to estuarine marshes and cypress swamps
(Kimber and Kollias 2000). The diet of the aquatic otter is comprised mostly of fish, however, amphibians (mostly frogs) and crustaceans (mainly crayfish) may constitute an important part of the diet and small mammals, mollusks, reptiles, birds, and berries are consumed opportunistically (Lariviere and Walton 1998). Lariviere and Walton (1998) note that otters may compete for resources with the American mink.
13 7 7771
K
Figure 4. Range of otter (Lontra canadensis) in North America (from Larivi&re and Walton 1998). Numbers refer to subspecies.
The definitive host of Dracunculus sp. becomes infected either by drinking water containing infected intermediate hosts or by ingesting a paratenic host that serves as a transport vector within the food chain (Anderson 2000). It has been established experimentally that frogs (Xenopus and Rana spp.) may be potential paratenic
(transport) hosts to D. insignis (Crichton and Beverley-Burton 1977, Eberhard and
Brandt 1995). It has also been suggested that mammals may be involved in paratensis of
D. insignis (Crichton and Beverley-Burton 1977).
Once ingested by the definitive host, infective larvae of D. insignis migrate through the intestinal wall, across the abdominal cavity, and into the connective tissues of the trunk,
14 abdomen, and inguinal regions (Crichton and Beverley-Burton 1974, 1975, 1977).
Mating occurs in the musculature of the back. Gravid females migrate to the limbs
(carpal and tarsal areas) where ulcerative skin lesions are formed. When the affected area comes into contact with water, the numerous larvae contained within the anterior end of the worm are expelled. Muller (1971) described the symptoms of guinea worm infection in humans, which included intense burning and itching at the time of blister formation. Crichton (1972) noted that raccoons were observed to scratch at areas of the skin where female worms were located. It is thought that once larvae are expelled, mature female worms die and become calcified and/or reabsorbed by the host tissue.
Male worms and unfertilized female worms are believed to remain in the musculature of the abdomen and back until they die and are eventually reabsorbed (Crichton and
Beverley-Burton 1974,1975,1977).
The prepatent period of D. insignis, the time interval from first infection of the definitive host by larvae until first-stage larvae are detected in a cutaneous lesion (Crichton and
Beverley-Burton 1975), has been determined in experimentally-inoculated ferrets
[Mustela putorius furo) (Brandt and Eberhard 1990a,b), raccoon (Crichton and Beverley-
Burton 1977), and mink (Crichton and Beverley-Burton 1974) to be approximately one year. Similarly, Muller (1971) estimated the prepatent period of D. medinensis, the human guinea worm, to also be approximately one year.
15 Classification of nematode parasites is typically based on the type of host and morphological features of the organism (Anderson 2000). In the case of the guinea worms, morphological classification of the members of the Dracunculus genus is based on key features that are only present in male worms. In the case of D. insignis (and the human guinea worm D. medinensis), males are rarely available for study (Brandt and
Eberhard 1990a). Consequently, morphology-based classification of Dracunculus spp. has proven rather complex (Muller 1971, Bimi et al. 2005, Wijova et al. 2005) and identification is frequently questioned because of inadequate descriptions and/or morphological similarities (Muller 1971, Wijova et al. 2005). It has been suggested that it is impossible to morphologically distinguish between D. medinensis and D. insignis, and based on cross-infection experiments, determined that D. medinensis and D. insignis may be conspecifics of different physiological 'strains' (Crichton and Beverley-Burton
1974,1975). However, Bimi et al. (2005) and Wijova et al. (2005) have recently shown that it is possible to distinguish D. medinensis from both D. insignis and from the reptile- infecting D. oesophageus based on 18S rRNA gene sequences.
D. lutrae is distinguished from the other mammal-infecting species of Dracunculus by the greater length of males, greater length of spicules and gubernaculum, presence of only three pairs of preanal papillae, and the arrangement of papillae in two transverse rows immediately posterior to the anus (Crichton 1972). Species identification of the
North America mammal-infecting guinea worms, D. lutrae and D. insignis is, therefore, usually accomplished by host identity. This study genetically differentiates between the two North American guinea worm species; D. lutrae and D. insignis, and investigates the possibility of host-related genetic divergence. In the case of D. insignis, significant host-associated genetic differences may shed light on the presence of host specialization or host races. If this generalist parasite has become specialized to a specific host, it is expected that it will show high levels of sequence divergence with respect to the host species.
The aims of this study were thus to investigate: (1) whether molecular characterization techniques such as mitochondrial cytochrome c oxidase I barcoding analysis can be used for reliable species identification of D. insignis and D. lutrae; (2) the parasite ecology of the generalist nematode D. insignis in sympatric host species and of the specialist parasite D. lutrae with a particular focus on possible host sex and age differences in guinea worm infection rates and intensities of infection; and (3) whether nematode parasitism, particularly guinea worm infection, has any relationship with host physical condition and/or fecundity.
17 CHAPTER 2 - Species identification of North American guinea worms (Nematoda: Dracunculus) with DNA barcoding
Introduction
The importance of accurately and efficiently identifying biological diversity is becoming ever more evident. However, morphological approaches to species identification can produce incorrect identifications due to phenotypic plasticity, genetic variability, morphologically cryptic taxa and differences in particular life stages or gender (Hebert et al. 2003a). Recent studies propose that DNA barcoding will often circumvent these problems (Hebert et al. 2003a, Savolainen et al. 2003). The barcoding method uses a short, standardized genetic marker as a tool for identification. A 658-bp region of the mitochondrial cytochrome c oxidase I (COI) gene is the barcode standard for the animal kingdom (Hebert et al. 2003a,b). Most eukaryotic cells contain mitochondria, and in comparison to the nuclear genome, mtDNA lacks introns, is less exposed to recombination, has a haploid mode of inheritance and has a relatively high mutation rate (Saccone et al. 1999). This results in significant variance in mtDNA sequences between species and a comparatively small variance within species.
Taxonomists have begun using DNA barcodes as a 'triage' tool for sorting specimens into groups, of which some will belong to known species and others will be new to science.
DNA barcoding also allows the identification of specimens in those cases where morphological features are missing (in the case of immature, partial or damaged specimens) or misleading (as in sexually dimorphic species) and as a supplement to other taxoriomic datasets to aid the delimitation of species boundaries (Schindel and
Miller 2005). For example, a recent DNA barcoding study (Hebert et al. 2004) found the common, neotropical skipper butterfly (Astraptesfulgerator), previously thought to be a single species, is a complex often species. This method may prove to be especially attractive for taxonomists identifying parasites whose morphological identifications are often difficult (Besansky et al. 2003, Powers 2004).
Currently, the identification of many nematode parasites is based on a combination of host use and morphology (Anderson 2000). For example, the guinea worms, members of the genus Dracunculus infecting mammals, are identified using morphological features that are only present in males, which are rarely available for study (Anderson
2000). Consequently, the identification of Dracunculus species based on morphology has proven complex (Muller 1971, Bimi et al. 2005, Wijova et al. 2005) and identifications are frequently questioned because of inadequate descriptions and/or morphological similarities (Muller 1971, Wijova et al. 2005). For example, Beverley-Burton and Crichton
(1973,1976) found that it was impossible to morphologically distinguish the human guinea worm, D. medinensis (Linnaeus, 1758), and D. insignis (Leidy, 1858) from North
American mammals such as mink (Neovison vison), raccoon (Procyon lotor) and fisher
[Martes pennanti). In fact, based on cross-infection experiments, they suggested that D. medinensis and D. insignis might simply be different physiological 'strains' of a single species. However, Bimi et al. (2005) and Wijova et al. (2005) have recently shown that it
19 is possible to distinguish D. medinensis from both D. insignis and from the reptile- infecting D. oesophageus based on 18S rRNA gene sequences.
To date, species identification of the North American mammal-infecting guinea worms,
D. lutrae and D. insignis, has only been achieved by morphology. D. lutrae has only been recovered from river otter (Lontra canadensis) and is distinguished from other mammal- infecting species of Dracunculus by the greater length of males, greater length of spicules and gubernaculum, presence of only three pairs of pre-anal papillae, and the arrangement of papillae in two transverse rows immediately posterior to the anus
(Crichton and Beverley-Burton 1973). However, in cases where males are lacking, specimens are identified by host alone (Bimi et al. 2005, Wijova et al. 2005).
Furthermore, the extent of phenotypic plasticity in these two species has not been examined.
Studies have found that due to the close association between parasites and their hosts, many 'generalist' parasites become specialized on different host species, often resulting in host race formation (Jaenike 1993, McCoy et al. 2001). Mayr (1970) defined host races as "non-interbreeding sympatric populations that differ in biological characteristics but not, or scarcely, in morphology" and suggested they were prevented from interbreeding by preference for different food plants or other hosts. Jaenike (1981) distinguished host races from sympatric host-associated sibling species by adding that
"if gene flow among two or more populations was restricted solely or primarily because of differential host preference, then these would constitute host races". He argues that if this basis for reproductive isolation were not present, host races would fuse into a single panmictic population, whereas sibling species would maintain their separate genetic identities. As Anderson and Jaenike (1997) point out, the understanding of the epidemiology of many parasitic diseases is severely hampered by the existence of morphologically identical cryptic species and host races. Recently, the DNA barcoding initiative is proving to be a useful tool when investigating host specialization or host races (Smith et al. 2007).
This study attempts to differentiate between the two North American guinea worm species, D. lutrae and D. insignis, and investigates possible host-related genetic divergence in D. insignis by sequencing a gene region that is less conserved than 18S rRNA. In the case of D. insignis, DNA barcoding of the COI gene may shed light on the presence of host specialization or host races if there are significant host-associated differences. If this generalist parasite has become specialized to a specific host, it is expected that the parasite will show high levels of sequence divergence with respect to the host species.
Methods
Fisher, mink, raccoon, and otter carcasses were collected during the fur harvests of
2005-2006 and 2006-2007 from licensed Ontario fur trappers on registered trap lines in
21 northeastern, southern and southeastern Ontario, Canada and frozen until dissection
(<6 months at -18°C).
During necropsies for guinea worm, external surfaces of all superficial muscles, intermuscular areas of the legs and feet, connective tissue beneath the latissimus dorsi and of the inguinal and axillary regions, and internal surfaces of the abdominal and pelvic cavities of each animal were examined. The prevalence of infection was high-
20/33 mink, 18/31 raccoon, 20/51 fisher and 20/21 otter were infected with guinea worm. Guinea worms were counted, sexed, measured with digital calipers (+0.1mm), weighed using an ACCULAB scale (+ 0.001 g), and preserved in 70% ethanol. Tissue from one specimen of D. lutrae from each of 20 individual otter hosts and one specimen of D. insignis from each of 20 individual fisher, 18 individual raccoon and 20 individual mink specimens were sequenced at the Canadian Centre for DNA Barcoding. As well, 7 additional guinea worms from one of the infected otters and 7 from one of the infected mink were included for within-host comparisons. Prior to submission, male specimens from fisher (n=2), mink (n=l), raccoon (n=2) and otter (n=ll) were identified to the species level following Crichton and Beverley-Burton (1973). Female guinea worms from fisher (n=18), mink (n=26), raccoon (n=15) and otter (n=16) and 1specimen of unknown sex from raccoon were classified to the species level according to their host. Specimens were deposited in the Canadian National Collection of Nematodes, Ottawa, Ontario,
Canada.
22 DNA extraction, amplification and sequencing
DNA extracts were prepared by placing 1-2 cm of ethanol preserved tissue from each specimen directly into 96-well plates containing lysis buffer and proteinase K. DNA extraction employed a glass fibre protocol (Ivanova et al. 2006). The 658-bp target region of COI was amplified by polymerase chain reaction (PCR); each 12.5 |iL PCR reaction included 6.25nL of 10% trehalose, 1.25|iL 10X PCR buffer, 0.625nL (2.5mM)
MgCI2,0.125|iL (10nM) each oligonucleotide primer, 0.625|iL (lOmM) dNTPs, 0.625nL
Taq polymerase and 4^L H20+template DNA (Hajibabaei et al. 2005). PCR reactions were run at the following thermal cycle conditions: 1min at 94°C followed by 5 cycles of
30 s at 94°C, 40 s at 50°C, and 1min at 72°C, followed by 35 cycles of 30 s at 94°C, 40 s at 55°C, and 1min at 72°C, and finally 10 min at 72°C. The standard invertebrate primer pair LC01490 (5'-GGTCAACAAATCATAAAGATATTGG-3') with HC02198 (5'-
TAAACTTCAGGGTGACCAAAAAATCA-3') was used (Folmer et al. 1994). Additionally, all primers were tailed with standard flanking sequences (M13F: 5'-
TGTAAAACGACGGCCAGT-3'; M13R: 5'-CAGGAAACAGCTATGAC-3') to allow subsequent sequencing with M13 primers. PCR products were visualized in a 96-well E-Gel
(Invitrogen), and sequenced (forward and reverse reads using M13F and M13R primers) on an ABI 3730 Genetic Analyzer (Hajibabaei et al. 2005). Sequences were assembled and edited using SeqScape software (Applied Biosystems) and entered into the Barcode of Life Data Systems (BOLD) database; sequences were also deposited in GenBank with the accession numbers EU646534-EU646615.
23 Data analyses
Sequences and original trace files are available in the 'Nematode Parasites of Canadian
Mammals' project on the Barcode of Life Data Systems (BOLD, www.barcodinglife.org) and on GenBank (http://www.ncbi.nlm.nih.gov/Genbank/). Sequences were aligned using the ClustalW (Thompson et al. 1994) module with MEGA version 4.0 (Tamura et al.
2007). The Kimura two-parameter (K2P) model of base substitution (Kimura 1980) was used as a simple measure of pairwise sequence distances. To visualize these distances, a neighbour-joining (NJ) tree with bootstrap analysis (500 replications) of K2P sequence distances showing intraspecific and interspecific variation was created using MEGA. K2P sequence divergences at both levels were determined using the 'Distance Summary' tool on BOLD.
Results
In total, 82 of 92 specimens were sequenced successfully; all reads were between 496 and 660 bp in length. The shorter sequences were due to a small number of chromatograms showing low-quality signal at one end or another; the ambiguous bases were trimmed so that only high-quality data were used. Alignment was straightforward as the sequences represent protein-coding genes containing no insertions or deletions, which would alter the reading frame. Aligned sequences are available directly as a download from the BOLD project file. A NJ tree of Kimura 2-Parameter (K2P) distances of COI Dracunculus sequences from male D. insignis and D. lutrae specimens shows that
24 the two species (n=15) are clearly separated with a mean interspecific sequence divergence of 9.82% (Figure 5).
25 65j RFNPM08&C7 Dracunculus lubae|Otter I RFNPM06&07 Dracunculus lutrae|Ctter — RFNPMO06-O7 DracuiaJus lutrae|Otter 59 RFISPM07&07 Dracunculus lutrae|Otter RFNPM078
0.01
Figure 5. Neighbour-joining tree of Kimura 2-Parameter (K2P) distances of COI Dracunculus sequences, with BOLD identifiers, from male specimens (n=15).
A NJ tree of Kimura 2-Parameter (K2P) distances of COI Dracunculus sequences from male and female specimens shows that, regardless of sex, the two species are clearly separated (Figure 6). When considering both sexes, the mean K2P sequence distance between Dracunculus insignis and D. lutrae was 9.64%. Two female specimens obtained from two separate otter hosts yielded sequences that group with specimens recovered from fisher, raccoon or mink rather than with other specimens from otter (Figure 6).
These two specimens are considered as D. insignis for further analyses.
26 FWMmxr OeimAe
orHPMmoro FFWe«7- WfUFMMfrqr Qwiaia • ww^wtw Hr*#»MCW4 DMialanoaM* IWMMffQwioMw^wllW lfmiiOIMVOaMO*ang«M>* *9*f*J€OMS/ Own** ny»Mi - WIKBMf Dwiato mfMDiurcMMounvMM* ifMMO/^DnRu PVfCMD»14V Dnnxu Figure 6. Neighbour-joining analysis of Kimura 2-Parameter (K2P) distances of COI Dracunculus sequences with BOLD identifiers (n=82). Asterisks indicate specimens from otter that yielded D. Insignis sequences. 27 Despite their recovery from four different hosts, specimens of D. insignis showed very little sequence variation; intraspecific divergence averaged just 0.02% and the maximum divergence value between any pair of individuals was 0.31%, representing just two nucleotide changes (Table 1). Table 1. COI sequence divergence (K2P) within Dracunculus species. COI Sequence Divergence (K2P) Mean SE Minimum Maximum N Dracunculus insignis 0.021 0.001 0 0.325 59 Dracunculus lutrae 0.327 0.013 0 0.621 23 Intra-specific variation was higher in D. lutrae, averaging 0.33% and reaching a maximum pair-wise divergence of 0.62%. The very low levels of sequence divergence in D. insignis prevented any analysis of the number of separate infections in single hosts. However, by analyzing multiple nematodes from one otter (Figure 7), it was possible to show that it had experienced at least two infections. 28 RFNPM06Q07 Dacunculus lutraejOtter RFNPM074-07 Dracunculus lutraejOtter 3 87 RFM M079-07 Dnacuroius lutraejOtter RFNPM071-07 Dracunculus lutraejOtter RFNPM07CWJ7 Dracunculus lutraejOtter RFNFM072-07 Dracunculus lutraejOtter RFNPM075-07 Draanaius lutraejOtter RFNPM077-07 Dracunculus lutraejOtter RFISPM06&C7 Dracunaius lutraejOtter RFNFMO0&O7 Dracunculus lutraejOtter RFNPM061-07 Dracunculus lutraejOtter RFNPM092-07 Dracunculus lutraejOtter * RR>PM062-07 Dracunaius lutraejOtter RFNPM067-07 Draarculus lutraejOtter RFNPM00007 Dracunculus lutraejOtter RFNPM094-07 Dracunaius lutraejOtter RFNPM06S 66 • RFNPMO09-O7 Dracunaius lutraejOtter I7I RFNPM06&07 Dracunculus lutraejOtter 0.0005 Figure 7. Neighbour-joining analysis of Kimura 2-Parameter (K2P) distances of COI Dracunculus lutrae sequences (n=23) with BOLD identifiers. Asterisks indicate multiple nematodes from one individual otter. 29 Although methods commonly associated with phylogenetics (e.g. K2P divergences, neighbour-joining trees) were used for analysis of patterns of sequence divergence, it should be noted that the purpose of this study was not phylogenetic reconstruction but species discrimination. DNA barcoding reliably differentiated D. insignis and D. lutrae as the two species were represented by distinct, non-overlapping clusters of sequences in the neighbor-joining (NJ) tree. Two specimens from separate otter hosts yielded sequences that grouped with D. insignis rather than with D. lutrae. Reports have indicated that D. insignis is capable of infecting otter, although none of these cases provide descriptions or accurate identification of the species (see Kimber and Kollias 2000). This study is the first to report D. insignis infections in otter with certainty. In each of these two otter hosts, only one female guinea worm was recovered. However, because both specimens were gravid females, these cases were viable infections. The fact that no males were present, though gravid females were, is not uncommon. From inoculation experiments of ferrets (Mustela putorius furo) with D. insignis, Brandt and Eberhard (1990b) found that the recovery of female worms was at least twice as great as for male worms. In one group of ten animals infected with approximately 100 third stage D. insignis larvae, 29 (66%) worms recovered were female and 15 (34%) were male (Brandt and Eberhard 1990b). In a second group of 21 ferrets inoculated with 50 third stage guinea worm larvae, 83 (90%) of the worms were female and 9 (10%) of the worms were male (Brandt and Eberhard 1990b). In a separate experiment, Brandt and Eberhard (1990b) found that of 19 D. insignis recovered from 7 of 10 infected ferrets, 74% were gravid females, 10% were immature females and 16% were males with only 2 animals harboring both male and gravid female worms. As Brandt and Eberhard (1990b) point out, mature females are large and difficult to overlook, whereas males are small and could easily be missed. It has also been suggested that Dracunculus males die after mating and are reabsorbed by the host (Crichton 1972), therefore the difference in recovery rates of male and female worms may be reflected by this. Eberhard et al. (1988) did, however, recover living male worms at 200 days of age in experimentally infected ferrets, indicating that not all male worms die shortly after mating. In this study, DNA barcoding revealed that otter are not only infected with D. lutrae, but also with D. insignis. Without male specimens present in an infected otter, it would not be possible to confidently determine species based on morphology, however, DNA barcoding would permit species assignment. Barcoding is also likely to be an effective tool for differentiating other members of the Dracunculus genus, such as D. medinensis, as well as potentially identifying other life cycle stages. For example, it is suggested that transmission of D. insignis and D. lutrae from the intermediate to definitive host may involve a paratenic host (Eberhard and Brandt 1995). However, neither naturally infected intermediate hosts nor naturally infected paratenic hosts have been identified. In addition, there have been occasional reports of human dracunculiasis in countries that are thought to be free of D. medinensis (Bimi et al. 2005). Bimi et al. (2005) suggested that these infections could have involved species of Dracunculus, other than D. medinensis, that had been acquired by humans when they inadvertently ingested a paratenic host of another species of the parasite; DNA barcoding may well answer such questions. Individuals of the D. insignis group showed little sequence divergence regardless of the source host, indicating that this species is a 'true' generalist that parasitizes at least four mammal species, including the otter. In contrast, despite its host specialization, D. lutrae included several separate mitochondrial lineages. Nematodes from a single otter had high sequence divergence, indicating that this individual was infected on several occasions. This result suggests that D. lutrae is both genetically diverse and panmictic; differing in this regard from the common presumption that genetic differentiation is occurring among parasites from different hosts (Price 1980). 32 CHAPTER 3 - Host Specialization and Exploitation Strategies of Dracunculus spp. Introduction Parasitism, a symbiotic relationship where one organism benefits at the expense of the other, is a very successful lifestyle; it has evolved independently in nearly every phylum of animals, and there are far more parasitic organisms than nonparasitic organisms in the world (Roberts and Janovy 2009). In order to achieve such widespread success, parasites have had to evolve different survival and reproductive strategies due to the complexity and heterogeneity of the parasites' environments. First, the host itself could have varying degrees of resistance, condition, be male or female, young or old (Thomas et al. 2002). Also, endoparasites have little option of changing hosts (Thomas et al. 2002). Furthermore, the ecosystem in which the host and the host population live could have varying levels of resources or host population density (Thomas et al. 2002). Host specialization and host exploitation strategies, such as sex-biased parasitism, are some of the adaptive features that help parasites to cope with their ever-changing environments (Ward 1992, Poulin 1996b, Thomas et al. 2002). A parasites' degree of host specialization, it's host specificity, is defined as the range of host species in which a parasite is capable of developing in (Roberts and Janovy 2009). Host specificity can arise in two distinct ways. Long-term associations of parasites and hosts that persist through speciation events can result in the restriction of sister parasite lineages to sister host lineages or a parasite lineage that is initially capable of utilizing 33 several species of hosts, may become restricted to a subset of them (Brooks 1979, 1988). A generalist parasite is capable of exploiting a wide range of host species, and by contrast, can persist following host extinction by continuing to exploit alternative hosts (Lymbery 1989, Poulin and Keeney 2008). A highly host-specific parasite, on the other hand is restricted to one host species, and its fate is closely linked to that of the host (Lymbery 1989, Poulin and Keeney 2008). The generalist versus specialist strategy for a parasite entails increasing or decreasing the range of hosts that the parasite can occupy (Sasal et al. 1999). Parasites typically have a narrow range of hosts and are viewed as being highly host specific, however, variation in specificity does occur (Poulin and Keeney 2008). The resource breadth hypothesis suggests that a species using a diversity of resources can attain a broad distribution and high local density (Brown 1984,1995). Krasnov et al. (2004) found that fleas exploiting many small mammal host species, or taxonomically unrelated hosts, achieved higher abundance than specialist fleas. Alternatively, the trade-off hypothesis proposes a negative relationship between how many host species a parasite can exploit and its reproductive potential (Poulin 1998). Here, it is assumed that generalism must come with a cost because of the need for adaptations against multiple defense systems (Whitlock 1996, Woolhouse et al. 2001). Because of the costs of these adaptations, parasites exploiting many host species are thought to be less successful in terms of abundance than more host-specific parasites (Poulin 1998, Woolhouse et al. 2001). Poulin (1998), for example, observed a negative relationship between the number of fish species used by 188 species of metazoan parasites and their average abundance in hosts. In some cases however, the rate and severity of parasitism was found to be independent of host specificity (Morand and Guegan 2000). In their study of 828 populations of adult parasitic gastrointestinal nematodes of 66 different terrestrial mammal species, Morand and Guegan (2000) found no relationship between host- specificity and either parasite prevalence or abundance. Furthermore, a generalist parasite with a wide host range may not exploit all of its host species equally (Thomas et al. 2002, Poulin and Keeney 2008, Leung and Poulin 2010). For example, Olstad et al. (2007) found natural variation in host preference of Gyrodactylus salaris. This monogenean ectoparasite, pathogenic to the usual host Atlantic Salmon (Salmo salar), was found naturally occurring in a population of Arctic Charr (Salvelinus alpinus), and was not occurring in nor pathogenic to sympatric Atlantic Salmon (Olstad et al. 2007). Parasites may also exploit hosts differently based on the hosts' sex (Scott 1988, Forbes and Baker 1991). Differences in parasitism rates have been observed in a variety of mammalian taxa, with male hosts generally carrying more parasites than females (Poulin 1996b, Schalk and Forbes 1997, Moore and Wilson 2002). Potential drivers for this male sex-biased parasitism may include male-biased sexual size dimorphism, the immunosuppressive effects of androgens, and the fact that in mammals, males often have larger home range sizes than females, increasing their probability of encountering parasites (Folstad and Karter 1992, Sheldon and Verhulst 1996, Moore and Wilson 2002). It has been suggested that this male-biased susceptibility to parasitism may be a contributor to male-biased mortality (Moore and Wilson 2002, Owens 2002). Parasites may also be more successful in hosts with reduced potential for immune response (Dare and Forbes 2008). Larger-sized helminths have been observed both in male (Dare and Forbes 2008) and in female hosts (Dare and Forbes 2009). For nematodes, larger size in females generally confers greater fecundity and therefore a fitness advantage (Poulin 1996a, Tompkins and Hudson 1999). Variation in parasite size between male and female hosts may simply be due to differences in host size (Dare and Forbes 2009), where a larger host individual offers more resources than a smaller one (Esch et al. 1990). Finally, disparity in parasite size between the sexes may also be due to different, sex-specific immune responses (Dare and Forbes 2009). Parasite prevalence, intensity, and size may also be influenced by host age (Scott 1988). Many parasites are common throughout the life of the host, whereas others decrease in prevalence and intensity as the host ages (Kisielewska and Zubczewska 1973, Scott 1988). These patterns may be influenced by factors including host behaviour and/or the adaptive immune response. The adaptive immune response allows identification of nonself intruders and results in increased resistance to them on repeated exposures (Roberts and Janovy 2009). 36 In this study, host specialization and host exploitation strategies of two naturally occurring and closely related species of the guinea worm, Dracunculus insignis (Leidy, 1858) and D. lutrae (Crichton and Beverley-Burton, 1973) were examined. Dracunculus spp. (Family Dracunculidae; Order Spirurida) are endoparasitic nematodes infecting the tissues and serous cavities of a variety of mammals (Chitwood 1933, Crichton 1972, Crichton and Beverley-Burton 1974). The generalist Dracunculus insignis is a parasite of American mink (Neovison vison), fisher (Martes pennanti), and raccoon (Procyon lotor), and has also been reported from short-tailed weasel (Mustela erminea), skunk (Mephitis spp.), opossum (Didelphis marsupialis), muskrat (Ondatra zibethica) and domestic dogs (Canis familiaris) in North America (Webster and Casey 1970, Crichton 1971, Gibson and McKiel 1972, Beyer et al. 1999). D. insignis infection has also been recently confirmed in the North American river otter (Lontra canadensis) (Elsasser et al. 2009). Elsasser et al. (2009) established that D. insignis is a 'true' generalist. D. lutrae is presumed to be a specialist as it has only been recovered from the river otter (Crichton and Beverley- Burton 1973). While both species are endemic to North America, the ranges of D. insignis and D. lutrae are not entirely known. Crichton and Beverley-Burton (1973) found that the distributions of D. insignis and D. lutrae overlap in the southern part of Ontario where D. insignis prevalence values were found to exceed 50% in both raccoon and mink. A higher prevalence of D. insignis in mink was found in southern Ontario, where raccoon are more abundant, compared to northern Ontario, where raccoon are seldom found 37 (Crichton and Beverley-Burton 1974). Crichton and Beverley-Burton (1974) suggested that raccoon may serve as a reservoir of D. insignis infection and that the range of D. insignis may be limited to that of raccoon. Crichton (1972) proposed that mink might be somewhat more refractory to D. insignis infection compared to raccoon. D. lutrae, on the other hand, has been found across the entire province of Ontario and, in most areas, the prevalence exceeds 75% (Crichton and Beverley-Burton 1974). D. lutrae is thought only to occur in otter (Crichton and Beverley-Burton 1974), therefore its range would likely be limited to that of otter. The definitive host of Dracunculus spp. becomes infected either by drinking water containing infected intermediate hosts or by ingesting an infected paratenic host which is an organism that serves to transfer a larval stage or stages from one host to another, but in which little or no development takes place (Anderson 2000). Neither naturally infected intermediate nor naturally infected paratenic hosts have been found. However, Crichton and Beverley-Burton (1975) determined experimentally that, similar to other dracunculoids, larvae of D. insignis are capable of developing in cyclopoid copepods such as Cyclops sp., while Eberhard and Brandt (1995) established experimentally that African clawed frogs (Xenopus iaevis) and bullfrogs (Rana catesbeiana) were capable of serving as paratenic hosts in dracunculid life cycles. Little is known about the transmission and development of D. lutrae; it is assumed to be similar to that of D. insignis (Anderson 2000). 38 In experimental studies with D. insignis, Crichton and Beverley-Burton (1975) found that once ingested, larvae migrated from the duodenum, across the abdominal cavity, and into the thoracic and abdominal musculature. Mating of Dracunculus spp. occurs in the musculature of the back (Crichton 1972). Once fertilized, gravid females migrate to the limbs, growing and filling with ova (Crichton and Beverley-Burton 1975). Once the host steps into a water source, the females release their larvae through lesions in the hosts' skin in the wrist and ankle regions (Crichton 1972). Male worms and unfertilized female worms die in the musculature of the back. It is thought that all worms are reabsorbed by the host (Crichton 1972). The prepatent period of D. insignis, the time interval from first infection of the definitive host by larvae until first-stage larvae are detected in a cutaneous lesion is approximately one year (Crichton and Beverley-Burton 1974,1975, 1977). To examine the host specialization and host exploitation strategies of Dracunculus spp., prevalence and intensity between hosts of D. insignis and hosts of D. lutrae were compared. According to the trade-off hypothesis, specialist parasite species, such as D. lutrae, should be more 'successful' in terms of prevalence and intensity than generalist parasite species such as D. insignis. This was also compared and contrasted to the fitness potential of the parasite, measured by parasite size, of the Dracunculus species. Crichton and Beverley-Burton (1973) reported lengths of 0. insignis from mink and raccoon, and D. lutrae from otter, however no comparisons were made. This study is the first to report guinea worm lengths from the host fisher. Sexual size dimorphism is exhibited in the definitive host species examined in this study; males are larger than females in mink (Eagle and Whitman 1987), fisher (Powell et al. 2003), raccoon (Ritke 1990), and otter (Lariviere and Walton 1998). This study examines possible effects of host sex on guinea worm infection rates, intensities of infection, and parasite size. Due to greater home range sizes, male-biased sexual size dimorphism, and/or immunosuppressive effects of androgens such as testosterone, males should be parasitized by guinea worm more frequently and at higher rates than females in all host species. Correspondingly, guinea worm from male hosts should also be larger because of the male-biased sexual size dimorphism and/or immunosuppressive effects of androgens. A sex bias in parasite infection could have evolutionary implications to the host because it could suggest that higher levels of parasitism might be a relative cost associated with that sex. This study also tests for host age effects on guinea worm parasitism. It has been shown that helminths typically evoke very little immune response despite their size and employ a variety of tactics for immune evasion, potentially down regulating the immune response of their host (Pryor et al. 2005). Dracunculus medinensis (Linnaeus, 1758), for example, have been found to contain substances such as morphine and its active opiate alkaloid metabolite morphine-6-glucoronide, possibly eliciting immunosuppression (Zhu et al. 2002). In addition, antigens resembling human albumin and human immunoglobins have been identified on the surface of D. medinensis (Bloch et al. 1999), possibly "cloaking" the parasite from the hosts immune response. If the host cannot build up the adaptive immune response due to immune evasion from the parasite, there would likely not be an effect of host age on guinea worm infection rates. In addition, intensities of infection would likely not be affected by age because of the one-year life span of the worm. However, it is possible that parasites reach larger sizes in adults compared to juveniles simply due to the fact that a larger host individual may simply present more available resources to the parasite (Esch et al. 1990). Methods In total, 223 mink, 126 fisher, 78 raccoon and 35 otter were collected from registered trap lines in northeastern, southern, and southeastern Ontario during the fur harvests of 2005/06,2006/07 and 2007/08, and frozen until time of dissection (<6 months at- 18°C). Carcasses were thawed in batches at room temperature and sexed by genital/gonadal examination. Hosts were weighed (+ 0.1g in mink; + 0.1kg in fisher, raccoon, and otter), and total lengths and body lengths (mm) were taken. Individuals were aged as either juvenile (< 1.0 year old) or adult (> 1.5 years old) based on the degree of temporal muscle coalescence in mink (Poole et al. 1994), occlusion of the pulp cavity of one lower canine in fisher (Kuehn and Berg 1981), and body weight in raccoon and otter (Grau et al. 1970, Stephenson 1977). Samples collected from trappers included individuals of both sexes and ages. As expected however, females were fewer than males, particularly with mink, fisher, and otter. Male-biased trapping bias is common in the Mustelidae, likely explaining the higher numbers of males (Buskirk and Lindstedt 1989). During necropsies for guinea worm, external surfaces of all superficial muscles, intermuscular areas of the legs and feet, connective tissue beneath the latissimus dorsi and of the inguinal and axillary regions, and internal surfaces of the abdominopelvic cavity of each animal were examined. Guinea worms were counted, sexed (by presence or absence of a spicule), measured for length (+ 0.1mm), weighed (+ 0.001 g), and preserved in 70% ethanol solution. Worms that could not be removed intact, and/or were too damaged to measure length were not included in analyses of guinea worm size. They were, however, included in the analyses of prevalence and intensity. Guinea worm prevalence was defined as the proportion of individuals in a sample that were infected with guinea worm, while guinea worm intensity was defined as the number of guinea worms per infected host (Bush et al. 1997). Statistical Analyses Fisher's exact tests, bootstrap t-tests, and confidence intervals of prevalence and intensity were performed using the Quantitative Parasitology 3.0 software (Reiczigel and Rozsa 2005). Other analyses were performed using SPSS 12.0 for Windows (2003). Significance for all analyses was accepted at a = 0.05. 42 Differences in guinea worm prevalence were determined using Fisher's exact tests (Rozsa et al. 2000). Differences in levels of infection intensity were evaluated using analysis of variance (ANOVA), analysis of covariance (ANCOVA) and bootstrap t-tests (Rozsa et al. 2000). Sterne's exact confidence intervals (95%) of infection intensity were determined by bootstrap t-tests (Rozsa et al. 2000). Tests for host, sex and age differences in Dracunculus spp. length were performed with factorial ANOVA or Mann- Whitney U. Relationships between guinea worm intensity and guinea worm size in each host species were investigated using linear regressions. To assess whether guinea worm infection was influenced by host body size, t-tests of host body length between guinea worm infected and uninfected hosts were conducted. Non-parametric correlation analyses (Spearman rho) were used to assess the relationship between host body length and guinea worm intensity for each sex-age cohort of each host species. Results Host exploitation strategies of D. insignis and D. lutrae Prevalence and intensity of guinea worm (Dracunculus spp.) infection were determined for mink, fisher, raccoon, and otter (Table 2). Prevalence of D. insignis was significantly lower than that of D. lutrae (Fisher's exact p < 0.001; Figure 8a). There was also a significant difference in guinea worm intensity between D. insignis and D. lutrae (t = 3.916, p = 0.006; Figure 8b). D. insignis infections were of lower intensity than D. lutrae infections. In male hosts only, guinea worm prevalence (Fisher's exact p < 0.001) and guinea worm intensity (t = 3.792, p = 0.009) were significantly lower in those infected 43 with D. insignis than those infected with D. lutrae. Similarly, guinea worm prevalence was significantly lower in female hosts infected with D. insignis than D. lutrae (Fisher's exact p = 0.001); however, intensity was not (t =1.131, p=0.402). Guinea worm prevalence (Fisher's exact p < 0.001) and intensity (t = 3.312, p = 0.024) were significantly lower in adult hosts infected with D. insignis than those infected with D. lutrae. Juvenile hosts of D. insignis were also less likely to be infected than juvenile hosts of D. lutrae (Fisher's exact p = 0.003). However, there was no significant difference in guinea worm infection intensity between juvenile hosts of D. insignis and those of D. lutrae (t = 0.013, p = 0.988). 44 Table 2. Summary statistics of Dracunculus spp. prevalence and intensity with associated 95% confidence limits for each sex and age cohort of each host species examined. Guinea worm length (mean (cm) ± standard deviation) in each sex/age cohort of each host species. Guinea Worm Guinea Worm (ND) Host Cohort N Prevalence 95% Conf. Limits Intensity 95% Conf. Limits Length (cm) ± SD Male/Total Male - juvenile 31 41.9 25.5 - 59.8 5.31 3.38 - 10.23 8.22 ± 7.93 14/65 © Male - adult 123 52.0 l 3.08 2.50 - 3.89 12.88 ± 7.91 22/194 Female - juvenile 47 38.3 25.4-53.2 2.39 1.78-3.06 12.19 ±5.86 2/49 Female - adult 22 11 <5 15.2-54.7 2.57 1.57-3.57 18.29 + 7.04 1/18 Fisher * 250 \::W • • Male - juvenile 25 36.0 19.6-56.1 2.56 IM-5M 18.79 ± 11.95 1/22 Male - adult 51 21.6 12.9-35.2 1.64 1.18-2.27 19.39 ±11.38 1/17 Female - juvenile 38 15.8 7.1-31.3 2.83 1.33-5.83 21.66 ±9.85 2/17 Female - adult 12 50.0 24.3 - 75.7 2.17 1.00-3.00 23.06 + 5.52 0/13 RaoixJOa M " \W: £§7 2A1-SM Male - juvenile 13 23.1 6.6 - 52.0 2.33 1.00-3.00 6.25 ± 2.47 0/6 Male - adult 27 25.9 12.4-46.2 3.29 1.71-6.14 3.04+1.31 9/21 Female - juvenile 21 57.1 35.4 - 76.7 5.25 2.83 - 10.00 5.82 ±6.13 17/56 Female - adult 17 47.1 25.3 - 75.8 1.75 1.13-2.63 12.46+13.94 1/12 Otter :35V ms nm Male - juvenile 3 100.0 36.9 - 100.0 4.67 1.00-7.67 7.96 ± 8.20 1/9 Male - adult 24 100.0 86.1 - 100.0 16.25 11.00-24.25 4.01 ±3.41 204/359 Female - juvenile 3 100.0 22.4 - 100.0 2.50 2.00 - 2.50 10.70 ±11.22 1/8 Female - adult 5 100.0 50.0 - 100.0 11.60 3.40-33.80 3.24 + 0.62 20/54 45 B 20 i 15 - 10 0)C 10 (Q ^* *rt ,s f<-» * ** •«. fc. if i„ '3 4 • )•: • a J* » •".. r? ^->--g-^- Si "Si -Hf, D. insignis D. lutrae D. insignis D. lutrae Figure 8. Prevalence (A) and intensity (B) of the generalist Dracunculus insignis and the specialist D. lutrae. Error bars represent 95% confidence interval. 46 With D. insignis, 14.3% of worms recovered were male, while 52.6% of D. lutrae were male. The ratio of female to male guinea worms found was significantly higher in D. insignis hosts than D. lutrae (Q2 = 153.73, p < 0.001). Lengths of Dracunculus spp. were significantly different between generalist and specialist host species (F[i,53o] = 128.62, p < 0.001). D. insignis (11.73 + 8.98, n = 310) were significantly larger than D. lutrae (4.33 + 4.25, n =220). Analyses were then restricted to fecund female worms found in the limbs. D. insignis were significantly longer than D. lutrae (F[i,2o9] = 4.64, p = 0.032). With respect to the generalist D. insignis, guinea worm prevalence was significantly higher in mink compared to fisher (Fisher's exact p < 0.001; Figure 9a), higher in raccoon compared to fisher (Fisher's exact p = 0.044; Figure 9a), but not significantly different between mink and raccoon (Fisher's exact p = 0.233; Figure 9a). Guinea worm intensity was higher in mink compared to fisher (t = 2.054, p = 0.038; Figure 9b) but was not significantly different between mink and raccoon (t = 0.464, p = 0.646; Figure 9b) or fisher and raccoon (t = 1.554, p = 0.152; Figure 9b). In terms of ratios of male and female guinea worms recovered, 12.0% of worms from mink were males, 5.8% from fisher were male, and 28.4% from raccoon were male. 47 Mink Fisher Raccoon Mink Fisher Raccoon Figure 9. Prevalence (A) and intensity (B) of Dracunculus insignis between host species. Error bars represent 95% confidence interval. Lowercase letters denote statistical differences. Mink and raccoon have significantly higher prevalence than fisher. 48 Lengths of Dracunculus spp. were significantly different between the host species (F[3,2io] = 13.89, p < 0.001; Figure 10), however there were significant interactions of age and sex (p < 0.001). An ANCOVA, controlling for sex and age, showed Dracunculus spp. lengths to be significantly different between mink and fisher (F[i,239] = 33.30, p < 0.001), mink and raccoon (F[i,27i] = 46.65, p < 0.001), fisher and raccoon (F[i,u0] = 96.87, p < 0.001), mink and otter (F[i,42o] = 165.67, p < 0.001), fisher and otter {F[i,259] = 158.00, p < 0.001), but not between raccoon and otter (F[i,29i) = 1.06, p = 0.305). Guinea worms from fisher were larger than those from mink, while those from raccoon and otter were smallest and not significantly different from one another. Intensity of guinea worm infection was not related to the size of guinea worms (Pearson correlations, all p>0.05). 32 a 28 | 24 i Mink Fisher Raccoon Otter Figure 10. Guinea worm (Dracunculus spp.) length (cm) of infected hosts of each host species. Error bars represent standard deviation of the mean. Letters above error bar denote statistical differences. Guinea worm from fisher were significantly larger than those from mink, while those from raccoon and otter were smallest and not significantly different from each other. 49 Host sex effect on Dracunculus spp. infection patterns Prevalence and intensity of guinea worm (Dracunculus spp.) infection were determined for each sex and age cohort of mink, fisher, raccoon, and otter (Table 2). There was no effect of sex on guinea worm prevalence or intensity in any of the species (Table 3; Figure lla&b). Table 3. Differences in prevalence and intensity between male and female Dracunculus spp. hosts of each age cohort (see Table 2 for prevalence, intensities, and confidence intervals). Prevalence Intensity Host Juvenile Adult Juvenile Adult P P t P t P Mink 0.748 0.081 1.793 0.220 0.754 0.450 Fisher 0.065 0.070 0.194 0.853 0.849 0.408 Raccoon 0.079 0.198 1.547 0.197 1.202 0.326 Otter 1.000 1.000 0.781 0.466 0.571 0.596 50 180 - 160 - •Juvenile male 0 Adult male • Juvenile female •Adult female Mink Fisher Raccoon Otter 20 -| •Juvenile male B Adult male • Juvenile female •Adult female Mink Fisher Raccoon Otter Figure 11. Prevalence (A) and intensity (B) of guinea worm (Dracunculus spp.) in each sex/age cohort of each host species. Error bars represent 95% confidence interval. 51 Guinea worm lengths were evaluated for differences due to host sex (Table 4). There was a significant difference in guinea worm length between the sexes in both juvenile and adult mink but not in any sex/age cohort of any other host species. In both age cohorts, female mink had larger worms than male mink. Table 4. Differences in guinea worm (Dracunculus spp.) length between male and female hosts of each age cohort (see Table 2 for mean lengths and standard deviations). Asterisks denote statistical differences. Juvenile Adult Host U P df U P df Mink 349.500 0.025* 65 439.500 0.021* 135 Fisher 52.000 0.605 23 30.000 0.834 16 Raccoon 62.000 0.861 47 38.500 0.520 24 Otter 26.000 0.814 15 1072.000 0.204 205 To exclude the potential influence of host body size on parasite infection patterns, host body lengths were tested between uninfected and infected individuals of each sex/age cohort. Host body size was not significantly different between the two groups (t-tests, all p>0.05). In addition, there was no relationship seen between host body length and guinea worm intensity (Spearman rho, all p>0.05). Larger individuals did not have higher levels of infection. Host age effect on Dracunculus spp. infection patterns There was no effect of age on guinea worm prevalence in any of the species except for female fisher (Table 5). Adult female fisher had a significantly higher prevalence than juvenile female fisher (p = 0.025). There was no effect of age on guinea worm intensity in any of the species except for male otter (Table 5). Adult male otter had significantly higher intensities of guinea worm infection than juvenile male otter (p = 0.025). It is important to note here, that the number of juvenile male otter sampled was small (n = 3). Table 5. Differences in prevalence and intensity between juvenile and adult Dracunculus spp. hosts of each sex cohort (see Table 2 for prevalence, intensities, and confidence intervals). Asterisks denote statistical differences. Prevalence Intensity Host Male Female Male Female P P t P t P Mink 0.422 0.789 1.368 0.331 0.274 0.796 Fisher 0.267 0.025* 0.998 0.373 0.529 0.631 Raccoon 1.000 0.745 0.690 0.545 1.933 0.134 Otter 1.000 1.000 2.662 0.025* 1.227 0.377 Guinea worm lengths were compared to evaluate differences due to host age (Table 6). Table 6. Differences in guinea worm (Dracunculus spp.) length between juvenile and adult hosts of each sex cohort (see Table 2 for mean lengths and standard deviations). Asterisks denote statistical differences. Male Female Mink 1486.000 0.001* 160 80.500 0.010* 40 Fisher 95.500 0.974 23 26.000 0.528 16 Raccoon 19.500 0.386 22 91.000 0.528 49 Otter 609.000 0.681 198 52.500 0.809 22 53 There was a significant difference in guinea worm length between juveniles and adults in both sexes of mink, but not in any other host species (Figure 12). In both males and females, adult mink had larger worms than juveniles. • Juvenile male E2 Adult male • Juvenile female • Adult female Raccoon Figure 12. Guinea worm (Dracunculus spp.) length (cm) of infected hosts in each sex/age cohort of each host species. Error bars represent standard deviation of the mean. 54 Discussion This study examined host specialization and host exploitation strategies of two closely related species of the guinea worm, Dracunculus insignis and D. lutrae. According to the trade-off hypothesis, specialist parasite species should be more 'successful' in terms of prevalence and intensity than generalist parasite species due to a trade-off between how many host species a parasite can exploit and how well it does in those hosts (Poulin 1998). As predicted, this study found that the specialist D. lutrae had higher levels of guinea worm infection rates and intensity than the generalist D. insignis, overall, supporting the trade-off hypothesis. In terms of parasite size, however, D. insignis were larger than D. lutrae. Although it was not possible to determine egg output or establish fecundity from female worms, it is reasonable to expect a positive relationship between fecundity and worm size (Poulin 1996a; Tomkins and Hudson 1999). Larvigerous females grow much larger than males or unfertilized females, and thus potentially extract more resources from the host. D. insignis and D. lutrae undergo only 1generational cycle in a year, with peak transmission of new larval infections in the spring and early summer (Anderson 2000). It is possible that transmission may not occur at the same time in all host species. Furthermore, since sampling was accomplished in the late fall and early winter, it is plausible that some already-mated males and unfertilized females have been reabsorbed by the host since mating is thought to take place 60 - 70 days post-infection (Crichton and Beverley-Burton 1975). 55 In nature, different factors may affect rates of parasitism. For example, the rate of transmission of helminth infections is known to be influenced by density-dependent mechanisms acting at various stages of the parasite's life-cycle (Churcher et al. 2005). It is possible that host densities of the intermediate, paratenic, and/or definitive hosts differ between the two species of guinea worm. Since neither naturally infected intermediate nor paratenic hosts of D. insignis and D. lutrae have been found, it cannot be speculated what these relationships might be. It is not known that there is overlap in habitat and diet between hosts of D. lutrae and D. insignis. For example, Lariviere and Walton (1998) note that otters may compete for resources with the mink. Although this study does not control for the impacts of host densities on measures of parasitism, it provides an indication of what is actually occurring in nature. The investigation of host exploitation strategies in the generalist D. insignis found that this species does not exploit all host species equally. Mink and raccoon were generally more likely to be infected than fisher. Crichton (1972) suggested that mink may be somewhat refractory to guinea worm infection compared to raccoon, and Crichton and Beverley-Burton (1974) added that raccoon may serve as a reservoir of guinea worm infection which passes over to mink. The results do not support this hypothesis, as guinea worm prevalence and intensity were not significantly different between mink and raccoon. 56 Although prevalence and intensity were found to be higher in mink and raccoon than fisher, guinea worm grew to a larger size in fisher. Studies have found a negative relationship between nematode size and parasite intensity in natural infections, however, these density-dependent reductions in nematode size are often only apparent at high parasite intensities (Dobson et al. 1990, Tompkins and Hudson 1999). This was not the case in this study, however, because no relationship was found between guinea worm intensity and guinea worm size. This may reflect a particular strategy that D. insignis has developed in order to increase its chances of transmission in the environment. The larger the number of fertile offspring that an individual produces during its lifetime and the number of offspring that survive to breed themselves, the higher the lifetime reproductive success (Thomas et al. 2002). Since prevalence rates are considerably lower in fisher, potentially due to lower host density or lower chances that the fisher will encounter the parasite, the worm potentially responds by increasing its fecundity. This could also be a function of density-dependent mechanisms. Diets of mink and raccoon consist more of aquatic organisms, likely the intermediate hosts of guinea worm, than the diet of fisher (Lotze and Anderson 1979, Powell 1981, Lariviere 1999). This study examined possible host sex effects on guinea worm infection rates, intensities of infection, and parasite size. Theoretically, larger hosts, regardless of sex, present more available resources for parasites (Esch et al. 1990). This study did not find evidence that worms in larger individuals will do better because of the larger amount of resources available to them. No relationship between host body size (body length) and guinea worm infection intensity was found in any of the sex-age cohorts. There was no difference in body size (body length) between uninfected and infected hosts. Despite the prediction of male-biased parasitism due to sexual size dimorphism in the definitive hosts, there was no strong evidence of sex bias in prevalence or intensity of either D. insignis or D. lutrae. The fitness potential of parasites might be expected to differ between parasites infecting male and female hosts because of the existence or expression of fewer defences in one sex of host (Dare and Forbes 2008). Furthermore, the data does not support the prediction that guinea worm from male hosts would be larger because of the male-biased sexual size dimorphism in the host species examined. This study tested for host age effects on guinea worm parasitism and, because of the approximately one year life span of the worm and that helminths evoke very little immune response and employ a variety of tactics for immune evasion, it was not predicted that there would be an effect of host age on guinea worm infection rates or intensities of infection. Adult and juvenile fisher, raccoon and otter were all equally likely to be infected, however, contrary to the prediction that there would be no age effect, adult female fisher were more likely to be infected than juvenile female fisher. It is possible that a particular food source, which may serve as the paratenic host for D. insignis, is more predominant in the adult female diet, than in the juvenile female diet. Regardless, it seems plausible that D. insignis and D. lutrae, like D. medinensis and other helminths, evoke very little immune response and employ a variety of tactics for immune evasion (Pryor et al. 2005). If the parasite did build up resistance in the host, prevalence rates might decrease with age since reinfection would be less likely. Because guinea worm has a one-year life span, infection intensity would not be expected to increase with host age, and this seemed to be the case in this study. If infected, adult and juvenile mink, fisher, and raccoon had the same intensities of infection, however, adult male otter had significantly higher intensity of infection than juvenile male otter. Here, however, more data is needed, as sample sizes were small. There may be an effect of host age on guinea worm size in female hosts. Adult female mink had larger guinea worms compared to juvenile female hosts of the same species. It is possible that guinea worm parasites reach larger sizes in adults compared to juveniles simply because a larger host individual may present more available resources to the parasite (Esch et al. 1990). These findings suggest new areas of investigation. Future work should further investigate guinea worm strategies in different host species in controlled lab settings to account for differences in parasite exposure levels. Furthermore, identification of intermediate and paratenic hosts of Dracunculus spp. in the wild may elucidate the infection patterns of guinea worm in definitive host species. Understanding of the parasite ecology of guinea worm could prove useful in management and conservation of host populations. CHAPTER 4 - The influence of guinea worm (Dracunculus insignis) infection on body condition and fecundity of fisher (Martes pennanti) Introduction In a host-parasite relationship, an ongoing evolutionary arms race occurs where each organism is adapting and counter-adapting to evade the defensive tactics of the other (Webster et al. 2004). Because being parasitic assumes that the parasite will receive benefits at the expense of the host, and due to the direct removal of nutrients or the allocation of energy towards defense against the parasite, the hosts' fitness may be reduced (Anderson and May 1978, Boonstra et al. 1980, Price 1980). This may result in fewer resources being available for the parasite, potentially reducing its fitness (Seppala et al. 2008). Ancillary infection may then result in further deterioration of the hosts' condition making it more susceptible to future infection - creating a 'vicious circle' (Beldomenico and Begon 2010). The negative effect of a parasite on its hosts' energy budget is occasionally detectable in the physiology of the host, including energy levels and condition (Irvine 2006). For example, by experimentally reducing the abundance of Trichostrongylus retortaeformis in mountain hares (Lepus timidus), Newey et al. (2004) increased host body condition. Although such trends are often difficult to identify in wild systems, Hughes et al. (2009) showed a significant decrease in female caribou (Rangifer tarandus) body weight with increasing numbers of abomasal nematodes. Neuhaus (2003) experimentally removed 60 ectoparasites from Columbian ground squirrels (Spermophilus columbianus) resulting in an increase in female body condition during lactation and weaning. Reduced reproductive success due to parasitism has also been identified in natural systems (Hughes et al. 2009, Gooderham and Schulte-Hostedde 2011, Patterson and Schulte-Hostedde 2011). For instance, high burdens of warble larvae (Hypoderma tarandi) were associated with significantly reduced pregnancy rates in caribou (Rangifer tarandus) (Hughes et al. 2009). Hudson et al. (1998,1999) demonstrated that the intestinal nematode (Trichostrongylus tenuis) reduces red grouse (Lagopus lagopus) fecundity. In the experiment in which ectoparasites were removed from Columbian ground squirrels, an increase in litter size was also seen (Neuhaus 2003). T. tenuis has been identified as the main factor contributing to red grouse (L lagopus) population crashes (Hudson et al. 1998,1999). Consequently, parasites have the potential to play an important regulatory role in host population dynamics (Anderson and May 1978, May and Anderson 1978, Irvine 2006). On the other hand, many studies have failed to detect any measurable negative parasite-mediated effect on fitness, especially in wild populations (lason and Boag 1988, Newey et al. 2004). For example, fitness of the keelback snake (Tropidonophis mairii) was found to be unaffected by haemogregarine blood parasites (Brown et al. 2006) even though detrimental effects of these parasites on fitness-related variables in the common lizard (Lacerta vivipara) were found (Oppliger et al. 1996). Many host-parasite associations may appear benign, suggesting that there may only be trivial consequences of parasitism for host fitness in certain natural populations (Brown et al. 2006). It is possible, however, that parasitized individuals may alter behaviour or life history strategies to compensate for costs incurred through parasitism, thus masking fitness effects (Harrison et al. 2010). Alternatively, the coevolution of parasites and hosts may actually weaken or eliminate fitness costs of parasitism with some species (Toft and Karter 1990, Toft 1991, Brown et al. 2006). Dracunculus spp. (Fam. Dracunculidae; O. Spirurida) are nematodes infecting tissues and serous cavities of mammals (Chitwood 1933, Crichton and Beverley-Burton 1974), including humans (Muller 1971). Endemic to North America, D. insignis is a generalist parasite occurring in fisher (Martes pennanti) (Crichton and Beverley-Burton 1974, Elsasser et al. 2009). It has been postulated that raccoon (Procyon lotor) may serve as a reservoir of D. insignis infection and that the range of D. insignis may be limited to that of raccoon (Crichton and Beverley-Burton 1974). Indeed, higher prevalence rates of D. insignis in mink (Neovison vison) have been found in southern Ontario, where raccoon are more abundant, compared to northern Ontario, where raccoon are seldom found (Crichton and Beverley-Burton 1974). No studies to date have reported prevalence rates or geographic distribution of guinea worm in fisher. After becoming infected by drinking water containing infected intermediate hosts or by ingesting a paratenic (reservoir) host that serves as a transport vector within the food chain (Anderson 2000), infective larvae migrate through the intestinal wall, across the abdominal cavity, and into the connective tissues of the trunk, abdomen, and inguinal regions (Crichton and Beverley-Burton 1975). Further development, growth, and mating of the worms occur in the musculature of the dorsum. Gravid females then migrate to the carpal and tarsal areas, reaching lengths upwards of 30 cm when mature. Ulcerative skin lesions are formed and when this affected area comes into contact with water, the numerous larvae are expelled. Expended female worms then die, become calcified, and/or reabsorbed by the host tissue (Crichton and Beverley-Burton 1975). Male worms and unfertilized female worms are thought to eventually be absorbed by the host (Crichton and Beverley-Burton 1975). In this study, the presence and effects of the nematode parasite D. insignis on one of its definitive host species, the fisher, is investigated. This study reports on the prevalence, intensity, and geographic distribution of 0. insignis in fisher and tests the predictions that body condition and fecundity should decline with guinea worm presence and increasing intensity. Methods Prevalence and intensity of guinea worm infections A total of 972 skinned fisher carcasses were collected during the fur harvests of 1999/2000 to 2007/2008 (except for 2000/2001) from registered trap lines in Ontario (see Table 7 for locations), Canada, and frozen until time of dissection (<6 months at - 63 18°C). Carcasses were retained for study if the carcass was in good condition. Gastrointestinal contents were emptied and body mass was taken. Fisher were sexed by genital/gonadal examination and aged as either juvenile (< 1year old) or adult (> 1 year old) based on the occlusion of the pulp cavity of one lower canine, and where necessary on temporal muscle coalescence (Kuehn and Berg 1981, Poole et al. 1994). The resulting sample consisted of 936 individuals: 248 juvenile males, 225 adult males, 236 juvenile females and 227 adult females (36 individuals were missing age data and thus were removed from analyses involving age). During necropsy examinations for guinea worm, external surfaces of all superficial muscles, connective tissue beneath the latissimus dorsi and of the inguinal and axillary regions, intermuscular areas of the legs and feet, and internal surfaces of the abdominal wall of each animal were examined. D. insignis specimens were counted, sexed (by presence or absence of spicule), measured in length (+ 0.1 mm), weighed (+ 0.001g), and preserved in 70% ethanol. Representative specimens of adult male and female D. insignis were deposited in the Canadian Museum of Nature, Ottawa, Ontario (catalogue No. CMNPA 2008-0004 and CMNPA 2008-0005). Body Condition Fisher from the fur harvests of 1999/2000, 2001/02 and 2002/03 from the above- mentioned study were used for the investigation of guinea worm infection on body condition. Body condition was estimated by a percent body fat index (PFAT) determined 64 from five individually discernable fat depots and the total body mass, according to procedures outlined in Robitaille and Jensen (2005). The fat depots, are known to accurately predict body fat levels (Robitaille and Jensen 2005), and thus used to calculate PFAT included; a) the popliteal fat mass found posterior to each femur, starting at the knee joint and bordered by the biceps femoris and the semi-tendinosus, b) the sternal fat mass, a v-shaped deposit of adipose tissue located at the base of the sternum beneath the abdominal muscles, c) the omental fat mass represented by the greater omentum, d) the mesenteric fat mass represented by the mesenteries attached to the intestines after removing the central vascular node, and e) the perirenal fat mass surrounding each of the kidneys and extending along the dorsal wall of the abdominal cavity into the pelvic region (Robitaille and Jensen 2005). Each fat depot was carefully excised and its fresh mass was weighed using a Sartorius™ scale (+0.01 g). Fecundity To examine the relationship between guinea worm infection and reproductive rates of the host, corpora lutea counts of 1234 adult (>1 year old) female fisher were assessed against guinea worm occurrence. Fisher were collected from central and southern Ontario by the Ontario Ministry of Natural Resources (MNR), Algonquin Region, between 1974 and 1985. Female fisher generally breed when 1year old and exhibit delayed implantation of 10 to 11 months (Wright and Coulter 1967, Strickland et al. 1982, Powell et al. 2003). Under favorable ecological conditions, the number of corpora lutea, indicative of the number of eggs ovulated, can be expected to show good 65 agreement with the number of resulting blastocysts or implanted embryos (Buskirk and Powell 1994). Consequently, pregnancy rates for fisher are commonly calculated as the percentage of adult females whose ovaries contain corpora lutea (Shea et al. 1985, Douglas and Strickland 1987, Crowley et al. 1990, Powell et al. 2003). Guinea worm- infected females had been identified by visual inspection (by M. Strickland and C. Douglas of the Ministry of Natural Resources, Algonquin Region) of the external surfaces for subcutaneous gravid female guinea worm, which are almost exclusively located on top of or intertwined among the muscles of the legs (Crichton 1971, Brandt and Eberhard 1990a). An additional study to explore the effects of guinea worm infection intensity on corpora lutea counts were also performed. This latter study employed adult (>1 year old) female fisher utilized in the initial study on prevalence and intensity of infections, and originated from the 2001/02 and 2002/03 fur harvests (n = 78). These specimens had been collected from Parry Sound, Bruce Peninsula, Muskoka District, and Simcoe County. The reproductive tracts were extracted, fixed in Bouin's solution and left to stand for a minimum of 3 months. Procedures for clearing, infiltrating, embedding, sectioning and staining ovaries were adapted from several sources (Kraus 2005). Serial sections of each ovary were examined under a microscope (40X) and corpora lutea were counted and totaled for each animal. 66 Statistical Analyses Differences in infection prevalence due to geographical location and trapping year were tested using Fisher's exact tests (Rozsa et al. 2000). Differences in levels of infection intensity among geographical location and trapping years were evaluated using bootstrap t-tests (Rozsa et al. 2000). T-tests were used for analyses involving body condition and corpora lutea counts. Where violations in assumptions of homogeneity of variance or skewness of the data were noted, the data were logarithmically transformed prior to analysis, and statistics for both raw and transformed data are presented. When necessary, Chi-square analyses, or Mann-Whitney U non-parametric tests were used where data did not conform to normal distributions or Levene's test for homogeneity of variances. Means of corpora lutea counts are shown with associated standard deviation. Correlational analyses were used to investigate relationships between parasite prevalence and intensity, and between infection intensity, host body condition and fecundity. Partial correlation analysis was performed to determine the relationship between intensity of infection and corpora lutea count when controlling for body condition. Fisher's exact tests, bootstrap t-tests of mean intensities, and determination of 95% confidence intervals (CI) were performed using the Quantitative Parasitology 3.0 software (Reiczigel and Rozsa 2005). Other analyses were performed using SPSS 12.0 for Windows (2003). Significance was accepted at an alpha level of 0.05. 67 Results Prevalence and intensity of guinea worm infections D. insignis infections were found in varying degrees throughout Ontario, with the exception of Timmins where no infections were noted (Table 7). Among the regions in which infections occurred, prevalence values differed significantly (x2[s] = 147.82, p <0.001; Table 8). Guinea worm intensities were also significantly different among regions (x2[6] = 28.392, p <0.001). There was a significant positive correlation between guinea worm prevalence and intensity (Spearman's p = 0.893, p = 0.007). Table 7. Prevalence, intensity, and associated confidence intervals (CI) of guinea worm (Dracunculus insignis) infection in fisher (Martes pennant!) from various regions of Ontario, Canada. Prevalence Intensity Region Location n % 95% CI Mean 95% CI Range Bancroft 45°N, 77°W 13 7.7 4.0 - 34.2 1.00 - 1 Bruce Peninsula 45°N, 81°W 64 18.8 10.7-30.4 2.33 1.50-3.33 1-6 French River 46°N, 80°W 29 13.8 4.86 - 30.8 1.75 1.00-2.50 1-4 0 n o 0 c Manitoulin Island 45°N, 82°W 125 1.6 1 2.50 1.00-2.50 1-4 Muskoka 44°N, 79°W 86 74.4 64.0 - 82.7 4.55 2.86-5.55 1 - 18 Parry Sound 45°N, 80°W 250 40.0 34.0 - 46.2 2.78 2.28 - 3.53 1 - 19 Simcoe County 42°N, 80°W 199 45.2 38.4 - 52.3 4.01 3.10-6.71 1-64 Sudbury 46°N, 81°W 78 39.7 29.4-51.3 2.61 2.06 - 3.29 1-8 Timmins 48°N, 81°W 30 0 0-11.2 - - - Prevalence of infection did not differ significantly between trapping years for any of the regions sampled in multiple years (Fisher's exact, ail p > 0.05; see Table 8 for data). Similarly, intensity of guinea worm infection did not change significantly from year to year in those regions which had multiple years of guinea worm infection; Bruce Peninsula, Parry Sound, Simcoe, or Sudbury (Bootstrap t-tests, all p > 0.05; see Table 8 for data). Table 8. Prevalence, intensity, and associated confidence intervals (CI) of guinea worm (Dracunculus insignis) infection in fisher (Martes pennant!) from various regions of Ontario, Canada. Prevalence Intensity n % 95% CI Mean 95% CI Range 1999-2000 French River 29 13.8 4.9 - 30.8 1.75 1.00 -2.50 1-4 Manitoulin Island 33 0 0-10.1 - - - 2001-2002 Bruce Peninsula 35 20.0 9.6 - 37.0 1.71 1.00 -2.29 1-3 Parry Sound 187 39.6 32.6 - 46.8 2.88 2.31 -3.76 1 -19 Simcoe County 109 45.0 35.7 - 54.6 2.80 2.18 -3.73 1 -12 2002-2003 Bruce Peninsula 29 17.2 7.1-36.0 3.20 1.00 -4.40 1-6 Muskoka 85 74.1 63.6 - 82.5 4.59 3.90 -5.59 1 - 18 Parry Sound 22 45.8 26.1 -66.2 3.10 1.50 -7.50 1 -16 Simcoe County 66 53.0 40.9 - 65.2 6.00 3.89 - 12.63 1-64 2003-2004 Manitoulin Island 15 6.7 0.4 - 30.2 4.00 - 4 Sudbury 24 50.0 31.0-69.0 2.75 1.92 -3.83 1-7 1 OS © 2004-2005 Manitoulin Island 77 1.3 v© 1.00 - 1 Sudbury 30 40.0 23.6 - 58.4 3.17 2.25 -4.50 1-8 Timmins 17 0 - - - - 2005-2006 Sudbury 13 38.5 16.6 - 65.8 1.60 1.00 - 1.80 1-2 2006-2007 Bancroft 13 7.7 0.4 - 34.2 1.00 - 1 Parry Sound 41 39.0 25.4 - 54.9 2.13 1.50 -3.44 1-8 2007-2008 Simcoe County 11 36.4 13.5-66.7 2.50 1.00 -3.25 1 -4 Timmins 13 0 0 - 22.5 - - - 69 Body Condition Body condition did not differ significantly between uninfected and infected animals in any of the sex-age cohorts: juvenile males (raw data: t = 0.035, p = 0.973); adult males (raw data: t = 0.838, p = 0.404); juvenile females (raw data: t = 0.276, p = 0.783; log- transformed data: t = 0.227, p = 0.821); or adult females (raw data: t = 0.649, p = 0.517; log-transformed data: t = 0.174, p = 0.862). To eliminate a possible geographic and/or year effect, these analyses were repeated using only the Parry Sound animals trapped during 2001- 2002, as it was the only region/trapping year with a sufficient sample size in each sex/age cohort. Body condition did not differ significantly between uninfected and infected animals from Parry Sound in any of the sex-age cohorts: juvenile males (raw data: t = 1.333, p = 0.190); adult males (raw data: t = 0.788, p = 0.436); juvenile females (raw data: t = 1.289, p = 0.204); or adult females (raw data: t = 0.426, p = 0.672). A significant negative correlation was found between body condition and infection intensity in infected animals (Spearman's p = - 0.164, p = 0.008; Figure 13). When each sex and age cohort was examined in more detail, it was determined that juvenile males were driving the significant correlation between body condition and infection intensity (Spearman's p = - 0.370, p = 0.001). This relationship between body condition and guinea worm intensity was not evident in adult male, adult female, or juvenile female fisher (Spearman's p, all p>0.05). 70 40- • • • II 'T 10 IS 20 Guinea worm intensity Figure 13. Relationship between body condition (PFAT) and guinea worm intensity of infected individuals (n = 257). One data point of a highly infected (64 worms) fisher has been removed for graphical purposes. Fecundity Of the 1234 adult female fisher visually examined by MNR personnel, 5.6% (CI = 4.4- 7.0%) were infected with gravid, female guinea worms (see Table 9). Corpora lutea counts ranged from 0 to 7. There was no significant difference in mean corpora lutea counts between uninfected (mean = 3.24 + 0.99) and infected (mean = 3.25 + 1.08) adult female fisher (Mann-Whitney U = 40172.00, p = 0.994). Table 9. Numbers of adult, female fisher uninfected and infected with D. insignis that were able to achieve pregnancy (corpora lutea >1) and that were not able to achieve pregnancy (corpora lutea = 0) for each age class. Infected Age Pregnant Non-Pregnant Pregnant Non-Pregnant 1 492 20 25 0 2 310 7 18 0 3 124 2 7 0 4 74 2 6 0 5 53 2 2 1 6 34 3 3 0 7 17 0 4 0 8 14 1 2 0 9 6 0 0 1 10 3 1 0 0 Total 1127 38 67 2 The proportion of females that were able to achieve pregnancy (corpora lutea >1) and those that were unable to achieve pregnancy (corpora lutea = 0) was compared between infected and non-infected animals (Table 9). Guinea worm parasitism did not appear to alter the proportion of females that were able to achieve pregnancy (x2[i] = 0.027, p = 0.869). No significant difference was found in corpora lutea counts between 72 uninfected and infected adult females (Mann-Whitney U = 40172, p = 0.994) even when controlling for age (all p > 0.05; Table 10). Table 10. Corpora lutea counts of uninfected and D. /ns/gn/s-infected adult, female fisher per age class. Corpora lutea (uninfected) Corpora lutea (infected) Age U P H + SD Range n (i ± SD Range n 1 2.94 ± 0.885 0-6 512 2.80 ±0.764 1-4 25 5751.00 0.322 2 3.33 ± 0.928 0-6 317 3.39 ±0.778 2-5 18 2776.00 0.835 3 3.48 + 0.855 0-6 126 4.00 ±1.00 3-6 7 321.50 0.189 4 3.71 ±1.017 0-7 76 3.50 ±1.378 2-6 6 172.00 0.280 5 3.76 ±1.088 0-5 55 2.67 ± 2.309 0-4 3 59.00 0.363 6 3.62 ±1.361 0-6 37 4.33 ±0.577 4-5 3 36.50 0.301 7 4.00 ±0.707 3-5 17 4 4 34.00 1.000 8 3.40 ± 1.298 0-6 15 3.50 ±0.707 3-4 2 14.00 0.870 9 3.17 ±1.941 1-6 6 0 1 0.00 0.127 10 3.50 ±2.380 0-5 4 n/a n/a 0 n/a n/a Total 3.24 + 0.991 0-7 1165 3.25 + 1.077 0-6 69 40172.00 0.994 In the study of the effects of guinea worm infection intensity on corpora lutea counts, 35 of 78 adult female fisher (examined for the above-mentioned body condition study) were infected with guinea worm (n= 4.38, CI = 3.24-6.35). Corpora lutea counts ranged from 0 to 5 (n = 2.47 + 0.14). No significant difference was found in corpora lutea counts between uninfected and infected adult females (Mann-Whitney U = 714.00, p = 0.685). However, there was a significant negative correlation found between the intensity of guinea worms and corpora lutea count (Spearman's p = -0.410, p = 0.016; Figure 14). A partial correlation analysis indicated a significant negative relationship between intensity of infection and corpora lutea count when controlling for body condition (semi- partial r = -0.498, p = 0.004). 73 5- i 1 1 1 r 0 5 10 15 20 Guinea worm intensity Figure 14. Relationship between corpora lutea counts and guinea worm intensity in infected female adult fisher. 74 Discussion The results of this study indicate that D. insignis may be considered a common parasite of fisher within central Ontario. Prevalence of D. insignis was positively correlated with infection intensity, a pattern common with other host-parasite interactions (Santi et al. 2006, Poulin 2007). A large variation in prevalence of guinea worm in fisher populations was found in this study ranging from 0 to 74.4%. No infected fisher were found in Timmins, the most northern area sampled. Previous studies have noted higher prevalences of D. insignis in mink and raccoon in southern Ontario compared to northern Ontario (Crichton and Beverley-Burton 1974). This pattern has been attributed to the distribution of raccoon and it has been hypothesized that raccoon may serve as a reservoir of D. insignis infection (Crichton and Beverley-Burton 1974). It is also plausible that the absence or low levels of D. insignis in northern Ontario might be due to the lack of suitable intermediate or paratenic hosts. Major prey of fisher are small to medium- sized mammals (including porcupine, snowshoe hare, squirrel, mice, and shrew), birds, and carrion (Powell 1981). D. insignis-'mfected intermediate or paratenic hosts have yet to be identified in the wild. It has been established experimentally that frogs (Xenopus and Rana spp.) are capable of being paratenic (transport) hosts to D. insignis (Crichton and Beverley-Burton 1977, Eberhard and Brandt 1995). It has also been suggested that mammals may be involved in paratensis of D. insignis (Crichton and Beverley-Burton 1977). Identification of intermediate and paratenic hosts of D. insignis in the wild would be valuable for future studies involving the ecology of this parasite. 75 This study did not find a difference in body condition between D. insignis-infected and uninfected fisher of any age or sex, potentially due to variation in energy intake and expenditure among individual fisher, which could conceal any negative effect of parasites. On the other hand, results showed that body condition of juvenile male fisher infected with guinea worm decreased with increasing infection intensity. No sex or age biases were found in previous studies of guinea worm prevalence and intensities in male fisher (see Chapter 3), therefore, higher levels of infection may be ruled out as a reason why juvenile male fisher condition would be more affected than any other sex/age cohort. Previous studies have found parasites to have different effects on host body condition depending on host age. Hawlena et al. (2006) found body mass of juvenile rodents was negatively affected by fleas at natural infestation levels, whereas adults were not. It is possible that condition of juvenile fisher may be more affected by guinea worm than adult fisher because juveniles require more energy for growth. Also, due to their larger surface to volume ratio, juveniles have higher energy requirements for maintaining unit body mass compared to adults (Kleiber 1961). These points are supported by studies which demonstrate that juvenile fisher of both sexes have higher fat levels than adults (Robitaille and Jensen 2005). Additionally, fisher exhibit sexual differences in development. For example, females may reach adult weight by 5.5 to 6 months of age whereas males may not reach full adult 76 weight until more than a year old (Powell 1981). By 7 months of age, the epiphyses of female fisher long bones have ossified, indicating that they have reached full size, whereas males do not reach this stage until 10 months (Dagg et al. 1975). Furthermore, although gender differences in diet have not been found, home ranges of males tend to be larger than those of females (Powell et al. 2003). Juvenile males, therefore, take longer to mature than females and have increased energetic requirements, thus may be more susceptible to negative effects of guinea worm parasitism than females because of their already depleted resources. Although a difference in corpora lutea counts between uninfected and infected adult female fisher was not found, and guinea worm parasitism did not appear to alter the proportion of females able to achieve pregnancy, a significant negative relationship between infection intensity and corpora lutea count was found. Reduced reproductive success due to parasitism has been identified in natural systems (Hudson et al. 1998, 1999, Hughes et al. 2009), demonstrating that parasites thus play a regulatory role in host population dynamics (Anderson and May 1978, May and Anderson 1978, Irvine 2006). Although corpora lutea counts have been found to increase with age (Shea et al. 1985), previous studies have shown that fisher do not seem to accumulate greater intensities of guinea worm with age (see Chapter 3). One reason may be that the lifespan of D. insignis is restricted to one year (Anderson 2000). In addition, it is plausible that the host does not acquire immunity with exposure to this nematode due to immune evasion or immunosuppressive tactics as seen in the closely related D. medinensis, and helminths in general (Bloch et al. 1999, Zhu et al. 2002, Pryor et al. 2005). This study highlights evidence that seemingly benign parasitic infections may have fitness consequences. D. insignis appears to have detrimental effects on host body condition and fecundity of fisher, and this is particularly important due to the current conservation status of the fisher. The fisher has marked fluctuations in its harvest numbers, indicative of a 9-10 year cycle in abundance (Bulmer 1974). Between 1800 and 1940, fisher populations declined or were extirpated in much of Canada and most of the United States due to overtrapping and habitat destruction (Powell 1981). Conservation efforts (closed trapping seasons, habitat recovery programmes and re-introductions) have allowed fisher to return to much of their former range (Powell et al. 2003). A more thorough understanding of potential influence of the guinea worm parasite on fisher and fisher populations should prove useful in decision-making regarding the conservation and management of fisher populations. The relationships between guinea worm and host physiology observed here are correlative in nature; therefore experimental approaches would help establish any causality between guinea worm, condition and fecundity. In addition, future studies investigating alternative measures of conditional status, including measures of immune activity (e.g. differential blood cell counts) in fisher infected with D. insignis may shed further light on the dynamics of this parasite-host relationship. 78 CHAPTER 5 - The influence of guinea worm (Dracunculus sp.) on host body condition Introduction Parasites survive on the tissues and metabolites of their hosts (Roberts and Janovy 2009). The persistent draw of a parasite on its hosts' energy budget has the potential to deplete the hosts energy and nutritional reserves of fat and protein (Neuhaus 2003, Newey et al. 2004, Hughes et al. 2009). The amount of energy available to a host, its body condition, has been shown to be negatively affected by parasitism (Whiteman and Parker 2004, Hughes et al. 2009), and this can have significant fitness consequences. For example, parasitized individuals may experience decreased growth rates, reproductive success, and/or survival (Boonstra et al. 1980, Forbes and Baker 1991, Albon et al. 2002, Hughes et al. 2009). Theoretical models suggest that host population dynamics may be influenced by parasites (Anderson and May 1978, May and Anderson 1978, Seilacher et al. 2007). This has been observed in natural systems (Hudson et al. 1998), however, because of the difficulty of conducting experiments at suitably large spatial scales, studies are few (May 1999). Furthermore, measurable negative parasite-mediated effects on fitness, especially in wild populations, are not always detected (lason and Boag 1988, Newey et al. 2004). In the case of the keelback snake (Tropidonophis mairii), fitness was found to be unaffected by haemogregarine blood parasites (Brown et al. 2006) even though 79 detrimental effects of these parasites on fitness-related variables were found in the common lizard (Lacerta vivipara) (Oppliger et al. 1996). Although they may seem innocuous, there may be sub-lethal consequences of parasitism for host fitness in natural populations (Brown et al. 2006). For example, Alzaga et al. (2008) found a negative relationship between parasitism and capacity to escape from predators in Iberian hares (Lepus granatensis). Furthermore, parasitized individuals may alter behaviour or life history strategies to compensate for costs incurred through parasitism, thus masking effects (Schwanz 2008, Harrison et al. 2010). The coevolution of parasites and hosts may actually weaken or eliminate fitness costs of parasitism with some species (Toft and Karter 1990, Toft 1991, Brown et al. 2006). There is the possibility that not only is the parasite imposing fitness costs on the host, but with infection, the host provides fewer resources for parasites, potentially reducing parasite fitness (Seppala et al. 2008). A 'vicious circle' is created when parasitism results in further deterioration of the host and ultimately influences other factors such as reproductive success (Beldomenico and Begon 2010). Parasites are often assumed to be more successful in hosts in poor condition. Infection occurrence and intensity may be more probable and more severe in individuals with an underlying poor condition (Beldomenico and Begon 2010), however, hosts in poor condition may provide less resources for the host (Seppala et al. 2008). 80 The effects of the guinea worm parasite (Dracunculus sp.) on its host have not been examined. Dracunculus insignis (Leidy, 1858) and D. lutrae (Crichton and Beverley- Burton, 1973) are nematode parasites of wildlife in North America. D. insignis is a generalist species infecting mink (Neovison vison), fisher (Martes pennant), and raccoon (Procyon lotor), while D. lutrae is a specialist infecting otter (Lontra canadensis) (Elsasser et al. 2009). This study investigates the effects of the nematode parasite on the common definitive host species mink, fisher, raccoon, and otter. Guinea worm infection occurs when the definitive host drinks water containing infected intermediate hosts or ingests a paratenic (reservoir) host that serves as a transport vector (Anderson 2000). Infective larvae migrate to the musculature of the back, further develop, and mate (Crichton and Beverley-Burton 1975). Gravid females migrate to the carpal and tarsal areas, filling with ova and reaching lengths of up to 30 cm. Ulcerative skin lesions are formed and the numerous larvae are expelled when the animal steps into water. In human hosts, emergence of the female guinea worm through the skin causes local itching and burning pain, and secondary bacterial infections frequently result in severe ulceration (Bloch et al. 1993). Expended female worms are then believed to die, become calcified, and/or are reabsorbed by the host tissue, and that male worms and unfertilized female worms are eventually absorbed by the host (Crichton and Beverley-Burton 1975). 81 This study tests the predictions that individuals infected with guinea worm will be in poorer body condition than uninfected individuals, and that host body condition will decrease with increasing guinea worm abundance and intensity. Previous studies have found a negative relationship between nematode size and parasite intensity for natural infections, however, these density-dependent reductions in nematode size are often only apparent at high parasite intensities (Dobson et al. 1990, Tompkins and Hudson 1999). Furthermore, in the same host-parasite systems where density-dependent reductions in nematode size were identified, a positive relationship between nematode size and parasite intensity below the density-dependence threshold was found (Tompkins and Hudson 1999). This study, therefore, considers the relationship between parasite intensity and size. Relationships between body condition and infection level might indicate that guinea worm infection has an effect on body condition of the host or that hosts in poorer condition are more likely to be infected with guinea worm. Previous studies have found no evidence of sex or age-biased parasitism in these hosts despite there being male biased sexual size dimorphism of the host species (see Chapter 3). Notwithstanding, this study considers sex and age cohorts separately due to marked differences in energetic budgets between sexes and age classes in the host species (Lotze and Anderson 1979, Powell 1981, Lariviere and Walton 1998, Lariviere 1999). 82 Methods Mink, fisher, raccoon and otter were collected from registered trap lines in northeastern, southern, and southeastern Ontario, Canada during the fur harvests of 2005 - 2007. Carcasses were kept frozen (-18°C) until time of dissection (<6 months). Individuals were retained for this study if carcass condition appeared to be in good condition (i.e. no signs of scavenging or decay). Body mass (+ 0.1 g in mink; + 0.1kg in fisher, raccoon, and otter), total body length including tail (+ 0.1cm), and tail length (+ 0.1cm) were measured. Body length was calculated by subtracting tail length from total body length. Animals were aged as either juvenile or adult based on the degree of temporal muscle coalescence in mink (Poole et al. 1994), occlusion of the pulp cavity of one lower canine in fisher (Kuehn and Berg 1981), and body weight in raccoon and otter (Grau et al. 1970, Stephenson 1977). Individuals were sexed by genital/gonadal examination. In this study sample, females are less represented than males, particularly with mink, fisher, and otter. Sex-biased trapping bias does occur in the Mustelidae, likely explaining this detail (Buskirk and Lindstedt 1989). External surfaces of all superficial muscles, intermuscular areas of the legs and feet, connective tissue beneath the latissimus dorsi and of the inguinal and axillary regions, and internal surfaces of the abdominopelvic cavity of each animal were examined for guinea worm. Guinea worms were counted, sexed (by presence or absence of spicule), measured in length (+ 0.01cm), and preserved in 70% ethanol. The overall guinea worm biomass was calculated by adding the total of all guinea worms lengths for each infected 83 host. Guinea worm prevalence is defined as the proportion of individuals infected, abundance is the mean number of guinea worms per host (infected and uninfected), and intensity is the mean number of guinea worms per infected host (Bush et al. 1997). Host body condition was estimated using standardized residuals of a linear regression of host body mass on host body size (length) for each sex/age cohort individually (Schulte- Hostedde et al. 2001). Relationships between host body condition and guinea worm infection status (infected or noninfected), abundance, and intensity were investigated in each host sex and age cohort of each host species. Logistic regressions were used to investigate the relationship between guinea worm infection status (infected or noninfected) and host body condition in each of the sex/age cohorts. Relationships between host body condition, and guinea worm abundance and intensity were investigated using linear regressions. To test for the effects of host body size on the probability of guinea worm infection, guinea worm abundance, and intensity, multiple regressions between each of these guinea worm parameters and both host body size and host body mass were conducted. The partial correlation coefficient for host body size was used to determine whether the probability of guinea worm infection, guinea worm abundance, or intensity were related to host body size. Prevalence and intensity and associated confidence intervals were performed using the Quantitative Parasitology 3.0 software (Reiczigel and Rozsa 2005). Other analyses were performed using SPSS 12.0 for Windows (2003). Significance for all analyses was accepted at a = 0.05. 84 Results Dracunculus sp. infections were common in this study area; D. insignis infection ranged from 25 - 46% in mink, fisher, and raccoon, while D. lutrae was found in 100% of otter examined (Table 11). Mean intensities of D. insignis infection ranged from 2.22 worms per infected fisher to 3.57 per infected raccoon, while mean D. lutrae was 13.43 worms per infected otter (Table 11). 85 Table 11. Summary statistics of Dracunculus sp. prevalence and intensity with associated 95% confidence limits for each host species examined. 0 Host Cohort N Prevalence 95% Conf. Limits Intensity 95% Conf. Limits W rm(ND) Male/Total M**k - may1"1'' - m Male-juvenile 31 41.9 25.5 - 59.8 5.31 3.38- 10.23 14/65 Male-adult 123 52.0 43.1-61.0 3.08 2.50 - 3.89 22/194 Female - juvenile 47 38.3 25.4 - 53.2 2.39 1.78 - 3.06 2/49 Female-adult 22 31.8 15.2-54.7 2.57 1.57-3.57 1/18 Male - juvenile 25 36.0 19.6 - 56.1 2.56 1.44 - 5.11 1/22 Male - adult 51 21.6 12.9 - 35.2 1.64 1.18 - 2.27 1/17 Female-juvenile 38 15.8 7.1 -31.3 2.83 1.33 - 5.83 2/17 Female-adult 12 50.0 24.3 - 75.7 2.17 1.00 - 3.00 0/13 Male-juvenile 13 23.1 6.6 - 52.0 2.33 1.00 - 3.00 0/6 Male-adult 27 25.9 12.4-46.2 3.29 1.71 -6.14 9/21 Female -juvenile 21 57.1 35.4 - 76.7 5.25 2.83-10.00 17/56 Female-adult 17 47.1 25.3 - 75.8 1.75 1.13 - 2.63 1/12 Male - juvenile 3 100.0 36.9- 100.0 4.67 1.00 - 7.67 1/9 Male-adult 24 100.0 86.1- 100.0 16.25 11.00 - 24.25 204/359 Female-juvenile 3 100.0 22.4- 100.0 2.50 2.00 - 2.50 1/8 Female-adult 5 100.0 50.0- 100.0 11.60 3.40 - 33.80 20/54 86 Logistic regressions of guinea worm presence on host body condition yielded no significant effects in mink, fisher, or raccoon (Table 12). Analyses could not be completed for otter due to the lack of uninfected individuals. Effects were marginally 2 nonsignificant in adult female fisher (F[i,9] = 4.47, r = 0.332, P = 0.064) and in juvenile 2 male raccoon (F[i,i0] = 4.65, r = 0.317, P = 0.057). There was no evidence to suggest that host body size had an effect on the probability of guinea worm infection. Larger host individuals were as likely to be infected as smaller individuals of their respective sex/age cohort (partial correlations, all P>0.05). Table 12. Results of logistic regression of host body condition on guinea worm infection status (infected or noninfected). Infection with guinea worm did not have an effect on body condition in mink, fisher, or raccoon. Analyses could not be completed for otter due to the lack of uninfected individuals. Host Condition x Guinea Worm Infection Status Host Species Sex Age 2 Degrees of F r P Freedom (total) Mink Male Juvenile 0.43 0.014 0.515 30 Adult 0.62 0.005 0.434 117 Female Juvenile 1.09 0.023 0.304 48 Adult 0.02 0.001 0.893 22 Fisher Male Juvenile 0.02 0.001 0.891 17 Adult 0.04 0.001 0.839 38 Female Juvenile 0.13 0.005 0.727 24 Adult 4.47 0.332 0.064 10 Raccoon Male Juvenile 4.65 0.317 0.057 11 Adult 0.76 0.039 0.393 20 Female Juvenile 1.45 0.079 0.245 18 Adult 0.26 0.023 0.623 12 Otter Male Juvenile Adult Female Juvenile - - - - Adult 87 There were no significant effects of guinea worm abundance on host body condition in any of the sex/age cohorts (Table 13), but this was marginally nonsignificant in juvenile 2 2 male mink (F[i,30] = 3.34, r = 0.103, P = 0.078) and juvenile male raccoon (F[i,u] = 4.15, r = 0.293, P = 0.069). Table 13. Linear regressions of host body condition on guinea worm abundance. Host Condition x Guinea Worm Abundance Host Species Sex Age 2 Degrees of F r P Freedom (total) Mink Male Juvenile 3.34 0.103 0.078 30 Adult 0.05 0.000 0.823 117 Female Juvenile 0.00 0.000 0.965 48 Adult 0.14 0.006 0.715 22 Fisher Male Juvenile 0.27 0.017 0.610 17 Adult 0.01 0.000 0.944 38 Female Juvenile 0.17 0.007 0.684 24 Adult 1.84 0.170 0.208 10 Raccoon Male Juvenile 4.15 0.293 0.069 11 Adult 0.01 0.000 0.937 20 Female Juvenile 1.86 0.099 0.190 18 Adult 1.28 0.104 0.282 12 Otter Male Juvenile - - - Adult 1.37 0.095 0.263 14 Female Juvenile - - - - Adult 1.56 0.609 0.430 2 88 There was a significant trend, however, toward smaller adult male fisher being more heavily parasitized than larger adult males (Body length: partial r= -0.486, P = 0.002; Figure 15). Repeating this analysis with one outlying data point removed (see Figure 15), shows that this one case was driving the significant relationship (Body length: partial r = -0.182, P = 0.280). Semi-partial r = -0.486, P = 0.002 -1 -2 — -3 -5 Guinea Worm Abundance Figure IS. Semi-partial correlation between residual host body length (corrected for host body mass) and guinea worm abundance for adult male fisher (n = 39). Smaller adult male fisher tended to have more guinea worms than larger males. Repeating this analyses with the one outlying data point removed showed that this one case was driving the relationship, as it was no longer significant. 89 Smaller juvenile female raccoon were more heavily parasitized than larger juvenile females (Body length: partial r = -0.479, P = 0.045; Figure 16). be a>c > "O O Semi-partial r = -0.479, P = 0.045 CO 1/1 o X "ro 3 QJin 0£ -2 T~ 10 15 20 25 Guinea Worm Abundance Figure 16. Semi-partial correlation between residual body length (corrected for body mass) and guinea worm abundance for juvenile female raccoon (n = 19) indicating that small females tended to have more guinea worms than large females. 90 There were no significant effects of guinea worm intensity on host body condition in any of the sex/age cohorts (Table 14). There were no effects of body size on guinea worm intensity in any of the sex/age cohorts (all P > 0.05). Table 14. Linear regressions of host body condition on guinea worm intensity. Host Condition x Guinea Worm Intensity Host Species Sex Age 2 Degrees of F r P Freedom (total) Mink Male Juvenile 3.80 0.257 0.077 12 Adult 0.05 0.001 0.825 58 Female Juvenile 1.68 0.085 0.212 19 Adult 1.18 0.191 0.327 6 Fisher Male Juvenile 0.38 0.158 0.602 3 Adult 0.10 0.016 0.765 7 Female Juvenile 0.55 0.156 0.511 4 Adult 0.10 0.033 0.769 4 Raccoon Male Juvenile - - - 1 Adult 0.56 0.122 0.497 5 Female Juvenile 0.26 0.031 0.627 9 Adult 1.03 0.205 0.368 5 Otter Male Juvenile - - - Adult 1.37 0.095 0.263 14 Female Juvenile - - - - Adult 1.56 0.609 0.430 2 91 Discussion In terms of physiological effects of guinea worm infection, it was presumed that the cost to the host would involve a loss of energy stores. Regardless of the high levels of guinea worm parasitism in this study, however, presence and abundance of infection did not have obvious effects on host body condition. The lack of an apparent energetic cost of guinea worm parasitism in this study might be explained by abundant food resources available to hosts, facilitating their hosts' ability to compensate for the energy exploited by guinea worm. Parasitized hosts have shown higher metabolic rates and thus have affected host energy needs (Booth et al. 1993, Khokhlova et al. 2002), however, it might be that the system is at the threshold of these effects being discernible. Conversely, a regulating mechanism may limit natural parasite densities to a point at which a negative effect on hosts is below the accuracy of measurement (Hawlena et al. 2006). Tompkins et al. (2011) point out that the effects of multiple and concomitant parasites are only beginning to be understood. Schulte-Hostedde and Elsasser (2011) found that spleen mass, an indicator of immunological activity, of male American mink was positively related to both body condition and parasite richness, but not associated with prevalence or intensity of any particular parasite. Body condition of male mink, however, was related to the intensities of individual parasites such as giant kidney worm (Dioctophyme renale) and sinus worm (Skrjabingylus nasicola), but not to parasite richness (Schulte-Hostedde and Elsasser 2011). 92 Under the assumption that larger nematodes would be more energetically demanding on the host, it was predicted that individuals harbouring higher overall guinea worm biomass would be in poorer body condition. This trend was not seen here. On the contrary, the reverse was found in one cohort; adult female mink in better condition had higher overall guinea worm lengths than adult female mink in poorer condition. These observations may be explained by the well-fed hypothesis which predicts that larger animals will be more parasitized because they represent a better nutritional source (Christe et al. 2003, Hawlena et al. 2005). It is also possible that individuals in better condition are better foragers, and because of this, encounter parasites more often. Furthermore, a growing body of evidence suggests that reduced host condition has no effect on host exploitation by parasites; rather deteriorating resources for the host can limit the amount of resources available for the parasite, thus directly affecting the parasite's fitness. In trematode-snail (Diplostomum spathaceum-Lymnaea stagnalis) systems, reduced host condition resulted in parasite producing fewer and poorer quality transmission stages (Seppala et al. 2008). It is also possible that individuals in better condition may simply be more capable of withstanding high parasite loads. Parasite fitness can increase with the amount of nutritive resources extracted from the host body and can decrease with host immune response (Bize et al. 2008). In general, helminths evoke very little immune response despite their size and employ a variety of tactics for immune evasion, potentially down regulating the immune response of their host (Pryor et al. 2005). Human guinea worms {Dracunculus medinensis) have been found to contain potentially immunosuppressive substances such as morphine (Zhu et al. 2002). In addition, antigens resembling human albumin and human immunoglobins that may serve to conceal the parasite from the hosts immune response have been identified on the surface of D. medinensis (Bloch et al. 1999). It is possible that guinea worm have evolved immunosuppressive strategies that allow them to evoke little immune response from their hosts. In this study, there was a trend for smaller male adult fisher to be more heavily parasitized than larger adult male fisher and for smaller juvenile female raccoon to be more heavily parasitized than larger juvenile female raccoon. There was also evidence that smaller adult female mink had greater overall guinea worms lengths than larger adult female mink. It is plausible that guinea worm may influence skeletal growth rates. This has been observed in other host-parasite systems (Merino and Potti 1995, Tschirren et al. 2003, Perez-Orella and Schulte-Hostedde 2005). Future studies on the effects of guinea worm infection should also consider the effects of other parasites. These results do not rule out that host body size might influence guinea worm infection intensities in some host species. Experimental studies are needed to elucidate any causal links between parasite and host condition. 94 CHAPTER 6 - General Discussion The basic ecology of many parasites and the effects that these parasites may have on their hosts are often not well known, even though they may occur commonly in nature. One such parasite is the North American guinea worm (Dracunculus sp.). In this research, I have established a genetic method to reliably differentiate between North American guinea worm species D. insignis and D. lutrae, explored the ecology of the these species, and investigated the influence of the guinea worm on its hosts. Using DNA barcoding, this study successfully and reliably differentiated between D. insignis, which infects the subcutaneous tissues of many mesocarnivores including raccoon, mink, and fisher, and D. lutrae, which has been recovered exclusively from otter. Using the barcoding technique, this study confirmed that D. insignis infects otter. Barcoding analyses demonstrated that D. insignis had little sequence divergence regardless of the source host, which implies that the species is a 'true' generalist. In contrast, D. lutrae included several separate mitochondrial lineages despite its host specialization of otter. Until now, species identification of Dracunculus sp. has been achieved by male morphology (Anderson 2000). Male guinea worm tend to remain in the musculature of the dorsal region of the host. Consequently, when male worms cannot be dissected from the host, such as in the case of live hosts, only female specimens may be available, 95 and thus accurate identification is not always possible. As shown in this study, barcoding may serve as an effective tool for differentiating other members of the genus Dracunculus. Cases of human dracunculiasis have been described from countries that have never been known to have the endemic disease (Hashikura 1927, Kobayashi et al. 1986, Wang et al. 1995). These infections may not be D. medinensis but some other species of Dracunculus acquired when the person ingested a paratenic host of the parasite. Barcoding may provide a straightforward and reliable method of resolving such issues. Future research may include validating the COI barcoding method on other dracunculoids, particularly the human guinea worm (D. medinensis). Barcoding may also serve as an effective tool for differentiating other life-cycle stages of Dracunculus spp. Intermediate and paratenic hosts of D. insignis and D. lutrae have yet to be found in the wild. Laboratory studies have established that frogs (Xenopus and Rana spp.) are capable of being paratenic hosts to D. insignis (Crichton and Beverley- Burton 1977, Eberhard and Brandt 1995). It has also been suggested that mammals may be involved in paratensis of D. insignis (Crichton and Beverley-Burton 1977). Future work may include employing the barcoding method on a suite of parasites infecting freshwater invertebrates and small mammals which may identify intermediate and paratenic hosts of North American guinea worm. Identification of these hosts in the wild would be valuable for future studies involving the ecology of this parasite. 96 In this research, I have also compared the prevalence, intensity, and guinea worm size between the generalist D. insignis and specialist D. lutrae. According to the trade-off hypothesis, generalist species should be less 'successful' in terms of parasitism than specialist species such that a negative relationship should exist between how many host species the parasite can exploit and its reproductive potential (Poulin 1998). As predicted, I found that hosts of the generalist D. insignis had lower rates of guinea worm infection and lower levels of guinea worm intensities than the specialist D. lutrae, overall, supporting the trade-off hypothesis. In terms of parasite size, D. insignis were larger than D. lutrae. This could be a particular strategy that D. insignis has evolved in order to enhance the probability of transmission in the environment because larger size in female nematodes generally confers greater fecundity and therefore a fitness advantage (Poulin 1996a, Tompkins and Hudson 1999). This is a potential area for future research. Future studies may include experimental manipulation of infection rates and intensities of guinea worm in laboratory animals, possibly farmed mink, to examine the relationship between these variables and parasite size. Experimental approaches would help to define relationships between host use and parasite size, and could validate if, in fact, lower prevalence rates are offset by larger guinea worm sizes. An investigation of host exploitation strategies of D. insignis found that this species does not exploit all host species to the same degree. Mink and raccoon were more likely to be infected than fisher. Previous studies have shown guinea worm parasitism to be common in otter, raccoon, and mink, however this study establishes that 0. insignis may be considered a common parasite of fisher, and is the first to report infection patterns of D. insignis in fisher. Crichton (1972) suggested that mink may be somewhat refractory to guinea worm infection compared to raccoon, and Crichton and Beverley-Burton (1974) added that raccoon may serve as a reservoir of guinea worm infection which passes over to mink. The results of this study do not support this hypothesis, as guinea worm prevalence and intensity were not significantly different between mink and raccoon. However, although prevalence and intensity were found to be higher in mink and raccoon than fisher, D. insignis grew to a larger size in fisher. This may be a strategy that D. insignis has evolved in order to enhance the probability of transmission in the environment. Future studies may include experimental manipulation of infection rates and intensities of guinea worm in the multiple host animals to further explain the factors influencing host exploitation strategies such as the hosts' susceptibility and immune responses. Despite the prediction of male-biased parasitism due to sexual size dimorphism in the definitive hosts, this study found no strong evidence of sex bias in prevalence or intensity, or guinea worm size of either D. insignis or D. lutrae. Furthermore, it was predicted that, if the host does not build up an adaptive immune response due to immune evasion from the parasite, there would likely not be an effect of host age on guinea worm infection rates. Similarly, intensities of infection would likely not be affected by age because of the one-year life span of the worm. This study found no 98 strong evidence of an age effect on guinea worm prevalence, intensity or guinea worm size. Because of the lack of an effect of host age on guinea worm parasitism, I can speculate that D. insignis and D. lutrae, like D. medinensis and other helminths, evoke very little immune response and employ a variety of tactics for immune evasion. It is possible that the North American guinea worm conceals itself from the hosts immune response similar to D. medinensis; which has been found to contain immunosuppressive substances such as morphine and its active opiate alkaloid metabolite morphine-6- glucoronide, and antigens resembling human albumin and human immunoglobins have been identified on its surface (Bloch et al. 1999, Zhu et al. 2002). Immunological composition studies of North American guinea worm may shed light on its immune defense tactics. In terms of physiological effects, it was presumed that the cost of guinea worm infection to the host would involve a decrease in body condition and fecundity. I found that condition of juvenile male fisher infected with guinea worm decreased with increasing infection intensity. Despite the prediction that individuals with larger guinea worms would be in poorer body condition simply because larger nematodes would require more resources and thus be more energetically demanding on the host, this was not found. Moreover, this study did not find a difference in corpora lutea counts between uninfected and infected adult female fisher, and guinea worm parasitism did not appear to alter the proportion of females able to achieve pregnancy, however, there was a significant negative relationship between infection intensity and corpora lutea count. Results of this study indicate that seemingly benign infection may have fitness consequences. Guinea worm appeared to influence host body condition and fecundity of fisher, potentially indicative of long-term implications for host fitness or host population health. Considering that fisher are considered to be rare or even extirpated in parts of their historic range in Canada and the United States (Gibilisco 1994, Thompson 2000), guinea worm infection may have implications for management and conservation of host populations. The relationships between guinea worm and host physiology observed in this study are correlative in nature; therefore experimental approaches would help establish any causality between guinea worm, condition and fecundity. Experimentation in controlled lab settings will also help to account for differences in parasite exposure levels. Future studies on the effects of guinea worm infection should also consider the effects of multiple and concomitant parasites, a factor that is recently being recognized as important (Tompkins et al. 2011). A more thorough understanding of the evolution and ecology of the guinea worm parasite in its hosts will enhance knowledge important for the conservation and management of host populations, and for human health. 100 Literature Cited Albon, S. D., A. Stien, R. J. Irvine, R. Langvatn, E. Ropstad, and O. Halvorsen. 2002. The role of parasites in the dynamics of a reindeer population. Proceedings of the Royal Society of London, B 269:1625-1632. Alzaga, V., J. Vicente, D. Villanua, P. Acevedo, F. Casas, and C. Gortazar. 2008. Body condition and parasite intensity correlates with escape capacity in Iberian hares (Lepus granatensis). Behavioral Ecology and Sociobiology 62:769 - 775. Anderson, R. C. 2000. Nematode parasites of vertebrates: their development and transmission, 2nd edition. CABI Publishing, Oxon, UK. Anderson, R. M., and R. M. May. 1978. Regulation and stability of host-parasite population interactions. I. Regulatory processes. Journal of Animal Ecology 47:219-247. Anderson, T. J. C., and J. Jaenike. 1997. Host specificity, evolutionary relationships and macrogeographic differentiation among Ascaris populations from humans and pigs. Parasitology 115:325-342. Atkinson, S. N., and M. A. Ramsay. 1995. The effects of prolonged fasting on the body composition and reproductive success of female polar bears (Ursus maritimus). Functional Ecology 9:559-567. Beldomenico, P. M., and M. Begon. 2010. Disease spread, susceptibility and infection intensity: vicious circles? Trends in Ecology and Evolution. 25(1): 21-27. Besansky, N. J., D. W. Severson, and M. T. Ferdig. 2003. DNA barcoding of parasites and invertebrate vectors: what you don't know can hurt you. Trends in Parasitology 19:545-546. Beverley-Burton, M., and V. F. J. Crichton. 1973. Letter: Identification of guinea-worm species. Transactions of the Royal Society of Tropical Medicine and Hygiene 6:152. Beverley-Burton, M., and V. F. J. Crichton. 1976. Attempted experimental cross infections with mammalian guinea-worms, Dracunculus sp. (Nematoda: Dracunculoidea). American Journal of Tropical Medicine and Hygiene 25:704 - 708. 101 Beyer, T., A., R. Pinckney, D., and J. Cooley, A. 1999. Massive Dracunculus insignis infection in a dog. Journal of the American Veterinary Medical Association 214:366-368. Bimi, L., A. R. Freeman, M. L. Eberhard, E. E. Ruiz-Tiben, and N. J. Pieniazek. 2005. Differentiating Dracunculus medinensis from D. insignis by the sequence analysis of the 18S rRNA gene. Annals of Tropical Medicine & Parasitology 99:511-517. Bize, P., C. Jeanneret, A. Klopfenstein, and A. Roulin. 2008. What makes a host profitable? Parasites balance host nutritive resources against immunity. American Naturalist 171:107-118. Bloch, P., M. Lund, B. J. Vennervald, and P. E. Simonsen. 1999. Human serum albumin and immunoglobin on Dracunculus medinensis. Acta Tropica 73:135 -141. Bloch, P., P. E. Simonsen, and B. J. Vennervald. 1993. The antibody response to Dracunculus medinensis in an endemic human population of northern Ghana. Journal of Helminthology 67:37 - 48. Boonstra, R., C. J. Krebs, and T. D. Beacham. 1980. Impact of botfly parasitism on Microtus townsendii populations. Canadian Journal of Zoology 58:1683-1692. Booth, D. T., D. H. Clayton, and B. A. Block. 1993. Experimental demonstration of the energetic cost of parasitism in free-ranging hosts. Proceedings: Biological Sciences 253:125-129. Bowman, J., A. G. Kidd, R. M. Gorman, and A. I. Schulte-Hostedde. 2007. Assessing the potential for impacts by feral mink on wild mink in Canada. Biological Conservation 139:12-18. Brandt, F. H., and M. L. Eberhard. 1990a. Distribution, behavior, and course of patency of Dracunculus insignis in experimentally infected ferrets. Journal of Parasitology 76:515-518. Brandt, F. H., and M. L. Eberhard. 1990b. Dracunculus insignis in ferrets: Comparison of inoculation routes. Journal of Parasitology 76:93-95. Brooks, D. R. 1979. Testing the context and extent of host-parasite coevolution. Systematic Zoology 28:299-307. Brooks, D. R. 1988. Macroevolutionary comparisons of host and parasite phytogenies. Annual Review of Ecology and Systematics 19:235-259. 102 Brown, G. P., C. M. Shilton, and R. Shine. 2006. Do parasites matter? Assessing the fitness consequences of haemogregarine infection in snakes. Canadian Journal of Zoology 84:668-676. Brown, J. H. 1984. On the relationship between abundance and distribution of species. American Naturalist 124:255-279. Brown, J. H. 1995. Macroecology. University of Chicago Press, Chicago. Bulmer, M. G. 1974. A statistical analysis of the 10-year cycle in Canada. Journal of Animal Ecology 43:701 - 718. Bush, A. 0., K. D. Lafferty, J. M. Lotz, and A. W. Shostak. 1997. Parasitology meets ecology on its own terms: Margolis et al. revisited. Journal of Parasitology 83:575-583. Buskirk, S. W., and S. L. Lindstedt. 1989. Sex biases in trapped samples of Mustelidae. Journal of Mammalogy 70:88-97. Buskirk, S. W., and R. A. Powell. 1994. Habitat ecology of fishers and American martens. Pages 283-296 in S. W. Buskirk, A. S. Harestad, M. G. Raphael, and R. A. Powell, editors. Martens, sables and fishers: Biology and conservation. Cornell University Press, Ithaca, NY. Chitwood, B. G. 1933. Does the guinea-worm occur in North America? Journal of the American Medical Association 100:802-804. Christe, P., M. S. Giorgi, P. Vogel, and R. Arlettaz. 2003. Differential species-specific ectoparasitic mite intensities in two ultimately coexisting sibling bat species: resource-mediated host attractiveness or parasite specialization? Journal of Animal Ecology 72:866 - 872. Churcher, T. S., N. M. Ferguson, and M.-G. Basanez. 2005. Density dependence and overdispersion in the transmission of helminth parasites. Parasitology 131:121 - 132. Crichton, V. F. J. 1971. Dracunculiasis in otter, mink and raccoon from Ontario. Journal of Parasitology 57:13. Crichton, V. F. J. 1972. The biology of Dracunculus spp. (Dracunculoidea: Dracunculidae) in wildlife from Ontario. Ph.D. University of Guelph, Guelph, Ontario. 103 Crichton, V. F. J., and M. Beverley-Burton. 1973. Dracunculus lutrae n. sp. (Nematoda: Dracunculoidea) from the otter, Lutra canadensis, in Ontario, Canada. Canadian Journal of Zoology 51:521-529. Crichton, V. F. J., and M. Beverley-Burton. 1974. Distribution and prevalence of Dracunculus spp. (Nematoda: Dracunculoidea). Canadian Journal of Zoology 52:163-167. Crichton, V. F. J., and M. Beverley-Burton. 1975. Migration, growth, and morphogenesis of Dracunculus insignis (Nematoda: Dracunculoidea). Canadian Journal of Zoology 53:105-113. Crichton, V. F. J., and M. Beverley-Burton. 1977. Observations on the seasonal transmission of Dracunculus insignis (Nematoda: Dracunculoidea) in raccoon (Procyon lotor(L.) in Ontario. Journal of Wildlife Diseases 13:273-280. Crowley, S. K., W. B. Krohn, and T. F. Paragi. 1990. A comparison of fisher reproductive estimates. Transactions of the Northeastern Section of the Wildlife Society 47:36-42. Dagg, A. I., D. Leach, and G. Sumner-Smith. 1975. Fusion of the distal femoral epiphyses in male and female marten and fisher. Canadian Journal of Zoology 53:1514 - 1518. Dare, 0. K., and M. R. Forbes. 2008. Rates of development in male and female Wood Frogs and patterns of parasitism by lung nematodes. Parasitology 135:385-393. Dare, O. K., and M. R. Forbes. 2009. Patterns of infection by lungworms, Rhabdias ranae and Haematoloechus spp., in northern leopard frogs: a relationship between sex and parasitism. Journal of Parasitology 95:275-280. Deviche, P., E. C. Greiner, and X. Manteca. 2001. Seasonal and age-related changes in blood parasite prevalence in dark-eyed juncos (Junco hiemalis, Aves, Passeriformes). Journal of Experimental Zoology 289:456-466. Dobson, A. P., P. J. Waller, and A. D. Donald. 1990. Population dynamics of Trichostrongylus colubriformis in sheep: the effect of infection rate on establishment of infective larvae and parasite fecundity. International Journal for Parasitology 20:347 - 352. Douglas, C. W., and M. A. Strickland. 1987. Fisher. Pages 511-530 in M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch, editors. Wild furbearer management and conservation in North America. Ontario Ministry of Natural Resources, Toronto. 104 Eagle, T. C., and J. S. Whitman. 1987. Mink. Pages 615-624 in M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch, editors. Wild furbearer management and conservation in North America. Ministry of Natural Resources. Eberhard M. L., E. Ruiz-Tiben, and S. V. Wallace. 1988. Dracunculus insignis: experimental infection in the ferret, Mustela putorius furo. Journal of Helminthology 62: 265-270. Eberhard, M. L., and F. H. Brandt. 1995. The role of tadpoles and frogs as paratenic hosts in the life cycle of Dracunculus insignis (Nematoda: Dracunculoidea). Journal of Parasitology 81:792-793. Elsasser, S. C., R. Floyd, P. D. N. Hebert, and A. I. Schulte-Hostedde. 2009. Species identification of North American guinea worms (Nematoda: Dracunculus) with DNA barcoding. Molecular Ecology Resources 9:707-712. Esch, G. W., A. 0. Bush, and A. M. Aho. 1990. Parasite communities: Patterns and processes. Chapman and Hall, London, U.K. Folmer, O., M. Black, W. Hoeh, R. Lutz, and R. Vrijenhoek. 1994. DNA Primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology 3:294- 299. Folstad, I., and A. J. Karter. 1992. Parasites, bright males, and the immunocompetence handicap. American Naturalist 139:603-622. Forbes, M. R. L. 1993. Parasitism and host reproductive effort. Oikos 67:444-450. Forbes, M. R. L., and R. L. Baker. 1991. Condition and fecundity of the damselfly, Enallagma ebrium (Hagen): the importance of ectoparasites. Oecologia 86:335- 341. Gehrt, S. D. 2003. Raccoon (Procyon lotor) and allies. Pages 611-613 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors. Wild mammals of North America. John Hopkins University Press, Baltimore, Maryland, USA. Gibilisco, C. J. 1994. Distributional dynamics of American martens and fishers in North America. Pages 59-71 in S. W. Buskirk, A. S. Harestad, and M. G. Raphael, editors. Martens, Sables, and Fishers: Biology and Conservation. Cornell University Press, Ithaca, NY. 105 Gibson, G. G., and D. A. McKiel. 1972. Dracunculus insignis (Leidy, 1858) and larval Eustrongylides sp. in a muskrat from Ontario, Canada. Canadian Journal of Zoology 50:897-901. Gooderham, K., and A. I, Schulte-Hostedde. 2011. Macroparasitism influences reproductive success in red squirrels (Tamiasciurus hudsonicus). Behavioral Ecology 22:1195-1200. Grau, G. A., G. C. Sanderson, and J. P. Rogers. 1970. Age determination of raccoons. Journal of Wildlife Management 34:364-372. Hajibabaei, M., J. R. deWaard, N. V. Ivanova, S. Ratnasingham, R. T. Dooh, S. L. Kirk, P. M. Mackie, and P. D. N. Hebert. 2005. Critical factors for assembling a high volume of DNA barcodes. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 360:1959-1967. Hamilton, G. D., and F. H. Bronson. 1985. Food restriction and reproductive development in wild house mice. Biology of Reproduction 32:773-778. Harrison, A., M. Scantlebury, and W. I. Montgomery. 2010. Body mass and sex-biased parasitism in wood mice Apodemus sylvaticus. Oikos 119:1099 -1104. Hashikura, T. 1927. One case of Filaria medinensis in Korea. Japanese Medical World 7:145-146. Hasu, T., D. P. Benesh, and E. T. Valtonen. 2009. Differences in parasite susceptibility and costs of resistance between naturally exposed and unexposed host populations. Journal of Evolutionary Biology 22:699-707. Hawlena, H., Z. Abramsky, and B. R. Krasnov. 2005. Age-biased parasitism and density- dependant distribution of fleas (Siphonaptera) on a desert rodent. Oecologia 146:200 - 208. Hawlena, H., I. S. Khokhlova, Z. Abramsky, and B. R. Krasnov. 2006. Age, intensity of infestation by flea parasites and body mass loss in a rodent host. Parasitology 133:187-193. Hebert, P. D. N., A. Cywinska, S. L. Ball, and J. R. deWaard. 2003a. Biological identifications through DNA barcodes. Proceedings of the Royal Society of London, B 270:313 - 321. 106 Hebert, P. D. N., S. Ratnasingham, and J. R. deWaard. 2003b. Barcoding animal life: cytochrome c subunit divergences among closely related species. Proceedings of the Royal Society of London, B (Supplement) 270:S96 - S99. Hebert, P. D. N., E. H. Penton, J. M. Burns, D. H. Janzen, and W. Hallwachs. 2004. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. Proceedings of the National Academy of Sciences 101:14812-14817. Hudson, P. J. 1986. The effect of a parasitic nematode on the breeding production of red grouse. Journal of Animal Ecology 55:85-92. Hudson, P. J., A. Dobson, and D. Newborn. 1998. Prevention of population cycles by parasite removal. Science 282:2256 - 2258. Hudson, P. J., A. Dobson, and D. Newborn. 1999. Population cycles and parasitism. Science 286:2425a. Hughes, J., S. D. Albon, R. J. Irvine, and S. Woodin. 2009. Is there a cost of parasites to caribou? Parasitology 136:253-265. lason, G. R., and B. Boag. 1988. Do intestinal helminths affect condition and fecundity of adult mountain hares? Journal of Wildlife Diseases 24:599-605. Irvine, R. J. 2006. Parasites and the dynamics of wild mammal populations. Animal Science 82:775-781. Ivanova, N. V., J. R. deWaard, and P. D. N. Hebert. 2006. An inexpensive, automation- friendly protocol for recovering high-quality DNA. Molecular Ecology Notes 6:998-1002. Jaenike, J. 1981. Criteria for ascertaining the existence of host races. American Naturalist 117:830-834. Jaenike, J. 1993. Rapid evolution of host specificity in a parasitic nematode. Evolutionary Ecology 7:103-108. Janzen, D. H. 1980. When is it coevolution? Evolution 34:611-612. Khokhlova, I. S., B. R. Krasnov, M. Kam, N. I. Burdelova, and A. A. Degen. 2002. Energy cost of extoparasitism: the flea Xenopsylla ramesis on the desert gerbil Gerbillus dasyurus. Journal of Zoology 258:349 - 354. 107 Kidd, A. G., J. Bowman, D. Lesbarreres, and A. I. Schulte-Hostedde. 2009. Hybridization between escaped domestic and wild American mink (Neovison vison). Molecular Ecology 18:1175-1186. Kimber, K. R., and G. V. Kollias. 2000. Infectious and parasitic diseases and contaminant- related problems of North American river otters (Lontra canadensis): a review. Journal of Zoo and Wildlife Medicine 31:452-472. Kimura, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111-120. Kisielewska, K., and Z. Zubczewska. 1973. Intestinal helminths as indexes of reproduction dynamics in the host population - common vole. Acta Theriologica 18:237-246. Kleiber, M. 1961. The Fire of Life: an Introduction to Animal Energetics. Wiley, New York. Kobayashi, A., A. Katatura, A. Hamada, T. Suzuki, Y. Hataba, N. Tashiro, and A. Yoshida. 1986. Human case of dracunculiasis in Japan. American Journal of Tropical Medicine and Hygiene 35:159-161. Krasnov, B. R., R. Poulin, G. I. Shenbrot, D. Mouillot, and I. S. Khokhlova. 2004. Ectoparasitic "Jack-of-AII-Trades": Relationship between abundance and host specificity in fleas (Siphonaptera) parasitic on small mammals. American Naturalist 164:506-516. Kraus, T. 2005. Ovulation rates in harvested fisher, Martes pennanti, populations of central Ontario, with special reference to habitat conditions. M.Sc. Laurentian University, Sudbury, Ontario. Krist, A. C., J. Jokela, J. Wiehn, and C. M. Lively. 2004. Effects of host condition on susceptibility to infection, parasite developmental rate, and parasite transmission in a snail-trematode interaction. Journal of Evolutionary Biology 17:33-40. Kuehn, D. W., and W. E. Berg. 1981. Use of radiographs to identify age-classes of fisher. Journal of Wildlife Management 45:1009-1010. Lariviere, S. 1999. Mustela vison. Mammalian species 608:1-9. Lariviere, S. 2004. Range expansion of raccoons in the Canadian prairies: review of hypotheses. Wildlife Society Bulletin 32:955-963. 108 Lariviere, S., and L. R. Walton. 1998. Lontra canadensis. Mammalian species 587:1-8. Leung, T. L. F., and R. Poulin. 2010. Infection success of different trematode genotypes in two alternative intermediate hosts: evidence for intraspecific specialization? Parasitology 137:321-328. Lotze, J.-H., and S. Anderson. 1979. Procyon lotor. Mammalian species 119:1-8. Lymbery, A. J. 1989. Host specificity, host range and host preference. Parasitology Today 5:298. May, R. M. 1999. Population Biology: crash tests for real. Nature 398:371 - 372. May, R. M., and R. M. Anderson. 1978. Regulation and stability of host-parasite population interactions. II. Destabilising processes. Journal of Animal Ecology 47:249-267. Mayr, E. 1970. Populations, Species, and Evolution. Harvard University Press, Cambridge, Massachusetts. McCoy, K. D., T. Boulinier, C. Tirard, and Y. Michalakis. 2001. Host specificity of a generalist parasite: genetic evidence of sympatric host races in the seabird tick Ixodes uriae. Journal of Evolutionary Biology 14:395-405. Merino, S., and J. Potti. 1995. Mites and blowflies decrease growth and survival in nestling pied flycatchers. Oikos 73:95 -103. Moore, S. L., and K. Wilson. 2002. Parasites as a viability cost of sexual selection in natural populations of mammals. Science 297:2015-2018. Morand, S., and J.-F. Guegan. 2000. Distribution and abundance of parasite nematodes: ecological specialisation, phylogenetic constraint or simply epidemiology? Oikos 88:563-573. Muller, R. 1971. Dracunculus and dracunculiasis. Advances in Parasitology 9:73-153. Neuhaus, P. 2003. Parasite removal and its impact on litter size and body condition in Columbian ground squirrels (Spermophilus columbianus). Proceedings of the Royal Society of London, B (Suppl.) 270:S213-S215. Newey, S., S. J. Thirgood, and P. J. Hudson. 2004. Do parasite burdens in spring influence condition and fecundity of female mountain hares Lepus timidus? Wildlife Biology 10:171 -176. 109 Nituch, L. A., J. Bowman, K. B. Beauclerc, and A. I. Schulte-Hostedde. 2011. Mink farms predict Aleutian disease exposure in wild American mink. PLoS ONE 6:e21693. Olstad, K., G. Robertsen, L. Bachmann, and T. A. Bakke. 2007. Variation in host preference within Gyrodactylus safaris (Monogenea): an experimental approach. Parasitology 134:589-597. Oppliger, A., M. L. Celerier, and J. Clobert. 1996. Physiological and behavioural changes in common lizards parasitized by haemogregarines. Parasitology 113:433-438. Owens, I. P. F. 2002. Sex differences in mortality rate. Science 297:2008-2009. Patterson, L. D., and A. I. Schulte-Hostedde. 2011. Behavioural correlates of parasitism and reproductive success in male eastern chipmunks (Tamias striatus). Animal Behaviour 81:1129-1137. Pedersen, A. B., and T. J. Greives. 2008. The interaction of parasites and resources cause crashes in a wild mouse population. Journal of Animal Ecology 77:370-377. Perez-Orella, C., and A. I. Schulte-Hostedde. 2005. Effects of sex and body size on ectoparasite loads in the northern flying squirrel (Glaucomys sabrinus). Canadian Journal of Zoology 83:1381-1385. Poole, K. G., G. M. Matson, M. A. Strickland, A. J. Magoun, R. P. Graf, and L. M. Dix. 1994. Age and sex determination for American martens and fishers. Pages 204- 223 in S. W. Buskirk, A. S. Harestad, M. G. Raphael, and R. A. Powell, editors. Martens, sables and fishers: Biology and conservation. Cornell University Press, Ithaca, NY. Poulin, R. 1996a. Helminth growth in vertebrate hosts: does host sex matter? International Journal for Parasitology 26:1311-1315. Poulin, R. 1996b. Sexual inequalities in helminth infections: a cost of being a male? American Naturalist 147:287-295. Poulin, R. 1998. Large-scale patterns of host use by parasites of freshwater fishes. Ecology Letters 1:118-128. Poulin, R. 2007. Are there general laws in parasite ecology? Parasitology 134:763 - 776. Poulin, R., and D. B. Keeney. 2008. Host specificity under molecular and experimental scrutiny. Trends in Parasitology 24:24-28. Powell, R. A. 1981. Martes pennanti. Mammalian species 156:1-6. 110 Powell, R. A., S. W. Buskirk, and W. J. Zielinski. 2003. Fisher and marten. Pages 635-649 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors. Wild mammals of North America: biology, management and conservation. The John Hopkins University Press, Baltimore and London. Powers, T. 2004. Nematode molecular diagnostics: from bands to barcodes. Annual Review of Phytopathology 42:367-383. Price, P. W. 1980. Evolutionary biology of parasites. Princeton University Press, Princeton, NJ. Pryor, S. C., S. Henry, and J. Sarfo. 2005. Endogenous morphine and parasitic helminthes. Medical Science Monitor 11:RA183 - 189. Reiczigel, J., and L. Rozsa. 2005. Quantitative Parasitology 3.0. Budapest. Distributed by the authors. Ritke, M. E. 1990. Sexual dimorphism in the raccoon (Procyon lotor): morphological evidence for intrasexual selection. American Midland Naturalist 124:342-351. Roberts, L. S., and J. J. Janovy. 2009. Gerald D. Schmidt & Larry S. Roberts' Foundations of Parasitology. McGraw-Hill, New York, NY. Robitaille, J.-F., and K. Jensen. 2005. Additional indices to estimate fat contents in fisher Martes pennanti population. Wildlife Biology 11:263-269. Rozsa, L., J. Reiczigel, and G. Majoros. 2000. Quantifying parasites in samples of hosts. Journal of Parasitology 86:228-232. Saccone, C., C. De Gorgi, C. Gissi, G. Pesole, and A. Reyes. 1999. Evolutionary genomics in Metazoa: the mitochondrial DNA as a model system. Gene 238:195-209. Santi, S. A., G. H. Parker, N. P. Schaffner, L. Capodagli, and M. A. Persinger. 2006. Prevalence, intensity, and geographic distribution of sinus worm (Skrjabingylus nasicola) infection in mink (Mustela vison) of central Ontario. Canadian Journal of Zoology 84:1011-1018. Sasal, P., S. Trouve, C. Muller-Graf, and S. Morand. 1999. Specificity and host predictability: a comparative analysis among monogenean parasites of fish. Journal of Animal Ecology 68:437-444. Savolainen, V., R. S. Cowan, A. P. Vogler, G. K. Roderick, and R. Lane. 2003. Towards writing the encyclopaedia of life: an introduction to DNA barcoding. Ill Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 360:1805-1811. Schalk, G., and M. R. Forbes. 1997. Male biases in parasitism of mammals: effects of study type, host age, and parasite taxon. Oikos 78:67-74. Schindel, D. E., and S. E. Miller. 2005. DNA barcoding a useful tool fortaxonomists. Nature 435:17. Schulte-Hostedde, A. I., and S. C. Elsasser. 2011. Spleen mass, body condition, and parasite load in male American mink (Neovison vison). Journal of Mammalogy 92:221 - 226. Schulte-Hostedde, A. I., J. S. Millar, and G. J. Hickling. 2001. Evaluating body condition in small mammals. Canadian Journal of Zoology 79:1021-1029. Schwanz, L. E. 2008. Chronic parasitic infection alters reproductive output in deer mice. Behavioral Ecology and Sociobiology 62:1351 -1358. Scott, M. E. 1988. The impact of infection and disease of animal populations: Implications for conservation biology. Conservation Biology 2:40-56. Seilacher, A., W.-E. Reif, and P. Wenk. 2007. The parasite connection in ecosystems and macroevolution. Naturwissenchaften 94:155 -169. Seppala, O., K. Liljeroos, A. Karvonen, and J. Jokela. 2008. Host condition as a constraint for parasite reproduction. Oikos 117:749-753. Shea, M. E., N. L. Rollins, R. T. Bowyer, and A. G. Clark. 1985. Corpora lutea numbers as related to fisher age and distribution in Maine. Journal of Wildlife Management 49:37-40. Sheldon, B. C., and S. Verhulst. 1996. Ecological immunity: costly parasite defences and trade-offs in evolutionary ecology. Trends in Ecology and Evolution 1:317-321. Smith, M. A., D. M. Wood, D. H. Janzen, W. Hallwachs, and P. D. N. Hebert. 2007. DNA barcodes affirm that 16 species of apparently generalist tropical parasitoid flies (Diptera, Tachinidae) are not all generalists. Proceedings of the Royal Society of London. Series B, Biological Sciences 104:4967-4972. Stein, A., R. J. Irvine, E. Ropstad, O. Halvorsen, R. Langvatn, and S. D. Albon. 2002. The impact of gastrointestinal nematodes on wild reindeer: experimental and cross- sectional studies. Journal of Animal Ecology 71:937-945. 112 Stephenson, A. B. 1977. Age determination and morphological variation of Ontario otters. Canadian Journal of Zoology 55:1577-1583. Strickland, M. A., C. W. Douglas, M. Novak, and N. P. Hunziger. 1982. Fisher. Pages 599- 612 in J. A. Chapman and G. A. Feldhamer, editors. Wild mammals of North America. John Hopkins University Press, Baltimore. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24:1596-1599. Thomas, F., S. P. Brown, M. Sukhdeo, and F. Renaud. 2002. Understanding parasite strategies: a state-dependent approach? Trends in Parasitology 18:387-390. Thompson, I. D. 2000. Forest vertebrates of Ontario: Patterns of distribution. Pages 54- 73 in H. A. Perera, D. L. Euller, and I. D. Thompson, editors. Ecology of a managed terrestrial landscape: patterns of processes of forest landscapes in Ontario. UBC Press, Vancouver. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTALW: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choices. Nucleic Acids Research 22:4673-4680. Toft, C. A. 1991. Current theory of host-parasite interactions. Pages 3-15 in J. E. Loye and M. Zuk, editors. Bird-parasite interactions. Ecology, evolution and behaviour. Oxford University Press, Oxford. Toft, C. A., and A. J. Karter. 1990. Parasite-host coevolution. Trends in Ecology and Evolution 5:326-329. Tompkins, D. M., A. M. Dunn, M. J. Smith, and S. Telfer. 2011. Wildlife diseases: from individuals to ecosystems. Journal of Animal Ecology 80:19 - 38. Tompkins, D. M., and P. J. Hudson. 1999. Regulation of nematode fecundity in the ring- necked pheasant (Phasianus colchicus): not just density dependence. Parasitology 118:417-423. Tschirren, B., P. S. Fitze, and H. Richner. 2003. Sexual dimorphism in susceptibility to parasites and cel^mediated immunity in great tit nestlings. Journal of Animal Ecology 72:839 - 845. Wang, Z., D. Jisuan, and X. Wang. 1995. Dracunculiasis discovered for the first time in China. Chinese Journal of Zoonosis 11:15-17. 113 Ward, S. A. 1992. Assessing functional explanations of host specificity. American Naturalist 139:883-891. Wauters, L. A., and A. A. Dhondt. 1995. Lifetime reprodcutive success and its correlates in female Eurasian red squirrels. Oikos 72:402-410. Webster, J. P., C. M. Gower, and L. Blair. 2004. Do hosts and parasites coevolve? Empirical support from the Schistosoma system. American Naturalist 164:S33- S51. Webster, W. A., and G. A. Casey. 1970. The occurrence of Dracunculus insignis (Leidy, 1858) Chandler, 1942 in a skunk from Ontario, Canada. Journal of Wildlife Diseases 6:71. Whiteman, N. K., and P. G. Parker. 2004. Body condition and parasite load predict territory ownership in the Galapagos Hawk. The Condor 106:915 - 921. Whitlock, M. C. 1996. The Red Queen beats the jack-of-all-trades: the limitations on the evolution of phenotypic plasticity and niche breadth. American Naturalist 148:S56-S77. Wijova, M., F. Moravec, A. Horak, D. Modry, and J. Lukes. 2005. Phylogenetic position of Dracunculus medinensis and some related nematodes inferred from 18S rRNA. Parasitology Research 96:133-135. Woolhouse, M. E. J., L. H. Taylor, and D. T. Haydon. 2001. Population biology of multihost pathogens. Science 292:1109-1112. Wright, P. L., and M. W. Coulter. 1967. Reproduction and growth in Maine fishers. Journal of Mammalogy 34:74-86. Zhu, W., G. Baggerman, W. E. Secor, F. Casares, S. C. Pryor, G. L. Fricchione, E. Ruiz- Tiben, M. L. Eberhard, L. Bimi, and G. B. Stefano. 2002. Dracunculus medinensis and Schistoma mansoni contain opiate alkaloids. Annals of Tropical Medicine & Parasitology 96:309 - 316. Zuk, M., and K. A. McKean. 1996. Sex differences in parasite infections: patterns and processes. International Journal for Parasitology 26:1009-1024. 114