Evolutionary Ecology of Host-Parasite Relationships: Role of Host Ecology, Phylogeny, and Demographics in Shaping Parasite Evolution." (2020)

University of New Mexico UNM Digital Repository

Biology ETDs Electronic Theses and Dissertations

Spring 4-13-2020 EVOLUTIONARY ECOLOGY OF HOST- PARASITE RELATIONSHIPS: ROLE OF HOST ECOLOGY, PHYLOGENY, AND DEMOGRAPHICS IN SHAPING PARASITE EVOLUTION Erika Taylor Gendron University of New Mexico

Follow this and additional works at: https://digitalrepository.unm.edu/biol_etds Part of the Animal Diseases Commons, Biology Commons, Evolution Commons, Genetics Commons, Genomics Commons, Other Ecology and Evolutionary Biology Commons, and the Parasitic Diseases Commons

Recommended Citation Gendron, Erika Taylor. "EVOLUTIONARY ECOLOGY OF HOST-PARASITE RELATIONSHIPS: ROLE OF HOST ECOLOGY, PHYLOGENY, AND DEMOGRAPHICS IN SHAPING PARASITE EVOLUTION." (2020). https://digitalrepository.unm.edu/ biol_etds/263

This Dissertation is brought to you for free and open access by the Electronic Theses and Dissertations at UNM Digital Repository. It has been accepted for inclusion in Biology ETDs by an authorized administrator of UNM Digital Repository. For more information, please contact [email protected]. i

Erika Taylor Gendron Candidate

Department of Biology Department

This dissertation is approved, and it is acceptable in quality and form for publication:

Approved by the Dissertation Committee:

Dr. Sara V. Brant, Chairperson

Dr. Eric S. Loker

Dr. Thomas Turner

Dr. Christopher Witt

Dr. Charles Criscione

EVOLUTIONARY ECOLOGY OF HOST-PARASITE RELATIONSHIPS: ROLE OF HOST ECOLOGY, PHYLOGENY, AND DEMOGRAPHICS IN SHAPING PARASITE EVOLUTION

Erika Taylor Gendron (Ebbs)

B.S. Biology, California State University, Fresno, 2011

DISSERTATION

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy Biology

The University of New Mexico Albuquerque, New Mexico

May 2018

iii

ACKNOWLEDGMENTS

I would like to first acknowledge my primary supervisor Dr. Sara Brant who has provided me with unlimited support and intellectual freedom over the past seven years. Her guidance and enthusiasm for Parasitology has shaped my academic development. I would also like to thank my co-supervisor Dr. Eric S. Loker for his guidance and generosity over the years, and specifically for creating a productive and supportive environment to study Parasitology. I would like to express my gratitude to my dissertation committee, Dr. Chris Witt, Dr. Tom Turner and Dr. Charles Criscione, for sharing their expertise and providing invaluable critiques of this work. I would like to thank my lab mates and members of the larger UNM Parasitology group: Ms. Martina Laidemitt, Dr. Sarah Buddenborg, Mr. Jon Schultz, Ms. Janeth Pena, Ms. Lijun Liu, Dr. Ramesh Devkota, Dr. Ben Hanelt, Ms. Rachel Swanteson-Franz, Dr. Coen Adema, Dr. Si-Ming Zang, Dr. Bruce Hofkin, Dr. Lijing Bu and Ms. Michelle Gordy. Thank you all for your feedback, encouragement and most of all your friendship. Several undergraduate students have contributed to this work, and my graduate experience as a whole, as such I would like to acknowledge;

Ms. D’Eldra Malone, Mr. Keith Keller, Ms. Emily Savris and Ms. Marissa Trujillo. I would like to acknowledge the Museum of Southwestern Biology and its Divisions for the many experiences and professional development opportunities it has afforded me as a curatorial assistant. I would also like to acknowledge the University of New Mexico’s Molecular Biology Facility, specifically Mr. George Rosenberg for his patience and helpful conversations. I would like to also thank my family for their constant support during my pursuit of this degree. Specifically, my husband Daniel and son Jude whom helped me maintain perspective during this process, my mother Sharon for her love and encouragement and my father Michael for instilling curiosity about the natural world.

EVOLUTIONARY ECOLOGY OF HOST-PARASITE RELATIONSHIPS: ROLE OF HOST ECOLOGY, PHYLOGENY, AND DEMOGRAPHICS IN SHAPING PARASITE EVOLUTION

by Erika Taylor Gendron (Ebbs)

B.S. Biology, California State University, Fresno, 2011 PhD. Biology, University of New Mexico, Albuquerque 2018

ABSTRACT Host-parasite systems exist across complex and ecologically heterogeneous

landscapes, and may occur across multiple taxonomically and ecologically disparate

host species. Under these conditions, mechanisms underlying microevolutionary

processes (i.e. gene flow, genetic drift) are not always clear, and may be mediated

by numerous co-occurring factors specific to individual hosts, host groups or life

stages. Host traits such as host immunology, demographics, phylogeny and ecology

are important factors, which may shape host-parasite relationships, and ultimately

evolutionary processes. Importantly, host traits acting at one life stage might impact

transmission at a different life stage. As such, the integrated nature of parasite life

cycles is an important consideration as host traits may act in concert to enhance or

constrain parasite evolution.

The research described herein used phylogeographic, phylogenomic, and

population genetic methods to further our understanding of how host traits impact the

evolutionary ecology of trematode systems, using avian schistosomes (Digenea:

Schistosomatidae) as a model. Much of this work specifically focused on the genus

Trichobilharzia and an important snail host Physa acuta. These studies in aggregate

v found that host habitat preference, migratory patterns and demographics were all important factors in mediating parasite transmission. Further within Trichobilharzia spp., which are parasites of ducks, two species which utilize P. acuta were found to have substantially different population genetic patterns. Data from this study suggest that duck habitat preference can increase (dabbling ducks) or dilute (diving ducks)

Trichobilharzia transmission. Further data suggest that certain combinations of hosts may facilitate transmission into a broader geographic range. At the macroevolutionary level this study finds support that host-switching has likely been an important mechanism of schistosome diversification, and did not find evidence suggesting a strong co-phylogenetic pattern among parasites and intermediate hosts. In total this research provides a framework for understanding how host traits may act to shape parasite transmission and subsequent evolution.

TABLES OF CONTENTS

TABLE OF CONTENTS ...... vi LIST OF FIGURES...... viii LIST OF TABLES ...... ix INTRODUCTION ...... 1 References ...... 7 CHAPTER I: SCHISTOSOMES WITH WINGS: HOW HOST PHYLOGENY AND ECOLOGY SHAPE THE GLOBAL DISTRIBUTION OF TRICHOBILHARZIA QUERQUEDULAE (SCHISTOSOMATIDAE) ...... 11 Abstract ...... 11 Introduction...... 12 Materials and Methods ...... 15 Results ...... 18 Discussion ...... 19 Acknowledgements ...... 27 References ...... 28 Figures ...... 35 Tables ...... 40 CHAPTER II: PHYLOGEOGRAPHY AND GENETICS OF THE GLOBALLY INVASIVE SNAIL PHYSA ACUTA DRAPARUND 1805 (= HAITIA ACUTA, TAYLOR 2003), AND ITS POTENTIAL AS AN INTERMEDIATE HOST TO LARVAL DIGENETIC TREMATODES ...... 46 Abstract ...... 46 Introduction...... 47 Materials and Methods ...... 51 Results ...... 58 Discussion ...... 63 Conclusions ...... 71 Acknowledgements ...... 72 References ...... 72 Figures ...... 82 Tables ...... 85 CHAPTER III: COMPARATIVE MICROEVOLUTIONARY PATTERNS OF CONGENERIC TREMATODE SPECIES REVEAL IMPORTANCE OF HOST ECOLOGY IN PARASITE EVOLUTION ...... 96 Abstract ...... 96 Introduction...... 97 Materials and Methods ...... 101 Results ...... 107 Discussion ...... 115 Acknowlegments ...... 121 References ...... 121

vii

Figures ...... 127 Tables ...... 132 CHAPTER IV: ILLUMINATING SCHISTOSOME DIVERSITY: PHYLOGENOMICS AND DIVERSIFICATION OF SCHISTOSOMATIDAE USING TARGETED SEQUENCE CAPTURE OF ULTRA-CONSERVED ELEMENTS ...... 143 Abstract ...... 143 Introduction...... 144 Materials and Methods ...... 148 Results ...... 151 Discussion ...... 153 References ...... 162 Figures ...... 170 Tables ...... 172 CONCLUSIONS ...... 175 APPENDIX A – Additional Files for Chapter II ...... 181 APPENDIX B – Additional Files for Chapter III ...... 188 APPENDIX C – Additional Files for Chapter IV ...... 212

viii

LIST OF FIGURES

CHAPTER I Figure 1 – cox1 phylogeny ...... 35 Figure 2 – ITS1 phylogeny ...... 36 Figure 3 – nd4 phylogeny ...... 37 Figure 4 – Blue wing group phylogeny ...... 38 CHAPTER II Figure 1 - Map of sampled populations ...... 82 Figure 2 – Relationships of recovered cox1 haplotypes ...... 83 Figure 3 – Bayesian skyline plots ...... 85 Figure 4 – Invasion timeline of Physa acuta ...... 87 CHAPTER III Figure 1 - cox1 phylogeny ...... 127 Figure 2 – Minimum spanning network ...... 129 Figure 3 – Bayesian skyline plots ...... 130 Figure 4 – Trichobilharzia summary tree ...... 131 CHAPTER IV Figure 1 - ML analyses of concatenated UCE loci ...... 170 Figure 2 – Host-switching within Schistosomatidae ...... 171

LIST OF TABLES

CHAPTER I Table 1 - Host association and locality origin of the specimens ...... 40 Table 2 – Prevalence of T. querquedulae ...... 44 Table 3 – Pairwise genetic distances ...... 45 CHAPTER II Table 1 - Summary of range-wide population statistics ...... 88 Table 2 – Analysis of Molecular Variance ...... 91 Table 3 – Pairwise genetic distances and ΦST values ...... 92 Table 4 – Reports of infection and trematode species richness ...... 93 Table 5 - Matrix of correlation coefficients ...... 95 CHAPTER III Table 1 - Summary of Trichobilharzia survey data ...... 132 Table 2 – Measures of host specificity ...... 139 Table 3 – Analyses of molecular variance ...... 141 Table 4 – Indices of genetic diversity ...... 142 CHAPTER IV Table 1 - Summary of sequence capture data ...... 172 Table 2 – Summary of UCE alignments ...... 174

INTRODUCTION

Theoretical Basis

Host-parasite (H-P) systems exist across complex and ecologically heterogeneous landscapes and are shaped by many co-occurring factors

(host immunology, host physiology, evolutionary constraints and ecology).

The persistence of host-parasite relationships over time and space often requires interaction among taxonomically and ecologically disparate host species, each having differential impacts on parasite transmission and evolution (Blouin et al., 1995; Criscione and Blouin, 2004; Steinauer et al.,

2007; Keeney et al., 2009; Archie and Ezenwa, 2011; Blasco-Costa and

Poulin, 2013) . As such, parasite ecology and evolution is inseparable from the ecology, behavior and demographics of their host(s). In this way host specific traits can act to enhance or constrain parasite population size

(Criscione and Blouin, 2005), geographic range (Blasco-Costa, 2012), transmission (Archie and Ezenwa, 2011), and adaptive potential (Nuismer,

2009). Teasing apart the multilayered impacts of host-use on parasite evolution requires a holistic understanding of evolutionary relationships of both the parasites and associated host species, across multiple time-scales.

To contribute to the understanding of parasite evolutionary ecology, specifically in relation to host specific traits this research integrated host- parasite phylogeography, population genetics and phylogenomics.

Empirical data resulting from evolutionary studies of parasite populations supports the idea that host specific traits (e.g. allogeneic/autogenic life stages,

2 dispersal and range) mediate microevolutionary forces (gene flow, genetic drift) and determine population structure within their associated parasites

(Blouin et al., 1995; Blouin et al., 1999; Criscione and Blouin, 2004; Blasco-

Costa, 2012; Blasco-Costa and Poulin, 2013). Yet beyond broad ecological distinctions (i.e. allogeneic/autogenetic, migratory/non-migratory), within multi- host systems little is known about what specific traits contribute to or determine gene flow, genetic drift and effective population size (Ne) (Nadler,

1995; Criscione et al., 2005; Blasco-Costa and Poulin, 2013). These parameters directly relate to, or can be predictive of, the evolution of drug resistance (Blouin et al., 1995), local adaptation (Dybdahl and Lively, 1996;

McCoy et al., 2002; Nuismer, 2009), probability of host-switching (Hoberg and

Brooks, 2008) and ultimately speciation of parasite lineages (Huyse et al.,

2005).

Study System

This research focused on the ecological and evolutionary relationships of digenetic trematodes and their hosts, primarily those within the family

Schistosomatidae (Weinland 1858). Schistosomatidae, is a medically and biologically important group which occurs globally. Unlike most trematodes schistosomes have a two-host life cycle (Cribb et al., 2003), which makes them more tractable for the types of studies described herein. Additionally, relative to many other trematode families, host-parasite associations within

Schistosomatidae are fairly well resolved and diverse (Snyder and Loker

2000; Brant and Loker, 2013), facilitating the study of how host-specific traits influence parasite evolution. Avian infecting lineages (known as avian

3 schistosomes) comprise the majority of schistosome diversity, and were the primary focus of this work.

Avian schistosomes represent 10 of the 14 named genera within

Schistosomatidae, though recent molecular survey’s have confirmed that this is an underestimate (Brant et al., 2006; Brant and Loker, 2013). Dioecious adult schistosomes colonize the vascular system of their vertebrate definitive host, either a mammal or a bird. There they mature and copulate, with the female producing eggs that eventually leave the body with the host’s feces, urine or nasal secretions, depending on the species. There is much variation in reproductive strategy and sexual development within Schistosomatidae

(Morand and Müller-Graf, 2000; Loker and Brant, 2006) , specifically in regards to the nature of the relationship between the male and female worms.

Adult body size, sexual dimorphism and life span also vary substantially across the family (Loker and Brant, 2006).

Liberated eggs hatch and release a motile larval stage known as a miracidium within an aquatic environment. Miracidia actively seek and penetrate a gastropod intermediate host. Schistosome species are remarkable with respect to the aggregate taxonomic range of intermediate hosts they utilize, especially so for avian schistosomes where an estimated 15 families of snails are employed as intermediate hosts (Brant and Loker, 2013).

Any particular schistosome species however typically successfully exploits a very limited range of gastropod host species. Prior studies (Brant et al., 2006) and research herein suggest that intermediate host switching has likely been an important mechanism of schistosome diversification at multiple time scales

(within and across genera). More interestingly however, is the fact that ‘host-

4 switches’ are not associated with deep nodes or long branches, indicating that host-switching events were recent (Brant and Loker, 2013) and a lack of strict co-phylogeny at deeper nodes. Schistosome species are considered to be more specific (Horak et al. 2015) to their snail intermediate host, relative to their definitive host, which is common among trematode families (Cribb et al.

2003). Jouet et al. (2010) found that European Trichobilharzia spp. are specific to single species of snails. Some evidence of co-phylogeny exists among Schistosoma spp. and snails within the family Planorbidae, suggesting a possible history of co-speciation (Morgan et al. 2002). This pattern does not appear to be supported within avian infecting genera like Trichobilharzia,

Dendritobilharzia or Bilharziella.

Within the snail intermediate host, asexual amplification occurs producing a second motile larval stage known as cercaria. Cercariae actively seek out their vertebrate definitive hosts, where they will penetrate the epidermis and migrate through the hosts’ vasculature until they arrive in the liver (generally), where they will undergo sexual maturation and sexual reproduction. Cercariae are attracted to general vertebrate fatty acids, and will often mistakenly attempt to penetrate unsuitable hosts (Horák et al., 2015).

The re-emerging zoonotic disease Human Cercarial Dermatitis (HCD), also known as “Swimmer’s Itch” (Horák et al., 2015), occurs when non-human schistosome cercariae accidentally penetrate human skin. HCD is of increasing concern from a public health and economic perspective (Horák et al., 2015). In relation to HCD and other snail-transmitted diseases, understanding the biology and distribution of snail intermediate hosts has broad significance.

One important host to Trichobilharzia, specifically within North America, is Physa acuta (Draparnaud, 1805). Physa acuta is endemic to North

America, but is now a globally invasive species, with populations established on six of the seven continents (Bousset et al., 2014; Lydeard et al., 2016). In general snails are adept invaders (Albrecht et al., 2009; 2008; Facon et al.,

2003; Hanson et al., 2013; Torchin et al., 2005) and in light of the fact that as group they host over 10,000 species of trematodes (Cribb et al., 2003), their biology is of great importance, specifically in regards to the potential for invasive snail populations to sustain trematode life cycles. Surveys of invasive snail populations suggest that there is a release of trematode burden following host invasion, for at least a period, in concordance with the ‘Enemy-Release hypothesis’ observed across host taxa (Keane, 2002; Torchin et al., 2003).

Snail-trematode surveys demonstrate that while trematode assemblages may change infection within the hosts invasive range does occur (Blakeslee et al.,

2012; Torchin et al., 2005). When local parasites establish within invasive hosts populations it is known as parasite spillback. It is unknown how, why or when spillback would be expected to occur, or if there are general determinants of spillback (Kelly et al., 2009). Free-living systems have shed light on possible relevant factors (Hayes and Barry, 2007; Keane, 2002), specifically in identifying the importance of demographic parameters such as effective population size and background genetic diversity which are expected to increase with time since invasion (Forsman, 2014; Rius and Darling, 2014).

Studies focusing on characterizing these parameters in combination with parasite data are ideal to address the underlying mechanisms of host-parasite

6 invasion dynamics and to identify factors relevant in modeling parasite spillback.

Overall Contributions

How genetic variation is distributed across species, space and time relates to persistance and diversification of parasites lineages. In total, all chapters (1-4) included herein sought to describe patterns of genetic variation within complex H-P systems. The H-P unit evolves at the population level, yet studying parasite populations is challenging as populations are fragmented across hosts, distinct life stages and ecological time and space (Prugnolle,

2005). Research presented herein provides a framework for understanding how populations of parasites within migratory hosts are distributed over ecological and evolutionary time.

With an emphasis on the biology schistosomes, much of this research reported here focuses on characterizing microevolutionary patterns, primarily genetic structure and diversity. These measures directly influence the adaptability of the host-parasite system, and how the unit can respond to biotic and abiotic changes such as increasing temperatures, chemotherapy pressure or habitat loss. Data from this work suggest both host and parasite demographic patterns may shape H-P interactions and influence transmission.

This body of research investigated patterns of host-use and parasite diversity (intra- and inter-specific) over ecological and evolutionary timescales and made substantial contributions to our understanding of the evolutionary ecology of parasitic multi-host helminths.

REFERENCES

Albrecht, C., Hauffe, T., Schreiber, K., Trajanovski, S., Wilke, T., 2009. Mollusc Biodiversity and Endemism in the Potential Ancient Lake Trichonis, Greece. Malacologia 51, 357–375. doi:10.4002/040.051.0209

Albrecht, C., Kroll, O., Moreno Terrazas, E., Wilke, T., 2008. Invasion of ancient Lake Titicaca by the globally invasive Physa acuta (Gastropoda: Pulmonata: Hygrophila). Biol Invasions 11, 1821–1826. doi:10.1007/s10530-008-9360-9

Archie, E.A., Ezenwa, V.O., 2011. Population genetic structure and history of a generalist parasite infecting multiple sympatric host species. International Journal for Parasitology 41, 89–98. doi:10.1016/j.ijpara.2010.07.014

Blakeslee, A.M.H., Altman, I., Miller, A.W., Byers, J.E., Hamer, C.E., Ruiz, G.M., 2012. Parasites and invasions: a biogeographic examination of parasites and hosts in native and introduced ranges. Journal of Biogeography 39, 609–622. doi:10.1111/j.1365-2699.2011.02631.x

Blasco-Costa, I., Poulin, R., 2013. Host traits explain the genetic structure of parasites: a meta-analysis. Parasitology 140, 1316–1322. doi:10.1017/S0031182013000784

Blasco-Costa, I.E.A., 2012. Swimming against the current: genetic structure, host mobility and the drift paradox in trematode parasites. Molecular Ecology 21, 207–217.

Blouin, M.S., Yowell, C.A., Courtney, C.H., Dame, J.B. 1995. Host Movement and the genetic structure of populations of parasitic nematodes. Genetics 141, 1007–1014.

Blouin, M., Liu, J., Berry, R., 1999. Life cycle variation and the genetic strucutre of nematode populations. Heredity 83, 253–259.

Bousset, L., Pointier, J.P., David, P., Jarne, P., 2014. Neither variation loss, nor change in selfing rate is associated with the worldwide invasion of Physa acuta from its native North America. Biol Invasions 16, 1769–1783. doi:10.1007/s10530-013-0626-5

Brant, S.V., Loker, E.S., 2013. Discovery-based studies of schistosome diversity stimulate new hypotheses about parasite biology. Trends in Parasitology 1–11. doi:10.1016/j.pt.2013.06.004

Brant, S.V., Morgan, J.A.T., Mkoji, G.M., Snyder, S.D., Rajapakse, R.P.V.J., Loker, E.S., 2006. An approach to revealing blood fluke life cycles, taxonomy, and diversity: provision of key reference data including DNA sequence from single life cycle stages. Journal of Parasitology 92, 77–88. doi:10.1645/GE-3515.1

Cribb, T., Bray, R., Olson, P., Timothy, D., Littlewood, J., 2003. Life Cycle Evolution in the Digenea: a New Perspective from Phylogeny, in: Advances in Parasitology Volume 54, Advances in Parasitology. Elsevier, pp. 197–254. doi:10.1016/S0065-308X(03)54004-0

Criscione, C., Blouin, M., 2004. Life cycles shape parasite evolution: Comparative population genetics of Salmon Trematodes. Evolution 58, 198–202.

Criscione, C.D., Poulin, R., Blouin, M., 2005. Molecular ecology of parasites: elucidating ecological and microevolutionary processes. Molecular Ecology 14, 2247–2257.

Criscione, C., Blouin, M., 2005. Effective sizes of macroparasite populations: a conceptual model. Trends in Parasitology 21.

Draparnaud, J., 1805. Histoire naturelle des mollusques terrestres et fluviatiles de la France. doi:10.5962/bhl.title.13180

Dybdahl, M., Lively C.M., 1996. The Geography of Coevolution: Comparative Population structures for a snail and its trematode parasite. Evolution 50, 2264–2275.

Facon, B., Pointier, J.P., Glaubrecht, M., Poux, C., Jarne, P., David, P., 2003. A molecular phylogeography approach to biological invasions of the New World by parthenogenetic Thiarid snails. Molecular Ecology 12, 3027– 3039. doi:10.1046/j.1365-294X.2003.01972.x

Forsman, A., 2014. Effects of genotypic and phenotypic variation on establishment are important for conservation, invasion, and infection biology. Proc. Natl. Acad. Sci. U.S.A. 111, 302–307. doi:10.1073/pnas.1317745111

Hanson, D., Cooke, S., Hirano, Y., Malaquias, M.A.E., Crocetta, F., Valdés, Á., 2013. Slipping through the Cracks: The Taxonomic Impediment Conceals the Origin and Dispersal of Haminoea japonica, an Invasive Species with Impacts to Human Health. PloS ONE 8, e77457. doi:10.1371/journal.pone.0077457

Hayes, K.R., Barry, S.C., 2007. Are there any consistent predictors of invasion success? Biol Invasions 10, 483–506. doi:10.1007/s10530-007- 9146-5

Hoberg, E.P., Brooks, D.R., 2008. A macroevolutionary mosaic: episodic host- switching, geographical colonization and diversification in complex host- parasite systems. Journal of Biogeography 35, 1533–1550. doi:10.1111/j.1365-2699.2008.01951.x

Horák, P., Mikeš, L., Lichtenbergová, L., Skála, V., Soldánová, M., Brant,

S.V., 2015. Avian schistosomes and outbreaks of cercarial dermatitis. Clin. Microbiol. Rev. 28, 165–190. doi:10.1128/CMR.00043-14

Huyse, T., Poulin, R., Théron, A., 2005. Speciation in parasites: a population genetics approach. Trends in Parasitology 21, 469–475. doi:10.1016/j.pt.2005.08.009

Jouet, D., Skírnisson, K., Kolářová, L., Ferte, H. 2010. Molecular diversity of Trichobilharzia franki in two intermediate hosts (Radix auricularia and Radix peregra): a complex of species. Infection, genetics and evolution 10, 1218-27.

Keane, R., 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology & Evolution 17, 164–170. doi:10.1016/S0169- 5347(02)02499-0

Keeney, D.B., King, T.M., Rowe, D.L., Poulin, R., 2009. Contrasting mtDNA diversity and population structure in a direct-developing marine gastropod and its trematode parasites. Molecular Ecology 18, 4591–4603. doi:10.1111/j.1365-294X.2009.04388.x

Kelly, D.W., Paterson, R.A., Townsend, C.R., Poulin, R., Tompkins, D.M., 2009. Parasite spillback: A neglected concept in invasion ecology? Ecology 90, 2047–2056. doi:10.1890/08-1085.1

Loker, E.S., Brant, S.V., 2006. Diversification, dioecy and dimorphism in schistosomes. Trends in Parasitology 22, 521–528. doi:10.1016/j.pt.2006.09.001

Snyder, S., Loker, E.S., 2000. Evolutionary relationships among the Schistosomatidae (Platyhelminhes: Digenea) and an Asian origin for Schistosoma. Journal of Parasitology 86, 283–288.

Lydeard, C., Campbell, D., Golz, M., 2016. Physa acuta Draparnaud, 1805 Should be Treated as a Native of North America, Not Europe. Malacologia 59, 347–350. doi:10.4002/040.059.0213

Morgan, J.A., DeJong, R.J., Jung, Y., Khallaayoune, K., Kock, S., Mkoji, G.M., Loker, E.S. 2002. A phylogeny of planorbid snails, with implications for the evolution of Schistosoma parasites. Molecular phylogenetics and evolution 25, 477-88.

McCoy, K.D., Thierry Boulinier, Schjørring, S., Michalakis, Y., 2002. Local adaptation of the ectoparasite Ixodes uriae to its seabird host. Evol Ecol Res 4, 441–456.

Nadler, S., 1995. Microevolution and the Genetic Structure of Parasite Populations. Journal of Parasitology 81, 395–403.

Nuismer, S.G.A.S., 2009. Interactions between Genetic Drift, Gene Flow and

Selection Mosaics Drive Parasite Local Adaptation. The American Naturalist 173.

Prugnolle, F., 2005. Population genetics of complex life-cycle parasites: an illustration with trematodes. International Journal of Parasitology 35, 255– 263.

Rius, M., Darling, J.A., 2014. How important is intraspecific genetic admixture to the success of colonizing populations? Trends in Ecology & Evolution 29, 233–242. doi:10.1016/j.tree.2014.02.003

Morand, S., Müller-Graf, C.D. 2000. Muscles or testes? Comparative evidence for sexual competition among dioecious blood parasites (Schistosomatidae) of vertebrates Parasitology 1–12.

Steinauer, M.L., Nickol, B.B., Ortí, G., 2007. Cryptic speciation and patterns of phenotypic variation of a highly variable acanthocephalan parasite. Molecular Ecology 16, 4097–4109. doi:10.1111/j.1365-294X.2007.03462.x

Torchin, M.E., Byers, J.E., Huspeni, T.C., 2005. Differential Parasitism of Native and Introduced Snails: Replacement of a Parasite Fauna. Biol Invasions 7, 885–894. doi:10.1007/s10530-004-2967-6

Torchin, M.E., Lafferty, K.D., Dobson, A.P., McKenzie, V.J., 2003. Introduced species and their missing parasites. Nature.

CHAPTER I

Schistosomes with wings: how host phylogeny and ecology shape the global distribution of Trichobilharzia querquedulae (Schistosomatidae)

AUTHORS: Erika T. Gendron (Ebbs), Eric S. Loker, Norm E. Davis, Veronica Flores, Aylen Veleizan, Sara V. Brant

ABSTRACT

Migratory waterfowl play an important role in the maintenance and spread of zoonotic diseases worldwide. An example is cercarial dermatitis, caused when larval stages of schistosomes that normally develop in birds penetrate human skin. Members of the genus Trichobilharzia

(Schistosomatidae), transmitted mainly by ducks, are considered to be major etiological agents of cercarial dermatitis globally. To better understand the diversity and distribution of Trichobilharzia spp., we surveyed ducks from

North America (United States and Canada), Argentina, South Africa and New

Zealand. To aid in species identification of the Trichobilharzia worms recovered, regions of the Cox1, ND4 and ITS1 were sequenced.

Furthermore, we provide molecular phylogenetic evidence for the cosmopolitan distribution and trans-hemispheric gene flow for one species,

Trichobilharzia querquedulae, previously thought to be restricted to North

America. These new samples from endemic non-migratory duck species indicate that T. querquedulae transmission occurs within each of the regions we sampled and that it is specific to the blue-winged + silver teal duck clade.

Prevalence within this host group is >95% across the known range of T. querquedulae, indicating that transmission is common. Genetic divergence is

12 evenly distributed among continents, and no phylogenetic structure associated with geography was observed. The results provide strong support for the global distribution and transmission of T. querquedulae and represent, to our knowledge, the first report of a cosmopolitan schistosome confirmed by genetic data. These data are the first known to support trans-hemispheric genetic exchange in a species responsible for causing cercarial dermatitis, indicating that the epidemiology of this group of poorly known zoonotic parasites is more complex than previously expected.

INTRODUCTION

Maintenance of complex life cycles over time and space is influenced by a myriad of both evolutionary and ecological forces. Determinants of host associations are essential in understanding the dynamics of any host- parasite system and for efforts to model the emergence of zoonotic parasites (Ostfeld and Keesing, 2000; Taylor et al., 2001; McKenzie, 2007).

For example, the degree to which a parasite has phylogenetic and/or ecological constraints with its host range will largely determine the predictability of transmission (Thompson, 2000; Hoberg and Brooks, 2008).

Within this context, we aimed to understand host associations and factors shaping the diversity and distribution of a common group of trematodes in waterfowl which are known for causing the globally re-emerging zoonotic disease human cercarial dermatitis (HCD) (de Gentile et al., 1996; Kolářová et al., 2010; Horak et al., 2015). This disease enters the human population when avian schistosome (Schistosomatidae) cercariae penetrate human skin, resulting in a severe, short-term allergic reaction (Horak et al., 2002;

Kourilova et al., 2004; Kolářová et al., 2013). Although the larvae are unable to establish patent infections within a human host, there are intense immune reactions that can be long lasting and painful (Kourilova et al.,

2004). While HCD is neither communicable nor fatal, it is still not without economic and epidemiologic importance and is considered a substantial public health problem (Horak and Kolarova, 2011; Soldánová et al., 2013;

Horak et al., 2015). Relatively little is known about the epidemiology of HCD due to the diversity of schistosome species, the lack of data on the preferred hosts and the timing of transmission, and the large geographical scale over which outbreaks occur.

An increase in the incidence of HCD outbreaks (de Gentile et al., 1996;

Larsen et al., 2004) has been linked to altered or managed habitats that support high densities of migratory and resident anseriform birds and permissive snail host species (Larsen et al., 2004; Horak and Kolarova, 2011) as well as to other forms of environmental degradation such as eutrophication of water bodies (Valdovinos and Balboa, 2008; Soldánová et al., 2013). These conditions have created an urgent need to better understand avian schistosome species diversity, distribution, host use, associated host ecology and life cycle maintenance, within a global framework. The avian schistosome genus Trichobilharzia (Skrjabin and Zakharov, 1920) is most frequently implicated in HCD outbreaks (Kolářová et al., 2010, 2013; Soldánová et al.,

2013; Horak et al., 2015) and is also the most speciose genus within the family (Blair and Islam, 1983; Brant and Loker, 2009). Adults of

Trichobilharzia spp. primarily infect ducks and as larvae develop within freshwater pulmonate snails in the families Physidae and Lymnaeidae (Blair

14 and Islam, 1983; Horak et al., 2002; Brant and Loker, 2009). Species diversity and host use of Trichobilharzia spp. are best known in Europe and North

America (Aldhoun et al., 2009; Brant and Loker, 2009; Jouet et al., 2009,

2010; Rizevsky et al., 2011; Christiansen et al., 2014; Lawton et al., 2014).

But even in these well-studied areas, the presence and dynamics of particular

Trichobilharzia spp. in other bird hosts and in other countries that share those bird hosts remain unknown. For example, are Trichobilharzia spp. shared across continents or does each continent have its own species, or both?

Brant and Loker (2009) investigated host-parasite associations among Trichobilharzia spp. within North America and found that one species, Trichobilharzia querquedulae (McLeod, 1937), has a clear association with a globally distributed clade of dabbling ducks (the ‘blue- wing’ group) that includes the holarctically distributed northern shovelers

(Anas clypeata); the cinnamon (Anas cyanoptera) and blue-wing teal (Anas discors) as well as several Southern Hemisphere endemic species

(Johnson and Sorenson, 1999). Species in the ‘blue-wing’ group share specific ecological preferences for shallow freshwater bodies across both their summer and winter grounds (Baldassarre et al., 1996; Kear, 2005).

Additionally, T. querquedulae infects Physa spp. snails, which are ubiquitous in North America and one common species, Physa acuta

(Draparnaud, 1805), is globally invasive (Bousset et al., 2014). Until recently, T. querquedulae had been reported from only the North American

‘blue-wing’ species (A. clypeata, A. cyanoptera and A. discors). Brant and

Loker (2009) sampled a phylogenetically diverse range of ducks (n = 299 individuals) within North America and found that T. querquedulae occurred

15 with a prevalence greater than 95%, exclusively within ‘blue-wing’ ducks.

Endemic, ecologically similar species of the ‘blue-wing’ group also occur in the Southern Hemisphere (Kear, 2005), but heretofore, records for T. querquedulae or other schistosome species are largely lacking for these species. Is the association between T. querquedulae and members of the

‘blue-wing’ group consistent across hemispheres, and if so, what factors contribute to the global maintenance of this association?

MATERIALS AND METHODS

Parasite collection

Ducks were collected from several locations in both the Northern and

Southern Hemispheres (Table 1); North America (United States and Canada),

South America (Argentina, 30° 5' 56.4'' S, 59° 29' 56.4'' W), Africa (South

Africa, 29° 12' 28.8'' S, 27° 11' 20.4'' E) and New Zealand (South Island, 44°

21' 3.6'' S, 170° 12' 32.4'' E). For the purposes of this paper, reasonable sampling of hosts and localities from Europe is represented in GenBank. The mesenteric and hepatic portal veins of ducks and geese were examined for species of Trichobilharzia. Mesenteric veins were inspected visually for adult worms, which were removed using Vanna’s micro scissors. Adult worms were removed from the hepatic portal vein and liver by either crushing and washing the liver in a series of decantation steps to isolate the worms or using a syringe to push saline through the hepatic portal vein to collect any worms that washed out. Schistosome samples were preserved in 95% ethanol for genetic, and 80% ethanol for morphological analyses. Parasites were deposited in the Museum of Southwestern Biology, Division of Parasites,

University of New Mexico, USA (Table 1). All work with vertebrate hosts was conducted with the approval of the Institutional Animal Care and Use

Committee at the University of New Mexico (IACUC # 11-100553-MCC,

Animal Welfare Assurance # A4023-01).

Genetic analysis

DNA from the samples was extracted using a DNeasy Tissue Kit

(Qiagen, Valenicia, California, USA) following the manufacturer’s protocol or by HotShot Lysis (Truett et al., 2000). We targeted two mitochondrial gene regions and one nuclear region for PCR amplification (Takara Ex Taq kit, Takara Biomedicals, Otsu, Japan), following protocols used by Brant and Loker (2009). The mitochondrial cytochrome oxidase subunit I (Cox1) gene (743 bp region, 5’ end) (primers in Brant and Loker, 2009) and the

NADH Dehydrogenase subunit 4 (ND4) gene (409 bp in the central portion of the gene), using the ND4 primers TSND4_F4 5’ -

AGTCCTTATCCGGAGCGTTA-3’, TSND4_R4 5’-

AACCAGCAACACACAAAAACA-3’. The ND4 gene was targeted due to high intraspecific variation relative to Cox1 and internal transcribed spacer

1 (ITS1) (Blouin et al., 1995; Webster et al., 2007), with a goal of assessing intraspecific relationships across hemispheres and host species.

The nuclear gene region of ITS1 (733 bp, including the 3’ portion of 18S rRNA and the 5’ portion of 5.8S) was amplified and sequenced using primers from Dvorak et al. (2002). Sequencing reactions were performed using the BigDye sequencing kit 3.1 (Applied Biosystems, Foster City,

California, USA). Sequences were edited using Sequencher 5.3 (Gene

Codes Corporation, Ann Arbor, MI, USA), aligned using ClustalW (Larkin et al., 2007) and then modified manually. Sequences generated during this study were submitted to GenBank (Table 1).

Reconstructing evolutionary relationships

Phylogenetic analyses included the samples collected herein as well as the available T. querquedulae samples and appropriate outgroups for

Clade Q (sensu Brant and Loker, 2009; Table 1). Each gene was treated as independent and the best models of nucleotide substitution was chosen based on the Akaike information criterion (AIC; Akaike, 1974) using

JModelTest (Darriba et al., 2012) as follows: TN93+G for Cox1 (Tamura and

Nei, 1993), K2+G for ITS1 (Kimura, 1980), HKY + I for ND4 (Hasegawa,

Kishino and Yano, 1985). These specific models were incorporated into

Maximum Likelihood (ML) analyses.

Gene trees were analyzed independently using ML methods in the program PAUP* 4.0 (Swofford, D. L. 2003. PAUP*. Phylogenetic Analysis

Using Parsimony (*and Other Methods). Version 4. Sinauer Associates,

Sunderland, Massachusetts, USA). Five hundred bootstrap replicates were preformed within each gene tree analysis to statistically assess the resulting topologies.

Bayesian Inference (BI) was performed using the program MrBayes

3.2.1 (Ronquist and Huelsenbeck, 2003), consisting of two replicated runs for each locus with four Markov chain Monte Carlo (MCMC) chains, one cold and three heated chains. Each analysis ran for 2,000,000 generations and was sampled every 1000 generations. The analysis was terminated when the S.D.

18 of the split frequencies was or fell below 0.01, supporting convergence.

Likelihood parameters and convergence between runs were assessed using the program Tracer v.1.5 (Rambaut et al., 2014 Tracer v1.6. http://beast.bio.ed.ac.uk/Tracer). The first 25% of trees from each analysis was discarded as burnin. Resulting phylogenetic trees were visualized and manipulated using Fig Tree v. 1.4 (Rambaut and Drummond, 2009, FigTree v1. 3.1. http://tree.bio.ed.ac.uk/software/figtree) and MEGA 6 (Tamura et al.,

2013).

Genetic distances were estimated, based on uncorrected p-distances between all sequences for each locus, and then averaged within lineages for comparison (Table 2), using the program MEGA 6 (Tamura et al., 2013).

RESULTS

Parasite infection and molecular identification

From our collections of endemic ducks in the three countries in the

Southern Hemisphere, T. querquedulae was found to be globally distributed in ducks from the ‘blue-wing’ clade as well as in the sister clade of silver teal ducks. In all three countries sampled, endemic ‘blue-wing’ and allied species were infected with a prevalence of 95% or greater, similar to what has been reported from the United States (Table 2).

Monophyly of T. querquedulae

Molecular phylogenetic analyses (ML, BI) show strong statistical support for the monophyly of T. querquedulae populations, regardless of the hemisphere sampled (Figs. 1 - 3). The resultant phylogenies show similar

19 gene trees among nuclear and mitochondrial genes, suggesting that

Southern Hemisphere samples belong to T. querquedulae and likely do not have a hybrid origin.

This study provides sequence data for the mitochondrial ND4 gene region of T. querquedulae, which had more genetic variation (Fig. 3, Table 3) relative to Cox1 and ITS1. Gene tree analysis (ML, BI) shows a well- supported monophyly of T. querquedulae and a lack of phylogenetic resolution for internal nodes, concordant with the Cox1 and ITS1 gene trees

(Figs. 1, 2).

The low nodal support within the T. querquedulae clade, consistent with

what has been reported previously (Brant and Loker, 2009), is suggestive of

a lack of geographic structure and is consistent with gene flow across

hemispheres (Bouzid et al., 2008; Goulding and Cohen, 2014). Genetic

distances within and between continents and hemispheres support low

intraspecific divergence (Table 3). On average, for all genes analyzed,

genetic divergence within North America (approximately 2%) is comparable

with genetic divergence between Northern and Southern Hemisphere

populations (approximately 2.7%).

DISCUSSION

The survey results reported here provide the first known genetic evidence for an avian schistosome, T. querquedulae, to have a global distribution spanning three continents and New Zealand. Our sampling indicates that the duck host range of T. querquedulae extends in the

Southern Hemisphere both to endemic members of the ‘blue-wing’ group

(Cape and New Zealand shovelers), and to members of a sister clade containing the silver teal (see Johnson and Sorenson, 1999). Southern

Hemisphere worms were recovered from endemic non-migratory species

(Kear, 2005), thus we can infer that the infections were acquired locally. For all loci analyzed, genetic divergence within and between Northern and

Southern Hemisphere individuals was low, suggesting their populations are connected by trans-hemispheric (specifically, between Northern and

Southern hemispheres) migration of particular duck species. The holarctically distributed northern shoveler (A. clypeata), for which annual migrations between the Northern and Southern Hemispheres is common

(Peters et al., 2014), is likely the primary host species facilitating gene flow across the range of T. querquedulae. It is also likely that populations of blue- wing and cinnamon teal (A. cyanoptera and A. discors) that breed in North

America and winter in the Southern Hemisphere facilitate T. querquedulae transmission, as they are seasonally sympatric with the endemic ‘blue- winged’ ducks and their allies (silver teal) (Wilson et al., 2011). Estimated genetic distances (Table 3) indicate that across its range T. querquedulae populations are connected and genetically homogenized, although increased population sampling is necessary to fully elucidate phylogeographic patterns.

More variable genetic markers, increased sampling and population genetic analyses are required to determine how much gene flow occurs across the hemispheres and the relative importance of different duck species in maintaining the global transmission of T. querquedulae.

Through targeted collections of Southern Hemisphere endemic host species, we find evidence to support the phylogenetic association between

T. querquedulae and the ‘blue-wing’ group uncovered by Brant and Loker

(2009) (host associations are summarized in Fig. 4). Our data suggest that the duck host range of T. querquedulae includes, in addition to the ‘blue- wing’ group, the sister clade that includes the silver teal. It is clear that the definitive hosts of T. querquedulae are not a random sample of Anas duck species and that definitive host phylogeny is important in life cycle maintenance. Below we discuss several aspects of the host and parasites biology which could maintain the observed phylogenetic association among

T. querquedulae and their specific duck hosts.

It is possible that these patterns are maintained as a consequence of specialization that has occurred over an evolutionary time frame based on host physiology or immune competence. It has been generally thought that definitive host specificity is low within species of Trichobilharzia (Blair and

Islam, 1983; Horak et al., 2002), however with increased taxonomic sampling of hosts and the inclusion of molecular data, patterns of duck host specificity are becoming increasingly apparent in some Trichobilharzia spp. (Brant and

Loker, 2009, 2013a; Jouet et al., 2015). Similar phylogenetic affinities are known to occur within other avian schistosome genera. For example,

Allobilharzia (Kolářová et al., 2006) in swans and Anserobilhariza (Brant et al., 2013b) in geese, are just two examples where avian schistosome transmission shows some phylogenetic constraints at the definitive host level. The physiological, genetic and/or immunological barriers mediating

Trichobilharzia infections within duck hosts is currently unknown (Horak et al., 2002) and, as suggested by the present results, in need of experimental study. Experimentally, non-‘blue wing’ ducks such as mallards (McLeod and

Little, 1942) can support egg-laying infections of T. querquedulae. However, natural infections of non-‘blue wing’ hosts appear to be rare, and our surveys over the last 10 years have recovered a single infection of an immature worm fragment within an American wigeon (Anas americana) in North

America (S.V. Brant, unpublished data).

It is likely that host ecological preferences, which are also influenced by the hosts’ evolutionary history, play a key role in the persistence of the association of T. querquedulae in one clade of ducks versus another. ‘Blue- wing’ ducks are ecological specialists (Baldassarre et al., 1996), with a preference for shallow marshy habitats. It can be hypothesized that shared habitat preferences specifically relating to nesting site selection and natal habitats can act as an encounter filter (Combes, 2001) among ‘blue-wing’ and allied species to restrain T. querquedulae host use. Nesting sites, where the majority of Trichobilharzia transmission is likely occurring (Rau et al.,

1975; Brant and Loker, 2009), are non-randomly chosen (Clark and Shutler,

1999). Blums et al. (2002) showed that over a 23-year period female northern shovelers (A. clypeata) exhibited greater than 88% nest site fidelity, significantly more so than other sympatric duck species measured. Further, northern shovelers and other ‘blue-wing’ species are known to select fragmented small water bodies similar to wetland islands for nesting (Blums et al., 2002), where the pair will live and breed. Males are known to guard these ‘islands’ and exhibit territorial behavior (Poston, 1974), so much so that availability of nesting sites limits population sizes (Vickery and Nudds,

1984). Generally, single to a few pairs occur at individual nesting sites, in contrast to other dabbling and diving duck species (Vickery and Nudds,

1984), which may have more concentrated nesting sites and may be sympatric with several different species of birds. Site and habitat fidelity exhibited by the duck host likely results in stable transmission dynamics for

T. querquedulae. This also assumes that the intermediate host is present and that at least peak cercarial shedding of T. querquedulae occurs during this time as well, when the shovelers and teal are isolated for nesting and raising of ducklings.

It could also be the case that the potential for temporal isolation is a mechanism for the maintenance of this global host-parasite association.

Members of the ‘blue-wing’ group are the first duck species to migrate in both the spring and the winter (Arzel et al., 2006), thus it would follow that their ducklings may be the first to encounter schistosome cercariae in the spring. This hypothesis may hinge on the importance of overwintering infected snails in sustaining a majority of the transmission (Crews and Esch,

1986; Goater et al., 1989). This idea is plausible for a snail host of T. querquedulae, Physa gyrina, as it is bivoltine, with overwintering snails producing an early spring generation (Brown, 1979), and it is likely an abundant permissive snail host within the duck host natal habitats (Pip,

1987). Data from this study support the hypothesis that migrating birds, in particular northern shovelers, transfer infections among continents suggesting that transmission may not be strictly confined to natal habitats.

However, habitat requirements by this particular clade of ducks would be the same among species. Nonetheless, these stimulating ideas require empirical studies documenting the dynamics of transmission at relevant spatial and temporal scales.

It is clear that for T. querquedulae to persist in the Southern

Hemisphere, suitable snail intermediate hosts must be present. Although

the identity of the intermediate host(s) for T. querquedulae in the Southern

Hemisphere remains unknown, several possibilities exist, some of which

can help account for the broad geographic range and even definitive host

associations observed. In North America, Physa spp. are confirmed hosts

(McLeod and Little, 1942; Brant and Loker, 2009; Brant et al., 2011).

Physid diversity is highest in North America (Wethingtion, 2007), but

several species are native to, or have invaded, the Southern Hemisphere.

South America harbors several endemic physids, which could sustain T.

querquedulae transmission, for example, Physa marmorata (Guilding,

1828), which has now invaded parts of South Africa (Appleton, 2003).

Physa acuta (Draparnaud, 1805), is a well-documented global invader

that has established in much of the Southern Hemisphere (Bousset et al.,

2014) and may well serve as a host for T. querquedulae. Physid snails

commonly colonize the shallow freshwater habitats favored by ‘blue wing’

duck species.

Generally, Trichobilharzia spp. are specific to a single snail species or a group of closely-related congeners (Kock, 2001; Rudolfova et al., 2005;

Brant and Loker, 2009; Jouet et al., 2009), and experimental and genetic confirmation that a species of Trichobilharzia can infect multiple genera of snails within a family has yet to be achieved. There is a report of an avian schistosome Dendritobilhariza pulverulenta (Cheatum, 1941) found in both

Gyraulus (North America, New Zealand) and Anisus (Europe) snails but the genetic identity of the worms from Anisus have not been confirmed as D.

25 pulverulenta, since there exists more than one species (Khalifa, 1976;

Vusse, 1980; Brant et al., 2011) and Dendritobilhariza loossi has been found experimentally to use Anisus vortex in Europe (Akramova et al., 2011). Snail surveys carried out in Europe (Rudolfova et al., 2005; Jouet et al., 2009b) and Africa (Appleton and Brackenbury, 1998; Laamrani et al., 2005), have failed to recover any Trichobilharzia spp. from native or invasive physids.

Two studies report schistosome cercariae from other physids in Europe

(Aplexa hypnorum, Gerard, 2004; Physa fontinalis, Rudolfova et al., 2005), but neither schistosome belongs to the genus Trichobilharzia. It should be noted, however, that even within North America the prevalence of T. querquedulae in Physa spp. is very low and snail surveys for trematodes frequently fail to recover T. querquedulae and other Trichobilharzia spp. (Loy and Haas, 2001) from their presumptive snail hosts. However, survey collections of physids at times of the year when these duck host have ducklings are lacking, and that might be peak time for cercarial shedding.

It is even less common to find a genetically confirmed schistosome conspecific infecting two different families of snails (Blair et al., 2001).

However one species, Trichobilharzia jequitibaensis (Leite et al., 1978), has been reported to infect both Physidae and Lymnaeidae. Miracidia obtained from naturally infected domestic Muscovy ducks (Cairina moschata domestica) successfully infected both P. marmorata and Lymnaea columella under experimental conditions. If such a situation were confirmed and shown to be true for T. querquedulae, then its broad geographic range would be more explicable since lymnaeid species diversity is much greater in the Eastern

Hemisphere than physid species diversity. It will be interesting to pursue and

26 verify these intriguing results with further experimentation, field collections and sequence analyses. These details of both definitive and intermediate host associations across a parasite’s range are necessary to understand transmission dynamics as well as to model zoonotic cycles.

The example of T. querquedulae globally distributed within a rather narrowly defined clade of ducks is instructive for those interested in the epidemiology of HCD, and more broadly the transmission of waterfowl zoonoses. The trans-hemispheric occurrence of T. querquedulae is certainly aided both by the migratory behavior (trans-hemispheric migration of the northern shoveler) and habitat preferences of definitive hosts as well as habitat preferences of the snail hosts. Perhaps the shared host ecology of duck hosts (habitat preference, nest site fidelity and timing of migration) across both the Northern and Southern Hemispheres sustain transmission and perpetuate the clear phylogenetic association between T. querquedulae and the ‘blue-wing’ group. Further research is necessary to partition the relative roles of host ecology and phylogeny in shaping the transmission of T. querquedulae over evolutionary time. For example, experimental infections evaluating permissiveness of non ‘blue-wing’ duck species and other snail host species would be a logical next step providing data vital to model transmission dynamics. The greater role duck host ecology plays, the more dynamic one would expect host associations and transmission to be in response to ecological change over evolutionary time, as opposed to a strict specialist model. Human-mediated habitat alteration and climate change have been implicated in affecting the transmission of other waterfowl zoonoses (Reed et al., 2003; Fuller et al., 2012; Dijk et al., 2014) and are

27 likely to alter Trichobilharzia host associations over time as well. Together with the fact that avian schistosomes of other genera infect different bird and snail lineages, we must remember that HCD is caused by a number of schistosome species, with markedly different patterns of host use, posing a challenge for those interested in understanding the epidemiology of HCD.

Thus future studies on the specific ecological relationships among hosts and worms, and on how permissive other species of ducks and snails are for T. querquedulae will be an exciting path forward to both understanding parasite- host biology as well as the ability to more specifically model zoonotic disease emergence.

ACKNOWLEDGMENTS

Thank you to the following for invaluable help collecting birds—South

Africa: Museum of Southwestern Biology Bird Division: Mr. Andy Johnson,

Dr. Ernest Valdez, Dr. Robert W. Dickerman and Dr. Chris Witt; National

Museum Bloemfontein: Mr. Rick Nuttall and Mr. Dawie H. de Swardt. We acknowledge technical support from the University of New Mexico Molecular

Biology Facility, which is supported by National Institutes of Health, USA grant P30GM110907 from the Institute Development Award program of the

National Center for Research Resources, USA. This work was primarily funded through a National Science Foundation, USA grant to S.V.B. (DEB

1021427) and a National Institutes of Health grant RO1 AI44913 to ESL. We thank the following funding agencies; Agencia de Promoción Científica y

Técnica, Argentina PICT 1288-2011 and CONICET, Argentina PIP No.:

11220110100550 to VF.

REFERENCES

Akaike, H. 1974. A new look at the statistical model identification. IEEE Trans. Automat. Cont. 19, 716-723.

Akramova, F.D., Azimov, D.A., Shakarboev, E.B. 2011. Morphology, biology and taxonomy of Dendritobilharzia loossi Skrjabin, 1924 (Trematoda: Bilharziellidae), a parasite of Pelecanus onocrotalus (Pelecanidae) and Anas platyrhynchos (Anatidae). Parasite. 18, 39-48.

Aldhoun, J.A., Kolárová, L., Horák, P., Skírnisson, K. 2009. Bird schistosome diversity in Iceland: molecular evidence. J. Helminthol. 83, 173–180.

Appleton, C.C., Brackenbury, T.D., Tonin, A.F.G. 1989. Physa mosambiquensis (Clessin,1886) rediscovered? South African J. Zool. 24, 340–344.

Appleton, C.C. 2003. Alien and invasive fresh water Gastropoda in South Africa. African J. Aquatic Sci. 28, 69-81.

Arzel, C., Elmberg, J., Guillemain, M. 2006. Ecology of spring-migrating Anatidae: a review. J. Ornithol. 147, 167-184.

Baldassarre, G. A., Bolen, E. G. 1994. Waterfowl ecology and management. John Willey & Sons. Inc. New York, USA.

Blair, D., Islam, K.S. 1983. The life cycle and morphology of Trichobilharzia australis n. sp. (Digenea: Schistosomatidae) from the nasal blood vessels of the black duck (Anas superciliosa) in Australia, with a review of the genus Trichobilharzia. Syst. Parasitol. 5, 89–117.

Blair, D., Davis, G. M., Wu, B. 2001. Evolutionary relationships between trematodes and snails emphasizing schistosomes and paragonimids. Parasitology. 123, 229-243.

Blouin, M. S., Yowell, C. A., Courtney, C. H., Dame, J. B. 1995. Host movement and the genetic structure of populations of parasitic nematodes. Genetics. 141, 1007-1014.

Blums, P., Nichols, J. D., Hines, J. E., Mednis, A. 2002. Sources of variation in survival and breeding site fidelity in three species of European ducks. J. Animal Ecology. 71, 438-450.

Bousset, L., Pointier, J. P., David, P., Jarne, P. 2014. Neither variation loss, nor change in selfing rate is associated with the worldwide invasion of Physa acuta from its native North America. Biol. Invasions. 16, 1769-1783.

Bouzid, W., Štefka, J., Hypša, V., Lek, S., Scholz, T., Legal, L., Kalthoum, O., Hassine, B., Loot, G. 2008. Geography and host specificity: two forces behind the genetic structure of the freshwater fish parasite Ligula intestinalis (Cestoda: Diphyllobothriidae). Int. J. Parasitol. 38, 1465-1479.

Brant, S.V., Loker, E.S. 2009. Molecular systematics of the avian schistosome genus Trichobilharzia (Trematoda: Schistosomatidae) in North America. J. Parasitol. 95, 941–963.

Brant, S. V., Bochte, C. A., Loker, E. S. 2011. New intermediate host records for the avian schistosomes Dendritobilharzia pulverulenta, Gigantobilharzia huronensis, and Trichobilharzia querquedulae from North America. J. Parasitol. 97, 946-949.

Brant, S. V., Loker, E. S. 2013a. Discovery-based studies of schistosome diversity stimulate new hypotheses about parasite biology. Trends Parasitol. 29, 449- 459.

Brant, S. V., Jouet, D., Ferte, H., Loker, E. S. 2013b. Anserobilharzia gen. n. (Digenea, Schistosomatidae) and redescription of A. brantae (Farr & Blankemeyer, 1956) comb. n. (syn. Trichobilharzia brantae), a parasite of geese (Anseriformes). Zootaxa. 3670, 193-206.

Cheatum, E.L. 1941. Dendritobilharzia anatinarum n. sp., a blood fluke from the mallard. J. Parasitol. 27, 165-170.

Christiansen, A. Ø., Olsen, A., Buchmann, K., Kania, P. W., Nejsum, P., Vennervald, B. J. 2014. Molecular diversity of avian schistosomes in Danish freshwater snails. Parasitol. Res. 1-11.

Clark, R.G., Shuter, D. 1999. Avian habitat selection: Pattern from process in nest- site use by ducks? Ecology. 80, 272-287.

Crews, A.E., Esch, G.W. 1986. Seasonal dynamics of Halipegus occidualis (Trematoda: Hemiuridae) in Helisoma anceps and its impact on fecundity of the snail host. J. Parasitol. 72, 646-651.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nature Meth. 9, 772. de Gentile, L., Picot, H., Bourdeau, P., Bardet, R., Kerjan, A., Piriou, M., LeGuennic, A., Bayssade-Dufour, C., Chabasse, D., Mott, K.E. 1996. Le dermatite cercarienne en Europe: un probleme de la sante´publique nouveau? Bulletin de l’Organisation Mondiale de la Sante´. 74, 159–163.

Dijk, J. G., Hoye, B. J., Verhagen, J. H., Nolet, B. A., Fouchier, R. A., Klaassen, M. 2014. Juveniles and migrants as drivers for seasonal epizootics of avian influenza virus. J. Anim. Ecol. 83, 266-275.

Draparnaud, J. P. R. 1805. Histoire naturelle des mollusques terrestres et fluviatiles de la France. D Colas, Paris, France.

Dvorak, J., Vanácová, Š., Hampl, V., Flegr, J., Horak, P. 2002. Comparison of European Trichobilharzia species based on ITS1 and ITS2 sequences. Parasitology. 124, 307-313.

Ferté, H., Depaquit, J., Carré, S., Villena, I., Léger, N. 2005. Presence of Trichobilharzia szidati in Lymnaea stagnalis and T. franki in Radix auricularia in northeastern France: molecular evidence. Parasitol. Res. 95, 150-154.

Fuller, T., Bensch, S., Muller, I. Novembre, J., Perez-Tris, J., Rickfels, R.E., Smith, T.B., Waldenstrom, J. 2012. The ecology of emerging infectious diseases in migratory birds: an assessment of the role of climate change and priorities for future research. Ecohealth. 9, 80-88.

Garvon, J. M., Fedynich, A. M., Peterson, M. J., Pence, D. B. 2011. Helminth community dynamics in populations of blue-winged teal (Anas discors) using two distinct migratory corridors. J. Parasitol. Res. 2011, Article ID 306257.

Gerard, C. 2004. First occurrence of Schistosomatidae infecting Aplexa hypnorum (Gastropoda, Physidae) in France. Parasite. 11, 231-234.

Goater, T.M., Shostak, A.W., Williams, J.A., Esch, G.W. 1989. A mark- recapture study of trematode parasitism in overwintered Helisoma anceps (Pulmonata), with special reference to Halipegus occidulis (Hemuridae). J. Parasitol. 75, 553-560. Goulding, T. C., Cohen, S.C. 2014. Phylogeography of a marine acanthocephalan: lack of cryptic diversity in a cosmopolitan parasite of mole crabs. J. Biogeogr. 41, 965–976.

Hasegawa, M., Kishino, H., Yano, T. A. 1985. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J. Mol. Evo. 22, 160-174.

Hoberg, E. P., Brooks, D. R. 2008. A macroevolutionary mosaic: episodic host‐ switching, geographical colonization and diversification in complex host–parasite systems. J. Biogeography. 35, 1533-1550.

Horák, P., Kolářová, L., Adema, C. M. 2002. Biology of the schistosome genus Trichobilharzia. Adv. Parasitol. 52, 155-233.

Horák, P., Kolářová, L. 2011. Snails, waterfowl and cercarial dermatitis. Freshwater Biol. 56, 779-790.

Horák, P., Mikes, L., Lichtenbergova, L., Skala, V. Soldanova, M., Brant, S.V. 2015. Avian schistosomes and outbreaks of Cercarial Dermatitis. Clin Microbiol. Rev. 28. 165-190.

Johnson, K.J., Sorenson, M.D. 1999. Phylogeny and biogeography of the dabbling ducks (Genus: Anas): A comparison of molecular and morphological evidence. The Auk. 116, 792-805.

Jouet, D., Ferté, H., Hologne, D., Kaltenback, M.L., Depaquit, J. 2009. Avian schistosomes in French aquatic birds: a molecular approach. J. Helminthol. 83, 181–189.

Jouet, D., Skirnisson, K., Kolarova, L., Ferte, H. 2010. Final hosts and variability of Trichobilharzia regenti under natural conditions. Parasitol. Res. 107, 923-930.

Jouet, D., Kolarova, L., Patrelle, C., Ferte, H., Skirnisson, K. 2015. Trichobilharzia anseri n. sp. (Schistosomatidae: Digenea), a new visceral species of avain schistosomes isolated from graylag goose (Anser anser L.) in Iceland and France. Inf. Gen. Evol. 34, 298-306.

Kear, J. 2005. Ducks, Geese and Swans: Species Accounts (Cairina to Mergus). Oxford University Press Inc. New York, USA.

Kimura, M. 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16, 111-120.

Kock, S. 2001. Investigations of intermediate host specificity help to elucidate the taxonomic status of Trichobilharzia ocellata (Digenea: Schistosomatidae). Parasitology. 123, 67-70.

Kolářová, L., Rudolfová, J., Hampl, V., Skírnisson, K. 2006. Allobilharzia visceralis gen. nov., sp. nov.(Schistosomatidae-Trematoda) from Cygnus cygnus (L.)(Anatidae). Parasitol. Int. 55, 179-186.

Kolářová, L., Horák, P., Skírnisson, K. 2010. Methodical approaches in the identification of areas with a potential risk of infection by bird schistosomes causing cercarial dermatitis. J. Helminthol. 84, 327-335.

Kolářová, L., Horák, P., Skírnisson, K., Marečková, H., Doenhoff, M. 2013. Cercarial dermatitis, a neglected allergic disease. Clin Rev Allergy Immunol. 45, 63-74.

Kourilova´, P., Hogg, K.G., Kolářová, L., Mountford, A.P. 2004. Cercarial dermatitis caused by bird schistosomes comprises both immediate and late phase cutaneous hypersensitivity reaction. J. Immunol. 172, 3766–3774.

Laamrani, H., Boelee, E., Madsen, H. 2005. Trematode infection among freshwater gastropods in Tessaout Amont irrigation system, Morocco. African Zool. 40, 77-82.

Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A., McWilliam, H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson,

J.D., Gibson, T.J., Higgins D.G. 2007. ClustalW and ClustalX version 2. Bioinformatics. 23, 2947-2948.

Larsen, A.H., Bresciani, J., Buchmann, K. 2004. Increasing frequency of cercarial dermatitis at higher latitudes. Acta Parasitologica. 49, 217–221.

Lawton, S. P., Lim, R. M., Dukes, J. P., Cook, R. T., Walker, A. J., Kirk, R. S. 2014. Identification of a major causative agent of human cercarial dermatitis, Trichobilharzia franki (Müller and Kimmig 1994), in southern England and its evolutionary relationships with other European populations. Parasites & Vectors. 7, 277.

Leite, A. C. R., de Araujo Costa, H. M., Costa, J. O. 1978. Trichobilharzia jequitibaensis sp. n. (Trematoda, Schistosomatidae) in Cairina moschata domestica (Anatidae). Revista Brazileira de Biologia. 38, 843-846.

Loy, C., Haas, W. 2001. Prevalence of cercariae from Lymnaea stagnalis snails in a pond system in Southern Germany. Parasitol. Res. 87, 878-882.

McKenzie, V.J. 2007. Human land use and patterns of parasitism in tropical amphibian hosts. Biol. Conserv. 137,102-116.

McLeod, J.A. 1937. Two new schistosomid trematodes from water birds. J. Parasitol. 23, 456-466.

McLeod, J. A., Little, G. E. 1942. Continued studies on Cercarial Dermatitis and the trematode family Schistosomatidae in Manitoba: Part I. Canadian J. Res. 20, 170-181.

Ostfeld, R.S., Keesing, F. 2000. Biodiversity and disease risk: the case of Lyme disease. Conservation Biol. 14, 722–728.

Peters, J. L., Winker, K., Millam, K. C., Lavretsky, P., Kulikova, I., Wilson, R. E., Zhuravlev, Y.N., McCracken, K. G. 2014. Mito‐ nuclear discord in six congeneric lineages of Holarctic ducks (genus Anas). Mol. Ecol. 23, 2961- 2974.

Poston, H.J. 1974. Home range and breeding biology of the Shoveler. Canadian Wildlife Serv. Rep. Ser. 25, 1–42.

Rau, M. E., Bourns, T. K. R., Ellis, J. C. 1975. Egg production by Trichobilharzia ocellata (Trematoda: Schistosomatidae) after initial and challenge infection in ducks. Canadian J. Zool. 53, 642-650.

Reed, K. D., Meece, J. K., Henkel, J. S., Shukla, S. K. 2003. Birds, migration and emerging zoonoses: West Nile virus, Lyme disease, influenza A and enteropathogens. Clin Med Res. 1, 5-12

Rizevsky, S. V., Cherviakovsky, E. M., Kurchenko, V. P. 2011. Molecular taxonomic identification of Schistosomatidae from Naroch Lake and Polonevichi Lake in Belarus. Biochem Sys Ecol. 39, 14-21.

Ronquist, F, Huelsenbeck, J.P. 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 19, 1572-1574.

Rudolfová, J., Hampl, V., Bayssade-Dufour, C., Lockyer, A. E., Littlewood, D. T. J., Horák, P. 2005. Validity reassessment of Trichobilharzia species using Lymnaea stagnalis as the intermediate host. Parasitol. Res. 95, 79-89.

Rudolfová, J., Littlewood, D. T. J., Sitka, J., Horák, P. 2007. Bird schistosomes of wildfowl in the Czech Republic and Poland. Folia Parasitologica 54, 88-93.

Skrjabin, K.I. and Zakharov, N.P. 1920. Zweineue Trematodengattugen aus den Blutgefassen der Volgel. Izvestnik Donskovo Veterinarnovo Instituta 2, 1-6.

Soldánová, M., Selbach, C., Kalbe, M., Kostadinova, A., Sures, B. 2013. Swimmer's itch: etiology, impact, and risk factors in Europe. Trends in Parasitol. 29, 65- 74.

Tamura, K., Nei, M. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol Evol. 10, 512-526.

Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S. 2013. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol. Bio. Evo. 30, 2725- 2729.

Taylor, L.H., Latham, S.M., Woolhouse, M.E.J. 2001 Risk factors for human disease emergence. Phil. Trans. R Soc. Lond. Biol Sci. 356, 983–989.

Thompson, R. A. 2000. Giardiasis as a re-emerging infectious disease and its zoonotic potential. Int. J. Parasitol. 30, 1259-1267.

Truett, G. E., Heeger, P., Mynatt, R. L., Truett, A. A., Walker, J. A., Warman, M. L. 2000. Preparation of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (Hot- Shot). Biotechniques. 29, 52-54.

Valdovinos, C., Balboa, C. 2008. Cercarial dermatitis and lake eutrophication in South- central Chile. Epidemiol. Infect. 136, 391–394.

Vickery, W.L. Nudds, T.D. 1984. Detection of density dependent effects in annual duck censuses. Ecology. 65, 96–104.

Vusse, F. J. V. 1980. A review of the genus Dendritobilharzia Skrjabin and Zakharow 1920 (Trematoda: Schistosomatidae). J. Parasitol. 1, 814-822.

Webster, B. L., Rudolfová, J., Horák, P., Littlewood, D. T. J. 2007. The complete mitochondrial genome of the bird schistosome Trichobilharzia regenti

(Platyhelminthes: Digenea), causative agent of cercarial dermatitis. J. Parasitol. 93, 553-561.

Wilson, R. E., Eaton, M. D., Sonsthagen, S. A., Peters, J. L., Johnson, K. P., Simarra, B., McCracken, K. G. 2011. Speciation, subspecies divergence, and paraphyly in the Cinnamon Teal and Blue-winged Teal. The Condor. 113, 747-761.

Figure 1. Maximum likelihood (ML) tree based on mitochondrial cytochrome oxidase subunit I (cox1) sequence data. Branch support values less than 50% were not shown. Values are reported as bootstrap support from ML analysis/ posterior probabilities from Bayesian analysis. The shaded box includes all Trichobilharzia querquedulae individuals analyzed, bolded taxa indicate samples collected from the

Southern Hemisphere. Accession numbers of other analyzed Trichobilharzia samples can be found in Table 1.

Figure. 2. Maximum likelihood (ML) tree based on internal transcribed spacer region

1 (ITS1) sequence data. Branch support values less than 50% are not shown; values are reported as bootstrap support from ML analysis/posterior probabilities from

Bayesian analysis. The shaded box includes all Trichobilharzia querquedulae individuals analyzed, bolded taxa indicate samples collected from the Southern

Hemisphere. Accession numbers of other analyzed Trichobilharzia samples can be found in Table 1.

Figure. 3. Maximum likelihood (ML) tree based on NADH Dehydrogenase subunit 4

(ND4) sequence data. Branch support values less than 50% were not shown; values are reported as bootstrap support from ML analysis/posterior probabilities from

Bayesian analysis. The shaded box includes all Trichobilharzia querquedulae individuals analyzed, bolded taxa indicate samples collected from the Southern

Hemisphere. Accession numbers of other analyzed Trichobilharzia samples analyzed can be found in Table 1.

Figure. 4. Simplified Anas phylogeny reconstructed from Johnson and Sorensen

(1999). (A) Tree depicts phylogenetic relationships within Anas. Clades are collapsed to include: pintails (A. acuta, A. georgica, A. bahamensis, A. erythrorhyncha, A. capensis), green-winged teals (A. flavirostris, A. carolinensis, A. crecca), Mallards (A. laysanensis, A. luzonica, A. platyrhynchos, A. poecilorhyncha,

A. zonorhyncha, A. diazi, A. rubripes, A. fulvigula, A. superciliosa, A. melleri, A. undulata, A. sparsa), gray teals (A. gibberifrons, A. castanea, A. bernieri), Brown teals (A. aucklandica, A. chlorotis), Wigeons (A. americana, A. sibilatrix, A. penelope,

A. strepera, A. falcata), silver teals (A. versicolor, A. puna, A. hottentota, A. querquedula), blue-winged teals (A. smithii, A. rhynchotis, A. clypeata, A. cyanoptera, A. discors, A. platalea), South American ducks (Speculanas specularis,

Amazonetta brasiliensis, Tachyeres pteneres, Lophonetta specularioides), outgroup

(Marmaronetta, Pteronetta, Cyanochen, Aythya, Asarcornis, Chenonetta, Callonetta,

Tadorna,Cairina, Aix, Sarkidiorni). (B) Tree depicts interspecific relationships within the ‘blue-winged’ teal clade (shaded in A). Host taxa are labeled by infection status with Trichobilharzia querquedulae ( experimental host (McLeod and Little, 1942);

☐, host not examined; , host examined/positive; , host examined/negative). All survey data summarized was performed by the authors with the exception of A. querquedulae (Kolarova et al., 1997).



Gene

Species Host Locality CO1 ND4 ITS MSB # Reference

Trichobilharzia W135 Anas clypeata LA FJ174497 KU057197 FJ174557 183 * querquedulae

W137 T. querquedulae A. discors LA FJ174498 - FJ174558 19497 *

W142 T. querquedulae A. discors LA - KU057196 KP788761 19502 *

W148.1 T. querquedulae A. cyanoptera NM FJ174499 KU057194 18584 * FJ174559

W148.2 T. querquedulae A. cyanoptera NM FJ174500 - - 18584 *

W153 T. querquedulae A. clypeata NM - KU057192 KP788763 18587 *

W155 T. querquedulae A. cyanoptera NM - - FJ174553 18589 *

W155.2 T. querquedulae A. cyanoptera NM - KU057200 - 18589 *

W155.3 T. querquedulae A. cyanoptera NM FJ174501 - KP788764 18589 *

W156 T. querquedulae A. discors NM FJ174502 KU057191 FJ174554 18590 *

W158 T. querquedulae A. clypeata NM FJ174503 - FJ174549 18592 *

W160 T. querquedulae A. discors NM - KU057201 KP788766 18594 *

W162 T. querquedulae A. clypeata NM FJ174504 KU057190 FJ174551 18602 *

W167 T. querquedulae A. discors NM - KU057202 - - *

W180 T. querquedulae A. cyanoptera CA FJ174505 - FJ174556 18573 *

W183 T. querquedulae A. clypeata CA FJ174506 KU057189 FJ174560 18575 *

W190 T. querquedulae A. discors CA FJ174507 - FJ174550 18582 *

W203 T. querquedulae A. clypeata AK FJ174508 - FJ174552 18636 *

W345 T. querquedulae A. clypeata MB FJ174509 KU057185 FJ174547 18626 *

W650 T. querquedulae A. smithii ZA - KU057205 KP788765 18990 This study

W664 T. querquedulae A. smithii ZA KU057180 - KP788762 19000 This study

W703 T. querquedulae A. rhynchotis NZ KU057181 KU057204 - 20792 This study

W704 T. querquedulae A. rhynchotis NZ KU057182 KU057203 - 20793 This study

Tshov T. querquedulae A. rhynchotis NZ KU057183 - KP788760 20794 This study

W833 T. querquedulae A. versicolor AR KU057184 KU057206 - - This study

E45 T. querquedulae A. discors FL FJ174510 - FJ174555 - *

E45.2 T. querquedulae A. discors FL - KU057199 - - *

E64 T. querquedulae A. discors FL KU057198 - - * FJ174511

SDS1006 T. querquedulae A. clypeata NE - - FJ174548 - *

Trichobilharzia sp. W206.2 A. americana AK - - KP788771 - * A

Trichobilharzia sp. W210 A. americana AK - - KP788772 - * A

Trichobilharzia sp. W213 A. americana AK FJ174527 - FJ174570 18636 * A

Trichobilharzia sp. W149 A. americana NM - KU057193 - 18585 * A

Trichobilharzia sp. W205 A. americana AK FJ174528 - KP788770 18638 * B

Lophodytes Trichobilharzia sp. W173 PA FJ174529 - FJ174576 - * C cucullatus

Trichobilharzia W175 A.platyrhynchos PA - - KP788768 - * physellae

Trichobilharzia W196 Aythya affinis NM - - KP788767 - * physellae

W263 T. physellae Physa gyrina NM FJ174523 * FJ174562 178 *

Ferte Trichobilharzia FR - - AY795572 - et al. 2005 RSFO1 franki Radix auricularia

Ferte RSFO3 T. franki R. auricularia FR HM131198 - AY795573 - et al. 2005

Lawton et HamRa6 T. franki R. auricularia GB - - - KJ775868 al. 2014

Lawton et al, HamRa7 T. franki R. auricularia GB - - - KJ775869 2014

Trichobilharzia Jouet et al. CYA18 Cygnus olor FR HM439500 - - regenti HM439497 2010

Trichobilharzia Webster et DQ859919 Radix peregra - GeneID: 5333425 - regenti - al. 2007

Table 1: Host association and locality origin of the specimens used in this study. *Brant and Loker 2009. Locality abbreviations: LA = Louisiana USA, NM = New Mexico USA, CA = California USA, AK = Alaska USA, FL = Florida USA, NE = Nebraska USA, PA = Pennsylvania USA, FR = France, GB = England United Kingdom, MB = Manitoba Canada, ZA = Freestate South Africa, NZ = South Island New Zealand, AR = Corrientes Argentina. All samples collected for this study have been deposited in the Museum of Southwestern Biology (MSB) Division of Parasites at the University of New Mexico; parasite and hosts records (including GPS coordinates) can be accessed via the Arctos database

Species Location # Examined # Infected Reference Anas clypeata North America 22 20 Brant and Loker 2009 Anas discors North America 20 20 Brant and Loker 2009 Anas discors North America 184 175 Garvon et al. 2011 Anas cyanoptera North America 12 12 Brant and Loker 2009 Anas smithii South Africa 2 2 This Study Anas hottentota Kenya 2 0 This Study Anas rhynchotis New Zealand 3 3 This Study Anas versicolor Argentina 3 3 This Study Anas platalea Argentina 1 0 This Study

Table 2: Prevalence of T. querquedulae within members of the ‘blue-wing’ group and their allies.



Gene

Species CO1 ND4 ITS

0.01 0.012 0.001 T. querquedulae within NA

T. querquedulae within SH 0.012 0.017 0.004

T. querquedulae between NA 0.012 0.015 0.003 and SH

T. querquedulae all sequences 0.011 0.012 0.002

T. querquedulae vs. T. franki 0.077 - 0.002

T. querquedulae vs. 0.082 0.11 0.002 Trichobilharzia sp. A

T. querquedulae vs. T. 0.082 0.13 0.002 physellae

Table 3. Genetic distances comparing ITS, CO1 and ND4 gene regions within and between global samples of Trichobilharzia querquedulae and between closely related Trichobilharzia spp. NA= North America, SH = Southern Hemisphere.



CHAPTER II

Phylogeography and genetics of the globally invasive snail Physa acuta Draparund 1805 (= Haitia acuta, Taylor 2003), and its potential as an intermediate host to larval digenetic trematodes.

AUTHORS: Erika T. Gendron (Ebbs), Eric S. Loker and Sara V. Brant

ABSTRACT

Background: Physa acuta is a globally invasive freshwater snail native to

North America. Prior studies have led to conflicting views of how P. acuta

populations are connected and genetic diversity is partitioned globally. This

study aims to characterize phylogeographic and population genetic structure

within the native range of P. acuta, elucidate its invasion history and assess

global patterns of genetic diversity. Further, using meta-analytic methods, we

test the ‘Enemy-Release hypothesis’ within the P. acuta – digenetic trematode

system. The ‘Enemy-Release hypothesis’ refers to the loss of native parasites

following establishment of their host within an invasive range. Population

genetic data is combined with surveys of trematode infections to map range-

wide trematode species richness associated with P. acuta, and to identify

relevant host-population parameters important in modeling host-parasite

invasion.

Results: Phylogenetic analyses using mtDNA uncovered two major clades (A

& B). Clade A occurs globally while clade B was only recovered from the

Western USA. All invasive populations sampled group within Clade A, where

multiple independent source populations were identified from across North

America. Significant population genetic structure was found within the native

range of P. acuta, with some evidence for contemporary geographic barriers

between western and eastern populations. Mito-nuclear discordance was

found suggesting historical isolation with secondary contact between the two

mitochondrial clades. Trematode species richness was found to differ

significantly between native and invasive populations, in concordance with the

‘Enemy-Release hypothesis’. Further, our data suggests a positive

relationship between nucleotide diversity of invasive populations and

trematode prevalence and richness.

Conclusions: This study includes a wider geographic sampling of P. acuta

within its native range that provides insight into phylogeographic and

population genetic structure, range-wide genetic diversity and estimation of

the invasion history. Meta-analysis of P. acuta – trematode surveys globally is

consistent with the ‘Enemy-Release hypothesis’. Additionally, results from this

study suggest that host demographic parameters, namely genetic diversity,

may play an essential role in how parasite communities assemble within

invasive host populations. This knowledge can be used to begin to construct a

framework to model host-parasite invasion dynamics over time.

INTRODUCTION

Invasive species are an important consequence of global change (Sax et al.,

2005), therefore understanding their origin and invasion history (Estoup and

Guillemaud, 2010), as well as how genetic variation is partitioned across their range

(Facon et al., 2008), is vital. Population genetics provides a powerful tool by which invasion processes can be revealed (Cristescu, 2015; Estoup and Guillemaud,

2010). Freshwater gastropods present an interesting case, as they are often

48 dispersed passively and distributed patchily (Jarne and Delay, 1991) in nature, but human activities (i.e. aquaria trade, boats) may greatly alter dispersal, population structure and consequently evolutionary dynamics (Facon et al., 2003; Jarne and

Delay, 1991).

One such invader is Physa acuta (Draparnaud, 1805)(= Haitia acuta, Taylor

2003), a globally invasive freshwater snail which is native to North America (Dillon et al., 2002; Lydeard et al., 2016), now occurs on all continents except Antarctica

(Bousset et al., 2014; Wethington and Lydeard, 2007). Remarkable reproductive plasticity has been reported within P. acuta, specifically in populations’ ability to alter the number of generations per year (Clampitt, 1970; Monsutti and Perrin, 1999), a characteristic that might contribute to the invasion success of P. acuta. Physa acuta is known to displace native gastropods to become the dominant species over very short periods of time (Appleton, 2003; Brackenbury and Appleton, 1993; Dobson,

2004). For example, doubling times as short as four weeks have been recorded from

European (invasive) P. acuta populations (Perrin, 1986).

The invasion history of P. acuta is muddled by two centuries of taxonomic confusion ( summarized by Wethington and Lydeard, 2007), due in large part to plasticity in shell phenotype and overestimation of nominal diversity. Consequently, the timing of invasion and the connectivity of native and invasive populations remain unclear. Interestingly, Physa acuta was first described from its invasive range in

France (Draparnaud, 1805) over two centuries ago and it was not until 12 years later that it was described by Say (1817) from Pennsylvania, USA, under a different name, Physa heterostropha. Nominal diversity of P. acuta-like snails reached at least eight species (Burch, 1989; Taylor, 2003; Te, 1978) prior to the inclusion of molecular genetic data (Pip and Franck, 2008; Wethington and Guralnick, 2004;

Wethington and Lydeard, 2007). Molecular genetic studies in addition to reproductive isolation experiments (Dillon et al., 2002; Dillon and Wethington, 2004) demonstrated the need to synonymize P. heterostropha (North America), P. virgata

(North America), P. integra (North America) and P. cubensis (Central and South

America) to a single species, Physa acuta (Draparnaud, 1805). Taylor (Taylor, 2003) placed P. acuta and closely allied species into the genus Haitia, based primarily on penial morphology, which has demonstrated phylogenetic utility (Taylor, 2003;

Wethington and Lydeard, 2007). However, current genetic studies support the use of penial morphology to define species or species groups rather than higher level taxonomy (Wethington and Lydeard, 2007), as such the use of Haitia has failed to become widely adopted. Molecular phylogenetic analyses have also supported that

P. acuta occurs globally (Wethington and Lydeard, 2007).

The vast distribution, invasiveness and ubiquity of P. acuta have garnered the interest of evolutionary biologists. Efforts have been made to elucidate the population genetic structure of P. acuta at both regional (Bousset et al., 2004; Dillon and Wethington, 2006; Escobar et al., 2007; Van Leeuwen et al., 2013) and global

(Bousset et al., 2014; 2004) scales. These studies reconstructed the genetic structure of P. acuta primarily in its invasive range in Europe, with limited sampling from native populations. Characterizing population structure within the native range is foundational to resolving invasion history (Ghabooli et al., 2011). Limited sampling from native populations limit possible hypotheses relative to a likely invasion history scenario and identification of source populations. Further, those prior studies have presented conflicting results of how genetic varation is partitioned globally, and specifically whether genetic diversity is homogenous among native and invasive populations (Bousset et al., 2014; 2004; Van Leeuwen et al., 2013). The primary

50 objective of this study was to address current gaps in our knowledge of the genetic distribution of P. acuta by chacterizing range-wide population genetic and phylogeographic patterns, and reconstruct its invasion history.

Apart from its remarkable distribution, P. acuta offers a unique lens to understand host-parasite invasion dynamics, which are complex and largely overlooked (Dunn, 2009; Kelly et al., 2009; Dunn and Hatcher, 2015). Physid snails are important intermediate hosts to digenetic trematodes (Platyhelminthes, Digenea) where they occur (Cichy et al., 2011; Gordy et al., 2016). More broadly, snails act as first and second intermediate host to a large proportion of trematodes, over 10,000 species (Cribb et al., 2003). Despite the several papers discussing parasites in invasive species, surprisingly little is known about how host population dynamics might influence the invasion processes of their associated parasites (Kelly et al.,

2009; Dunn and Hatcher, 2015). Following invasion, host species are expected to lose their associated trematode assemblages (Torchin et al., 2003) based on the

‘Enemy-Release hypothesis’ (Keane, 2002). This hypothesis posits that release of regulation by natural enemies contributes to the rapid establishment of invasive populations. It is also expected that invasive hosts may acquire a new indigenous parasite assemblage within their invasive range. What shapes the assembly of the new parasite community is unclear (Kelly et al., 2009). Here we measure P. acuta- trematode species richness globally to test for release of digenetic trematodes within invasive populations and secondarily test for correlations among trematode infections and host population genetic parameters to identify factors that may be relevant to host-parasite invasion, and discuss a framework to use these data to being to model invasion dynamics.

MATERIALS AND METHODS

Specimen collection

Physa acuta populations were sampled opportunistically, between 1998–

2015, to document the biodiversity of their trematode parasites. The specimens used in this study were collected from across North America (native range) as well as several invasive localities (locality details in Additional Files 1 & 2) and represent P. acuta over a 17-year period (Figure 1).

Snails were collected primarily from the waters edge using a mesh sieve or plucked from vegetation. The snails were first screened for trematode infection by leaving an individual snail in the well of a cell culture plate submerged in artificial spring water over night. Snails were considered infected upon visual conformation of mobile trematode larva (cercariae) via a dissection microscope. The cercariae and their snail hosts, as well as the uninfected snails, were isolated and stored in 95% ethanol and deposited into the Museum of Southwestern Biology (MSB), Division of

Parasites (Additional Files 1 & 2).

To improve geographic sampling and ensure we were working with a monophyletic species, specimens from this study were combined in phylogenetic analyses with known P. acuta samples published in the NCBI GenBank sequence database (Additional File 1). Sequences were published between 2004 –2016, from both native and invasive populations. 159 physid sequences were downloaded and included in this study.

DNA extraction and PCR amplification

A small piece of tissue was taken from the head foot of individual snails. DNA was extracted using the E.Z.N.A. Mollusc DNA kit (Omega Biotek) following the manufacture’s protocol.

We sequenced two mitochondrial DNA loci (cox1 and 16S) as they are the most abundant genetic marker available in the NCBI database and have been used successfully to recover population genetic variation within P. acuta (Wethington and

Guralnick, 2004; Wethington and Lydeard, 2007). We also sequenced a nuclear locus (ITS1) to assess mito-nuclear discordance of the recovered subclades.

We sequenced 509 bases of cox1 gene using the universal primers,

LCO11490: 5’-GGTCAACAAATCATAAAGATATTGG-3’ and LCO2198: 5’-

TAAACTTCAGGGTGACCAAAAAATCA-3’ (Folmer et al., 1994) and 469 bp of 16S rDNA using Brh: 5’–CCGGTCTGAACTCAGATCACGT-3’ and Arl: 5’–

CGCCTGTTTAACAAAAACAT -3’ (Palumbi, 1991). Additionally we sequenced the complete (495 bp) ITS1 internal transcribed spacer region (ITS-1S: 5’-

CCATGAACGAGGAATTCC CAG–3’, 5.8AS: 5’ –TTAGCAAACCGACCCTCAGAC –

3’) of a subset of the P. acuta individuals chosen to represent the recovered mitochondrial clades. Sequencing reactions were performed using the BigDye sequencing kit 3.1 (Applied Biosystems, Foster City, California, USA). Sequences were edited using Sequencher 5.3 (Gene Codes Corporation, Ann Arbor, MI, USA), and then aligned using ClustalW (Larkin et al., 2007) and further aligned manually if necessary. GBlocks (Castresana, 2000) was used to identify and remove difficult to align regions present within both the ITS1 and 16S alignments. Sequences generated during this study were submitted to GenBank under the accession numbers MF694400-MF694492, MG001330-MG001338 (Additional Files 1 & 2). All alignments used in this study were deposited into TreeBase (www.treebase.org).

Phylogenetic analyses

Phylogenetic analyses of Physa species were performed for the purposes of determining the in-group in all subsequent phylogeographic and population genetic

53 analyses. Aplexa elongata was used as an out-group (Wethington and Lydeard,

2007). Physa phylogenetic datasets were analyzed separately as cox1 (509 bp), 16S

(611 bp) and cox1+16S (1,120 bp). Relatively few ITS1 (525 bp) sequences exist within the NCBI database (Additional Files 1 & 2), and were therefore unable to be used in a concatenated analyses of Physa species. The T92+G model of nucleotide substitution was used to model the cox1, 16S, ITS1 datasets, and HKY+G was used to model the cox1+16S dataset. Models were chosen based on the Akaike information criterion (AIC, Akaike, 1974) using JModelTest2 (Darriba et al., 2012).

Phylogenetic analysis of Physa species did not recover a supported sister clade for P. acuta, and so, Physa spelunca was chosen as an out-group based on

Wethington and Lydeard (Wethington and Lydeard, 2007) and was used to root subsequent in-group analyses. Samples accessioned in NCBI and/or were identified as P. acuta via BLASTn search but had a genetic distance greater than 5% (distance between P. acuta and P. spelunca) were excluded from in-group analyses. In-group analyses included 151 individual snails and were performed to evaluate phylogeographic patterns within P. acuta. Each locus was treated as an independent dataset and analyzed separately. Optimal models of nucletotide substitiution were found to be the same per gene region as used for phylogenetic analysis of Physa.

Maximum Likelihood (ML), Minimum Evolution (ME) and Bayesian Inference

(BI) were preformed on each phylogenetic dataset. ME and ML analyses were performed in the program PAUP* 4.0 (Swofford, 2003). One thousand bootstrap replicates were preformed within each gene tree analysis to statistically assess the resulting topologies.

Bayesian Inference (BI) was performed using the program MrBayes v. 3.2.6

(Ronquist and Huelsenbeck, 2003), consisting of two replicated runs for each locus

54 with four Markov chain Monte Carlo (MCMC) chains, one cold and three heated chains. Each analysis ran for 10,000,000 generations and was sampled every 1000 generations. The analysis was terminated when the standard deviation of the split frequencies was or fell below 0.01, supporting convergence. Likelihood parameters and convergence between runs were assessed using the program Tracer v.1.6

(Rambaut et al., 2014), based on EES values greater than 200. The first 2,500 trees from each analysis were discarded as burnin. Resulting phylogenetic trees were visualized and manipulated using Fig Tree v. 1.3 (Rambaut and A. Drummond, 2009) and MEGA 6 (Tamura, 2011).

Recovered topologies were statistically compared to a null phylogenetic hypothesis (HO) using an approximately unbiased (AU) test (Shimodaira, 2002), where per site likelihood scores were bootstrapped (10,000 replications) using the program CONSEL (Shimodaira and Hasegawa, 2001) to generate p-values. To robustly test the ‘single source’ hypothesis we constrained all invasive individuals as

1 monophyletic, with an Eastern USA origin (HO ), using the cox1 dataset. Secondly, to test for mito-nuclear discordance within P. acuta, we constrained the ITS1 topology

2 for complete concordance with recovered mitochondrial topologies (HO ).

Constrained trees were created in Mesquite (Maddison and Maddison, 2015). Per site likelihood scores were calculated in PAUP* (Swofford, 2003) .

Genetic diversity and population structure

To characterize range-wide population genetic diversity and structure we grouped native populations according to the freshwater ecological complex (FWEC,

Abell, 2000) from which they were collected. FWECs are based on delineations of freshwater ecoregions determined by fish distributions (Maxwell et al., 1995).

FWEC’s are a subset of ecoregions and are designated based on major habitat type and biological distinctiveness. We sampled 8 of the 10 North American FWECs

(Abell, 2000). We selected between 2-10 individual P. acuta from each population to be used for genetic analysis.

Genetic diversity summary statistics of P. acuta populations were estimated using DNAsp v. 5 (Librado and Rozas, 2009) based on the cox1 and 16S gene regions; number of unique haplotypes (h), number of segregating (polymorphic) sites

(S), average number of nucleotide differences (K), haplotype diversity (Hd), nucleotide diversity (π) and Watterson’s estimator (Θw, per site) were calculated. The dataset of 151 individual P. acuta cox1 sequences consisted of 88 unique haplotypes. A minimum spanning network was constructed to view connections among haplotypes using the program PopART (Leigh and Bryant, 2015), http://popart.otago.ac.nz/). Relationships among haplotypes (185 bases) were also evaluated using BI, to determine ancestral relationships among haplotypes.

Parameters for BI were identical to those used in analyses described previously.

To estimate overall range-wide genetic structuring of P. acuta (cox1) a one- way Analysis of Molecular Variance test (AMOVA (Excoffier et al., 1992)) was performed in Arlequin v. 3.5 (Excoffier and Lischer, 2010). Populations were partitioned by five geographical regions (North America, Eurasia, South America,

Africa, and Australasia): ΦCT estimates variation among the five regions sampled,

ΦSC estimates variation among localities within the regions and ΦST estimates variation from all samples across all localities sampled. A second AMOVA was preformed to assess population genetic structure within the native range of P. acuta partitioned by sampling localities and the eight FWECs sampled.

To further evaluate population genetic structure within P. acuta, pairwise ΦST

(Reynolds et al., 1983) values were estimated from the cox1 dataset. Significance of

ΦST values (p < 0.05) was determined by permutation tests of 10,000 random permutations, calculated in Arlequin v. 3.5 (Excoffier and Lischer, 2010).

Pairwise uncorrected p-distances were calculated for cox1, 16S and ITS1 datasets within and between FWEC, countries within the invasive range, within and between the invasive and native ranges, and within and between recovered clades identified through MSN and phylogenetic analysis using MEGA 6 (Tamura, 2011).

Demographic analysis

We estimated changes in the effective population size (Ne) over time by fitting a Bayesian Skyline demographic model (BSP, Drummond et al., 2005) for each

FWEC and/or geographic region, recovered clades and the total cox1 dataset in

*BEAST v. 1.7.0 (Drummond et al., 2012). The HKY substitution model was used for two simultaneous MCMC runs for between 3,000,000 and 80,000,000 iterations sampling between every 3,000 and 10,000 steps. Analyses were run using three mutation rates of 1e-4 (0.04%), 1e-6 (1%) and 4e-6 (4%) per million year, chosen to reflect a range of possible rates of cox1 evolution based on other gastropod systems

(Reid et al., 1996; Crandall et al., 2008; Panova et al., 2011; Harris et al., 2013).

Convergence was checked (ESS of 200 or greater) and results visualized using

Tracer v.1.6 (Rambaut et al., 2014). BSP data from each analyzed dataset generated in Tracer v. 1.6 was exported and visualized in Microsoft Excel for comparison.

Invasion history and meta-analysis of the trematodes of P. acuta

In conjunction with the molecular data generated by this study, a literature search was carried out to compile reports of P. acuta outside of North America.

ISI Web of Knowledge was searched for 1) all reports of P. acuta outside of North

America and 2) surveys of trematode assemblages of P. acuta within both its native and invasive range. Data complied was published from 1805 up to the end of 2016, and recovered using combinations of keywords; Physa acuta, Physa, Physella,

Costella, Haitia, trematode, parasites, invasive gastropod, malacological survey, trematode survey and cercarial survey. Studies cited by recovered articles were also included.

In total, the final dataset included 196 published reports of Physa acuta outside of North America and 7 trematode surveys from the native range of P. acuta and 15 within its invasive range. Physa acuta-trematode surveys carried out by the authors for the purposes of this study were also added to the meta-dataset. From each P. acuta–trematode survey, observed species richness (So) and prevalence

(pr, percentage of patent infections) were calculated. So is defined as the number of different trematode species occurring within a surveyed host population (Poulin,

1998). Because of the difficulties associated with identifying larval trematodes a conservative approach to calculating So was taken, ‘species’ richness was determined by number of trematode families recovered. Consequently, reported measures of So from the native range are likely underestimates (Gordy et al., 2016;

Laidemitt et al., 2017). Only patent infections, where trematode cercariae were being shed, were considered as infected. Evidence of metacercaria (trematode cysts) within P. acuta were not included as the relationships between trematodes and invertebrate second intermediate hosts are generally less specific (Evans and

Gordon, 1983; McCarthy and Kanev, 1990). Genetic results from this study were used to assign population parameters as regional estimates for each study used within the meta-analysis. For example, population genetic estimates from African

58 populations were assigned to all P. acuta–trematode surveys undertaken in Africa.

Australasian samples were excluded from these analyses due to small sample size.

Estimates of haplotype diversity (Hd) and nucleotide diversity (π), were assigned as population parameters. Time since invasion (tsi) was estimated by assigning each surveyed population into an ‘invasion phase’ category: phase 4: < 80 years since invasion, phase 3: 80-180 years since invasion, phase 2: 180-280 years since invasion and phase 1: 280 + (native range). ‘Invasion phases’ were determined based on published reports of P. acuta invasions and genetic evidence provided by this study. Correlation analyses of So, tsi and pr with Hd, π, and Ɵ were evaluated using Pearson’s correlation coefficient. Probability (p) values were adjusted for multiple comparisons using both Bonferroni and Benjamini-Hochberg corrections

(Benjamini and Hochberg, 1995).

RESULTS

Phylogenetic analyses

Cox1, 16S, 16S+cox1 and ITS1 gene tree analysis (Additional Files 3-6) of

Physa species demonstrates that samples included in this study form a clade with P. acuta sequences published in the NCBI database (Additional Files 1&2), and represent a single globally distributed species. All phylogenetic analyses failed to resolve sister relationships to P. acuta, similar to the findings of Wethington and

Lydeard (Wethington and Lydeard, 2007).

In-group analyses demonstrate phylogenetic structuring (Figure 2, Additional

File 6) and revealed two clades, referred to hereafter as clade A and clade B.

Clades were moderately supported statistically for ME and ML analyses (cox1,

89/82; 16S, 89/88) though posterior probabilities from BI were strong (cox1, 1; 16S,

1). Clade A (n=123) is the largest, containing all invasive populations in addition to populations recovered across the native range. While some individuals from clade A and B were found sympatrically, clade B (n=26) was only recovered west of the

Rocky Mountains. Clades A and B were not recovered from the ITS1 phylogeny, possibly suggesting mito-nuclear discordance (Additional File 5).

Phylogenetic analyses of the P. acuta in-group largely lacked resolution for terminal nodes across all genes samples. We therefore employed approximately unbiased topology tests to more robustly assess phylogenetic hypotheses. Using

1 this method we rejected the null hypotheses, of a single invasive origin (HO , p= 2e-

7) and strict concordance of mitochondrial clades A and B within the ITS1 dataset

2 (HO , p= 3e-5).

Genetic diversity and population structure

Indices of genetic diversity (cox1 and 16S) for each population sampled are summarized in Table 1. In comparing native to invasive range, greater genetic diversity (Hdnative = 0.97, Hdinvasive = 0.615; πnative = 0.0289, πinvasive = 0.0098) was found within the native range. Haplotype diversity is consistently high among FWECs sampled, often reaching 100% (Colorado, Coastal, Great Basin, Atlantic).

Within the cox1 and 16S datasets 88 and 43 unique haplotypes were recovered, respectively. More genetic variability was recovered from the cox1 dataset and haplotype relationships were estimated by constructing a Minimum

Spanning Network (MSN; Figure 2A). The MSN corroborates the split of clades A and B, however it also suggests that haplogroups may be further substructured.

FWEC’s do not appear to be a primary factor in shaping the genetic structure of P. acuta. Clade A contains all of the invasive haplotypes. Many of the haplogroups within clade A (I-V) are connected by haplotypes recovered from the Eastern United

States (Mississippi and Atlantic FWECs), indicating that these regions are likely a source of invasions from the native range of P. acuta.

Clade B shows greater nucleotide diversity than clade A (even when invasive haplotypes are removed, πCladeA= 0.019, πCladeB= 0.024). All haplotypes were recovered from the Western United States (west of the Rocky Mountains). While sampling is not sufficient to know if Clade B is truly geographically restricted, our data suggests (1) clade B is a Western USA clade and (2) FWEC’s may play a greater role in shaping western populations of P. acuta (Figure 2A, haplgroups VI and VII).

Relationships among haplotypes were also examined using BI analysis

(Figure 2B). Haplotype relationships are identical to those observed from the MSN, and the divergence of clade A and B is supported with 100% posterior probability.

Invasive haplotypes group with multiple source populations, from across the native range. Together the MSN and BI analyses of haplotypes suggest that there have been several distinct invasion events to the Eastern Hemisphere from multiple sources across North America.

Overall population genetic structure of cox1 was evaluated by AMOVA (Table

2). Both range-wide and native-range AMOVA’s suggest a significant ΦCT (p<0.0001 and p<0.0001, respectively). In the native range 78.55% of genetic variation is attributed to within population variation, indicating that overall FWEC’s, and the biogeographic regions tested in general, are not important in structuring P. acuta populations.

Pairwise uncorrected p-distances within and between populations of P. acuta were calculated from the cox1 dataset (Table 3). Within the native range within population genetic distance averages 0.0292 (~3%), relative to 0.0102 (~1%) within

61 the invasive range which are significantly lower (p=0.0006). Within population genetic distances between the Western USA and Eastern USA (native range) vary significantly (p=0.05), and average 0.0346 (~3.5%) and 0.022 (2.2%) respectively.

Overall western populations exhibit the highest within population genetic distances while Africa and Asia demonstrate the lowest (x= 0.004, 0.4%).

Pairwise genetic distances and ΦST values were calculated (Table 3) and assessed statistically (only significance values for ΦST are shown). The Mississippi

FWEC, the largest complex sampled, shows significant (p<0.05) genetic differentiation from four of the six other FWECs (Rio Grande, Coastal, Colorado and

Atlantic) as well from Africa and Middle East. Similarly, the Atlantic FWEC shows significant (p<0.05) genetic differentiation from four of the six other FWECs (Rio

Grande, Coastal, Colorado and Mississippi) as well from Europe, Australasia, Asia and the Middle East.

Demographic and invasion history analyses

Bayesian Skyline Plots (Figure 3, Additional File 7) estimated changes in effective population sizes over coalescent time. Results based on a 1% change per million year rate of cox1 evolution (Figure 3) largely matched estimates using 0.04% and 4% change per million year (Additional File 7). Regionally (Figure 3A) Ne is greatest within North America and has been stable over the past 60,000 years.

Populations from the invasive range suggest much smaller effective sizes, congruent with previous analyses of genetic diversity. Eurasian and African/Australasian populations indicate a possible recent population expansion. When Clades A and B are compared (Figure 3B), Clade A and the total P. acuta cox1 dataset suggests a recent population decline which is likely an artifact of population subdivision. Heller et al. (2013) showed subdivision can confound coalescent estimators of Ne and

62 result in an erroneous signal of population decline. There is no evidence (Figure 3C) of population decline within the native range. There is evidence to suggest that

Clade B has recently undergone a demographic expansion (within the past 10,000 years), both clades A and B have comparatively similar contemporary Ne.

Invasion timeline and meta-analysis of the trematodes of P. acuta

Based on reports from the literature an invasion timeline is summarized in

Figure 4 (full bibliographic information can be found in Additional File 8). Findings from this literature search informed subsequent time since invasion analysis.

Physa acuta – trematode surveys from both the native (n=7) and invasive

(n=15) range were compiled to estimate range-wide species richness (Table 4).

Average So within the native range (So = 3.25) was significantly greater (p=0.0016) than within the invasive range (So = 0.3125). Western European (n=2) and Iranian

(n=1) populations of P. acuta were the only invasive populations surveyed where patent infections were reported. Prevalence varied significantly (p=0.04) among invasive (0.00319 %) and native ranges (0.2%). Variables were analyzed using

Pearson’s correlation coefficient (ρ), summarized in Table 5. All comparisons between So, tsi, pr and the demographic variables (Hd, π, Ɵ) resulted in significant unadjusted p-values, except for pr and Hd. To correct for multiple comparisons, p- values were adjusted using BH and Bonferroni methods (Table 5). Following adjustment, So, tsi and pr were found to significantly correlate with π and Ɵ (pBH=

0.003, pBH= 0.000, pBH= 0.033, respectively). π and Ɵ were found to have a correlation coefficient of 1 (Table 5). There is a strong positive relationship between tsi, π (ρ = 0.97) and So (ρ = 0.8) and a moderate correlation with pr (ρ = 0.57), supporting the prediction that genetic diversity increases with time since invasion. In total, these results suggest a significant relationship between demographic

63 parameters, namely background genetic variation, and the rate and diversity of trematode infections.

DISCUSSION

Range wide population genetic patterns

Our results propose discordant population genetic patterns between the native and invasive range of P. acuta. While haplotype diversity within the invasive range is not low relative to other invasive snails, Melanoides tuberculata (Facon et al., 2003), Haminoea japonica (Hanson et al., 2013) (within their invasive range), by comparison there is an approximately 20% reduction in Hd relative to the native range. Our analyses reveal that invasive populations have substantially less mtDNA diversity relative to native populations (Table 1), a finding supported by numerous studies of invasive gastropods (Facon et al., 2003; Hanson et al., 2013). Interestingly however, this result is in contrast to recently published work on P. acuta. Bousset et al. (Bousset et al., 2014) used 9 microsatellite loci to estimate range-wide patterns of genetic diversity and reported no difference in neutral genetic variation between the native and invasive ranges. These contradictory results are likely explained by the fact that the Bousset et al.(2014) study was limited to five native populations, four of the five were from the Eastern USA where many invasive haplotypes likely originated. Consequently, Bousset et al.(2014) were unable to capture similar amounts of genetic diversity revealed in this study. Discordance of mtDNA and microsatellite data is not uncommon (Bayha et al., 2015) and both reveal aspects of invasion history. Mitochondrial markers can often better detect subtle genetic subdivision, due to reduced effective population size and lack of recombination

(Bayha et al., 2015; Larmuseau et al., 2010; Lukoschek et al., 2008). Consequently,

64 a genealogical approach utilizing mtDNA markers is frequently used to characterize population genetic structure and identify source populations (Cristescu et al., 2001;

Facon et al., 2003; Hanson et al., 2013). Our findings based largely on mtDNA, combined with those of Bousset et al. (2014) suggest P. acuta is overall a genetically diverse species, relative to other invasive gastropods (Facon et al., 2003; Lounnas et al., 2017; Städler et al., 2005). However, this study reports significant genetic structuring within the native range and between some native and invasive populations that has not been previously reported by other investigators. Further, previously unknown mito-nuclear discordance (Figure 2, Additional Files 5 & 6) suggests historical isolation followed by secondary contact has likely occurred among native western and eastern populations.

Genetic structure within the native range and biogeographic considerations

The Eastern USA is hypothesized as the origin of all P. acuta (Bousset et al.,

2014; Dillon et al., 2002). Our data suggest that sampled genetic diversity is comparable (marginally greater in Western USA) across the native range with evidence of geographic structuring between the Eastern and Western USA populations. Phylogenetic analyses suggest the existence of two clades (A & B) within the native range; interestingly these clades are sympatric in the Western USA, while clade B appears to be restricted to Western USA. Field surveys and subsequent molecular analysis reveal that individuals belonging to clade A and/or B can be collected side by side from the same water body in New Mexico, Colorado,

California and Nevada. Phylogenetic analysis (Additional Files 3-6) did not result in sufficient resolution to reveal the geographic origin of P. acuta, however increased sampling of western populations fails to support the Eastern origin hypothesis (Dillon et al., 2002; Bousset et al., 2014; Lydeard et al., 2016). Several individuals (S94 &

S22) collected from California (Coastal FWEC) and one (S17) from New Mexico (Rio

Grande FWEC) were close matches genetically to P. acuta, but did not fall in the P. acuta clade. Additionally the presumed sister species P. spelunca has only been recovered from Wyoming (Turner and Clench, 1974) and did not have support as the sister species to P. acuta from our analyses. Anecdotally, recent molecular genetic studies of Physa (Rogers and Wethington, 2007; Gates et al., 2013; Moore et al.,

2015) may demonstrate a greater number of independent taxonomic units from western waterways relative to the Eastern USA, suggesting that biogeographically speaking, the west may have been important in physid diversification.

This study is the first to compare and assess discordance among nuclear and mitochondrial markers within P. acuta. Discordance between ITS1 and mitochondrial phylogenies could suggest incomplete lineage sorting, though more likely a recent history of genetic exchange between once isolated populations. Clades A & B were not recovered from the ITS1 gene tree (Additional File 5), and when the largely unresolved ITS1 phylogeny was constrained to match the mitochondrial topology a significantly (p=3e-5) worse phylogeny was recovered. Gene trees of the non- recombining mitochondrial markers may suggest historical isolation between the

Western and Eastern USA. It may be hypothesized that the Rocky Mountains, or associated waterways, have played a role in restricting gene-flow, which has been shown relevant in other freshwater systems (Forister et al., 2004; Finn et al., 2006;

Hughes et al., 2009). It can also be hypothesized that anthropogenic dispersal within

North America or subsequent re-introduction from the invasive range into the

Western USA, has led to reticulation of the two clades (Turner, 2002), and contemporary admixture within the Western USA. Another possible line of evidence supporting this hypothesis is the recent discovery of two distinct mitochondrial

66 genomes, varying in size (Isolate A 14,490 vs Isolate B, 14,314 bp) and 37 indels, collected from the same pond in New Mexico (Nolan et al., 2014). cox1, 16S and

ITS1 sequence from Nolan et al. (Nolan et al., 2014) was included in our phylogenetic dataset and consistently parse out with clade A (Isolate A) and B

(Isolate B). At this time there has been no investigation into the biological relevance or functional consequences of Isolate A versus B. Mitochondrial genome divergence may in fact be a relic of once isolated populations conferring no functional differences. The fact that invasive haplotypes have only been recovered from clade

A is an intriguing pattern warranting further investigations.

Invasion history of Physa acuta

Figure 4 summarizes records of P. acuta globally (see Additional File 8 for full bibliographic information). These records, combined with population genetic inference, lead to a hypothesized route of invasion. The first record of P. acuta outside of North America was its initial description in France in 1805 (Draparnaud,

1805), over 200 years ago. It is hypothesized that these were the first founding populations of P. acuta. Anderson (Anderson, 2003) was the first to posit that the P. acuta discovered by Draparnaud (Draparnaud, 1805), in the Bordeaux region of

France, were introduced via the cotton trade from Mississippi USA during the 18th century. Trade between France and the USA ceased during the Napoleonic wars and began between USA and England, where records of P. acuta appear in the mid

1800’s (Forbes, 1853; Jeffreys, 1862). Western Europe (England and France) likely maintained the first founding populations of P. acuta. However, in less than 100 years, populations were reported from Eastern Europe, Africa, Asia, South America, the Middle East, the Caribbean, New Zealand and Australia. Prior studies have suggested that the subsequent global invasion resulted from the spread of early

Western European colonizers (Anderson, 2003; Oscoz et al., 2010). If this was the case we would expect 1) clustering of all invasive populations with Western

European haplotypes and 2) a stepping stone like model of haplotype relationships

(Reusch et al., 2010; Simon et al., 2011). This model was not supported by any of the preformed population genetic analyses and further, an approximately unbiased

(AU, Shimodaira, 2002) topology test specifically rejected the monophyly of invasive populations (p=2e-7).

Our data support that there have been multiple independent invasions into

Western Europe, likely from different sources (Figure 2). Data from this study suggests that some European populations are closely allied with populations in Asia and the Middle East and share the most common invasive haplotype, we infer that this shared haplotype (Posada and Crandall, 2001) is related to the initial invasive populations of P. acuta in Western Europe roughly 200 years ago, and has subsequently spread east. It is more parsimonious to hypothesize that genetically depauperate European populations expanded east than to assume that identical haplotypes from a genetically diverse source (North America) founded three distinct invasive regions.

Samples from Africa, South America and the Caribbean are related and are likely the result of an independent invasion event, though it is currently not possible to determine the directionality without more samples. Physa acuta invasions into

Africa have been more recent relative to European invasions. Extensive malacological surveys across the African continent suggest the African invasion occurred in the range of 1940-50’s, with populations reported from South Africa (De

Kock and Wolmarans, 1998; De Kock et al., 2002; Appleton, 2003), Namibia (Curtis,

1991), Morocco (Mohamed and Najat, 1998), Nigera (Fashuyi, 1990), and Kenya

(Brown, 1980). Records suggest a similar time frame for the South American and

Caribbean invasion (Paraense and Pointier, 2003; Collado, 2016). Surveys by

Albrecht et al. (2008) suggest that Lake Titicaca (Peru & Bolivia) was colonized 20 years ago.

Our data suggests haplotype connections between invasive populations and individuals collected in the Western and Eastern USA and Canada. While our data supports the initial invasions from the Southeastern USA to Western Europe, we also find evidence of haplotype connections between the Coastal FWEC (California) and the Middle East and Eastern Europe. Currently, based on our sampling it is unclear if there are in fact Western source populations, or if this relationship is the result of the re-introduction of invasive P. acuta from other continents into the Western USA. Our data supports that all source populations have originated from Clade A, however, it remains unclear if Western haplotypes of Clade A originated from the Eastern USA or are the result of secondary contact from invasive populations, though these scenarios are not mutually exclusive. Increased sampling within the western range is necessary to fully elucidate the source of Clade A members recovered from the

Western USA. Founding events, within and from the native range, are likely the result of the aquaria or aquatic plant trade as posited by multiple investigators

(Anderson, 2003; Appleton, 2003; Bousset et al., 2014; Hui Ng et al., 2015).

Host-Parasite invasion dynamics

Combining published reports of P. acuta occurrence with our genetic data we found a significant relationship between genetic diversity (π and Ɵ) and trematode prevalence (pr), and richness (SO). Nucleotide diversity (π) is a better measure of background genetic variation than haplotype diversity, as it is more robust to the stochastic effects of drift (Avise, 2000). Higher values of haplotype diversity relative

69 to nucleotide diversity were recovered from the invasive range suggesting rapid expansion from founding populations (Avise, 2000).

We also found that time since invasion (tsi) significantly correlated with π, Ɵ as well as pr and SO. Of all the invasive populations screened for trematodes and included in this study there were only three reports of P. acuta shedding larval trematodes (cercariae) within its invasive range (Table 4); France (Dreyfuss et al.,

2002), Spain (Toledo et al., 1998), Iran (Athari et al., 2006). Populations from

Western Europe (presumably the first invasive populations) and allied populations from the Middle East have the greatest π of invasive populations sampled.

These data suggest that the rate and richness of trematode infection increases positively with background genetic variation and time since invasion, and leads to the prediction that over time as invasive P. acuta populations become more genetically diverse their role in local trematode transmission will increase. Meta- analysis was limited by population genetic sampling within the invasive range, which will be necessary to more robustly model the relationship between genetic diversity and trematode prevalence.

Overall results from meta-analyses support the ‘Enemy-Release’ hypothesis

(Keane, 2002; Torchin et al., 2003). There are numerous explanations for this within the P. acuta – trematode system. Firstly, most founding populations likely consisted of eggs or young snails, as common with the aquaria trade, and were therefore uninfected. Secondly, there might be incompatibility among native parasites and invasive host populations. An additional ecological explanation is that invasive P. acuta populations often establish in highly modified habitats (i.e. drainage ditches, man-made water-bodies (Anderson, 2003; Brackenbury and Appleton, 1993), which may not be suitable to sustain a diversity of trematode lifecycles. As invasive

70 populations spread over time into a broader range of habitat types, one would expect interaction with wildlife+trematode fauna to become more frequent with a greater potential for native parasites to establish in P. acuta. For example, in countries such as England and Germany, where the invasion history of P. acuta is well documented, initial reports of P. acuta are from habitats such as botanical gardens that are known to import aquatic vegetation, but less than a decade later, P. acuta is reported from natural habitats (summarized by Vinarski, 2017).

Apart from directly hosting trematodes, invasive P. acuta populations might play a role in diluting or enhancing transmission of indigenous parasites in several ways. Physa acuta may act as a decoy host (Yousif et al., 1998; Chipev, 2009;

Munoz-Antoli et al., 2003) to indigenous trematode species, as it is an efficient competitor to native snails often displacing them to become the dominant gastropod within the community (Brackenbury and Appleton, 1993; Dobson, 2004). Dobson

(Dobson, 2004) showed that following its establishment in Mozambique, P. acuta overtook Bulinus forskalii, a primary transmitter of human schistosomiasis

(Schistosoma hematobium), as the dominant snail. It has been shown experimentally

Schistosoma hematobium can infect P. acuta in Nigeria, though the parasite cannot reach patency (Akinwale et al., 2011). There are also numerous reports of invasive

P. acuta acting as a reservoir for larval parasites, specifically as a second intermediate host to trematodes. Similarly, non-host specific larvae of the invasive nematode Angiostrongylus (=Parastrongylus) cantonensis were reported in Saudi

Arabia (Bin Dajem, 2009) and Egypt (Abo-Madyan et al., 2005) from P. acuta.

Host-parasite invasion dynamics are often underappreciated when considering the impact of invasive species, especially when the associated parasites are not causing notable pathology. Most commonly it seems hosts acquire a distinct

71 assemblage of indigenous parasites within their invasive range (Torchin et al., 2003) over time, and the consequences to local transmission and biodiversity are not clear.

While the primary goal of this study was to characterize the population genetic structure and invasion history of P. acuta, we aim to begin to use this system to address the role of host demographics in the acquisition of local parasite assemblages specifically, to identify potential predictors of parasite invasion success for the purposes of modeling host-parasite invasion over time.

CONCLUSIONS

This study provides improved sampling of Physa acuta within its native range, where mtDNA analyses reveal two phylogenetic clades (A & B) and previously unknown population structure between the Eastern and Western USA. Increased sampling of native populations allowed the estimation of the invasion history of P. acuta. We report evidence for the occurrence of multiple independent invasion events from different source populations across the USA and Eastern Canada. The initial invasion into Western Europe occurred over 200 years ago, with more recent invasions into the Southern Hemisphere (South America, Africa) occurring within the last 20-80 years.

We also examined the relationship between tsi, pr, So and host-population genetic parameters (Hd, π, Ɵ) to identify parameters of relevance for modeling host- parasite invasion dynamics over time. We found a significant positive relationship among π with trematode pr and So. These data suggest that, (1) host demographic parameters might be integral in the assembly of parasite communities within invasive host populations and (2) background genetic diversity may be an important parameter to consider in future efforts to model host-parasite invasion dynamics.

Overall, our data supports predictions based on the ‘Enemy Release hypothesis’

(Keane, 2002; Torchin et al., 2003). Hosts, particularity snail intermediate hosts, are substantially easier to sample and monitor over time and space relative to their associated trematode assemblages and may prove a valuable and tractable tool to model parasite invasion.

ACKNOWLEDGEMENTS

The authors would like to thank the following people for their help in the collection and/or screening of the snails used in this study: Dr. Ramesh Devkota, Dr.

Ben Hanelt, Dr. Norm Davis, Ms. Martina Laidemitt, Mr. John Schlutz. The authors would also like to thank Dr. Charles Criscione and three anonymous reviewers for their helpful comments regarding this manuscript, and Dr. Thomas Turner for helpful suggestions regarding data analysis. We acknowledge technical support from the

University of New Mexico Molecular Biology Facility, which is supported by National

Institutes of Health, USA grant P30GM110907 from the Institute Development Award program of the National Center for Research Resources, USA. This work was funded through a National Science Foundation, USA grant to SVB (DEB 1021427) and a National Institutes of Health grant RO1 AI44913 to ESL.

REFERENCES

Abell, R.A., W.W.F., 2000. Freshwater Ecoregions of North America. Island Press.

Abo-Madyan, A.A., Morsy, T.A., Motawea, S.M., Garhy, El, M.F., Massoud, A.M.A., 2005. Spot light survey on fresh-water snails of medical importance in Al Fayoum Governorate, Egypt. J Egypt Soc Parasitol 35, 49–58.

Akaike, H., 1974. A new look at the statistical model identification. IEEE Trans. Automat. Contr. 19, 716–723. doi:10.1109/TAC.1974.1100705

Akinwale, O.P., Kane, R.A., Rollinson, D., Stothard, J.R., Ajayi, M.B., Akande, D.O., Ogungbemi, M.O., Duker, C., Gyang, P.V., Adeleke, M.A., 2011. Molecular approaches to the identification of Bulinus species in south-west Nigeria and observations on natural snail infections with schistosomes. Journal of Helminthology 85, 283–293. doi:10.1017/S0022149X10000568

Anderson, R., 2003. Physella (Costatella) acuta Draparnaud in Britain and Ireland-its taxonomy, origins and relationships to other introduced Physidae. Journal of Conchology.

Appleton, C.C., 2003. Alien and invasive fresh water Gastropoda in South Africa. African Journal of Aquatic Science 28, 69–81. doi:10.2989/16085914.2003.9626602

Athari, A., Gohar-Dehi, S., Rostami-Jalilian, M., 2006. Determination of definitive and intermediate hosts of cercarial dermatitis-producing agents in northern Iran. Arch Iran Med 9, 11–15.

Avise, J.C., 2000. Phylogeography. Harvard University Press.

Bayha, K.M., Chang, M.H., Mariani, C.L., Richardson, J.L., Edwards, D.L., DeBoer, T.S., Moseley, C., Aksoy, E., Decker, M.B., Gaffney, P.M., Harbison, G.R., McDonald, J.H., Caccone, A., 2015. Worldwide phylogeography of the invasive ctenophore Mnemiopsis leidyi (Ctenophora) based on nuclear and mitochondrial DNA data. Biol Invasions 17, 827–850. doi:10.1007/s10530-014-0770-6

Benjamini, Y., Hochberg, Y., 1995. Controlling the False Discovery Rate - a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B-Methodological 57, 289–300.

Bin Dajem, S.M., 2009. Freshwater snails in Asser region, Saudi Arabia with special refernce to the zoonotic trematode. J Egypt Soc Parasitol 39, 551–557.

Bousset, L., Henry, P.Y., Sourrouille, P., Jarne, P., 2004. Population biology of the invasive freshwater snail Physa acuta approached through genetic markers, ecological characterization and demography. Molecular Ecology 13, 2023–2036. doi:10.1111/j.1365-294X.2004.02200.x

Brackenbury, T.D., Appleton, C.C., 1993. Recolonization of the Umsindusi river. African Journal of Zoology.

Brown, D.S., 1980. Freshwater snails of Africa and their medical importance. Taylor and Francis.

Burch, J.B., 1989. North American freshwater snails. Malacological Publications, Hamburg, Michigan.

Castresana, J., 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution 17, 540–552. doi:10.1093/oxfordjournals.molbev.a026334

Chipev, N.H., 2009. Decoy effect and host infection by miracidia within snail communities. Parasitology 106, 265–276. doi:10.1017/S0031182000075089

Cichy, A., Faltýnková, A., Żbikowska, E., 2011. Cercariae (Trematoda, Digenea) in European freshwater snails - a checklist of records from over one hundred years. Folia Malacologica 19, 165–189. doi:10.2478/v10125-011-0023-6

Clampitt, P.T., 1970. The comparative ecology of the snails Physa gyrina and Physa integra (Basommatophora: Physidae). Malacologia 10, 113–151.

Collado, G.A., 2016. Unraveling cryptic invasion of a freshwater snail in Chile based on molecular and morphological data. Biodiversity and Conservation 26, 567– 578. doi:10.1007/s10531-016-1255-y

Crandell, E.D., Frey, M.A., Grosberg, R.K., Barber, P.H., 2008. Contrasting demographic history and phylogeographical patterns in two Indo‐Pacific gastropods. Molecular Ecology 17, 611–626. doi:10.1111/j.1365- 294X.2007.03600.x

Cribb, T., Bray, R., Olson, P., T.D., Littlewood, J., 2003. Life Cycle Evolution in the Digenea: a New Perspective from Phylogeny, in: Advances in Parasitology Volume 54, Advances in Parasitology. Elsevier, pp. 197–254. doi:10.1016/S0065-308X(03)54004-0

Cristescu, M.E., 2015. Genetic reconstructions of invasion history. Molecular Ecology 24, 2212–2225. doi:10.1111/mec.13117

Cristescu, M.E.A., Hebert, P.D.N., Witt, J.D.S., MacIsaac, H.J., Grigorovich, I.A., 2001. An invasion history for Cercopagis pengoi based on mitochondrial gene sequences. Limnology and Oceanography 46, 224–229. doi:10.4319/lo.2001.46.2.0224

Curtis, B.A., 1991. Freshwater macro-invertebrates of Namibia. Madoqua 17, 164– 187.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models, new heuristics and parallel computing. Nature methods.

De Kock, K.N., Wolmarans, C.T., 1998. A re-evaluation of the occurrence of freshwater molluscs in the Kruger National Park. Koedoe 41, 1–8.

doi:10.4102/koedoe.v41i1.240

De Kock, K.N., Wolmarans, C.T., Preez, Du, L.H., 2002. Freshwater mollusc diversity in the Kruger National Park: a comparison between a period of prolonged drought and a period of exceptionally high rainfall. Koedoe 45, 1–11. doi:10.4102/koedoe.v45i2.23

Dillon, R.T., Jr, Wethington, A.R., 2004. No-choice mating experiments among six nominal taxa of the subgenus Physella (Basommatophora: Physidae). Heldia 69–78.

Dillon, R.T., Wethington, A.R., 2006. The Michigan Physidae revisited: A population genetic survey. Malacologia 48, 133–142.

Dillon, R.T., Wethington, A.R., Rhett, J.M., Smith, T.P., 2002. Populations of the European freshwater pulmonate Physa acuta are not reproductively isolated from American Physa heterostropha or Physa integra. Invertebrate Biology 121, 226– 234. doi:10.1111/j.1744-7410.2002.tb00062.x

Dobson, M., 2004. Replacement of native freshwater snails by the exotic Physa acuta (Gastropoda: Physidae) in southern Mozambique; a possible control mechanism for Annals of Tropical Medicine & Parasitology.

Draparnaud, J.P.R., 1805. Histoire naturelle des mollusques terrestres et fluviatiles de la France.

Dreyfuss, G., Vignoles, P., Abrous, M., Rondelaud, D., 2002. Unusual snail species involved in the transmission of Fasciola hepatica in watercress beds in Central France. Parasite 9, 113–120. doi:10.1051/parasite/2002092113

Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian Phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution 29, 1969–1973. doi:10.1093/molbev/mss075

Drummond, A., Shapiro B. Ho, S. 2005. Bayesian Coalescent Inference of Past Population Dynamics from Molecular Sequences. Molecular Biology and Evolution 22, 1185–1192.

Dunn, A.M., 2009. Chapter 7 Parasites and Biological Invasions, in: Natural History of Host-Parasite Interactions, Advances in Parasitology. Elsevier, pp. 161–184. doi:10.1016/S0065-308X(08)00607-6

Dunn, A.M., Hatcher, M.J., 2015. Parasites and biological invasions: parallels, interactions, and control. Trends in Parasitology 31, 189–199.

Ellis, M.B., 1926. British Snails (1969 reprint). Claredon Press, Oxford. doi:10.1016/S0007-1536(69)80174-9

Estoup, A., Guillemaud, T., 2010. Reconstructing routes of invasion using genetic data: why, how and so what? Molecular Ecology 19, 4113–4130.

doi:10.1111/j.1365-294X.2010.04773.x

Evans, N.A., Gordon, D.M., 1983. Experimental observations on the specificity of Echinoparyphium recurvatum toward second intermediate hosts. Zeitschrift fur Parasitenkunde 69, 217–222. doi:10.1007/BF00926956

Excoffier, L., Lischer, H.E.L., 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources 10, 564–567. doi:10.1111/j.1755-0998.2010.02847.x

Excoffier, L., Smouse, P.E., Quattro, J.M., 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics.

Facon, B., Pointier, J.-P., Jarne, P., Sarda, V., David, P., 2008. High Genetic Variance in Life-History Strategies within Invasive Populations by Way of Multiple Introductions. Current Biology 18, 363–367. doi:10.1016/j.cub.2008.01.063

Fashuyi, S.A., 1990. Freshwater gastropod molluscs in Ondo State, Nigeria. J Afr Zool 104, 165–170.

Finn, D.S., Theobald, D.M., Black, W.C., Poff, N.L., 2006. Spatial population genetic structure and limited dispersal in a Rocky Mountain alpine stream insect. Molecular Ecology 15, 3553–3566. doi:10.1111/j.1365-294X.2006.03034.x

Folmer, O., Hoeh, W.R., Black, M.B., Vrijenhoek, R.C., 1994. Conserved primers for PCR amplification of mitochondrial DNA from different invertebrate phyla. Molecular Marine Biology and ….

Forbes, E., 1853. A history of British Mollusca and of their shells Vol. IV. John Van Voorst, London.

Forister, M.L., Fordyce, J.A., Shapiro, A.M., 2004. Geological barriers and restricted gene flow in the holarctic skipper Hesperia comma (Hesperiidae). Molecular Ecology 13, 3489–3499. doi:10.1111/j.1365-294X.2004.02349.x

Gates, K.K., Kerans, B.L., Keebaugh, J.L., Kalinowski, S., Vu, N., 2013. Taxonomic identity of the endangered Snake River physa, Physa natricina (Pulmonata: Physidae) combining traditional and molecular techniques. Conservation Genetics 14, 159–169. doi:10.1007/s10592-012-0439-6

Ghabooli, S., Shiganova, T.A., Zhan, A., Cristescu, M.E., Eghtesadi-Araghi, P., MacIsaac, H.J., 2011. Multiple introductions and invasion pathways for the invasive ctenophore Mnemiopsis leidyi in Eurasia. Biol Invasions 13, 679–690. doi:10.1007/s10530-010-9859-8

Gordy, M.A., Kish, L., Tarrabain, M., Hanington, P.C., 2016. A comprehensive survey of larval digenean trematodes and their snail hosts in central Alberta, Canada. Parasitology Research 115, 3867–3880. doi:10.1007/s00436-016-5152- 9

Harris, D.J., Ferreira, A.F., De Frias Martins, A.M., 2013. High levels of mitochondrial DNA diversity within oxychilid land snails (subgenus Drouetia Gude, 1911) from São Miguel island, Azores. J Molluscan Stud 79, 177–182. doi:10.1093/mollus/eyt009

Heller, R., Chikhi, L., Siegismund, H.R., 2013. The Confounding Effect of Population Structure on Bayesian Skyline Plot Inferences of Demographic History. PloS ONE 8, e62992. doi:10.1371/journal.pone.0062992

Hughes, J.M., Schmidt, D.J., FINN, D.S., 2009. Genes in Streams: Using DNA to Understand the Movement of Freshwater Fauna and Their Riverine Habitat. BioScience 59, 573–583. doi:10.1525/bio.2009.59.7.8

Hui Ng, T., Kiat Tan, S., Yeo, D.C.J., 2015. Clarifying the identity of the long- established, globally-invasive Physa acuta Draparnaud, 1805 (Gastropoda: Physidae) in Singapore. BioInvasions Records 4, 189–194.

Jarne, P., Delay, B., 1991. Population genetics of freshwater snails. Trends in Ecology & Evolution 6, 383–386. doi:10.1016/0169-5347(91)90158-T

Jeffreys, J.G., 1862. British Conchology. Van Voorst, London.

Keane, R., 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology & Evolution 17, 164–170. doi:10.1016/S0169-5347(02)02499-0

Kelly, D.W., Paterson, R.A., Townsend, C.R., Poulin, R., Tompkins, D.M., 2009. Parasite spillback: A neglected concept in invasion ecology? Ecology 90, 2047– 2056. doi:10.1890/08-1085.1

Kerney, M.P., 1976. Atlas of the non-marine Mollusca of the British Isles. Conchological Society of Great Britain and Ireland and the Biological Records Centre, Huntingdon.

Laidemitt, M.R., Zawadzki, E.T., Brant, S.V., Mutuku, M.W., Mkoji, G.M., Loker, E.S., 2017. Loads of trematodes: discovering hidden diversity of paramphistomoids in Kenyan ruminants. Parasitology 144, 131–147. doi:10.1017/S0031182016001827

Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., 2007. Clustal W and Clustal

X version 2.0.

Larmuseau, M.H.D., Raeymaekers, J.A.M., Hellemans, B., Van Houdt, J.K.J., Volckaert, F.A.M., 2010. Mito-nuclear discordance in the degree of population differentiation in a marine goby. Heredity 105, 532–542. doi:10.1038/hdy.2010.9

Leigh, J.W., Bryant, D., 2015. popart: full‐feature software for haplotype network construction. Methods in Ecology and Evolution 6, 1110–1116. doi:10.1111/2041-210X.12410

Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics.

Lounnas, M., Correa, A.C., Vázquez, A.A., Dia, A., Escobar, J.S., Nicot, A., Arenas, J., Ayaqui, R., Dubois, M.P., Gimenez, T., Gutiérrez, A., González Ramírez, C., Noya, O., Prepelitchi, L., Uribe, N., Wisnivesky Colli, C., Yong, M., David, P., Loker, E.S., Jarne, P., Pointier, J.P., Hurtrez Boussès, S., 2017. Self‐fertilization, long‐distance flash invasion and biogeography shape the population structure of Pseudosuccinea columella at the worldwide scale. Molecular Ecology 26, 887– 903. doi:10.1111/mec.13984

Lozek, V., 1964. Quartarmollusken der Tschechoslwakie Rozpravy ustredniho. Ustavu geolickeho 31, 1–374.

Lukoschek, V., Waycott, M., Keogh, J.S., 2008. Relative information content of polymorphic microsatellites and mitochondrial DNA for inferring dispersal and population genetic structure in the olive sea snake, Aipysurus laevis. Molecular Ecology 17, 3062–3077. doi:10.1111/j.1365-294X.2008.03815.x

Lydeard, C., Campbell, D., Golz, M., 2016. Physa acuta Draparnaud, 1805 Should be Treated as a Native of North America, Not Europe. Malacologia 59, 347–350. doi:10.4002/040.059.0213

Maddison, W.P., Maddison, D.R., 2015. Mesquite: a modular system for evolutionary analysis. Version 3.04. 2015.

Maxwell, J.R., Edwards, C.J., Jensen, M.E., Paustian, S.J., Parrott, H., Hill, D.M., 1995. A Hierarchical Framework of Aquatic Ecological Units in North America (Nearctic Zone).

Mohamed, F., Najat, M., 1998. Co-occurrence de Physa acuta Draparnaud et vecteurs intermédiaires de la bilharziose au Maroc. Riv Idrobiol 37, 169–177.

Monsutti, A., Perrin, N., 1999. Dinucleotide microsatellite loci reveal a high selfing rate in the freshwater snail Physa acuta. Molecular Ecology 8, 1076–1078.

Moore, A.C., Burch, J.B., Duda, T.F., 2015. Recognition of a highly restricted freshwater snail lineage (Physidae: Physella) in southeastern Oregon: convergent evolution, historical context, and conservation considerations.

Conservation Genetics 16, 113–123. doi:10.1007/s10592-014-0645-5

Munoz-Antoli, C., Trelis, M., Gozalbo, M., Toledo, R., Haberl, B., Esteban, J.G., 2003. Interactions related to non-host snails in the host-finding process of Euparyphium albuferensis and Echinostoma friedi (Trematoda : Echinostomatidae) miracidia. Parasitology Research 91, 353–356. doi:10.1007/s00436-003-0968-5

Nolan, J.R., Bergthorsson, U., Adema, C., 2014. Physella acuta: atypical mitochondrial gene order among panpulmonates (Gastropoda). Journal of Molluscan Studies, 80, 388-399.

Oscoz, J., Tomás, P., Durán, C., 2010. Review and new records of non-indigenous freshwater invertebrates in the Ebro River basin (Northeast Spain). Aquatic Invasions 5, 263–284. doi:10.3391/ai.2010.5.3.04

Palumbi, S., 1991. simple fool's guide to PCR.

Panova, M., Blakeslee, A.M.H., Miller, A.W., Mäkinen, T., Ruiz, G.M., Johannesson, K., André, C., 2011. Glacial History of the North Atlantic Marine Snail, Littorina saxatilis, Inferred from Distribution of Mitochondrial DNA Lineages. PloS ONE 6, e17511. doi:10.1371/journal.pone.0017511

Paraense, W.L., Pointier, J.-P., 2003. Physa acuta Draparnaud, 1805 (Gastropoda: Physidae): a study of topotypic specimens. Memórias do Instituto Oswaldo Cruz 98, 513–517. doi:10.1590/S0074-02762003000400016

Perrin, N., 1986. The life history parameters of Physa acuta (Gastropoda, Mollusca) in experimental conditions. Revue Suisse Zoologique 93, 725–736.

Pip, E., Franck, J.P.C., 2008. Molecular phylogenetics of central Canadian Physidae (Pulmonata: Basommatophora). Canadian Journal of Zoology 86, 10–16. doi:10.1139/Z07-112

Posada, D., Crandall, K., 2001. Intraspecific gene genealogies: trees grafting into networks. Trends in Ecology & Evolution 16, 37–45.

Poulin, R., 1998. Comparison of three estimators of species richness in parasite component communities. Journal of Parasitology 84, 485–490.

Rambaut, A., Drummond, A., 2009. FigTree v1. 3.1.

Rambaut, A., Suchard, M., Xie, D., Drummond, A., 2014. Tracer v1. 6 http://beast. bio. ed. ac. uk.

Reid, D.G., Rumbak, E., Thomas, R.H., 1996. DNA, Morphology and Fossils: Phylogeny and Evolutionary Rates of the Gastropod Genus Littorina. Phil. Trans. R. Soc. Lond. B 351, 877–895. doi:10.1098/rstb.1996.0082

Reusch, T.B.H., Bolte, S., Sparwel, M., Moss, A.G., Javidpour, J., 2010.

Microsatellites reveal origin and genetic diversity of Eurasian invasions by one of the world’s most notorious marine invader, Mnemiopsis leidyi (Ctenophora). Molecular Ecology 19, 2690–2699. doi:10.1111/j.1365-294X.2010.04701.x

Reynolds, J., Weir, B.S., Cockerham, C.C., 1983. Estimation of the coancestry coefficient: basis for a short-term genetic distance. Genetics.

Rogers, D.C., Wethington, A.R., 2007. Physa natricina Taylor 1988, junior synonym of Physa acuta draparnaud, 1805 (Pulmonata: Physidae). Zootaxa.

Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572–1574. doi:10.1093/bioinformatics/btg180 Sax, D., Kinlan, B., Smith, K., 2005. A conceptual framework for comparing species assemblages in native and exotic habitats. OIKOS 108, 457–464.

Say, T., 1817. Conchology. In: William Nicholson. American Edition. British Encyclopedia, Philadelphia.

Sebastián Escobar, J., Epinat, G., Sarda, V., David, P., 2007. No correlation between inbreeding depression and delayed selfing in the freshwater snail Physa acuta. Evolution 61, 2655–2670. doi:10.1111/j.1558-5646.2007.00223.x

Shimodaira, H., 2002. An Approximately Unbiased Test of Phylogenetic Tree Selection. Systematic Biology 51, 492–508. doi:10.1080/10635150290069913

Shimodaira, H., Hasegawa, M., 2001. CONSEL: for assessing the confidence of phylogenetic tree selection. Bioinformatics 17, 1246–1247.

Simon, A., Britton, R., Gozlan, R., van Oosterhout, C., Volckaert, F.A.M., Hänfling, B., 2011. Invasive cyprinid fish in Europe originate from the single introduction of an admixed source population followed by a complex pattern of spread. PloS ONE 6, e18560. doi:10.1371/journal.pone.0018560

Städler, T., Frye, M., Neiman, M., Lively, C.M., 2005. Mitochondrial haplotypes and the New Zealand origin of clonal European Potamopyrgus, an invasive aquatic snail. Molecular Ecology 14, 2465–2473. doi:10.1111/j.1365-294X.2005.02603.x

Stelfox, A.W., 1911. Lists of land and freshwater molluscs of Ireland. Proceedings of the Royal Irish Academy 29B, 65–164. doi:10.5962/bhl.title.111171

Swofford, D.L., 2003. PAUP*: phylogenetic analysis using parsimony (*and Other Methods), version 4.0, Sinauer Associates. Sunderland Massachusetts, USA. doi:10.1234/12345678

Tamura, K.E.A., 2011. MEGA5: Molecular Evoluionary Genetic Analysis Using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods. Molecular Biology and Evolution 28.

Taylor, D.W., 2003. Introduction to Physidae (Gastropoda: Hygrophila);

biogeography, classification, morphology. Rev. Biol. Trop.

Te, G.A., 1978. A Systematic Study of the Family Physidae (Basommatophora: Pulmonata). Doctor of Philosophy Dissertation. University of Michigan. Ann Arbor.

Toledo, R., Munoz-Antoli, C., Perez, M., Esteban, J.G., 1998. Larval trematode infections in freshwater gastropods from the Albufera Natural Park in Spain. Journal of Helminthology 72, 79–82. doi:10.1017/S0022149X00016473

Torchin, M.E., Lafferty, K.D., Dobson, A.P., McKenzie, V.J., 2003. Introduced species and their missing parasites. Nature.

Turner, G.F., 2002. Parallel speciation, despeciation and respeciation: implications for species definition. Fish and Fisheries 3, 225–229. doi:10.1046/j.1467- 2979.2002.00085.x

Turner, R.D., Clench, W.J., 1974. A new blind Physa from Wyoming with notes on its adaptation to the cave environment. The Nautilus.

Van Leeuwen, C.H.A., Huig, N., Van der Velde, G., Van Alen, T.A., Wagemaker, C.A.M., Sherman, C.D.H., Klaassen, M., Figuerola, J., 2013. How did this snail get here? Several dispersal vectors inferred for an aquatic invasive species. Freshwater Biology 58, 88–99. doi:10.1111/fwb.12041

Vinarski, M.V., 2017. The history of an invasion: phases of the explosive spread of the physid snail Physella acuta through Europe, Transcaucasia and Central Asia. Biol Invasions 19, 1299–1314. doi:10.1007/s10530-016-1339-3

Welch, R.J., 1930. Foregin Physa, Ards, County Down. Irish Naturalists Journal 3, 91.

Wethington, A.R., Guralnick, R., 2004. Are populations of physids from different hot springs distinctive lineages? American Mala. Bulletin

Wethington, A.R., Lydeard, C., 2007. A molecular phylogeny of Physidae (Gastropoda: Basommatophora) based on mitochondrial DNA sequences. J Molluscan Stud 73, 241–257. doi:10.1093/mollus/eym021

Yousif, F., el-Emam, M., el-Sayed, K., 1998. Effect of six non-target snails on Schistosoma mansoni miracidial host finding and infection of Biomphalaria alexandrina under laboratory conditions. J Egypt Soc Parasitol 28, 559–568.

Figure 1. Map of sampled populations. Colors denote the Freshwater Ecological

Complexes from which P. acuta populations were collected.

Figure 2. Relationships of recovered cox1 haplotypes. A. Minimum spanning network of 88 P. acuta haplotypes. The network is shaded to depict the split between clade A (pink) and B (green), within each clade haplogroups are denoted as I –VII.

B. BI Phylogeny of all 88 haplotypes. Nodes with a posterior probability of ≥ 95% are denoted by a black circle. Haplotypes are colored by the FWEC they were collected from.

Figure 3. Bayesian Skyline Plots. Effective size over time plots estimated in

BEAST using a Bayesian Skyline prior using a mutation rate of 1% (1e-6) per million years. Mean values are shown and confidence values were excluded for clarity. The x-axis indicates time in ‘thousand years ago’ (kya). A. Range-wide estimates for

North America (dark grey, n=79), Central and South America (lime green, n=13),

Eurasia (pink, n=49) and Africa and Oceania (blue, n=7). North America dataset ran for 50,000,000 generations, Eurasia 20,000,000 generations, Central/South America and Africa/Australasia ran for 5,000,000 generations. B. Estimates of recovered

Clades, Clade A (aqua, n=116) and B (dark green, n=33) compared to total (light grey, n=149). The total dataset and Clade A ran for 80,000,000 generations, Clade B ran for 10,000,000 generations. C. Estimates by FWECs sampled: Rio Grande (red, n=16), Coastal (blue, n=13), Colorado (purple, n=11), Atlantic (orange, n=9),

Mississippi (green, n=26), the Great Basin and St. Lawrence FWECs were excluded because of low sample size. All FWEC datasets were run for 5,000,000 generations, with the exception of Atlantic (3,000,000). Bayesian skyline plots were constructed in

TRACER, the data was exported and visualized using Excel.

Figure 4. Invasion timeline of Physa acuta. 1. Absence of Physa acuta shells from

European fossil deposits older than 18th Century (Lozek, 1964). 2. Described by

Draparnaud (Draparnaud, 1805) near Bordeaux, France. Anderson (Anderson,

2003) posits likelihood of introduction from Mississippi USA via the cotton trade. 3.

Trade between USA and France ends during Napoleonic wars, trade with England and USA begins. 4. First reports of P. acuta in England (Forbes, 1853; Jeffreys,

1862). 5. Reports of P. acuta all over United Kingdom by the latter half of 20th

Century (Ellis, 1926; Kerney, 1976; Stelfox, 1911; Welch, 1930) 6-12. Primary citations for these reports are listed in Additional File 7.

Region/Population FWEC N H Hd S K Π Θw Native Range USA Gila River, AZ COL 3 Denver Area, CO COL 2 Fremont Co., CO COL 2 Chaffee Co., CO COL 1 Gafield Co., CO COL 1 Mesa Co., CO COL 1 Rio Blanco Co., CO COL 1 Yuma Co., CO COL 1 Total 12 11 0.98 54 17.16 0.0337 0.035 Woodward Park, CA COA 8 Los Banos WMA, CA COA 2 Coyote Creek, CA COA 3 Total 13 13 1 69 18.45 0.0363 0.044 New Harmony, IN MIS 2 Okoboji Lake, IA MIS 1 Reynolds County, MO MIS 1 Medicine Lake, MN MIS 1 Bench Creek Pond, MT MIS 10 Mormon Lake, NE MIS 12 Big Horn River, WY MIS 1 Stillwater, OK MIS 5 Total 24 21 0.989 86 13.93 0.02736 0.0453 Stillwater NWR, NV GRB 2 Total 2 2 1 20 20 0.0393 0.0393 Carson Nat. Fr., NM RIO 3

Bosque Del Apache, NM RIO 9 Shady Lakes, NM RIO 2 Bitter Lake, NM RIO Total 13 12 0.987 49 14.65 0.0278 0.029 Philadelphia, PA ATL 2 Charles Towne Landing SP, SC ATL 7 Total 9 9 1 33 10.58 0.021 0.024 Canada Point Peele NP, ON STL 1 Niagra River, ON STL 3 Total 4 3 0.833 14 8.167 0.016 0.015 NR Total 73 68 0.973 45.29 14.74 0.0289 0.033 Invasive Range Cuba IR 3 Chile IR 10 Lake Titicaca, Peru IR 1 C&S AM Total 14 6 0.604 13 2.879 0.0056 0.008 Kisumu, Kenya IR 4 AF Total 4 3 0.833 5 2.5 0.00493 0.005 France IR 4 Greece/Macadonia IR 23 Netherlands IR 1 EU Total 28 7 0.696 35 8.132 0.016 0.0177 New Zealand IR 1 Australasia IR 2 NZ/OZ 3 3 1 9 6 0.0118 0.012

Total South Korea IR 1 Thailand IR 1 Singapore IR 3 Maylasia IR 3 AS Total 7 2 0.286 6 1.714 0.0034 0.005 Iraq IR 2 Iran IR 12 MD Total 14 7 0.813 29 8.66 0.017 0.018 IR Total 70 28 0.615 16.16 4.98 0.0098 0.011 Global Total 143 96 0.794 30.725 9.86 0.0193 0.023

Table 1. Summary of range-wide population statistics, cox1. Regional abbreviations: NR, native range, IR, invasive range, C&S AM, Central and South America, AF, Africa, EU, Europe, NZ/OZ, New Zealand and Australia, MD, Middle East, FWEC refers to Freshwater Ecological Complexes; COL = Colorado FWEC, COA = Coastal FWEC, MISS = Mississippi FWEC, GRB = Great Basin FWEC, RIO = Rio Grande FWEC, ATL = Atlantic FWEC and STL = St. Lawrence FWEC. Population statistic abbreviations: N, number of individuals sampled, H, number of haplotypes, Hd, haplotype diversity, S, segregating sites, K, average sequence divergence, π, nucleotide diversity, Θw, Wattersons estimator per site, Italicized numbers refer to population averages. (*) Indicates a statistically significant value (P<0.05).

% Fixation Source of Variation d.f. p value Variation indices Range Wide Among Regions 6 3.19 ΦCT= 0.093 0.183 Among Populations within Regions 28 27.3 ΦSC = 0.282 < 0.001 Within Populations 110 69.52 ΦST= 0.305 < 0.001 Native Range Among Regions 6 4.27 ΦCT= 0.043 0.228 Among Populations within Regions 14 19.18 ΦSC = 0.20 < 0.001 Within Populations 55 76.55 ΦST= 0.235 < 0.001

Table 2. Analysis of Molecular Variance

1 2 3 4 5 6 7 8 9 10 11 12 13 1. Rio Grande 0.03 0.0673 0.0167 0.1571 0.2818 0.17422 0.0706 0.2668 0.2396 0.2926 0.377 0.2613 0.0288 2. Coastal 0.0366 0.0386 0.05412 0.08 0.1557 0.0269 0.2006 0.1841 0.1509 0.1573 0.2607 0.129 -0.0085 3. Colorado 0.0338 0.0398 0.0363 0.1694 0.3058 0.1857 0.0897 0.3418 0.24643 0.35611 0.4045 0.3038 0.0623 4. Mississippi 0.34 0.0364 0.0387 0.0288 0.0969 0.01138 0.022 0.0983 0.0745 0.00036 0.245 0.0662 -0.0125 5. Atlantic 0.0356 0.0364 0.0418 0.0283 0.0214 0.0773 0.2032 0.1869 0.17405 0.1216 0.0944 0.1189 0.0199 6. St. Lawrence 0.0295 0.0305 0.0348 0.0241 0.0211 0.0166 0.1199 0.0586 0.2199 0.2518 0.4667 -0.006 -0.0158 7. Great Basin 0.031 0.0328 0.0346 0.0327 0.0337 0.0281 0.0333 0.2469 0.2587 0.4986 0.4948 0.2333 -0.1357 8. Europe 0.0298 0.0325 0.0366 0.0251 0.023 0.0175 0.0288 0.0165 0.1726 0.0994 0.3923 0.0291 0.0932 9. Australasia 0.031 0.035 0.0373 0.025 0.022 0.0186 0.0311 0.0184 0.012 0.4158 0.6129 0.1178 -0.0355 10. Asia 0.0249 0.027 0.0318 0.017 0.0143 0.0116 0.0233 0.012 0.01 0.003 0.7128 0.0567 0.13 11. Africa 0.0343 0.036 0.0414 0.0277 0.016 0.02 0.0327 0.0221 0.02 0.01375 0.005 0.383 0.2029 12. Middle East 0.0317 0.0332 0.0378 0.0258 0.0228 0.0178 0.031 0.0181 0.019 0.0124 0.0228 0.0187 0.06245 13. S. & C. America 0.0311 0.0323 0.038 0.0243 0.0146 0.0173 0.0286 0.019 0.0174 0.0103 0.0056 0.0195 0.0058

Table 3. Pairwise genetic distances and ΦST values. Pairwise within and between FWEC/region genetic distances are shown in the lower diagonal. Within genetic distances are italicized. Pairwise ΦST values are shown in the upper diagonal, values with a p- value ≤0.05 are bolded.

# Individuals # Patent Country Citation Prevalence So Surveyed Infections Native Range New Mexico (Bosque del Apache) This study 1,200 78 0.065 5 New Mexico (Eagles Nest) This study 65 5 0.07 2 Nebraska This study 120 16 0.13 4 Montana This study 165 3 0.018 2 Wyoming This study 26 8 0.31 1 Colorado This study 300 5 0.017 3 Mexico Barragán-Sáenz et al. 2009 496 109 0.22 5

Invasive Range New Zeland Mitchell & Leung 2015 810 0 0 0 Benin Ibikounlé et al. 2009 345 0 0 0 Czech Republic Faltýnková 2005 57 0 0 0 Egypt Abo-Madyan et al. 2005 na 0 0 0 Egypt Bin Dajem 2009 704 0 0 0 Egypt Dahesh & Farid na 0 0 0 France Gerard et al. 2003 413 1 0.0024 1 Germany Faltýnková and Hass 2006 141 0 0 0 Iceland Skinnerson et al. 2008 737 0 0 0 Iran Athari et al. 2006 3,560 8 0.0022 1 Loker & Laidemitt, Kenya unpublished data 589 0 0 0 Morocco Laamrani et al. 2005 na 0 0 0 New Zeland This study 1,480 0 0 0 Spain Toledo et al. 1998 2,717 1 0.00037 1 Zambia Phiri et al. 2007 9 0 0 0

Table 4. Reports of infection and trematode species richness of Physa acuta as first intermediate. na, data not given, SO, observed species richness calculated by the number of unique trematode families observed.

SO tsi pr Hd π Ɵ

SO 0.8 0.77 0.54 0.76 0.77 tsi 0.0025 0.57 0.46 0.97 -0.98 pr 0.0039 0.032* 0.44 0.55 0.56 Hd 0.0614 0.06 0.096 0.5 0.5 π 0.003 0 0.033* 0.061 1 Ɵ 0.003 0 0.033* 0.061 0

Table 5. Matrix of correlation coefficients. Pearson’s correlation coefficients are shown in the upper diagonal and corresponding p-values, adjusted using Benjamini-Hochberg (BH) correction, are shown in the lower diagonal. p-values ≤0.05 are bolded. *Indicates p-values that significant under the BH correction, but not significant under the more conservative Bonferroni correction.

CHAPTER III

Comparative microevolutionary patterns of congeneric trematode species reveal importance of host ecology in parasite evolution

AUTHORS: Erika T. Gendron (Ebbs), D’Eldra Malone, Eric S. Loker, Norm E. Davis, Sara V. Brant

ABSTRACT

Identifying factors that shape microevolutionary forces in natural populations is a fundamental goal of evolutionary biology. In multi-host parasite systems identification of these factors can be confounded by disparate host ecologies and evolutionary histories. One approach to teasing apart the relative influence of ecology verses evolutionary history on observed microevolutionary patterns is to compare congeners, which are assumed to have evolved under similar immunological constraints. This research evaluates comparative population genetic patterns across three congeneric Trichobilharzia species, sampled range-wide, to determine ecological factors important in shaping parasite genetic diversity, genetic effective size and gene flow. As adults Trichobilharzia spp. infect anatids (ducks, geese and swans) and then use freshwater snails (Physidae and Lymnaeidae) as intermediate hosts. Additionally Trichobilharzia survey data across 73 anatid species is presented from which measures of host specificity were estimated, in an effort to characterize anatid-Trichobilharzia host associations. We found higher levels of host specificity than expected based on published experimental work, possibly mediated by host ecology. Further comparative population genetic investigations revealed substantial differences in the population structure and genetic diversity of the three species of interest. In total, results from this study shed light on how specific host

97 ecological traits impact parasite distribution, demography and microevolutionary processes.

INTRODUCTION

Identifying factors that shape microevolutionary forces in natural populations is a fundamental goal of evolutionary biology (Slatkin, 1985). In multi-host helminth systems very little is known about what factors contribute to or determine gene flow, genetic drift and effective size (Ne), (Nadler, 1995; Criscione et al., 2005). These parameters relate directly to the evolution of drug resistance (Blouin et al., 1995), local adaption (Dybdahl and Lively, 1996; Nuismer and Gandon, 2009), probability of host-switching (Hoberg and Brooks, 2008) and ultimately speciation of parasite lineages (Huyse et al., 2003). Considering the parasitic lifestyle is ubiquitous

(Weinstein and Kuris, 2016), this lack of knowledge represents a significant gap in our understanding of microevolutionary processes within natural populations.

In parasite systems microevolutionary forces are mediated by one or more host(s) and relate largely to the ecology of the individual host species (Nadler, 1995;

Criscione and Blouin, 2004; Criscione et al., 2005; Nieberding et al., 2006; Whiteman et al., 2007; Barrett et al., 2008). From these studies and others it is acknowledged broadly that host traits (i.e. life-history, phylogenetic history, host range, host vagility) determine parasite genetic structure (Gorton et al., 2012; Blasco-Costa and Poulin,

2013).

Yet, it is unclear how parasite species partition themselves among hosts with overlapping geographic and ecological ranges. Specifically, it is unknown to what extent subtle ecological differences among closely related, sympatric and/or parapatric, host species impact microevolutionary forces. This scenario facilitates

98 encounter between the parasite species and a broad range of compatible hosts, creating an arena of potential host-parasite interactions. Under these conditions generalism for at least one life cycle stage might be expected, where the parasite species utilizes the range of hosts available to it. Alternatively, strict specialization may develop where a parasite species can only exploit a narrow range of hosts. A third alternative, though not mutually exclusive, can be predicted, where parasite species might develop ecological specificity to a host species or a subset of possible host species, such that there is a reduction in realized host range due to predictable host ecological traits or behaviors. Our definition of ecological specificity is similar to that of ‘faux specialists’ proposed by Brooks and McLeenan (2002), though we are specifically considering parasites whose range is narrowed by the ecology of one or more host species. Like ‘faux specialists’, under a model of ecological specificity parasite species may maximize the short-term gains of specialization without compromising the long-term advantages of generalism (Hoberg and Brooks, 2008).

In terms of population structure, ecological specificity might act to reduce parasite gene flow by narrowing infections to a subset of actual hosts among a broad range of potential host species, leading to genetic differentiation and population genetic patterns that co-vary with realized host range (Nadler, 1995; Poulin and Keeney,

2008; Archie and Ezenwa, 2011).

This research sought to ask how host ecological traits impact parasite microevolution and to characterize host-parasite associations (generalists, specialists, ecological specialists) when parasite species co-occur with a broad ecological range of compatible hosts. To achieve this we took a comparative population genetic approach among congeneric parasite species. Comparing congeneric species is ideal as it is assumed that species evolved under similar

99 immunological and evolutionary constraints (Dawson et al., 2002; Benton 2009).

Consequently, microevolutionary differences observed are therefore likely to result from ecological interrelationships.

This research focuses on species within the genus Trichobilharzia

(Schistosomatidae), specifically three North American taxa within the species complex known as Clade Q (sensu Brant and Loker 2009). As adults, these worms develop and reproduce sexually in a duck definitive host (Anatidae) and release parasite eggs in their feces. A free-swimming larval stage (miracidium) hatches from the egg to find the freshwater snail (Pulmonata) intermediate host where asexual amplification occurs. Free-swimming larvae (cercariae) are released from the snail to find a duck host, thus completing the life cycle. Trichobilharzia is a globally distributed, species rich genus which infects a broad ecological and taxonomic range of definitive anatid hosts (ducks, geese and swans) (Brant and Loker, 2009; 2013).

Observational (Blair and Islam, 1983; Brant and Loker, 2009; Jouet, 2009) and experimental (McMullen and Beaver, 1945; Horak et al., 1999) data largely suggest duck host use is not constrained by physiological and/or immunological filters, suggesting a generalist association with anatid hosts. However, surveys in combination with molecular systematic data (Brant and Loker, 2009; Ebbs et al.,

2016) suggest some Trichobilharzia-host associations appear more specialized than expected based on host compatibility alone. Results from field collections show that

Trichobilharzia species often preferentially infect certain duck species or duck ecological groups. Here we compare phylogeographic, population genetic and demographic patterns of three North American congeners where with differing patterns of duck host use; T. physellae (TP), T. querquedulae (TQ), and

Trichobilharzia sp. A (TA).

100

We sampled TP, TQ and TA across the ranges of their hosts. TP occurs across

North America and is primarily associated with diving ducks, an ecological group of ducks (e.g. Aythya, Bucelphala, Mergus) characterized by their preference of large stable water bodies and diving feeding behavior (Baldassarre and Bolen, 1984). TQ occurs globally infecting a genus of dabbling ducks, Spatula (Gonzalez et al., 2009) specific to shallow water bodies. Based on current sampling, the host range of TA

(Brant and Loker, 2009) seems limited relative to TQ and TP, and thus far has been recovered only from the American wigeon (Anas americana), which is described as a generalist dabbling duck (Johnson and Greir, 1988).

Importantly, TP and TQ both utilize Physa spp. (Physidae) as their snail intermediate host, which unites these parasites in space, time and in contact with a similar sampling of duck host species. Physa spp. are abundant among different habitats across North America, and parts of South America (Wethington and

Lydeard, 2007). One species, P. acuta is globally invasive occurring on 6 of the 7 continents (Bousset et al., 2014; Ebbs et al. In press). Physa spp. are also more likely to be found in the preferred habitat of the duck host than not and thus increases their probability of transmission. TA utilizes Stagnicola elodes

(Lymnaeidae), which is more restricted ecologically and geographically relative to

Physa, and achieves highest densities at higher latitudes (or higher elevations) within North America. Generally, within digenetic trematodes there is a pattern of greater specificity in regards to the first intermediate host (snail), which seems to consistent within Trichobilharzia (Brant and Loker, 2009; Jouet, 2010). As such this study will focus on adult worms and their associations among duck host species and ecological groups.

101

We surveyed Trichobilharzia spp. extensively within North America and some

Southern Hemisphere localities to assess duck host associations and evaluate potential impacts on observed population genetic and demographic patterns.

Additionally, we discuss the microevolutionary consequences associated with different levels of ecological specificity, specifically in relation to Trichobilharzia persistence and diversification.

METHODS

Host-Parasite Survey Duck hosts were sampled globally (Table 1, Additional File 1). For the purposes of this study the authors examined 346 avian, primarily anatid, hosts

(Table 1). Additionally, we compiled results from several published Trichobilharzia – duck surveys to more robustly calculate host-range and make determinations of host specificity (Table 1). Combined survey (this study + published) data totaled 1,358 examined birds, across 94 species (73 anatids) and 7 countries. From these data we calculated several descriptive measures to capture duck host associations within

Trichobilharzia: prevalence, the proportion of infected hosts (Bush et al., 1997), observed host range, the number of host species infected by a parasite species and several specificity indices were calculated. As it can be assumed that parasites capable of infecting a broad taxonomic range tend towards generalism, relative to those infecting a single host genus the STD measures the taxonomic distinctness of all hosts species used by a single parasite species (Poulin and Mouillot, 1999). As rare hosts may disproportionately contribute a high STD (increased generalism) we also used the weighted specificity index (STD*) (Poulin and Mouillot, 2005). STD* is weighted by the parasites prevalence within each host. As members of

Trichobilharzia almost exclusively parasitize hosts within the family Anatidae, total

102 taxonomic distinctiveness among hosts is limited and these measures fail to capture the potential for ecological specificity, as such we developed a modified specificity index (STD-modified). Based on the Poulin and Mouillot (2005) index, we distinguished anatid hosts by ecological group rather than taxonomic group. Ecological groups were assigned following determinations from The Sibley Guide to Birds (Sibley,

2000). We approximated the degree of ecological specificity, calculated as the proportion of infections within each ecological group. Intensity and consequently abundance are important parameters and are included in most attempts to quantify specificity (Poulin and Mouillot 2005). Due to inherent difficulties of sampling

Trichobilharzia infrapopulations it is not feasible to obtain accurate measures of intensity, as it is unlikely that entire infrapopulations are ever recovered and intact worms are rare. As such, our measures are limited in this way.

To assess microevolutionary impacts of host specificity, correlation between measures of specificity and species wide estimates of nucleotide diversity (π) and Θ was tested by calculating Pearson’s correlation coefficients, as implemented in R

(http://www.R-project.org).

Hosts were either donated by hunters or obtained from the Museum of

Southwestern Biology, Bird Division (MSB) at the University of New Mexico

Albuquerque. All work with vertebrate hosts was conducted with the approval of the

Institutional Animal Care and Use Committee (IACUC) at the University of New

Mexico, USA (IACUC # 11- 100553-MCC, Animal Welfare Assurance # A4023-01).

Species of Trichobilharzia are parasites of the venous system and primarily reside in the mesenteric and hepatic portal veins. Mesenteric veins were inspected for adult worms with the aid of a dissecting microscope, and were removed using Vanna’s microscissors. Adult worms were recovered by perfusing the hepatic portal vein with

103 saline and then crushing the liver. Worms were then isolated in a series of decantation steps. Trichobilharzia samples were preserved in 95% ethanol for genetic and 80% ethanol for morphological assessment. In addition to the schistosome species recovered from the anatid hosts, co-occurring parasites (liver, kidney, heart and gastrointestinal tract) were also collected. All samples were deposited as vouchers in the MSB, Division of Parasites (Additional File 1).

DNA Extraction, PCR Amplification and Sequencing

Trichobilharzia DNA was extracted using the QIAmp DNA Micro Kit (Qiagen,

Valenicia, California, USA), the DNeasy Blood and Tissue Kit (Qiagen, Valenicia,

California, USA) or by HotShot Lysis (Truett et al., 2000). We amplified two mitochondrial and one nuclear gene region using primarily the Takara Ex Taq kit

(Takara Biomedicals, Otsu, Japan). For samples that were difficult to amplify, we used either the GoTaq Flexi (Promega) or the Platinum Taq kit (Invitrogen) using 0.4 ul of Taq and 4ul of 2.5 mM MgCl2 per reaction. We sequenced a 743 bp (5’ end) region of the cytochrome oxidase I gene (cox1), a 409 bp (5’ end) region of the

NADH 4 gene (nad4) and 733 bp of the internal transcribed spacer region 1 (ITS1, including the 3’ end of 18S rRNA and the 5’ end of 5.8S). PCR protocols and primers used were identical to those used by Brant and Loker (2009) (cox1, ITS1) and Ebbs et al. (2016) (nad4).

Sequencing reactions were performed using the BigDye 3.1 sequencing kit

(Applied Biosystems, Foster City, California, USA). Sequences were edited using

Sequencher 5.3 (Gene Codes Corporation, Ann Arbor, MI, USA) and aligned using

ClustalW (Larkin et al., 2007) and adjusted manually as needed. Sequences generated from this study are accessioned in the NCBI Gen Bank database.

Phylogenetic Analyses

104

To evaluate intraspecific relationships and population structure phylogenetic analyses of Clade Q (sensu Brant and Loker, 2009) were carried out. Phylogenetic datasets were analyzed by gene (cox1, nad4, ITS1) and concatenated (cox1+ nad4

+ITS1). Mitochondrial genes (cox1 and nad4) are well suited for intraspecific studies as there is no recombination and consequently have a higher sorting rate(Avise,

1987). Part of the nuclear rRNA complex, ITS1 was chosen to evaluate potential mito-nuclear discordance. Appropriate models of nucleotide substitution were selected using JModelTest2 (Darriba et al., 2012)based on the Akaike information criterion (AIC (Akaike, 1974)).

Pairwise uncorrected p-distances were calculated in MEGA 6 (Tamura et al.,

2013)within and between the three species of Trichobilharzia to determine the in- group for all subsequent intra-specific analyses. A cut off of ≤5% divergence for cox1 was chosen to delineate each in-group. In-group datasets included; TQ ncox1= 63, nnad4=41, TP ncox1= 27, nnad4=14, TA ncox1= 12, nnad4=9.

We assessed haplotype relationships in our mitochondrial genes for the in- group of TP, TQ and TA using minimum spanning network analysis in the program

POPart ((Leigh and Bryant, 2015) http://popart.otago.ac.nz/). Haplotypes were coded by migratory flyway and host species. Migratory flyways are common way of partitioning bird populations (Swanson et al., 1985) and their symbionts (Pearce et al., 2009; Lam et al., 2012; Muñoz et al., 2013). Four North America flyways (Pacific,

Central, Mississippi and Atlantic) have been defined; each were sampled for adult

Trichobilharzia for this study. This study chose to focus primarily on adult worms, as they represent the point of gene flow within the Trichobilharzia life cycle. However, cercarial stages of TP and TA were included in some analyses to increase sample size. Unique haplotypes differed by at least one base pair. All samples collected

105 outside of North America were collected from the Southern Hemisphere and were pooled together because of the small sample size.

Genetic diversity and population structure

To estimate the genetic diversity and population genetic structure of the TP,

TQ and TA infrapopulations, samples were grouped by the state or province the host was collected from and then regionally by the migratory flyway from which the host was collected. Indices of genetic diversity for each species and their respective populations were estimated in DNAsp v. 5 (Librado and Rozas, 2009) for both cox1 and nad4; number of polymorphic sites (S), average number of nucleotide differences (K) and nucleotide diversity (π). Uncorrected p-distances were calculated between flyways, regions and host species in Arlequin 3.5 (Excoffier and Lischer,

2010)

One-way AMOVAs (Excoffier et al., 1992) were performed in Arelquin v. 3.5

(Excoffier and Lischer, 2010) to estimate overall population genetic structure (cox1).

A set of population genetic hypotheses was tested for each species. (1) Where infrapopulations were grouped by host locality, and then regionally by flyway. For the

TQ dataset AMOVAs were run both with the inclusion and exclusion of Southern

Hemisphere populations. (2) We tested for the effect of host species on population structure. (3) We tested for the effect of latitude by separating populations into high vs. low latitude groups rather than the longitudinal groups designated by flyway. ΦST estimates variation from all samples across all localities sampled, ΦCT estimates variation among flyways, host species, or latitudinal group and ΦSC estimates variation among localities within the flyways, host species, and latitudinal groups. We also estimated pairwise ΦST (Reynolds et al., 1983) by locality, flyway and host

106 species. Significance (P<0.05) was determined by permutation tests of 10,000 random permutations.

To test for isolation by distance patterns we constructed a geographic distance matrix by converting geographic coordinates to Euclidean distances between localities as implemented in Primer 7 (Clarke and Gorley, 2015). Correlation between geographic distances and both uncorrected p-distances and pairwise ΦST matrices were calculated using Spearman’s rank correlation also in Primer 7.

Demographic analyses

To assess contemporary and historical demographic patterns, Watterson’s estimator (Θw) and Tajima’s D (D), were calculated (cox1) and assessed statistically using coalescent simulations with a 95% confidence interval and 10,000 permutations (Table 1) in DNAsp (Librado and Rozas, 2009).

Changes in Ne over time were estimated by fitting a Bayesian Skyline demographic model (BSP ( Drummond, et al., 2005)) for each of the three species of interest based on cox1 *BEAST v. 1.7.0 (Drummond et al., 2012). The HKY substitution model was used for two simultaneous MCMC runs for 10,000,000 iterations sampling every 1,000 steps. Mutation rates of 2% and 4% change per million year were used to estimate mitochondrial genome evolution for schistosomes

(Schistosoma mansoni, (Morgan et al., 2005)). As substantial divergence within TA, between the two flyways it was recovered from (Pacific and Central), was found these populations were analyzed separately. Convergence was checked (ESS of

200 or greater) and results visualized using Tracer v.1.5 (Rambaut et al., 2014

Tracer v1.6. http://beast.bio.ed.ac.uk/Tracer). BSP data from each analyzed datasets generated in Tracer v. 1.6 was exported and visualized in Microsoft Excel for

107 comparison.

Genetic Diversity of Trichobilharzia

Genus wide genetic diversity was estimated for the purposes of assessing the relationship between diversity (π, K and Θw) and measures of host specificity. The cox1 sequences of other Trichobilharzia species published in the NCBI database were used to estimate phylogenetic relationships (TN93+G+1) and patterns of genetic diversity across the genus, using methods described previously.

Anserobilharzia brantae and Allobilharzia visceralis were selected as out groups

(Brant and Loker, 2013). Pairwise uncorrected p-distances were calculated within and between each species in MEGA 6 (Tamura et al., 2013).

RESULTS

Trichobilharzia survey data

Schistosomes from a total of 1,358 birds were included in this study.

Specimens representing Trichobilharzia were only recovered from hosts within the family Anatidae. Based on our examinations and those within the literature we found

464 hosts (out of 1,358, 34.2%) to be infected with at least one species of

Trichobilharzia. Co-infections were common, (Table 1), and were probably underestimated as not all individuals were assessed genetically. Of the 73 anatid species examined, 29 were found to be infected with Trichobilharzia (Table 1)

Using three different measures of host specificity, we found that TQ, TA, T. stagnicolae, T. mergi, and T. anseri all showed evidence of duck host specificity (STD values of 1). Using conventional indices based on taxonomic distinctiveness (Poulin and Mouillot, 1999; 2005) TP and T. regenti had intermediate to high measures of

108

STD, indicating generalism. When a modified indexing method was applied to capture ecological distinctiveness, TP has a lower STD-modified value while T. regenti reached the maximum value (STD-modified = 3, despite having a fewer number of observed host

* species. It is important to note that STD and STD were calculated under the recommendations of (Poulin and Mouillot, 1999; 2005) and only incorporate two taxonomic levels, consequently STD values can only vary between 1 and 2, where two is maximum generalism. The STD-modified value vaired across four ecological groups (divers, dabblers, sea ducks, geese/swans) and can vary between 1 and 4, therefore the STD-modified of TP is moderate. Only STD was calculated for T. franki as prevalence values were ambiguously reported due to co-infection with other

Trichobilharzia spp. and could not be included within these calculations. Further there are known cryptic lineages within the T. franki complex (Jouet et al., 2010), which could skew our measure of host specificity, for these reasons we did not make any determinations regarding host associations within T. franki.

Approximating ecological specificity as the proportion of infections occurring within a single duck ecological group, we determined 6 species to have ecological specificity (TQ, TP, TA, T. stagnicolae, T. mergi, and T. anseri), as over 90% of their infections were associated with one duck species or ecological group, while T. regenti was found to be a generalist among anatids. It is likely that T. szidati also falls within this category, based on species reports, however a lack of published prevalence data (from duck hosts) did not allow its inclusion within these calculations. We estimate that 93%, 100% and 100% of TP, TQ and TA, respectively, infections occur within a single duck ecological group, and consequently utilize a narrow selection of their potential host range (Anatidae). The underlying mechanisms of this observed specificity are yet to be determined and will

109 require experimental infections to confirm the physiological boundaries of

Trichobilharzia host use. Several undescribed Trichobilharzia lineages resulted from our survey data but were excluded from our calculations of host range and ecological specificity due to low sampling.

For the species of interest, our data supports that TP infects multiple genera of diving ducks (Aythya, Bucephala, Mergus) and was recovered from 8 species within 5 genera. The highest prevalence’s were recorded from Aythya (A. affinis,

27%, A. collaris, 26%, A. valisineria, 20%), which were determined to be core host species, as 93% of all TP infections were found within these three host species

(Bush et al., 1997). The lowest prevalence values were found within Anas platyrhynchos (12.5%), Clangula hyemalis (10%), Mergus merganser (10%),

Bucephala albeola (7.7%), and Anas strepera (2.8%), which were determined to be satellite (Bush et al., 1997) host species. In concordance with previous studies

(Brant and Loker, 2009; Ebbs et al., 2016), TQ was found almost exclusively (98.3%) from Spatula, or the ‘blue-wing’ duck group (Johnson and Sorenson, 1999), with a single report (2.7%) from an American wigeon (Anas americana). Prevalence was highest within Spatula discors (90%), but all Spatula species examined were infected at a rate of ≥74%. Trichobilharzia sp. A was recovered exclusively (100%) from the

American wigeon (Anas americana), having the narrowest host range of the three species, with a prevalence 27%. Anas americana was found to harbor the highest

Trichobilharzia species diversity of any of the host species examined, 6

Trichobilharzia species were recovered, compared to an average of 1.5

Trichobilharzia species per host species within North America (1.8 per host species globally).

110

Correlation of observed host range, indices of host specificity and species wide estimates of genetic diversity (π) and Θ was evaluated by calculating Pearson’s correlation coefficients. There was no correlation with observed host range and π (R

= -0.165) or Θ (R=0.147). All indices of host specificity were at least moderately supported with π (R, STD= -0.571; R, STD*= -0.4995; R, STD-modified = -0.6077) and to a lesser extent Θ (R, STD= -0.4375; R, STD*= -0.329; R, STD-modified = -0.4424).

Suggesting that ecological specialists were correlated with higher values of π and Θ.

Phylogeographic patterns among congeners

TRICHOBILHARZIA QUERQUEDULAE: Phylogenetic analysis of TQ supports its global distribution (Ebbs et al. 2016, Figure 1) within Spatula.

Mitochondrial data suggests that several individuals (W345.3, W750.2 and W771,

~4.7% divergent) fall basal to the main TQ clade (Figure 1, Additional File 3).

Genetic distances were low enough to support the inclusion of these samples within

TQ in-group analyses, however this subclade may warrant further investigation as a putative cryptic species, or divergent subpopulation. Interestingly, all individuals were recovered from S. clypeata and primarily from northern latitudes. Apart from the three samples in question, minimal phylogenetic structure was observed within TQ with no pattern suggesting structure by geography or host use based on either ML or

BI analyses (Figure 1, Additional Files 3-5).

Minimum spanning networks (Figure 2) of cox1 revealed high amounts of haplotype diversity within TQ, suggesting genetically diverse and well-connected populations. Infrapopulations (within host) haplotype diversity was similar to component population (across hosts) haplotype diversity, suggesting infrapopulation recruitment occurs randomly over time and space (Nadler, 1995; Sire et al., 2001).

TQ haplotypes did not partition according to flyway, and Southern Hemisphere

111 haplotypes did not group together and were not more or less divergent than Northern

Hemisphere haplotypes. No evidence of genetic structure in association with host species by was found. An AMOVA testing for structure by latitude was found to be significant, indicating possible population genetic differentiation between breeding and wintering range.

TRICHOBILHARZIA PHYSELLAE: Two clades (TP1, TP2) were recovered within TP based on cox1 analysis (Figure 1), taxa grouped within each subclade for nad4 and cox1 + nad4 + ITS1 analyses (Additional Files 3-5), but subclades were not supported statistically. Despite limited geographic sampling, representation within the two subclades might suggest a distinction between eastern and western populations of TP. Similar to TQ, minimum spanning networks of cox1 suggest genetically diverse and well-connected populations (Figure 2). For TP two haplogroups were recovered, equivalent to clades TP1 and TP2 in the mitochondrial gene tree analyses. Haplotypes within TP1 were recovered from 3 of the 4 flyways.

TP2 was also recovered 3 of the 4 flyways, however 12/16 individuals were recovered from the Central flyway. Networks were additionally coded by high versus low latitude and host species, and host species, which provided no structure to haplotype relationships. An AMOVA testing for structure by flyway was found to be significant, indicating population genetic differentiation in association with migratory flyway.

TRICHOBILHARZIA SP. A: Mitochondrial and concatenated datasets support the monophyly of TA and an affinity to group with T. franki and Trichobilharzia sp. B

(Additional File 5). However, ITS1 analysis (Additional File 4) suggests that TA is paraphyletic with TP, T. franki, and Trichobilharzia sp. B. This result could suggest mito-nuclear discordance within the TA dataset that was not found within either TP or

112

TQ. A single specimen (W249) grouped with support within Trichobilharzia sp. C, rather than TP in cox1 (Figure 1). Geographical sampling was too limited to draw phylogeographic inferences for a minimum spanning network of TA haplotypes.

No supported sister group relationships within Clade Q were found. BI of mtDNA supported Trichobilharzia sp. B as sister to Trichobilharzia sp. A, and both lineages infect American wigeon (Anas americana), though the snail host to

Trichobilharzia sp. B is currently unknown. This however was not found in the analysis of ITS1 or for the concatenated dataset, cox1+nad4+ITS1. The cox1 analysis showed a South American Trichobilharzia lineage (W701) transmitted by a physid snail (Pinto et al., 2014) as a well-supported sister clade to T. physellae, however this relationship was not found in the ITS1 gene tree (Figure 1, Additional

File 4). No gene tree supported T. regenti or any other Trichobilharzia lineage as the root of Clade Q (Figure 4).

Comparative population genetic structure

TRICHOBILHARZIA QUERQUEDULAE: Hierarchal AMOVA was used to test multiple hypothesis of population genetic structure (Table 3). When structure by flyway was tested, ΦSC (among populations within flyways) and ΦST (within populations) were found to be marginally significant (p= 0.0528 ± 0.00619 and p=

0.0439 ± 0.00594, respectively) while ΦCT was found to be non-significant indicating population structure is not mediated by migratory flyway. A second AMOVA was performed excluding non-North American populations, to determine if Southern

Hemisphere populations were driving observed structure. Results from this analysis

(results not shown) were concordant with the total AMOVA. Instead of flyways, populations were partitioned by latitudinal group (high vs. low) as a proxy for genetic structure associated with breeding vs. wintering range. Analyses including (ΦCT=

113

0.0441, p= 0.00196) and excluding (ΦCT= 0.0072, p=0.0332, results not shown)

Southern Hemisphere populations support latitude in structuring TQ populations.

Further significant population structure was found at all levels (Table 3) when latitude was considered. No significant structure by host species was found. Pairwise ƟST values among flyways are summarized in Additional File 9. When all populations within a flyway were pooled no significant differentiation between flyway was found within TQ. However, all North American flyways were found to be distinct from the

Southern Hemisphere populations. Pairwise ƟST between localities within flyways and by host species were also measured (results not shown) which identified TQ collected from Manitoba and New Zealand as the most distinct populations.

Interestingly, TQ populations collected from different host species (Spatula clypeata vs. S. cyanoptera) from the same locality (New Mexico) were found to be significantly distinct (p=≤0.01). A Spearman rank test was performed and correlation between geographic distance and both pairwise ƟST and uncorrected p-distance was found to be weak (ρ= -0.11 and ρ= 0.002, respectively) suggesting no relationship between geographic distance and genetic diversity/structure.

TRICHOBILHARZIA PHYSELLAE: AMOVA by flyway indicated significant differentiation among flyways was found (ΦCT= 0.55592, p= 0.02737) (Table 3), supporting the population genetic hypothesis that flyways structure TP populations.

When populations were grouped according to latitudinal group no significant population genetic structure was observed. When TP individuals were pooled by host species they were collected from and then partitioned by host genera a significant ΦST was recovered (p= 0.00293), however, neither the ΦSC nor ΦCT were significant suggesting no real influence of host species. The TP dataset is more limited in terms of sample size, yet significant differentiation was found between

114

several flyways (Additional File 9). Pairwise ΦST between localities within flyways and by host species were also measured (results not shown) which identified that TP recovered from New Mexico were distinct from both Alaskan (ƟST= 0.2952, p<0.001) and Michigan (ΦST= 0.24159, p<0.001) populations. The Mississippi and Atlantic flyways harbored the highest within flyway divergence values, 1.2% and 0.95% respectively in comparison to the Central (0.01%) and Pacific (0.048%). There was no evidence that host species influenced population genetic structure.

Correlation between geographic distance and uncorrected p-distance was moderate (ρ= 0.473) indicating a relationship between genetic diversity and geography. When geographic distances were correlated with pairwise ΦST values no relationship was found (ρ= -0.025).

TRICHOBILHARZIA SP. A: Since TA was only recovered from two flyways a hierarchal AMOVA was not performed. Pairwise ΦST between the two flyways sampled (Pacific and Central) was significant (ΦST=0.3283, p <0.001). High intra- flyway divergence was recovered within the Central flyway (1.4%) relative to the

Pacific (0.36%) with an average p-distance of 1.38% between the two flyways. Due to limited geographic sampling a Spearmen rank test was not performed.

Genetic diversity and demographic analyses

Haplotype diversity was similar across TP, TQ and TA, however nucleotide diversity (π) and Ɵ were substantially different across the three species (Table 4).

Historical effective size was estimated by fitting a Bayesian Skyline model

(Drummond). Figure 3 shows estimates of Ne over time (kya) for TP, TQ, and TA based on a 4% change/million years (Morgan, 2005), a lower rate of 2% was also tested (Additional File 8). For TP and TQ estimates of Ne were almost double under the 2% rate. Both TQ and TP indicate a recent population expansion, further

115 supported by significant positive Tajima’s D values (Table 4). Over coalescent time the effective size of TQ was approximately 3X greater than TP, suggesting this species has a history of maintaining larger genetically diverse populations. Despite having greater π than TQ, TA exhibited the smallest effective population size of the three species with no evidence for notable historical demographic changes.

Genetic diversity within Trichobilharzia.

Using our data combined with published cox1 sequence for other

Trichobilharzia spp. in the NCBI database we estimated species wide genetic diversity across Trichobilharzia (Table 4). Of the five additional species included, none has comparable nucleotide diversity to the three species of interest, with the exception of T. stagnicolae (π = 0.01326). Similarly, estimates of Ɵ were highest among TQ, TP, TA and T. stagnicolae. Observed genetic diversity was highest among species with high measures of ecological specificity. Species with the broadest host range in terms of number of species and taxonomic distinctness

(Poulin and Mouillot, 2005) had the lowest estimates of π and Ɵ (Figure 4).

DISCUSSION

A primary goal of this study was to identify ecological factors (habitat preference, migratory behavior, and host range) that are important to the microevolution of

Trichobilharzia, specifically under conditions of partially overlapping host ecology.

Remarkably little is known regarding the ecological determinates of Ne, π and Ɵ or how genetic drift operates within complex, widespread parasite systems (Criscione et al., 2005). These parameters have vital implications for disease ecology and parasite control, as they determine how alleles (e.g. resistance alleles) move within and between populations. Ɵ is often described as a measure for adaptive potential

(Criscione and Blouin, 2004; Gandon et al., 2008; Archie and Ezenwa, 2011) as

116 such it may indicate a species genetic flexibility. In the case of Trichobilharzia spp. it can be hypothesized that TQ has a greater adaptive potential than TP, and theoretically populations may be more robust to environmental change (e.g. host extinction, climate change).

Comparative population genetic analyses of TQ, TP and TA revealed differences in population structure, genetic diversity and demographic histories.

Specifically, the results suggest that TQ is a common and genetically diverse schistosome of Spatula spp. across four continents, with population structure likely associated with duck migratory ranges, such that parasite population structure is associated with latitude as a proxy for breeding and wintering grounds. No deeper phylogeographic structure was found within TQ despite its global distribution.

Historical effective population sizes (Ne) of TQ were at least 3X greater than TP and

TA. Contemporary regional measures of π and Ɵ were high across the range of TQ, including Southern Hemisphere populations, and largely matched species wide and infra-population measures. In contrast, species wide TP maintains lower π and Ɵ.

Overall geography plays a more important role in the genetic structure of TP relative to TQ. Phylogenetic analysis suggests the existence of two clades, possibly separated by the Eastern and Western part of the range. Hierarchal AMOVA (Table

3) and pairwise ΦST values (Additional File 9) support geographic structuring, potentially in association with migratory flyway, within TP, which would support an east/west divide. It is important to remember, both TQ and TP rely on the same snail intermediate hosts (Physa spp.), and are consequently united ecologically in time and space, as such it is at the definitive host level where TQ and TP partition themselves among different ecological host groups. We report a 63% difference in the prevalence of TQ and TP across all hosts, and a 64% difference between the

117 single most common host of each species, suggesting a stark different in transmission success. It can be hypothesized that as both species infect Physa snails, the habitats that Spatula spp. (TQ) confer a greater probability of transmission relative to Aythya spp. (TP). While experimental infections are necessary to rule out inherent differences in host susceptibility, these results may imply that certain combinations of hosts and ecologies facilitate the maintenance of high parasite diversity and Ne.

This study sought to characterize microevolutionary patterns among three congeneric trematode species, to evaluate how hosts with partially overlapping ecologies might impact parasite population genetic structure and diversity.

Substantial differences in population genetic structure and demographic parameters among TQ, TP and TA were found, and further patterns of genetic diversity and ecological specificity varied across Trichobilharzia. In total this data found that parasite microevolutionary patterns can vary with host-ecology among ecologically specific and overlapping parasite species, specifically in terms of host habitat preference and site fidelity. Here we discuss several potential reasons why and associated consequences.

Infrapopulation diversity of TQ was equal to overall component population diversity and the number of unique haplotypes within infrapopulations increased linearly with the number of individuals sampled. This suggests that infrapopulation recruitment (transmission) is random, likely occurring across time and space and a well-mixed component population. Notably, the majority of worms sampled were males due to the difficulty of recovering females from the duck host. The high level of infrapopulation diversity observed suggests that duck hosts sampled maintain a genetically diverse pool of male worms.

118

In contrast to TQ, infrapopulations of TP appear to be less divergent, possibly suggesting that recruitment is biased regionally, perhaps associated with the overall great population subdivision observed within TP relative to TQ. This result warrants further investigation as it might indicate the potential for differential intraspecific transmission dynamics and ultimately local adaption within TP. Infrapopulation sampling of TA was insufficient to address population recruitment, however notably multiple infrapopulations were sequenced from American wigeon (Anas americana) which revealed Trichobilharzia species diversity unequaled by any of the other host species sampled. Our data suggests that A. americana hosts at least 5

Trichobilharzia species, 4 of which are currently undescribed. Anas americana is considered to be one of the most generalists species within Anas (Johnson and

Grier, 1988) in terms of its habitat preferences and thus it may be more likely to encounter a variety of different snail species and their associated parasites. It can be hypothesized that these generalist behaviors drive the high species diversity we observe.

The hosts of TQ (Spatula clypeata, S. discors and S. cyanoptera in North

America) have a broad migratory range and are known to be locally opportunistic.

Habitat quality is the primary determinant of site selection for these species; consequently ducks move locally to find optimal habitat, increasing their probability of interacting with infected snails among different habitats (Johnson and Grier, 1988;

Osborn et al., 2017). Additionally, these ducks prefer shallow marsh habitat (Osborn et al., 2017). These behaviors result in Spatula spp. spending a majority of their time in Physa rich habitat, increasing the likelihood of encountering Trichobilharzia spp. In contrast, North American Aythya spp. are highly philopatric, and occupy large stable water bodies, where they are less likely to encounter snails. Habitat quality is not a

119 primary factor in site selection and consequently local movement of Aythya spp. is limited (Johnson and Grier, 1988; Austin et al., 2016;). These behaviors in combination decrease the likelihood of Aythya spp. to encounter locally available

Trichobilharzia. It can also be predicted that philopatric host species are more likely to facilitate population genetic structure among their parasites, specifically in relation to flyway. Which is a pattern we observed in TP.

Interestingly the American wigeon (A. americana), a dabbling duck, is one the most generalist species within Anas (Johnson and Grier, 1988; Sibley, 2000). Anas americana are known to spend the most time on land grazing relative to other Anas spp., they can also be found in flowing river waters and brackish estuarine habitats.

Seasonally, A. americana individuals may occur in a diversity of habitat types, which likely increases the probability of encounter to different species of Trichobilharzia transmitting snails. We hypothesize the habitat generalism described for A. americana drives the remarkably high Trichobilharzia lineage diversity recovered within this species.

A secondary goal of this study was to attempt to quantify the nature of duck host associations within Trichobilharzia (i.e. specialists= one host genera; generalists= multiple host genera; ecological specialists= multiple host genera within an ecological group; ecological generalists= multiple host genera that span multiple ecological groups). The host range of a parasite (i.e. host specificity) is conceptually determined by two filters the encounter and/or compatibility filter (Combes, 1991).

Prior studies suggest that Trichobilharzia are largely compatible with anatids

(waterfowl) and even some non-anatids. McMullen and Beaver (1945) demonstrated experimentally TP could develop successfully in non-anatid birds (e.g canaries and pigeons). Consensus regarding Trichobilharzia definitive host specificity would

120 suggest, under a traditional definition of host specificity, generalism. However, a more systematic approach to experimentally defining duck host compatibility is needed. In contrast, extensive field surveys and molecular systematic investigations

(Brant and Loker 2009, Ebbs et al. 2016, this study) have made clear that several

Trichobilharzia species (TQ, TP, TA and T. stagnicolae) are abundant in one or more core duck species (or ecological group), and remarkably rare in other potential

(satellite) hosts species. Based the published experimental work discussed above it might be predicted that immunological or evolutionary constrains (compatibility) are unlikely, which would suggest that the encounter filter is narrowed by features specific to hosts ecology (i.e ecological partitioning, temporal partitioning). Our data suggests host specificity of Trichobilharzia spp. with either dabbling or diving ducks is common. Divers and dabblers represent ecologically distinct groups which may be co-distributed geographically but partition themselves ecologically by habitat preferences and feeding behaviors (Baldassarre and Bolen, 1984), making them at least seasonally allopatric (Austin et al., 2016). Juvenile and adult diving ducks spend the majority of their time in deep water, however nesting sites occur near shore putting breeding pairs and chicks in parapatry with dabbling ducks (Austin et al., 2016).

In sum we investigated population genetic patterns of three congeneric

Trichobilharzia species which have a broad range of potential, sympatric or parapatric, anatid hosts. We found evidence suggesting ecological specificity among some Trichobilharzia species, and that specific host ecological traits may act within conspecifics to reduce gene flow across potential host species resulting in population genetic structure that co-varies with host traits. Species exhibiting ecological specificity had on average higher measures of π and Ɵ relative to species infecting a

121 broader ecological range of potential hosts. Our data suggests that host ecological traits mediate substantial differences in Trichobilharzia population genetic structure and dynamics and relates to the adaptive potential of individual species.

ACKNOWLEDGMENTS Thank you to the following for invaluable help collecting birds—Museum of

Southwestern Biology Bird Division, Mr. Andy Johnson, Dr. Ernest Valdez, Dr.

Robert W. Dickerman and Dr. Chris Witt; National Museum Bloemfontein: Mr. Rick

Nuttall and Mr. Dawid H. de Swardt.: Illinois Natural History Survey, Dr. Heath Hagy and Mr. Josh Osborne; Dr. Vasyl Tkatch, Dr. Stephen Greimen; Dr. Sean Locke.

Individual duck hunters: Mr. Brad Bothner, Mr. Josh Butler, Mr. Tony Gilyard, Dr.

James Mckeehen and Mr. Bob Mckeehen, Mr. Matthew Snyder. Dr. Veronica Flores.

Individuals who assisted with host necropsies: Mr. Keith Keller and Ms. Emily Savris.

We acknowledge technical support from the University of New Mexico Molecular

Biology Facility, which is supported by National Institutes of Health, USA grant

P30GM110907 from the Institute Development Award program of the National

Center for Research Resources, USA. This work was primarily funded through a

National Science Foundation, USA grant to S.V.B. (DEB 1021427) and a National

Institutes of Health grant RO1 AI44913 to E.S.L.

REFRENCES

Akaike, H., 1974. A new look at the statistical model identification. IEEE Trans. Automat. Contr. 19, 716–723. doi:10.1109/TAC.1974.1100705

Bush, A.E., 1997. Parasitology meets ecology in its own terms: Margolis et. al. revisited. Journal of Parasitology 84, 575–583.

122

Austin, J.E., O'Neil, S.T., Warren, J.M., 2016. Habitat selection by postbreeding female diving ducks: influence of habitat attributes and conspecifics. Journal of Avian Biology 48, 295–308. doi:10.1111/jav.01063

Avise, J., 1987. Intraspecific Phylogeography: The Mitochondrial-DNA Bridge Between Population Genetics And Systematics. Annual Review of Ecology and Systematics 18, 489–522. doi:10.1146/annurev.ecolsys.18.1.489

Baldassarre, G.A., Bolen, E.G., 1984. Field-Feeding Ecology of Waterfowl Wintering on the Southern High Plains of Texas. The Journal of Wildlife Management 48, 63. doi:10.2307/3808453

Barrett, L.G., Thrall, P.H., Burdon, J.J., Linde, C.C., 2008. Life history determines genetic structure and evolutionary potential of host–parasite interactions. Trends in Ecology & Evolution 23, 678–685. doi:10.1016/j.tree.2008.06.017

Blair, K.S., Islam, 1983. The life-cycle and morphology of Trichobilharzia australis n.sp. (Digenea: Schistosomatidae) from the nasal blood vessels of the black duck (Anas superciliosa) in Australia, with a review of the genus Trichobilharzia. Systematic Parasitology 5, 89–117.

Blasco-Costa, I., Poulin, R., 2013. Host traits explain the genetic structure of parasites: a meta-analysis. Parasitology 140, 1316–1322. doi:10.1017/S0031182013000784

Blouin, M.S., Yowell, C.A., Courtney, C.H., Dame, J.B., 1995. Host Movement and the genetic structure of Populations of parasitic nematodes. Genetics 141, 1007–1014.

Brant, S.V., Loker, E.S., 2013. Discovery-based studies of schistosome diversity stimulate new hypotheses about parasite biology. Trends in Parasitology 1–11. doi:10.1016/j.pt.2013.06.004

Brant, S.V., Loker, E.S., 2009. Molecular Systematics of the Avian Schistosome Genus Trichobilharzia (Trematoda: Schistosomatidae) in North America. Journal of Parasitology 95, 941–963. doi:10.1645/GE-1870.1

Clarke, K.R., Gorley, R.N., 2015. PRIMER v7 user manual. PRIMER-E: Plymouth.

Combes, C., Théron, A., 2000. Metazoan parasites and resource heterogeneity: constraints and benefits. International Journal for Parasitology 30, 299–304.

Criscione, C., Blouin, M., 2004. Life cycles chape parasite evolution: Comparative population genetics of Salmon Trematodes. Evolution 58, 198–202.

123

Criscione, C.D., Poulin, R., Blouin, M., 2005. Molecular ecology of parasites: elucidating ecological and microevolutionary processes. Molecular Ecology 14, 2247–2257.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models, new heuristics and parallel computing. Nature methods.

Dawson, M.N., Louie, K.D., Barlow, M., Jacobs, D.K., Swift, C.C., 2002. Comparative phylogeography of sympatric sister species, Clevelandia ios and Eucyclogobius newberryi (Teleostei, Gobiidae), across the California Transition Zone. Molecular Ecology 11, 1065–1075.

Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian Phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution 29, 1969–1973. doi:10.1093/molbev/mss075

Drummond, R.A., Shapriro, B., Ho, S. 2005. Bayesian Coalescent Inference of Past Population Dynamics from Molecular Sequences. Molecular Biology and Evolution 22, 1185–1192.

Dybdahl, C.M, Lively, C. 1996. The Geography of Coevolution: Comparative Population strucutres for a snail and its trematode parasite. Evolution 50, 2264– 2275.

Ebbs, E.T., Loker, E.S., Davis, N.E., Flores, V., Veleizan, A., Brant, S.V., 2016. Schistosomes with wings: how host phylogeny and ecology shape the global distribution of Trichobilharzia querquedulae (Schistosomatidae). International Journal for Parasitology 46, 669–677. doi:10.1016/j.ijpara.2016.04.009

Gandon, S., Buckling, A., Decaestecker, E., Day, T., 2008. Host-parasite coevolution and patterns of adaptation across time and space. Journal of Evolutionary Biology 21, 1861–1866. doi:10.1111/j.1420-9101.2008.01598.x

Gonzalez, J., Düttmann, H., Wink, M., 2009. Phylogenetic relationships based on two mitochondrial genes and hybridization patterns in Anatidae. Journal of Zoology 279, 310–318. doi:10.1111/j.1469-7998.2009.00622.x

Gorton, M.J., Kasl, E.L., Detwiler, J.T., Criscione, C.D., 2012. Testing local-scale panmixia provides insights into the cryptic ecology, evolution, and epidemiology of metazoan animal parasites. Parasitology 139, 981–997. doi:10.1017/S0031182012000455

124

Hoberg, E.P., Brooks, D.R., 2008. A macroevolutionary mosaic: episodic host- switching, geographical colonization and diversification in complex host-parasite systems. Journal of Biogeography 35, 1533–1550. doi:10.1111/j.1365- 2699.2008.01951.x

Horak, P., Dvořák, J., Kolarova, L., trefil, L., 1999. Trichobilharzia regenti, a pathogen of the avian and mammalian central nervous systems. Parasitology 119, 577–581. doi:10.1017/s0031182099005132

Huyse, T., Audenaert, V., Volckaert, F.A.M., 2003. Speciation and host-parasite relationships in the parasite genus Gyrodactylus (Monogenea, Platyhelminthes) infecting gobies of the genus Pomatoschistus (Gobiidae, Teleostei). International Journal for Parasitology 33, 1679–1689.

Johnson, D.H., Grier, J.W., 1988. Determinants of Breeding Distributions of Ducks. Wildlife Monographs 1–37.

Jouet D, Skírnisson K, Kolářová L, Ferte H., 2010. Molecular Diversity of Trichobilharzia franki in two intermediate hosts (Radix auricularia and Radix peregra): A complex of species. Infection, Genetics and Evolution 10, 1218– 1227.

Jouet, D.E.A., 2009. Avian schistosomes in French aquatic birds: a molecular approach. Journal of Helminthology 83, 181–189.

Johnson, K., Sorenson, M.,1999. Phylogeny and biogeography of dabbling ducks (genus: Anas): A comparison of molecular and morphological evidence. The Auk 116, 792–805.

Lam, T.T.Y., Ip, H.S., Ghedin, E., Wentworth, D.E., Halpin, R.A., Stockwell, T.B., Spiro, D.J., Dusek, R.J., Bortner, J.B., Hoskins, J., Bales, B.D., Yparraguirre, D.R., Holmes, E.C., 2012. Migratory flyway and geographical distance are barriers to the gene flow of influenza virus among North American birds. Ecol. Lett. 15, 24–33. doi:10.1111/j.1461-0248.2011.01703.x

Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., 2007. Clustal W and Clustal X version 2.0.

Leigh, J.W., Bryant, D., 2015. POPART: full‐feature software for haplotype network construction. Methods in Ecology and Evolution 6, 1110–1116. doi:10.1111/2041-210X.12410

Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics.

McMullen, D.B., Beaver, P.C., 1945. Studies on schistosome dermatitis. American Journal of Epidemiology 42, 128–154. doi:10.1093/oxfordjournals.aje.a119030

Morgan, J.A., Dejong, R.J., Adeoye, G.O., Ansa, E., Barbosa, C.S., Brémond, P.,

125

Cesari, I.M., Charbonnel, N., Corrêa, L.R., Coulibaly, G., D’andrea, P.S, Loker, E.S. 2005. Origin and Diversification of the the Human parasite Schistosoma mansoni. Molecular Ecology 14, 3889–3902.

Muñoz, J., Amat, F., Green, A.J., Figuerola, J., Gómez, A., 2013. Bird migratory flyways influence the phylogeography of the invasive brine shrimp Artemia franciscana in its native American range. PeerJ 1, e200. doi:10.7717/peerj.200

Nadler, S., 1995. Microevolutoin and the Genetic Structure of Parasite Populations. Journal of Parasitology 81, 395–403.

Nieberding, C., Morand, S., Libois, R., Michaux, J.R., 2006. Parasites and the island syndrome: the colonization of the western Mediterranean islands by Heligmosomoides polygyrus (Dujardin, 1845). Journal of Biogeography 33, 1212–1222. doi:10.1111/j.1365-2699.2006.01503.x

Nuismer, M., Gandon S. 2009. Interactions between Genetic Drift, Gene Flow and Selection Mosaics Drive Parasite Local Adaptation. The American Naturalist 173.

Osborn, J.M., Hagy, H.M., Mcclanahan, M.D., Davis, J.B., Gray, M.J., 2017. Habitat selection and activities of dabbling ducks during non‐breeding periods. The Journal of Wildlife Management 81, 1482–1493. doi:10.1002/jwmg.21324

Pearce, J.M., Ramey, A.M., Flint, P.L., Koehler, A.V., Fleskes, J.P., Franson, J.C., Hall, J.S., Derksen, D.V., Ip, H.S., 2009. Avian influenza at both ends of a migratory flyway: characterizing viral genomic diversity to optimize surveillance plans for North America. Evol Appl 2, 457–468. doi:10.1111/j.1752- 4571.2009.00071.x

Pinto, H.A., Brant, S.V., de Melo, A.L., 2014. Physa marmorata (Mollusca: Physidae) as a natural intermediate host of Trichobilharzia (Trematoda: Schistosomatidae), a potential causative agent of avian cercarial dermatitis in Brazil. Acta Tropica 138, 38–43. doi:10.1016/j.actatropica.2014.06.002

Poulin, R., Keeney, D.B., 2008. Host specificity under molecular and experimental scrutiny. Trends in Parasitology 24, 24–28. doi:10.1016/j.pt.2007.10.002

Poulin, R., Mouillot, D., 2005. Combining phylogenetic and ecological information into a new index of host specificity. Journal of Parasitology 91, 511–514. doi:10.1645/GE-398R

Poulin, R., Mouillot, D., 1999. Parasite specialization from a phylogenetic perspective: a new index of host specificity. Parasitology 126, 473–480. doi:10.1017/S0031182003002993

Reynolds, J., Weir, B.S., Cockerham, C.C., 1983. Estimation of the coancestry coefficient: basis for a short-term genetic distance. Genetics.

Sire, C., Durand, P., Pointier, J.P., Théron, A., 2001. Genetic diversity of Schistosoma mansoni within and among individual hosts (Rattus rattus):

126

infrapopulation differentiation at microspatial scale. International Journal for Parasitology 31, 1609–1616.

Slatkin, M., 1985. Gene Flow in Natural Populations. Annual Review of Ecology and Systematics 16, 393–430. doi:10.1146/annurev.ecolsys.16.1.393

Swanson, G.A., Hawkins, A.S., Hanson, R.D., Nelson, H.K., Reeves, H.M., 1985. Flyways: Pioneering Waterfowl Management in North America. The Journal of Wildlife Management 49, 832. doi:10.2307/3801721

Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Molecular Biology and Evolution 30, 2725–2729. doi:10.1093/molbev/mst197

Weinstein, S.B., Kuris, A.M., 2016. Independent origins of parasitism in Animalia. Biol. Lett. 12, 20160324. doi:10.1098/rsbl.2016.0324

Wethington, A.R., Lydeard, C., 2007. A molecular phylogeny of Physidae (Gastropoda: Basommatophora) based on mitochondrial DNA sequences. J Molluscan Stud 73, 241–257. doi:10.1093/mollus/eym021

Whiteman, N.K., Kimball, R.T., Parker, P.G., 2007. Co‐phylogeography and comparative population genetics of the threatened Galápagos hawk and three ectoparasite species: ecology shapes population histories within parasite communities. Molecular Ecology 16, 4759–4773. doi:10.1111/j.1365- 294X.2007.03512.x

127

W137 W756.2 W752 W925 W157 W748 W162 W743 W767.4 W203 W137.2 W190 W183 .99 W154.2 W750 W138.3 W160 W180 W159 SDS1006 W753 W777 W156 W153 W161 W345.2 W746 .99 W756 W135 W142 W155.3 Tshov .99 W755.2 W754 W148 W767.3 W158 W2135 W664 W698 W166.2 VVT3023 .95 W135.2 W833 W259 W747 W148.2 W154 W778 E45 .96 W774 E64 W191 W152.4 .97 1 W704 W181 W831 1 W749 W345.2 W703 W750.2 W345.3 Trichobilharzia sp. C Trichobilharzia franki W701 W604 1 W194 W497 W234 1 W235 W510 W608 I W193 W602 W249 1 W230 W211 W146 W147 W765 W765.5 W196 W765.6 W765.3 W256 .96 W745 SAL92-09 II W171 W236 W175 W765.4 W255 1 W263 Trichobilharzia sp. B 1 1 W149.2 W511.2 1 W212 W182 Pacific W149 .99 W192 Central VVT3043 W609 Mississippi W601 .99 W600 Atlantic W611 .98 W511 S. Hemisphere W607 Trichobilharzia regenti 0.02

128

Figure 1. cox1 phylogeny. BI inference of TQ, TP and TA, only posterior probabilities ≥.95 are presented. Branches are colored by species; TQ= orange,

TP=green and TA=blue. Individual worms are labeled by the migratory flyway they were collected from, denoted by a colored box. The two TP clades are denoted at I and II.

129

A B

Figure 2. Minimum spanning network of cox1 haplotypes. (A) Trichobilharzia querquedulae haplotypes, (B) T. physellae and (C)

Trichobilharzia sp. A. Haplotypes are colored by the migratory flyway the host was collected from.

130

300000

250000

200000

e 150000 N

100000

50000

0 0 500 1000 1500 2000 2500 Time (kya)

TQ TP TA

Figure 3. Bayesian skyline plot of cox1 datasets. Estimates of Ne (y-axis) over time (kya, x-axis). Lines are colored by species; TQ= orange, TP=green and

TA=blue.

131

A. DEG PD SH B.

Trichobilharzia NA * physellae * W701 SA * Trichobilharzia franki EU * Trichobilharzia sp. A NA

* 0 0.005 0.01 0.015 0.02

Trichobilharzia GL C. * querquedulae

Trichobilharzia sp. C NA * * Trichobilharzia regenti EU * Trichobilharzia mergi IC * Trichobilharzia szidati EU * Trichobilharzia stagnicolae NA * 0 0.01 0.02 0.03 0.04 * Trichobilharzia anseri EU Allobilharzia visceralis * Anserobilharzia brantae

Figure 4. Trichobilharzia summary tree. (A) Bayesian phylogeny of cox1 gene region, posterior probabilities ≥0.95 are denoted with an asterisk (*). DEG=Duck ecological group, PD =Parasite distribution, SH=Snail host. Under snail host, blocks with a white background indicate physid transmitted, while blocks with a black background indicate lymnaeid transmitted. (B) Bar graph of species wide estimates of nucleotide diversity (π) and (C) species wide estimates of Ɵ.

132

No. No. % Species of Avian Host examined infected infected Trichobilharzia Locality Source Anseriformes: Anatidae FL, LA, Trichobilharzia NC, Aix sposa 8 3 0.38 sp. C NM, PA Brant and Loker; This study Alopochen aegyptiaca 9 0 0 ZA This study Amazonetta brasiliensis 3 0 0 AR This study AK, CA, Trichobilharzia LA, NV, Anas acuta 27 2 0.07 sp. E NM, MB This study Trichobilharzia sp. A,B,D,E & AK, CA, T. NC, Anas americana 37 10 0.27 querquedulae NM, WA Brant and Loker; This study Anas bahamensis 1 0 0 AR This study T. szidati, T. Anas crecca (Europe) 15 9 0.6 franki FR Jouet et al. 2009 Anas crecca (Iceland) 2 0 0 IC Skirnisson and Kolarova 2008 AK, CA, LA, MN, Trichobilharzia NE, NC, Anas crecca (North America) 80 6 0.075 sp. NM, MB Brant and Loker 2009; This study Trichobilharzia Anas erythrorhyncha 5 2 0.4 sp. ZA This study Anas flavirostris 10 0 0 AR This study Anas fulvigula 5 0 0 LA Brant and Loker 2009 Anas georgica 6 0 0 AR This study Anas penelope 2 0 0 IC Skirnisson and Kolarova 2008 Anas platyrhynchos (Europe, 196 108 0.55 T. regenti FR Jouet et al 2009

133

nasal) Anas platyrhynchos (Europe, T. szidati, T. visceral) 31 26 0.84 franki FR Jouet et al 2009 Anas platyrhynchos (Iceland, nasal) 15 11 0.733 T. regenti IC Skirnisson and Kolarova 2008 Anas platyrhynchos (Iceland, Trichobilharzia visceral) 15 10 0.667 sp. IC Skirnisson and Kolarova 2008 AK, LA, Anas platyrhynchos (North MI, NM, America) 24 3 0.125 T. physellae PA Brant and Loker 2009; This Study Trichobilharzia NC, Anas rubripes 4 1 0.25 sp. NM, PA Brant and Loker 2009; This Study Trichobilharzia Anas sibilatrix 5 1 0.2 sp. AR This study FL, LA, NC, Anas strepera 36 1 0.028 T. physellae NM, PA Brant and Loker 2009; This Study Trichobilharzia Anas undulata 11 6 0.55 sp. ZA This study Anser anser (nasal) 75 6 0.08 T. regenti FR, IC Jouet et al. 2015 Jouet et al 2009; Skirnisson and Kolarova 2008; Jouet et Anser anser(visceral) 133 60 0.45 T. anseri FR, IC al. 2015 AK, CA, FL, IL, LA, NC, NM, NY, ON, PA, Aythya affinis 72 19 0.26 T. physellae WA Brant and Loker 2009; This study CA, LA, Aythya americana 4 0 0 NM Brant and Loker 2009; This study

134

CA, FL, LA, NC, Aythya collaris 8 2 0.25 T. physellae NM Brant and Loker 2009; This study Aythya ferina 18 2 0.11 T. regenti FR Jouet et al 2008; Jouet et al 2009 Aythya fuligula (nasal) 10 5 0.5 T. regenti Jouet et al 2008; Jouet et al 2009 Aythya fuligula (visceral) 1 1 1 T. franki FR Jouet et al 2009 AK, FL, PA, MB, Brant and Loker; Skirnisson and Kolarova 2008; This Aythya malraia 11 0 0 IC study Trichobilharzia Aythya novaeseelandia 20 18 0.9 sp. NZ Davis 2006 Aythya valisineria 5 1 0.2 T. physellae NM, NV Brant and Loker 2009 NC, NM, NV, Branta canadensis 9 0 0 MB Brant and Loker 2009 Branta hutchinsii 2 0 0 NM This study FL, NC, NM, NV, Bucephala albeola 13 1 0.077 T. physellae PA Brant and Loker; This study Bucephala clangula 3 0 0 CA, NM Brant and Loker 2009 Bucephala islandica 1 0 0 IC Skirnisson and Kolarova 2008 Cairina moschata 2 0 0 PE This study Callonetta leucophrys 4 0 0 AR This study LA, NM, Chen caerulescens 11 0 0 MB Brant and Loker 2009; This study Chen rossii 1 0 0 NM This study Clangula hyemalis 10 1 0.1 T. physellae AK This study NC, NV, Cygnus columbianus 18 0 0 NM Brant and Loker 2009; This study Cygnus cygnus 30 0 0 IC Skirnisson and Kolarova 2008 Cygnus olor (visceral) 9 5 0.55 T. franki FR Jouet et al 2009

135

Cygnus olor (nasal) 40 13 0.325 T. regenti FR Jouet et al 2009 Dendrocygna viduata 6 0 0 AR, ZA This study Histrionicus histrionicus 3 0 0 AK Brant and Loker 2009 Trichobilharzia LA, NC, Lophodytes cucullatus 6 2 0.33 sp. C PA Brant and Loker 2009; This study Melanitta americana 1 0 0 AK This study Melanitta fusca 5 0 0 AK ,MB Brant and Loker 2009 Melanitta perspicillata 3 0 0 AK Brant and Loker 2009 Mergus merganser (North T. stagnicolae, America) 10 3 0.3 T. physellae MI, NM Brant and Loker 2009; This Study Mergus merganser (Europe) 1 1 1 T. regenti FR Jouet et al. 2009 Mergus serrator (Iceland) 20 19 0.95 T. mergi IC Skirnisson and Kolarova 2008 Mergus serrator (North America) 3 0 0 CA, NM This study Netta erythrophthalmia 3 0 0 ZA This study CA, FL, NC, Oxyura jamaicensis 9 0 0 NM, NV Brant and Loker 2009; This Study Plectropterus gambensis 1 0 0 ZA This study Somateria mollissima 1 0 0 MB This study Somateria spectabilis 3 0 0 AK This study Spatula clypeata (Europe, nasal) 5 2 0.4 T. regenti FR Jouet et al. 2009 Spatula clypeata (Europe, visceral) 4 2 0.5 T. szidati FR Jouet et al. 2009 AK, CA, FL, LA, T. NC, NE, Spatula clypeata (North America) 53 39 0.74 querquedulae NM, MB Brant and Loker; This study T. AR, CA, Spatula cyanoptera 14 11 0.79 querquedulae NM Brant and Loker 2009; This study

136

CA, FL, T. LA, MN, Spatula discors 49 44 0.9 querquedulae NM, PA Brant and Loker 2009; This study Spatula platelea 1 0 0 AR This study T. Spatula smithii 2 2 1 querquedulae ZA This study T. Spatula versicolor 10 6 0.6 querquedulae AR This study Tadorna cana 1 0 0 ZA This study Thalassornis leuconotus 1 0 0 ZA This study Non-Anatids Agelaius phoeniceus (Passeriformes:Icteridae) 4 0 0 NM This study Alle alle (Charadriiformes: Alcidae) 1 0 0 DE This study Anhinga anhinga (Suliformes: Anhingidae) 10 0 0 FL, LA Brant and Loker 2009 Aramides ypecaha (Gruiformes: Rallidae) 1 0 0 AR This study Aramus guarauna (Gruiformes: Aramidae) 1 0 0 FL Brant and Loker 2009 Ardea herodias (Pelecaniformes: Ardeidae) 2 0 0 NM This study Bostrychia hagedash (Pelecaniformes: Threskiornithidae) 1 0 0 ZA This study Egretta intermedia (Pelecaniformes: Ardeidae) 1 0 0 ZA Brant and Loker 2009 Egretta thula (Pelecaniformes: Ardeidae) 1 0 0 LA Brant and Loker 2009 Egretta tricolor (Pelecaniformes: Ardeidae) 1 0 0 LA Brant and Loker 2009

137

Eudocimus albus (Pelecaniformes: Threskiornithidae) 6 0 0 LA Brant and Loker 2009 Euplectes sp. (Passeriformes: Ploceidae) 4 0 0 ZA This study Fulica americana (Gruiformes: Rallidae) 8 0 0 NC, NM This study Gavia immer (Gaviiformes: Gaviidae) 1 0 0 NM This study Larus californicus (Charadriiformes: Laridae) 1 0 0 CA Brant and Loker 2009 Larus delawarensis (Charadriiformes: Laridae) 15 0 0 CA, LA Brant and Loker 2009 Larus fuscus (Charadriiformes: Laridae) 1 0 0 LA Brant and Loker 2009 Patagioenas leucocephala (Columbiformes: Columbidae) 2 0 0 FL This study Pelecanus erythrorhynchos (Pelecaniformes: Pelecanidae) 1 0 0 NM This study Phalacrocorax africanus (Pelecaniformes: Phalacrocoracidae) 3 0 0 ZA This study Phalacrocorax auritus (Pelecaniformes: Phalacrocoracidae) 2 0 0 LA Brant and Loker 2009 Plegadis chihi (Pelecaniformes: Threskiornithidae) 6 0 0 LA Brant and Loker 2009 Pluvialis dominica (Charadriiformes: Charadriidae) 1 0 0 LA Brant and Loker 2009 Podiceps auritus (Podicipediformes: Podicipedidae) 1 0 0 NM This study

138

Podiceps nigricollis (Podicipediformes: Podicipedidae) 1 0 0 NM This study Podilymbus podiceps (Podicipediformes: Podicipedidae) 2 0 0 NM Brant and Loker 2009 Scopus umbretta (Pelecaniformes: Scopidae) 2 0 0 ZA This study Sterna maxima (Charadriiformes: Laridae) 1 0 0 LA Brant and Loker 2009 Tachybaptus ruficollis (Podicipediformes: Podicipedidae) 1 0 0 ZA This study Threskiornis aethiopicus (Pelecaniformes: Threskiornithidae) 1 0 0 ZA This study Uria aalge (Charadriiformes: Alcidae) 2 0 0 AK This study Xanthocephalus xanthocephalus (Passeriformes:Icteridae) 1 0 0 NM This study Table 1. Summary of Trichobilharzia survey data. Note that the majority of hosts were co-infected with other helminths and/or many with other Schistosomatid spp., these worms were accessioned with the associated Trichobilharzia records. Locality abbreviations are as follows: AK=Alaska, AR=Argentina, CA= California, DE=Delaware, FL= Florida, FR=France, IC=Iceland, LA=Lousiana, MB= Manitoba Canada, MI= Michigan, MN=Minnesota, NC= North Carolina, NE= Nebraska, NM = New Mexico, NV= Nevada, NZ= New Zealand, ON= Ontario Canada, PA=Pennsylvania, PE= Peru, PR= Puerto Rico, WA= Washington, ZA= South Africa.

139

Obs. STD- Host Ecological Species Host Species Prevalence STD STD* MOD Range Group % Eco Group TQ Spatula discors 0.9 dabbling duck Spatula cyanoptera 0.79 dabbling duck Spatula clypeata 0.74 dabbling duck Spatula versicolor 0.6 dabbling duck Spatula smitthi 1 dabbling duck Anas americana 0.03 Total 1.05 1.02 1 6 1 (dabbling ducks) TP Aythya affinis 0.26 diving duck Aythya collaris 0.25 diving duck Aythya valisineria 0.2 diving duck Bucephala albeola 0.077 diving duck Mergus merganser 0.1 diving duck 0.93 (diving ducks) Clangula hyemalis 0.1 sea duck Anas strepera 0.028 dabbling duck Anas platyrynchos 0.125 dabbling duck 0.07 (other ducks) Total 1.86 1.82 1.9 8 TA Anas americana 0.27 ─ ─ ─ 1 dabbling duck 1 (dabbling ducks) T. regenti Spatula clypeata 0.4 dabbling duck Anas platyrynchos 0.64 dabbling duck 0.82 (dabbling ducks) Aythya fuligula 0.5 diving duck Aythya ferina 0.11 diving duck 0.05(diving ducks)

140

Anser anser 0.08 swans/geese 0.13 (swans and Cygnus olor 0.325 swans/geese geese) Total 1.87 1.59 3 6 T. mergi Mergus serrator 0.95 ─ ─ ─ 1 diving duck 1 (diving ducks) T. anseri Anser anser 0.45 ─ ─ ─ 1 swans/geese 1 (swans/geese) Mergus T. stagnicolae merganser 0.2 ─ ─ ─ 1 diving duck 1 (diving ducks)

Table 2. Measures of host specificity.

141

% Fixation Source of Variation d.f. p value Variation indices T. querquedulae - Flyway Among Flyways 4 -4.64 ΦCT= -0.0464 0.4858 Among Populations within ΦSC = 8 11.04 0.0528* Flyways 0.10547 Within Populations 46 93.6 ΦST= 0.0639 0.0439* T. querquedulae - Latitude Among LG 2 4.41 ΦCT= 0.0441 0.00196 ΦSC = Among Populations within LG 10 4.42 0.0508* 0.04621 ΦST= Within Populations 46 91.17 0.0616 0.08827 T. querquedulae - Host spp. ΦCT= - Among HS 6 -0.47 0.4594 0.00472 ΦSC = Among Populations within HS 13 8.34 0.0675 0.08303 Within Populations 35 92.13 ΦST= 0.0787 0.0577 T. physellae – Flyway ΦCT= Among Flyways 3 55.59 0.02737 0.55592 Among Populations within ΦSC = 2 -42.73 0.8495 Flyways 0.96217 ΦST= Within Populations 20 87.14 0.1564 0.12864 T. physellae - Latitude ΦCT= Among LG 1 2.71 0.50733 0.02714 ΦSC = Among Populations within LG 4 7.12 0.52884 0.07319 ΦST= Within Populations 20 90.17 0.15934 0.09835 T. physellae - Host spp. ΦCT= Among HS 4 20.13 0.12219 0.20128 ΦSC Among Populations within HS 3 10.46 0.26588 =0.13098 Within Populations 17 69.41 ΦST= 0.3059 0.00293

Table 3. Analyses of molecular variance results. For both TQ and TP population structure was tested by flyway, latitude and host species. Significant p-values are denoted in bold, and (*) indicates values that were within one standard deviation of significance and were considered marginally significant.

142

Taxa n Prevalence Pi K Theta Tajima's D 1 Trichobilharzia querquedulae 63 81% (90%) 0.017 11.9 0.04542 -2.4223** 2 Trichobilharzia physellae 27 18% (26%) 0.00696 4.868 0.01387 -2.25017** 3 Trichobilharzia sp. A 15 27% 0.01175 4.348 0.01342 -0.77396 4 Trichobilharzia franki 7 ─ 0.00503 3.524 0.0064 - 5 Trichobilharzia regenti 8 34% (64%) 0.00236 1.75 0.0026 - 6 Trichobilharzia anseri 6 45% 0.00251 1.867 0.00252 - 7 Trichobilharzia stagnicolae 8 20% 0.01326 10.93 0.01825 - 8 Trichobilharzia szidati 8 * 0.00731 5.429 0.00934 -

Table 4: Indices of genetic diversity across Trichobilharzia

143

CHAPTER IV

Illuminating Schistosome Diversity: Phylogenomics and diversification of Schistosomatidae using targeted sequence capture of ultra-conserved elements

AUTHORS: Erika T. Gendron (Ebbs), Lijing Bu, Eric S. Loker, Vasyl Tkach, Veronica Flores and Sara V. Brant

ABSTRACT

Schistosomatidae Stiles and Hassall 1898 is a medically significant family of helminth parasites (Digenea: Trematoda), which infects mammals, birds (definitive hosts) and aquatic snails (intermediate hosts). Currently there are 14 named genera, of which interrelationships are largely unresolved. The lack of a resolved phylogeny has encumbered our understanding of the schistosome evolution, specifically in regard to patterns of host-use and the role of host-switching in diversification. This research used targeted sequence capture of ultra-conserved elements (UCE) to generate a phylogenomic dataset for estimation of schistosome interrelationships.

This dataset represents the largest phylogenetic effort within Schistosomatidae in terms of genetic loci; we recovered an average of 500 and 1,500 UCE loci (of ~4,000 targeted) between two sequencing runs. Relative to published UCE datasets our capture success was low, reasons for which are discussed. As such, UCE alignments contained a substantial amount of missing data, which limited downstream analyses. In light of this, we report increased resolution of two pivotal genera Heterobilharzia and Schistosomatium, which supports a secondary host- capture of mammals in North America. Support of other pivotal genera,

Macrobilharzia and Ornithobilharzia, varied among phylogenetic analyses. In total application of sequence capture technology within Schistosomatidae increased our

144 understanding of schistosome relationships. The implications of these results in regard to host-use and schistosome diversification are discussed.

INTRODUCTION

Schistosomatidae Stiles and Hassall 1898 represent a biologically diverse family of trematodes infecting birds or, mammals (definitive hosts) and aquatic snails

(intermediate hosts). Schistosomes are of great significance to human health as three species within the genus Schistosoma Weinland, 1858 are the major etiological agents of human schistosomiasis, one of the world’s most recalcitrant Neglected

Tropical Diseases still infecting over 250 million people globally (Tchuem Tchuenté et al., 2017). Non-human schistosomes, particularly those associated with avian definitive hosts, also cause Human Cercarial Dermatitis (swimmer’s itch) now considered a re-remerging zoonotic pathogen (Horák et al., 2015). From the perspective of parasite biology, schistosomes are remarkable in that they are the only exclusively dioecious (Platt, 1997) family of digenetic trematodes. This derived state has resulted in interesting patterns of dimorphic adult traits and reproductive strategies across the family, including substantial variation in adult body size and morphology (Morand and Müller-Graf, 2000). There is a remarkably broad range of intermediate hosts employed by schistosomes, 15 families including members of

Caenogastropoda and Heterobranchia, with evidence for numerous intermediate host-switching events (Brant and Loker, 2013).

Because of the medical importance, there has long been substantial interest in understanding interrelationships, patterns of host-use and character evolution within the Schistosomatidae (Weinland, 1858; Stiles and Hassall, 1898; Farley 1971;

Khali, 2002). However efforts were constrained by a lack of informative

145 morphological characters and adequate adult parasite material for description, particularly among avian infecting lineages. The advent of molecular makers (Snyder and Loker, 2000; Lockeyer et al., 2003; Brant et al., 2006) greatly aided efforts to elucidate schistosome diversity, particularly within avian infecting genera which proved to harbor cryptic diversity (i.e. Trichobilharzia, (Brant and Loker, 2009; Ebbs et al., 2016), and supported the existence of 14 named genera and just over 100 named species (Brant and Loker, 2013). This diversity count is likely an underestimate based on more recent taxon sampling of larval schistosomes (Jouet et al., 2010; Flores et al., 2015; Brant et al., 2017; Pinto et al., 2017, this study).

Historically Schistosomatidae was divided into three subfamilies (McMullen and

Beaver 1945, Lu and Bai 1976, Khali 2002). However, molecular phylogenetic methods combined with increased taxon sampling (Lockyer et al., 2003; Brant et al.,

2006) have not provided any support for the previous subfamily designations nor reflected schistosome interrelationships. In fact schistosome interrelationships have remained relatively unresolved, specifically at the deeper nodes (Snyder and Loker,

2000; Lockyer et al., 2003; Brant et al., 2006; Brant and Loker, 2013; Pinto et al.,

2017).

Genetic relationships within Schistosomatidae have thus far been based on few loci, primarily markers within the ribosomal RNA complex (28S, 18S, ITS1 & 2) and single mitochondrial gene, cox1. While these studies have made great strides in advancing our knowledge of species diversity, they have failed to resolve deeper nodes within the family (Snyder and Loker, 2000; Lockyer et al., 2003; Brant et al.,

2006; Brant and Loker, 2013; Pinto et al., 2017). Specifically, there is no resolution regarding the phylogenetic placement of pivotal genera: Heterobilharzia,

Schistosomatium and Macrobilharzia. Both Heterobilharzia and Schistosomatium

146

(herein referred to as HS clade) are monotypic mammalian infecting genera and represent the only mammalian lineage found in North America. The major mammalian clade (Schistosoma, Bivitellobilharzia, or the SB clade), which includes those species that infect humans, is found in tropical and sub-tropical latitudes. Most genetic studies have demonstrated an affinity between the HS clade with the derived avian clade (herein referred to as the DAS clade; Trichobilharzia, Allobilharzia,

Anserobilharzia, Dendritobilharzia, Gigantobilharzia and Bilharziella) rather than the

SB mammalian clade. Such a relationship may not be surprising given that cercaria within the DAS and HS have eyespots, whereas Schistosoma cercariae lack eyespots (Khali, 2002). Only a single species of Macrobilharzia, M. macrobilharzia infecting Anhinga anhinga in North America, is known from genetic data.

Macrobilharzia baeri (Fain 1955) has been reported from cormorants in Rwanda, however no genetic data currently exists. Macrobilharzia has consistently failed to group with any support with other schistosome lineages. Since schistosomes are found in both mammalian and avian hosts, resolution of the phylogenetic position of

Macrobilharzia and the HS clade is necessary to know how many times schistosomes have undergone definitive host-switches. Currently, it is hypothesized that the HS clade is the result of a secondary host-capture within North America

(Snyder and Loker, 2000). Most current molecular phylogenies place the AO clade as basal (though weakly supported) within Schistosomatidae (Brant et al., 2006;

Brant and Loker, 2013). Definitive resolution of basal genera is necessary to understand ancestral definitive and intermediate host relationships among schistosomes, and to develop hypotheses regarding diversification time.

This study sought to resolve phylogenetic relationships within

Schistosomatidae in an effort to understand patterns of host-use and host-switching,

147 in addition to character evolution. A resolved phylogeny is required to enable testing of phylogenetic hypotheses regarding schistosome diversification, and can be applied to address questions such as: have specific traits aided schistosome diversification? How common is intermediate and definitive host-switching within

Schistosomatidae? Did the major schistosome clades radiate simultaneously?

Additionally, we expect a resolved phylogeny to provide direction for future collection efforts of taxa that are under-sampled. For example, the intermediate host of

Bivitellobilharzia has remained a mystery despite extensive sampling (Devkota et al.,

2014a); definitive placement of the genus might suggest potential intermediate host families for more targeted search efforts. Why some interrelationships have proved particularly difficult to resolve despite increased taxon sampling is unclear. Brant et al. (2006) suggests that early schistosomes underwent a rapid radiation from a marine environment (AO clade) into freshwater snails and definitive hosts, suggesting the main mammalian (SB clade) and avian (DAS clade) clades diversified around the same time. Under this scenario substantial incomplete lineage sorting

(ILS) might be expected (Maddison and Knowles, 2006).

Recent developments in phylogenomic methods have proven effective in resolving difficult groups (McCormack et al., 2013; Grégoir et al., 2015), particularly where ILS was expected. One such method uses targeted sequence capture of ultra conserved elements (UCEs) (Faircloth et al., 2012) to generate thousands of nuclear loci for phylogenomic analysis. The use of UCEs as phylogenomic markers has been successful at resolving historically problematic vertebrate groups (Esselstyn et al.,

2017; Faircloth et al., 2013; Prum et al., 2015) as well as an increasing number of invertebrate groups (Blaimer et al., 2015; 2016a; Branstetter et al., 2016; Faircloth et al., 2015). We applied similar methods within Schistosomatidae, representing the

148 first application of this technology within a parasitic helminth system, to resolve schistosome interrelationships and address long standing questions of schistosome evolution.

MATERIALS AND METHODS

Taxon sampling

Specimens used for this study are vouchered at the Museum of Southwestern

Biology, Parasite Division in Albuquerque New Mexico USA. Locality information, host data and museum accession numbers are summarized in Additional File 1.

Schistosome samples were collected using standard parasitological procedures as described in Brant and Loker (2009) and Ebbs et al. (2016). Specimens used for this study were collected between 1999- 2017, and primarily preserved in 95% ethanol, but some were also saved in RNAlater. For the sequence capture experiments, 13 of the 14 named genera (and 3 as yet unnamed) were selected.

Sequence capture of ultra-conserved elements

19 samples were selected for targeted sequence capture of UCEs, specifically to address relationships across Schistosomatidae and resolve deeper nodes that have evaded phylogenetic placement in previous studies (i.e. Macrobilharzia, HS clade).

A custom bait set (18,550 baits, 120 nucleotides in length, 2X tiling density) was designed using Schistosoma mansoni as a reference genome. Approximately 4,000

UCE loci were targeted. The bait set was manufactured by Arbor Biosciences

(www.arborbiosci.com).

Obtaining template DNA of sufficient quantity and quality proved to be a significant challenge and limiting factor. Thus, several different extraction protocols and modifications to sample preparation were used. Most samples were extracted

149 using silica-based columns from either the QIAamp DNA Micro Kit (Qiagen), which is optimized for small amounts of tissue, or the DNeasy Blood and Tissue kit (Qiagen).

Additionally, organic extraction (E.Z.N.A. Mollusc DNA kit, Omega) and Hot Shot

Lysis (Truett et al., 2000) were also attempted. Different amounts of tissues and worm body regions were also extracted to optimize extraction efficacy and reduce host contamination. Samples were quantified using Qubit Fluorometric quantification

(Thermo Fisher) using the manufacturers buffers and following recommended protocol. Samples selected for sequence capture ranged from 0.31- 2.36 ng of DNA, generally lower then the 1 ng recommendation from MyBaits (MYbaits® Manual v.

3.02). Sample quality was assessed using a Bioanalyzer, however DNA quantity was generally too low to accurately gauge fragment size. Across samples, fragment size was not equal, with some samples showing significant amounts of degradation and others relatively little. Measures were taken to avoid further degradation (i.e. excluding vortexing, minimizing thawing/re-freezing of samples).

Library enrichment procedures for the MYcroarray MYBaits kit (MYbaits® Manual v. 3.02) were followed, but with several modifications to the standard sequence capture and library preparation protocols to accommodate low amounts of DNA and variable fragment size. Our samples were divided into two distinct sequence capture experiments and sequencing runs (R1 and R2), methods varied across R1 and R2 allowing for the comparison (Table 1). R1 used a standard sticky-end library preparation coupled with standard amplification polymerase. R2 employed blunt end library preparation chemistry and a uracil non-stalling amplification polymerase. This step was done to reduce adapter dimers, which were abundant in R1 samples. For all runs a size-selection step following library preparation was not preformed due to such low DNA quantity. Between R1 and R2, hybridization temperatures were

150 modified (62C or 65C, respectively). Post-capture libraries were amplified for 12 cycles. Sequencing of paired-end, 100 bp read was conducted on a HiSeq 2000.

Processing and alignment of UCE data

Quality control of the raw reads included trimming adapter contamination and low quality bases from reads, using the program Trimmomatic (Bolger et al., 2014) with slide window size 4 bp and quality score 20, minimum read length 36 bp. Clean reads were processed following the PHYLUCE software package (Faircloth, 2016).

The PHYLUCE workflow includes contig assembly using Trinity (Grabherr et al.,

2011), assembled contigs were then matched to the probe set using lastz (Harris,

2007). For species that have genome sequences, their genome sequences were downloaded from NCBI ftp site. Where species had no draft genome assembled, but

SRA data available, we downloaded the Sequence Read Archive (SRA)

(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3013647/) using the NCBI SRA- toolkit (https://trace.ncbi.nlm.nih.gov/Traces/sra/sra.cgi?view=toolkit_doc). The SRA data were converted to fastq format (fastq-dump in SRA toolkit) and trimmed following same steps described above. Draft genome sequences were assembled using MEGAHIT program (Li et al., 2015). UCE loci were mined from genome sequences following the PHYLUCE protocol. An incomplete data matrix was generated including mined UCE loci and those generated by this study. Samples with low capture success, where fewer than 100 UCE loci were recovered

(n=4) were excluded from subsequent analyses. 13 alignments of variable matrix completeness (~82%-36%) were created using MAFFT v. 2.2.7 (Liu et al. 2011) for

28 schistosome species and 5 out-group taxa. Only a single UCE locus was shared among all samples, as such the final trimmed and aligned datasets all incorporated various amounts of missing data. Missing data refers to the proportion of taxa not

151 represented for a given loci. For example, the “~27% missing” dataset allows inclusion of loci where only 24 of the 33 taxa are represented.

Phylogenetic analyses

Shared UCE loci were mined from published genomes of 5 out-group taxa within Digenea; Echinostoma caproni, Fascioloides gigantica, Opisthorchis viverrini,

Paragonimus westermani and Dicrocoelium dendriticum (Additional File 2). Gene trees of the 113 best-represented UCE loci were analyzed in RaxML (Stamatakis,

2008) using ML and 500 thorough bootstrap replicates. Individual gene trees were almost entirely unresolved at deeper nodes, and individual loci contained minimal pairwise divergence. This, combined with the substantial amount of missing data within our dataset, did not make our dataset ideal for robust species tree estimation

(Bayzid and Warnow, 2012; Hosner et al., 2016). We instead took a super matrix approach and analyzed concatenated UCE loci (Faircloth, 2012; Faircloth et al.

2015, Warnow, 2015, Meiklejohn et al., 2016). ML analyses were conducted in

RaxML, using the GTRGAMMA substitution model (Faircloth, 2012; Faircloth et al.

2015). Nodal support was evaluated using thorough bootstrapping (500 replicates).

Supermatrices varied in data completeness from 82%-36% complete.

RESULTS

UCE enrichment and sequencing

Generally, we did not see a direct positive linear relationship between starting tissue amount and recovered quantities of DNA. Multiplexed sequencing of enriched libraries resulted in an average 6,939,218 (23,988–24,435,956) reads per sample, with an average sequencing depth of 350X (58-1001X). An average of 44,060 (12-

442,043) contigs with a mean length of 325 bp were assembled. On average, we

152 recovered ~500 and ~1,500 UCE loci in R1 and R2 respectively. Only a single locus was shared among all taxa. This is a much smaller recovery rate relative to published sequence capture datasets of vertebrates (Faircloth et al., 2013; 2015;

Esselstyn et al., 2017) and the few existing studies on invertebrates (Blaimer et al.,

2016a). The unevenness of capture success among samples is likely related to the poor starting DNA quantity and quality (Blaimer et al., 2016b).

Phylogenetic analysis

Individual alignments (n=13) of concatenated UCE loci across a spectrum of data completeness (~82% - 36.4%) were analyzed and their likelihood scores calculated in RaxML. Log likelihood scores decreased linearly with the amount of missing data within our alignments (Table 2). The best score was recovered from the

82% UCE alignment (-249,913.44), based on 113 UCE loci (41,615 bp). ML estimation of this dataset showed the following results (Figure 1A): a) statistical support for the main Schistosoma clades (Barker and Blair, 1996; Lawton et al.,

2011; Devkota et al., 2016); b) lack of statistical support for the position of

Macrobilharzia other than that it is a sister genus to Schistosomatid W688; c) lack of support for Ornithobilharzia (AO clade) as basal; and d) statistical support for the HS clade basal to the DAS clade and Bilharziella basal within the DAS clade. ~79% (258

UCE loci) and ~76% (470 UCE loci) complete alignments were congruent with the

82% alignment. The resultant phylogeny from the 73% UCE alignment (Figure 1B), based on 756 UCE loci (265,282 bp) had a much lower likelihood score relative to the 82% alignment (-1,487,900.85, Δln = 1,237,987.41) but did resolve those relationships not supported in the above analyses. The 73% UCE alignment recovered the following relationships: a) Schistosoma was supported as sister to the

DAS+HS+ Macrobilharzia clade; b) Macrobilharzia and Schistosomatid W688 were

153 basal to a HS+DAS clade providing evidence for non- monophyly avian and mammalian clades. In other words, the switch into mammals happened at least twice; c) Bilhaziella polonica was basal to the DAS clade; d) Ornithobilharzia was supported (AO clade, 100 bootstrap support) as a basal genus to remaining

Schistosomatidae. Alignments with ~70%-36% missing data were concordant with the initial 82% trees in failing to resolve the placement of the AO clade and

Macrobilharzia, though the topologies were not all identical, however were not used in subsequent treatments of the data increasing missing data may confound phylogenetic accuracy (Wiens, 2006).

DISCUSSION

This is the first study to use UCE data and phylogenomic methods within

Schistosomatidae to address long-standing questions regarding the evolution and interrelationships of any helminth group. This study reconstructed phylogenies ranging from 113-2,179 UCE loci whereas prior phylogenetic investigations were based on a maximum of four loci (Snyder and Loker, 2000; Lockyer et al., 2003;

Brant et al., 2006; Brant and Loker, 2013). UCE loci recovery and data completeness was much reduced relative to vertebrate systems (McCormack et al.,

2013; Longo et al., 2017) and the few published invertebrate studies (Blaimer et al.,

2016a; Branstetter et al., 2016). Application of sequence capture methods (Faircloth et al., 2012) proved challenging in a system limited because of diminutive size of the organisms, availability of fresh material and few reference genomes. As such, the resulting datasets contained extensive missing data, which limited downstream phylogenetic analyses. However, despite these challenges the recovered UCE loci were able to resolve some of the pivotal phylogenetic relationships within

154

Schistosomatidae, and shed light on long standing questions regarding schistosome evolution. All alignments (82%-36% complete) support the HS clade as sister to the

DAS clade, supporting the probability of a secondary host capture of mammals

(Snyder and Loker, 2000). Bilharziella was consistently supported as basal within the

DAS clade which included Dendritobilharzia, Gigantobilharzia, Avian Schistosome C

& D, Allobilharzia and Trichobilharzia. The Schistosoma clade was consistently supported including three the main clades; S. mansoni, S. hematobium and S. japonicum. The other Asian Schistosoma were not included within this study.

Between UCE datasets there was inconsistency in regards to the placement of

Ornithobilharzia (AO clade) and the Macrobilharzia clade. The 73% UCE alignment alone supported the AO clade as basal within Schistosomatidae, a result that has been commonly recovered using 28S and 18S+28S+cox1 data (Snyder, 2004; Brant et al., 2006). Secondly, only the 73% UCE alignment supported the Macrobilharzia clade (Macrobilharzia macrobilharzia + Schistosomatid W688).

Patterns of host-use

Marine spirochiids (Spirochiidae) are the sister group to Schistosomatidae based on traditional genetic markers (Snyder, 2004; Brant et al., 2006), suggesting ancestral proto-schistosomes evolved from lineages infecting ectothermic marine turtles. Patterns of host-use are discussed relative to the 73%-UCE topology, as it was the most resolved and has general support from previously published phylogenies (Brant and Loker, 2013). The 73%-UCE ML analysis supports

Ornithobilharzia (AO clade) as a basal, suggesting that proto-schistosomes switched into marine birds from marine turtles, likely diversifying within Chardiiformes (Figure

2), from which they are most commonly reported (Brant and Loker, 2013). However, members of the AO clade can also be found in species of marine Anseriformes

155

(Brant and Ebbs, unpublished data). It is likely that basal schistosome lifecycles occurred entirely within marine environments, specifically within marine caenogastropods. Including intermediate hosts within the families: Potamididae,

Batillariidae, Nassariidae, and Littorinidae. Brant and Loker (2013) point out that marine habitats have been under sampled historically and may continue to be a source of novel schistosome lineages. Our data would suggest that schistosomes then colonized mammals diversifying into the SB clade (Figure 2). Contemporary members of the SB clade are known to infect a range of mammalian groups:

Elephantidae, Rhinocerotidae, Bovidae, Rodentia, Hippopotamidae and Hominidae.

Intermediate host use of Bivitellobilharzia remains unknown, however proto-

Schistosoma is considered to have co-diversified with the amphibious caenogastropod pomatiopsid snails (Davis, 1993; Snyder and Loker, 2000). Because there are three species of Schistosoma that infect humans, the SB clade has been well sampled relative to other schistosome clades. The SB clade represents a substantial radiation within mammals (26 species) and relationships have been largely resolved (Devkota et al., 2014a;2016; Lawton et al., 2011).

The 73%- UCE dataset places Macrobilharzia at the base of the DAS + HS clade (Figure 1B). Suggesting avian host use is ancestral to the DAS+HS clade, however it should be noted that the definitive host associations of Schistosomatid

W688 (Devkota et al., 2014b) have not been uncovered. This lineage is likely a sister genus to Macrobilharzia, and cercaria have distinctive eyespots which is largely an avian schistosome trait (except for the HS clade). Macrobilharzia baeri has been described from the old world cormorants, however currently no genetic data exists.

The HS clade was supported as a basal clade to the DAS clade. Resolution of the phylogenetic placement of Macrobilharzia and the North American mammalian

156 schistosomes (HS clade) provides evidence that a secondary host-capture of mammals occurred among more derived genera, as first postulated by Snyder and

Loker (2000). Interestingly Heterobilharzia americana Price 1929 is known to have a very broad definitive host range as it has been recovered from dogs, horses, raccoons, nutria and rabbits (Lee, 1962a; 1962b). Further, species level investigations are required to verify if H. americana is a single species and not a complex of cryptic species.

Bilharziella polonica, which broadly infects water birds (Anseriformes,

Gruiformes, Ciconiformes, Podicipeformes) was supported as the basal genus of the

DAS clade. Bilharziella utilizes planorbid snails suggesting the ancestral host of the

DAS clade. Planorbids are important hosts within the DAS clade as well utilized by:

Dendritobilharzia, Avian Schistosome C (Pinto et al., 2017), Avian Schistosome D

(this study) and Anserobilharzia. There have been at least 8 intermediate host- switching events (Brant and Loker, 2013) within the DAS clade (discussed in the following section), and a stronger association with Anseriformes (Dendritobilharzia,

Avian Schistosome C, Avian Schistosome D, Anserobilharzia, Allobilharzia and

Trichobilharzia). Species diversity within the DAS clade, particularly Trichobilharzia

(~35 species, Brant and Loker, 2013), is likely in large part due to colonization of

Anseriformes. Paradoxically, Anserobilharzia and Dendritobilharzia also utilize anatid hosts, in addition to common intermediate hosts (Gyraulus spp. and related snails), yet genera are species poor (based on genetic data), relative to Trichobilharzia.

Intermediate host switching

The breadth of taxonomic diversity of intermediate host use within

Schistosomatidae, particularly among members of the DAS clade, is remarkable and

157 has been discussed as integral in schistosome diversification (Brant and Loker,

2013). Avian schistosomes alone are known to infect 15 families of snails (4 from mammalian lineages), including representatives from Caenogasotropoda and

Heterobranchia. Intermediate host switching events are known to (potentially) result in vicariance sufficient for speciation (Huyse et al., 2003; 2005). Interestingly among schistosomes, most obviously within the DAS clade, intermediate host switching events are marked by short inter-nodes and do not appear to be associated with deep divergence events. This could indicate that 1) the breadth of intraspecific snail use is not well understood, particularly within the DAS clade, 2) that schistosomes diversified very recently and genetic differences among lineages (especially for nuclear loci) have not had time to accumulate or 3) that schistosome lineages maintain a certain flexibility in regards to intermediate host use that facilitates intermediate host switching. Population level studies of schistosomes’ species

(Schistosoma japonicum; (Rudge et al., 2009; Da-Bing Lu et al., 2011),

Trichobilharzia spp. (Ebbs et al. 2016), and to a lesser extent Schistosoma mansoni

& S. hematobium (Sire et al., 1999; Theron et al., 2004; Crellen et al., 2016; Červená et al., 2016) suggest that species maintain high genetic diversity and large effective population sizes. For avian schistosomes this is likely the result of infecting highly vagile hosts, which are likely to shape the demographics of individual schistosome lineages. As a group, schistosomes were likely ancestrally parasites of migratory marine birds (Charadriiformes: Laridae), as such they likely have a history of large effective sizes and high genetic diversity (and thereby adaptive potential) (Blouin et al., 1995; Blasco-Costa, 2012), which may have facilitated long-distances host- switching (Huyse et al., 2005).

Schistosome Diversification

158

Recent efforts to describe schistosome biodiversity have demonstrated that the DAS clade contains several unnamed genera (Brant and Loker, 2013, Brant et al., 2017; Pinto et al., 2017). Notably many of the putative genera have been found to have a broad geographic range and infect Anseriformes. For example, a study of the putative new genus ‘Avian Schistosome C’ (Figure 1-3) was published (Pinto et al., 2017) connecting populations collected from North America, South America and

Europe, in association with ducks and planorbid snails. Studies like these demonstrate that there is still much to learn in terms of schistosome diversity.

However, as the majority of new lineages fall within the DAS clade it is unlikely that increased taxon sampling will greatly alter phylogenetic inferences of Macrobilharzia and the HS clade.

Studies describing novel schistosome lineages (Flores et al., 2017) also demonstrate that intermediate host switching is ubiquitous within schistosomes, specifically within the DAS clade, and has been an important factor in schistosome diversification. Why some schistosome genera are species rich while others are monotypic or species poor is an interesting question that remains unanswered. For example, why has the HS clade failed to diversify in a similar manner, as

Schistosoma in the old world? Why did it not colonize humans? Heterobilharzia and

Schistosomatium, both monotypic genera, occur in common snail hosts

(Lymnaeidae) and broadly infect mammalian hosts. One potential explanation could be that hominids and early members of the HS clade do not interact sufficiently in time and/or space. Alternatively contemporary species diversity might be a poor reflection of historical diversity, meaning that members of the HS clade might have been reached maximum lineage diversity some time in the past. With these considerations in mind, increased sampling at the population level of these genera

159 could inform on historical demographic patterns, species diversity and even hybridization (Crellen et al., 2016). UCE loci have been shown to be effective at resolving shallow time relationships, even when compared to ddRAD sequencing methods (Leaché et al., 2015).

Without the availability of fossil data, for schistosomes and other helminth parasites, no primary calibration points exist to estimate divergence times. As such,

Parasitologists often turn to host divergence data to derive secondary calibration points (Jorge et al., 2018), under the assumption that parasites must have evolved within the bounds of their hosts. This assumption might be sound where parasites and hosts show evidence of co-phylogeny as the result of a long co-evolutionary history, for example, among directly transmitted eco-parasites (Huyse and Volckaert,

2005; Whiteman et al., 2007; Hahn et al., 2015). However, as evident within

Schistosomatidae (Brant and Loker, 2013, this study) long distance host-switching at the definitive and intermediate host level is common, eroding co-evolutionary signature at the deeper nodes (Ellis et al., 2015). As such, the use of host- divergence data should be used with great caution when applied to

Schistosomatidae. With this in mind it might be possible to provide some hard bounds on particular nodes.

Marine spirorchiids (Spirochiidae) are the well supported sister clade to

Schistosomatidae (Snyder, 2004; Brant et al., 2006), and are known to infect marine turtles. Based on 539 UCE loci it was estimated that living turtles (Shaffer et al.,

2017) originated during the mid-late Triassic. The snail family Nassariidae, one of the common intermediate host families to Austrobilharzia (AO clade) is estimated to have diverged roughly 120 mya (Galindo et al., 2016). Extant members which host contemporary Austrobilharzia are thought to have diverged 30-40 mya.

160

Charadiiformes a primary definitive host group of the AO clade is thought to have diverged between 23-16 mya (Baker et al., 2007). Correa et al. (2010) reconstructed phylogenetic relationships among Lymnaeidae, and demonstrated a distinct split among North American and Old World lymnaeids, and an accurate date for this split could provide a hard bound on when the HS clade evolved in North America. This does not assume that the HS clade co-diverged with North American lymnaeids, but rather that they were unlikely to predate their hosts. Similarly the diversification of modern ducks has been estimated to have occurred between (25-5 mya) (Sun et al.,

2017), which might be used as a hard bound on the divergence of Trichobilharzia.

Quite a bit is known, regarding the timing and origins of Schistosoma. Snyder and Loker (2000), hypothesize Schistosoma (including S. japonicum) likely originated in the mid-Miocene (16-11.6 mya), based largely on estimates by Davis

(1993) using fossil data to estimate the divergence of Pomatiopsidae snails, hosts to the extant S. japonicum clade. The timing of Pomatiopsidae divergence was later supported with genetic data which estimated divergence between 12-5 mya (Liu et al., 2014). Using whole genome sequencing of several S. mansoni and S. rodhaini individuals Crellen et al. (2016) estimated that these two sister species diverged

147.6-107.5 kya.

Phylogenomic considerations and future directions

Schistosomes posed several unique challenges to our phylogenomic efforts.

Firstly, across all genera, acquisition of adequate material presents inherent difficulties. Adult worms are diminutive in size, reside in the hosts venous system and are very fragile, often recovered as small fragments. Some species proved more difficult than others in terms of acquiring an adequate sample of DNA and subsequent capture success. For example Austrobilharzia and Anserobilharzia

161 yielded consistently poor DNA and had the lowest UCE capture rates, despite having access to recently preserved tissue. In general adult worms performed the best for recovered DNA, average number of reads, and UCE capture success. However, among cercaria samples none of these measures improved on average by increasing the number of individual cercaria extracted, suggesting this pattern is not solely related to amount of starting tissue. Notably, whole adult worms often yielded lower quantities of DNA relative to fragments of worms (Table 1).

Modifications of our sequence capture and library preparation protocols greatly increased the recovery rate of UCE loci (a 3-fold increase). Future studies preforming sequence capture on organisms with low quality or poor quality starting tissues should incorporate these modifications. Additionally moving forward it would be advisable to design a new bait kit to specifically target the most common 2,000

UCE loci recovered and increase bait tilling density to improve recovery of UCE loci.

While we were able to resolve some pivotal interrelationships within

Schistosomatidae, UCE loci proved less effective at resolving relationships than expected based published phylogenies (Faircloth et al., 2015; Blaimer et al., 2016a).

There is no doubt that starting with DNA of low quantity and quality contributed to our relatively poor sequence capture results, resulting in alignments that relied heavily on missing data. However, this likely does not entirely explain the difficulties in resolving schistosome relationships, as published studies with similarly low and uneven recovery of UCE loci have resulted in well-resolved phylogenies (Blaimer,

2016b). Large amounts of sequence data are fairly robust to missing data

(Maddison, 1993; Wiens, 2006), and it has been demonstrated that incomplete taxon sampling is more unfavorable to phylogenetic inference than missing data (Streicher et al., 2016). As such, several known but unnamed genera were not included in this

162 dataset; their future inclusion may prove to increase resolution. As suggested previously, the radiation resulting in contemporary schistosome diversity may be particularity refractory to resolution for two possible reasons. Firstly, contemporary genera might have radiated simultaneously and rapidly, resulting in high amounts of incomplete lineage sorting (Maddison and Knowles, 2006), and among some genera a hard polytomy (Maddison, 1989; Whitfield and Lockhart, 2007). Secondly, colonization-extinction dynamics are common within parasites (Hoberg and Brooks,

2008), as such contemporary lineages might be more closely related to one or many extinct lineages than to their most closely related extant taxa. A combination of these factors might work in concert to make the resolution of Schistosomatidae, and potentially other helminth taxa difficult, if not impossible to resolve.

REFERENCES

Correa, A.C., Escobar, J.S., Durand, P., Renaud, F., David, P., Jarne, P., Pointier, J.P., Hurtrez-Boussès, S., 2010. Bridging the gaps in the molecular phylogeny of the Lymnaeidae (Gastropoda: Pulmonata), vectors of Fascioloasis. BMC Evolutionary Biology 10.p. 381.

Baker, A.J., Pereira, S.L., Paton, T.A., 2007. Phylogenetic relationships and divergence times of Charadriiformes genera: multigene evidence for the Cretaceous origin of at least 14 clades of shorebirds. Biol. Lett. 3, 205–210. doi:10.1098/rsbl.2006.0606

Barker, S.C., Blair, D., 1996. Molecular Phylogeny of Schistosoma Species Supports Traditional Groupings within the Genus. The Journal of Parasitology 82, 292. doi:10.2307/3284163

Bayzid, M.S., Warnow, T., 2012. Estimating optimal species trees from incomplete gene trees under deep coalescence. J. Comput. Biol. 19, 591–605. doi:10.1089/cmb.2012.0037

Blaimer, B.B., Brady, S.G., Schultz, T.R., Lloyd, M.W., Fisher, B.L., Ward, P.S., 2015. Phylogenomic methods outperform traditional multi-locus approaches in resolving deep evolutionary history: a case study of formicine ants. BMC Evolutionary Biology 15, 271. doi:10.1186/s12862-015-0552-5

Blaimer, B.B., LaPolla, J.S., Branstetter, M.G., Lloyd, M.W., Brady, S.G., 2016a.

163

Phylogenomics, biogeography and diversification of obligate mealybug-tending ants in the genus Acropyga. Molecular Phylogenetics and Evolution 102, 20–29. doi:10.1016/j.ympev.2016.05.030

Blaimer, B.B., Lloyd, M.W., Guillory, W.X., Brady, S.G., 2016b. Sequence Capture and Phylogenetic Utility of Genomic Ultraconserved Elements Obtained from Pinned Insect Specimens. PloS ONE 11, e0161531. doi:10.1371/journal.pone.0161531

Blasco-Costa, I.E.A., 2012. Swimming against the current: genetic structure, host mobility and the drift paradox in trematode parasites. Molecular Ecology 21, 207– 217.

Blouin, M.S., Yowell, C.A., Courtney, C.H., Dame, J.B., 1995. Host Movement and the genetic structure of Populations of parasitic nematodes. Genetics 141, 1007–1014.

Bolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. doi:10.1093/bioinformatics/btu170

Branstetter, M.G., Danforth, B.N., Pitts, J.P., Faircloth, B.C., Ward, P.S., Buffington, M.L., Gates, M.W., Kula, R.R., Brady, S.G., 2016. Phylogenomic Analysis of Ants, Bees and Stinging Wasps: Improved Taxon Sampling Enhances Understanding of Hymenopteran Evolution. bioRxiv 068957. doi:10.1101/068957

Brant, S.V., Loker, E.S., 2013. Discovery-based studies of schistosome diversity stimulate new hypotheses about parasite biology. Trends in Parasitology 1–11. doi:10.1016/j.pt.2013.06.004

Brant, S.V., Loker, E.S., Casalins, L., Flores, V., 2017. Phylogenetic Placement of a Schistosome from an Unusual Marine Snail Host, the False Limpet (Siphonaria lessoni) and Gulls (Larus dominicanus) from Argentina with a Brief Review of Marine Schistosomes from Snails. The Journal of Parasitology 103, 75–82. doi:10.1645/16-43

Carstens, B.C., Knowles, L.L., 2007. Estimating Species Phylogeny from Gene-Tree Probabilities Despite Incomplete Lineage Sorting: An Example from Melanoplus Grasshoppers. Systematic Biology 56, 400–411. doi:10.1080/10635150701405560

164

Crellen, T., Allan, F., David, S., Durrant, C., Huckvale, T., Holroyd, N., Emery, A.M., Rollinson, D., Aanensen, D.M., Berriman, M., Webster, J.P., Cotton, J.A., 2016. Whole genome resequencing of the human parasite Schistosoma mansoni reveals population history and effects of selection. Sci Rep 6, 20954. doi:10.1038/srep20954

Červená, B., Brant, S.V., Fairet, E., Shirley, M.H., Petrželková, K.J., Modrý, D., 2016. Schistosoma mansoni in Gabon: Emerging or Ignored? Am. J. Trop. Med. Hyg. 95, 849–851. doi:10.4269/ajtmh.16-0446

Lu, D.B., Wang, T.P., Rudge, J.W., Donnelly, C.A., Fang, G.R., Webster, J.P., 2011. Genetic Diversity of Schistosoma japonicum miracidia from individual rodent hosts. International Journal of Parasitology 41, 1371–1376.

Davis, G.M., 1993. Evolution of prosobranch snails transmitting Asian Schistosoma; coevolution with Schistosoma: a review. Prog Clin Parasitol 3, 145–204.

Devkota, R., Brant, S.V., Loker, E.S., 2016. A genetically distinct Schistosoma from Radix luteola from Nepal related to Schistosoma turkestanicum: A phylogenetic study of schistosome and snail host. Acta Tropica 164, 45–53. doi:10.1016/j.actatropica.2016.08.015

Devkota, R., Brant, S.V., Thapa, A., Loker, E.S., 2014a. Sharing schistosomes: the elephant schistosome Bivitellobilharzia nairi also infects the greater one-horned rhinoceros (Rhinoceros unicornis) in Chitwan National Park, Nepal. Journal of Helminthology 88, 32–40. doi:10.1017/S0022149X12000697

Devkota, R., Brant, S.V., Thapa, S., Loker, E.S., 2014b. Two avian schistosome cercariae from Nepal, including a Macrobilharzia-like species from Indoplanorbis exustus. Parasitology International 63, 374–380. doi:10.1016/j.parint.2013.12.009

Ellis, V.A., Collins, M.D., Medeiros, M.C.I., Sari, E.H.R., Coffey, E.D., Dickerson, R.C., Lugarini, C., Stratford, J.A., Henry, D.R., Merrill, L., Matthews, A.E., Hanson, A.A., Roberts, J.R., Joyce, M., Kunkel, M.R., Ricklefs, R.E., 2015. Local host specialization, host-switching, and dispersal shape the regional distributions of avian haemosporidian parasites. Proc. Natl. Acad. Sci. U.S.A. 112, 11294– 11299. doi:10.1073/pnas.1515309112

Esselstyn, J.A., Oliveros, C.H., Swanson, M.T., Faircloth, B.C., 2017. Investigating Difficult Nodes in the Placental Mammal Tree with Expanded Taxon Sampling and Thousands of Ultraconserved Elements. Genome Biol Evol 9, 2308–2321. doi:10.1093/gbe/evx168

Faircloth, B.C., 2016. PHYLUCE is a software package for the analysis of conserved

165

genomic loci. Bioinformatics 32, 786–788. doi:10.1093/bioinformatics/btv646

Faircloth, B.C., Branstetter, M.G., White, N.D., Brady, S.G., 2015. Target enrichment of ultraconserved elements from arthropods provides a genomic perspective on relationships among Hymenoptera. Molecular Ecology Resources 15, 489–501. doi:10.1111/1755-0998.12328

Faircloth, B.C., McCormack, J.E., Crawford, N.G., Harvey, M.G., Brumfield, R.T., Glenn, T.C., 2012. Ultraconserved elements anchor thousands of genetic markers spanning multiple evolutionary timescales. Systematic Biology 61, 717– 726. doi:10.1093/sysbio/sys004

Faircloth, B.C., Sorenson, L., Santini, F., Alfaro, M.E., 2013. A Phylogenomic Perspective on the Radiation of Ray-Finned Fishes Based upon Targeted Sequencing of Ultraconserved Elements (UCEs). PloS One 8, e65923. doi:10.1371/journal.pone.0065923

Farley, J., 1971. A review of the family Schistosomatidae: excluding the genus Schistosoma from mammals. Journal of Helminthology 45, 289–320.

Galindo, L.A., Puillandre, N., Utge, J., Lozouet, P., Bouchet, P., 2016. The phylogeny and systematics of the Nassariidae revisited (Gastropoda, Buccinoidea). Molecular Phylogenetics and Evolution 99, 337–353. doi:10.1016/j.ympev.2016.03.019

Grabherr, M.G., Haas, B.J., Yassour, M., Levin, J.Z., Thompson, D.A., Amit, I., Adiconis, X., Fan, L., Raychowdhury, R., Zeng, Q., Chen, Z., Mauceli, E., Hacohen, N., Gnirke, A., Rhind, N., di Palma, F., Birren, B.W., Nusbaum, C., Lindblad-Toh, K., Friedman, N., Regev, A., 2011. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 29, 644–652. doi:10.1038/nbt.1883

Grégoir, A.F., Hablützel, P.I., Vanhove, M.P.M., Pariselle, A., Bamps, J., Volckaert, F.A.M., Raeymaekers, J.A.M., 2015. A link between host dispersal and parasite diversity in two sympatric cichlids of Lake Tanganyika. Freshwater Biology 60, 323–335. doi:10.1111/fwb.12492

Hahn, C., Weiss, S.J., Stojanovski, S., Bachmann, L., 2015. Co-Speciation of the Ectoparasite Gyrodactylus teuchis (Monogenea, Platyhelminthes) and Its Salmonid Hosts. PloS ONE 10, e0127340–20. doi:10.1371/journal.pone.0127340

Hoberg, E.P., Brooks, D.R., 2008. A macroevolutionary mosaic: episodic host- switching, geographical colonization and diversification in complex host-parasite systems. Journal of Biogeography 35, 1533–1550. doi:10.1111/j.1365- 2699.2008.01951.x

Horák, P., Mikeš, L., Lichtenbergová, L., Skála, V., Soldánová, M., Brant, S.V., 2015. Avian schistosomes and outbreaks of cercarial dermatitis. Clin. Microbiol. Rev. 28, 165–190. doi:10.1128/CMR.00043-14

166

Hosner, P.A., Faircloth, B.C., Glenn, T.C., Braun, E.L., Kimball, R.T., 2016. Avoiding Missing Data Biases in Phylogenomic Inference: An Empirical Study in the Landfowl (Aves: Galliformes). Molecular Biology and Evolution 33, 1110–1125. doi:10.1093/molbev/msv347

Huyse, T., Poulin, R., Théron, A., 2005. Speciation in parasites: a population genetics approach. Trends in Parasitology 21, 469–475. doi:10.1016/j.pt.2005.08.009

Huyse, T., Volckaert, F.A.M., 2005. Comparing host and parasite phylogenies: gyrodactylus flatworms jumping from goby to goby. Systematic Biology 54, 710– 718. doi:10.1080/10635150500221036

Jorge, F., Perera, A., Poulin, R., Roca, V., Carretero, M.A., 2018. Getting there and around: Host range oscillations during colonization of the Canary Islands by the parasitic nematode Spauligodon. Molecular Ecology 65, 31. doi:10.1111/mec.14458

Khalil, L.F., 2002. Family Schistosomatidae Stiles & Hassall, 1898., in: Keys to the Trematoda: Volume 1. CABI, Wallingford, pp. 419–432. doi:10.1079/9780851995472.0419

Lawton, S.P., Hirai, H., Ironside, J.E., Johnston, D.A., Rollinson, D., 2011. Genomes and geography: genomic insights into the evolution and phylogeography of the genus Schistosoma. Parasites & Vectors 4, 131. doi:10.1186/1756-3305-4-131

Leaché, A.D., Chavez, A.S., Jones, L.N., Grummer, J.A., Gottscho, A.D., Linkem, C.W., 2015. Phylogenomics of phrynosomatid lizards: conflicting signals from sequence capture versus restriction site associated DNA sequencing. Genome Biol Evol 7, 706–719. doi:10.1093/gbe/evv026

Lee, H.F., 1962a. Life history of Heterobilharzia americana Price 1929, a schistosome of the raccoon and other mammals in southeastern United States. Journal of Parasitology 48, 728–739.

Lee, H.F., 1962b. Susceptibility of mammalian hosts to experimental infection with Heterobilharzia americana. Journal of Parasitology 48, 740–745.

Li, D., Liu, C.-M., Luo, R., Sadakane, K., Lam, T.-W., 2015. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31, 1674–1676. doi:10.1093/bioinformatics/btv033

Liu, L., Huo, G.-N., He, H.-B., Zhou, B., Attwood, S.W., 2014. A phylogeny for the pomatiopsidae (Gastropoda: Rissooidea): a resource for taxonomic,

167

parasitological and biodiversity studies. BMC Evolutionary Biology 14, 29. doi:10.1186/1471-2148-14-29

Lockyer, A.E., Olson, P.D., Østergaard, P., Rollinson, D., Johnston, D.A., Attwood, S.W., Southgate, V.R., Horak, P., Snyder, S.D., Le, T.H., Agatsuma, T., Mcmanus, D.P., Carmichael, A.C., Naem, S., Littlewood, D.T.J., 2003. The phylogeny of the Schistosomatidae based on three genes with emphasis on the interrelationships of Schistosoma Weinland, 1858. Parasitology 126, 203–224. doi:10.1017/S0031182002002792

Loker, E.S., Brant, S.V., 2006. Diversification, dioecy and dimorphism in schistosomes. Trends in Parasitology 22, 521–528. doi:10.1016/j.pt.2006.09.001

Snyder, S. and Loker, E.S., 2000. Evolutionary relationships among the Schistosomatidae (Platyhelminhes: Digenea) and an Asian origin for Schistosoma. Journal of Parasitology 86, 283–288.

Longo, S.J., Faircloth, B.C., Meyer, A., Westneat, M.W., Alfaro, M.E., Wainwright, P.C., 2017. Phylogenomic analysis of a rapid radiation of misfit fishes (Syngnathiformes) using ultraconserved elements. Molecular Phylogenetics and Evolution 113, 33–48. doi:10.1016/j.ympev.2017.05.002

Maddison, W., 1989. Reconstructing character evolution on polytomous cladograms. Cladistics 5, 365–377. doi:10.1111/j.1096-0031.1989.tb00569.

Maddison, W.P., 1993. Missing Data versus Missing Characters in Phylogenetic Analysis. Systematic Biology 42, 576–581. doi:10.1093/sysbio/42.4.576

Maddison, W.P., Knowles, L.L., 2006. Inferring phylogeny despite incomplete lineage sorting. Systematic Biology 55, 21–30. doi:10.1080/10635150500354928

McCormack, J.E., Harvey, M.G., Faircloth, B.C., Crawford, N.G., Glenn, T.C., Brumfield, R.T., 2013. A phylogeny of birds based on over 1,500 loci collected by target enrichment and high-throughput sequencing. PloS One 8, e54848. doi:10.1371/journal.pone.0054848

Meiklejohn, K.A., Faircloth, B.C., Glenn, T.C., Kimball, R.T., Braun, E.L., 2016. Analysis of a Rapid Evolutionary Radiation Using Ultraconserved Elements: Evidence for a Bias in Some Multispecies Coalescent Methods. Systematic Biology 65, 612–627. doi:10.1093/sysbio/syw014

Morand, S., Müller-Graf, C.D.M., 2000. Muscles or testes? Comparative evidence for sexual competition among dioecious blood parasites (Schistosomatidae) of vertebrates. Parasitology, 120, 45-56.

Pinto, H.A., Pulido-Murillo, E.A., de Melo, A.L., Brant, S.V., 2017. Putative new genera and species of avian schistosomes potentially involved in human cercarial dermatitis in the Americas, Europe and Africa. Acta Tropica 176, 415– 420. doi:10.1016/j.actatropica.2017.09.016

168

Platt, T.R., 1997. Evolution of Schistosomes ( Digenea: Schistosomatoidea): The origin of dioecy and colonization of the venous system. Journal of Parasitology 83, 1035–1044.

Prum, R.O., Berv, J.S., Dornburg, A., Field, D.J., Townsend, J.P., Lemmon, E.M., Lemmon, A.R., 2015. A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA sequencing. Nature 526, 569–573. doi:10.1038/nature15697

Rudge, J.W., Lu, D.-B., Fang, G.-R., Wang, T.-P., Basáñez, M.-G., Webster, J.P., 2009. Parasite genetic differentiation by habitat type and host species: molecular epidemiology of Schistosoma japonicum in hilly and marshland areas of Anhui Province, China. Molecular Ecology 18, 2134–2147. doi:10.1111/j.1365- 294X.2009.04181.x

Shaffer, H.B., McCartney-Melstad, E., Near, T.J., Mount, G.G., Spinks, P.Q., 2017. Phylogenomic analyses of 539 highly informative loci dates a fully resolved time tree for the major clades of living turtles (Testudines). Molecular Phylogenetics and Evolution 115, 7–15. doi:10.1016/j.ympev.2017.07.006

Sire, C., Durand, P., Pointier, J.-P., Théron, A., 1999. Genetic Diversity and Recruitment Pattern of Schistosoma mansoni in a Biomphalaria glabrata Snail Population: A Field Study Using Random-Amplified Polymorphic DNA Markers. The Journal of Parasitology 85, 436. doi:10.2307/3285775

Snyder, S.D., 2004. Phylogeny and paraphyly among tetrapod blood flukes (Digenea: Schistosomatidae and Spirorchiidae). International Journal for Parasitology 34, 1385–1392. doi:10.1016/j.ijpara.2004.08.006

Stamatakis A, Hoover P, Rougemont J. 2008. A Rapid Bootstrap Algorithm for the RAxML Web Servers. Renner S, editor. Systematic Biology.57:758–71.

Streicher, J.W., Schulte, J.A., Wiens, J.J., 2016. How Should Genes and Taxa be Sampled for Phylogenomic Analyses with Missing Data? An Empirical Study in Iguanian Lizards. Systematic Biology 65, 128–145. doi:10.1093/sysbio/syv058

Sun, Z., Pan, T., Hu, C., Sun, L., Ding, H., Wang, H., Zhang, C., Jin, H., Chang, Q., Kan, X., Zhang, B., 2017. Rapid and recent diversification patterns in Anseriformes birds: Inferred from molecular phylogeny and diversification analyses. PloS ONE 12, e0184529. doi:10.1371/journal.pone.0184529

Tchuem Tchuenté, L.-A., Rollinson, D., Stothard, J.R., Molyneux, D., 2017. Moving from control to elimination of schistosomiasis in sub-Saharan Africa: time to change and adapt strategies. Infect Dis Poverty 6, 42. doi:10.1186/s40249-017- 0256-8

Theron A, E.A., 2004. Molecular ecology of Schistosoma mansoni transmission inferred from the genetic composition of larval and adult infrapopualtions within intermediate and definitive hosts. Parasitology 129, 571–585.

169

Warnow, T., 2015. Concatenation Analyses in the Presence of Incomplete Lineage Sorting. PLoS Currents. doi:10.1371/10.1371/currents.tol.8d41ac0f13d1abedf4c4a59f5d17b1f7

Whitfield, J.B., Lockhart, P.J., 2007. Deciphering ancient rapid radiations. Trends in Ecology & Evolution 22, 258–265. doi:10.1016/j.tree.2007.01.012

Wiens, J.J., 2006. Missing data and the design of phylogenetic analyses. J Biomed Inform 39, 34–42. doi:10.1016/j.jbi.2005.04.001

170

A. B. Dicrocoelium dendriticum Dicrocoelium dendriticum Opisthorchis viverrini 95 Paragonimus westermani Echinostoma caproni Opisthorchis viverrini Fascioloides gigantica Fascioloides gigantica Paragonimus westermani Echinostoma caproni Schistosoma rhodani Ornithobilharzia canaliculata Schistosoma mansoni Schistosoma mansoni Schistosoma mansoni Schistosoma mansoni Schistosoma margebowi Schistosoma rhodani Schistosoma mathetti Schistosoma margebowi Schistosoma hematobium Schistosoma mathetti Schistosoma curssoni Schistosoma hematobium Schistosoma japonicum Schistosoma curssoni Schistosoma japonicum Schistosoma japonicum Schistosomatid W688 Schistosoma japonicum Macrobilharzia macrobilharzia Schistosomatid W688 Ornithobilharzia canaliculata 98 Macrobilharzia macrobilharzia Bilharziella polonica Bilharziella polonica Avian Schistosome C Avian Schistosome C Avian Schistosome D Avian Schistosome D Dendritobilharzia pulverulenta Dendritobilharzia pulverulenta Dendritobilharzia pulverulenta Dendritobilharzia pulverulenta 98 Gigantobilharzia huronensis Gigantobilharzia huronensis Gigantobilharzia huronensis Gigantobilharzia huronensis Allobilharzia visceralis Allobilharzia visceralis Trichobilharzia regenti Trichobilharzia szidati Trichobilharzia querquedulae Trichobilharzia sp E. 98 Trichobilharzia physellae Trichobilharzia stagnicolae Trichobilharzia sp. A Trichobilharzia regenti Trichobilharzia sp E. Trichobilharzia sp. A Trichobilharzia stagnicolae Trichobilharzia querquedulae Trichobilharzia szidati Trichobilharzia physellae Schistosomatium douthitti Heterobilharzia americana 0.05 Heterobilharzia americana Schistosomatium douthitti

Figure 1: ML analyses of concatenated UCE loci. Phylogenies were estimated in

RaxML, GTRGAMMA+I. Black squares indicate a bootstrap support value of 100%.

(A) 82% complete (18% missing data) UCE dataset based on 110 loci (41,615 bp).

(B) 73% complete (27% missing data) UCE based on 756 loci (265,282 bp).

171

Figure 2: Host-switching within Schistosomatidae. Important traits mapped onto the the resolved 73% UCE phylogeny. Nodes are numbered as follows: 1= Switch into avian hosts from an ectothermic reptilian ancestral host, 2= Switch from

Caenogastropoda into Heterobranchia, 3= Secondary host-capture of mammalian hosts within North America, 4= Diversification within Anseriformes. Red triangles indicate intra-host habitat shifts. Green triangles indicate switches between

Caenogastropoda into Heterobranchia. All branches have 100% bootstrap support unless otherwise noted

172

Life Raw No. of Run Sample ID Taxa DNA [ug] No. of bp UCEs stage Reads contigs 2 W352/W340 Anserobilharzia brantae C 0.562 668,282 1,158 330,978 48 2 W396/359 Austrobilharzia variglandis A 0.704 124,242 145 35,487 6 2 W930 Bilharziela polonica A 0.31 5,572,448 17,081 5,897,771 1677 Dendritobilharzia 2 W813 pulverulenta A 1.32 7,352,118 26,978 8,405,837 1737 Dendritobilharzia 25,041,97 2 W926 pulverulenta A 2.36 2 47,498 14,794,680 1631 14,030,53 2 W414/513 Gigantobilharzia huronensis C 0.701 2 242,326 72,500,699 1650 16,867,49 2 W678 Gigantobilharzia huronensis C 0.65 6 48,623 16,046,122 1582 16,016,29 139,396,85 2 W805 Heterobilharzia americana A 2.21 1 442,043 2 1784 Macrobilharzia 2 W931 macrobilahrzia A 0.944 2,182,575 3,508 1,580,643 1590 15,222,38 2 W393 Ornithobilharzia canaliculata A 2.5 5 31,569 8,873,305 1441 2 W510/511 Trichobilharzia sp. A C 0.494 2,777,802 6,715 2,136,811 1964 1 W606 Trichobilharzia sp. E A - 9,124,167 22,342 16,933 592 2 W604 Trichobilharzia physellae A 0.224 1,133,317 2,683 952,777 1753 13,523,27 2 W929 Trichobilharzia querquedulae A 0.674 1 70,532 21,183,008 1576 2 W203 Trichobilharzia querquedulae A 0.408 68,006 75 19,185 2 10,207,44 1 W233 Trichobilharzia stagnicolae C 0.0531 0 18,675 13,662 658 1 W607 Avian Schistosome C C 0.212 4,784,485 34,656 53,303 1599 2 W342 Avian Schistosome D C 1.7 3,066,693 13,126 3,804,186 171 2 W216 Schistosomatid W216 C 0.63 23,988 12 3,050 0 2 W688 Schistosomatid W688 C 0.448 6,626,793 8,006 3,890,904 1428

173

Table 1. Summary of sequence capture data. Methodologies between run 1 and run 2 are detailed within the methods section. Sample ID’s with a (/) indicates that samples were pooled in an attempt to increase DNA quantities. Schistosome life stage is indicated as A (adult) or C (cercaria).

174

% missing Alignment data No. Loci No. bp -logLN 1 18.19% 113 41615 -249,913.44 2 21.22% 258 95215 -552,102.19 3 24.25% 470 167917 -958,990.67 4 27.28% 756 265282 -1,487,900.85 5 30.31% 1,070 372577 -2,061,422.86 6 33.34% 1,364 474081 -2581373.332 7 39.40% 1,819 627118 -3350563.131 8 42.43% 1,970 627118 -3586512.511 9 48.90% 2,006 677378 -3809408.073 10 51.50% 2,082 736137 -3841912.875 11 54.55% 2,144 743745 -3866958.204 12 60.60% 2,159 753058 -3903325.447 13 63.60% 2,179 755774 -3912154.447

Table 2. Summary of UCE alignments.

175

CONCLUSIONS

Host traits are important factors in shaping host-parasite interactions, including, but not limited to; host immunology, physiology, demographics, range, phylogeny and ecology. These factors are co-occurring, often across multiple taxonomically and ecologically disparate host species, and can act to enhance or constrain parasite population size, geographic range, transmission, and adaptive potential. Additionally, host traits at one life stage may facilitate or reduce transmission at another life stage, and are therefore integrative and inseparable in shaping parasite evolution. This integrative nature is likely a characteristic of trematode lifecycles and possibly multi-host parasites in general. This research sought to further our understanding of how host traits impact the evolutionary ecology of trematode systems, using schistosomes (Digenea: Schistosomatidae) and associated hosts as a model. Research described here integrated phylogeographic, phylogenomic, and population genetic methods, to investigate host-parasite relationships across multiple timescales.

Much of this research focused on the schistosome genus Trichobilharzia.

Molecular phylogenetic methods were used to connect populations of Trichobilharzia querquedulae from three continents (North America, South America, Africa) and New

Zealand. These data confirmed that T. querquedulae was a single globally distributed species, with evidence for trans-hemispheric transmission. Further extensive helminthological surveys of potential duck hosts revealed that T. querquedulae maintained a specific association with Spatula (= Anas, ‘Blue-wing’ ducks) species. This relationship is likely maintained by a combination of duck host phylogeny and ecology. There are several reasons why ecological traits specific to

Spatula species might facilitate the global transmission of T. querquedulae. Spatula

176 spp. prefer shallow, marsh habitats where they feed at the surface. These habitats are snail rich, increasing the likelihood of transmission between the snail intermediate host (Physa spp.) and Spatula spp. Further one member of Spatula (S. clypeata) is Holarctic in its distribution, and is known to migrate (at a lower frequency) into the Southern hemisphere as part of its wintering range. Molecular data confirms that endemic, non-migratory, Spatula spp. transmit T. querquedulae within the Southern hemisphere, and may suggest that S. clypeata migration facilities trans-hemispheric gene flow. The global transmission of T. querquedulae is likely made possible by (though not confirmed) the global invasion of its snail host

Physa acuta. Data from this study suggest that certain definitive and intermediate host combinations may act in concert to facilitate Trichobilharzia transmission, supporting the integrative nature of schistosome lifecycles.

The observation of specificity between T. querquedulae and Spatula spp. was unexpected as Trichobilharzia species were thought to non-specifically infect anatid hosts (Anatidae: ducks, geese and swans). A closely related species T. physellae is known to infect Passeriformesand Columbiformes under experimental conditions and other Trichobilharzia species are reported from a broad range of anatid hosts. Using the terminology of Combes (1999), these reports suggest that Trichobilharzia maintains a wide compatibility filter. As such, the broad potential host range of T. querquedulae indicates that the encounter filter is narrowed. Host specificity observed between T. querquedulae and its definitive hosts led to more directed comparative studies of North American Trichobilharzia species and their host associations.

Surveys of Trichobilharzia and their duck hosts resulted in the finding that some level of host specificity was common among Trichobilharzia species and their

177 duck hosts. It was found that measures of host specificity including duck ecological group often better captured duck host associations than considering host taxonomy alone. One of the most interesting findings was that T. physellae, which was considered a generalist using traditional measures of host specificity, was found to be specific to ‘diving ducks’, when duck ecological group was considered. Diving ducks represent an ecological group within Anatidae, characterized by feeding behavior and habitat preferences. Trichobilharzia physellae was found to infect its most common host (Aythya affinis) at a much lower rate (26%) relative to T. querquedulae within its most common host (Spatula discors, 90%). This is likely the result of decreased transmission success within the deeper water bodies (less interaction with intermediate host) that diving ducks occupy.

This research also compared population genetic and phylogeographic patterns among North American Trichobilharzia species found to be specific to either dabbling or diving ducks; T. physellae (diving ducks), T. querquedulae (dabbling ducks, Spatula spp.) and Trichobilharzia sp. A (dabbling ducks, Anas americana).

Importantly both T. physellae and T. querquedulae infect Physa spp. as their intermediate host, which unites these species in ecological space and time.

Comparison of congeneric Trichobilharzia species revealed that different host species structured parasite populations in different ways. Trichobilharzia physellae populations appear to be structured in relation to the migratory flyways used by their duck hosts, which is a common pattern found among symbionts of philopatric ducks.

Spatula spp. are less philopatric relative to Aythya as such migratory flyway does not seem to impact the population structure of T. querquedulae. Instead we found an association with latitude, which is likely a proxy for breeding versus wintering ground.

Genetic diversity among the three species was variable, the highest values of which

178 were found in T. querquedulae and Trichobilharzia sp. A, possibly suggesting an advantage to concentrating numbers within a narrow range of hosts (specialization).

The effective size of T. querquedulae was found to be over 3X greater than the other two species. In sum this work demonstrated that Trichobilharzia-duck associations are more specific than previously thought, and that individual duck hosts or host groups can differentially impact Trichobilharzia microevolutionary patterns.

Along a similar vein phylogeographic, population genetic and demographic patterns of an important intermediate host Physa acuta were estimated. Physa acuta transmits at least three schistosome species (T. querquedulae, T. physellae and

Gigantobilharzia huronensis) as well as numerous other trematode species in North

America. This snail is of particular importance as it is now global in distribution, as the result of a 200+ year invasion process. This research had two parts, 1) characterize phylogeographic and population genetic patterns to reconstruct the invasion history of P. acuta and 2) test the ‘Enemy-release hypothesis. Increased sampling from the native range of P. acuta (North America), particularly from the

Western USA, provided greater insight into the phylogeography of P. acuta allowing for identification of source populations. Using published parasitological surveys of P. acuta in combination with those carried out for the purposes of this research we found limited evidence of trematodes infecting invasive P. acuta (as first intermediate host), supporting the ‘Enemy-release’ hypothesis. The few cases of trematode infection that were found were in association with invasive populations in the oldest and most genetically diverse regions of the invasive range. A significant positive correlation, between time since invasion, genetic diversity, and being colonized by local parasites was found. These results have significant implications for modeling host-parasite invasion dynamics over time, and more broadly understanding how

179 host traits might mediate parasite assembly. This data suggests that host demographic parameters (nucleotide diversity, effective population size) are important predictors of parasite assembly within invasive populations.

Apart for understanding host-parasite associations at the microevolutionary level, this research also sought to characterize patterns at the macroevolutionary level within Schistosomatidae. Evolutionary relationships within Schistosomatidae have historically lacked resolution, impeding a clear appreciation for patterns of host use. This research applied sequence capture technology to target ultra-conserved elements (UCE loci) for phylogenomic analyses, to specifically resolve phylogenetic placement of pivotal genera and identify host-switching events. Increased resolution in regards to the North American mammalian schistosomes was found, which supported that a secondary host capture occurred. Additionally, within the derived avian clade where the majority of species diversity occurs, we find evidence for high amounts of intermediate host switching. Suggesting that, 1) intermediate host switching was an important mechanism of schistosome diversification and 2) an overall lack of co-phylogenetic structure among schistosomes and their snail hosts.

Largely phylogenetic resolution was poorer than expected based on published

UCE phylogenies of vertebrate and some invertebrate taxa. There are likely several reasons for this, which require further investigation. Firstly, sequence capture success reported within this research is on average lower, and substantially more uneven, relative to published studies. Decreased capture success was likely due to poor DNA starting quantity and quality, which likely affected downstream phylogenomic analyses. There is also the possibility, though not mutually exclusive, that some nodes within Schistosomatidae may never be resolved. Previous authors have suggested that Schistosomatidae underwent a rapid radiation, which might

180 have resulted in a hard polytomy. Generally UCE loci recovered contained very few informative sites, and suggested high amounts of incomplete lineage sorting.

This research investigated different host traits across different life stages and time scales to determine their role shaping host-parasite relationships. We found evidence for specificity within Trichobilharzia in regards to definitive host use, and more specifically that duck habitat preferences may substantially impact

Trichobilharzia transmission and demographic patterns. Similarly host demographics may play an integral role in parasite assembly under invasion conditions. In total these data support that host traits are important mediators of host-parasite relationships, and that host traits across life-stages are integrated often acting in concert to enhance or constrain parasite transmission.

181

APPENDIX A

CHAPTER II ADDITIONAL FILES

Additional File 1. cox1 phylogeny of Physa estimated using Bayesian inference. Posterior probability values ≥.95 are shown. Taxon names are reflective of the names assigned to sequences in the original NCBI records.

182

Additional Figure 2. Concatenated 16+cox1 phylogeny. Branch support values

≥.95/70 (PP/BS) are shown. Taxon names are reflective of the names assigned to sequences in the original NCBI records.

183

Additional File 3. ITS1 phylogeny estimated using Bayesian inference pdf.

Physa acuta samples are colored according to the clade they were recovered from based on mitochondrial gene tree analysis: pink = Clade A, green = Clade B. Physa acuta taxa labeled as black did not group within P. acuta based on mitochondrial gene trees, and were excluded from all in-group analyses (S17, S120 and all P. zionis samples). ITS1 data only exists for HQ283259 and Pacuta_18S/ITS and were therefore not included in mitochondrial gene tree analyses. Posterior probabilities are illustrated as branch support values.

184

185

Additional Figure 4. 16S in-group analysis. Taxa are colored according to the

FWEC they were collected from. Phylogeny (A) was generated using ML methods and phylogeny (B) was generated using BI. Branch support values ≥ 70 bootstrap

(A) and ≥ .95 posterior probabilities (B) are denoted by a black circle

186

A. 100000 NAM 1%

NAM 0.04% 10000 NAM 4%

EUR 1% 1000 EUR 0.04%

EUR 4% 100 SAM 1%

SAM 0.04% 10 SAM 4%

1 AFOC 1% 0 10 20 30 40 50 60 AFOC 0.04% B. 10000 AFOC 4%

ATL 1% ATL 0.04% ATL 4% 1000 COA 1% COA 0.04% COA 4% COL 1% 100 COL 0.04% COL 4% MISS 1% MISS 0.04% 10 MISS 4% RIO 1% RIO 0.04% RIO 4% 1 C. 0 5 10 15 20 25 30 1000 A 1%

A 0.04%

A 4 %

B 1%

100 B 0.04%

B 4%

TOT 1%

TOT 0.04% 10 Total 4%

1 0 10 20 30 40 50 60 70 Time (kya)

187

Additional File 5. Bayesian Skyline Plots. Effective size over time plots estimated in BEAST using a Bayesian Skyline prior, mean values are shown and confidence values were excluded for clarity. Three mutation rates (per million year): 1% (solid line), 0.04% (dashed line) and 4%(circle line) were tested. (A) Range-wide estimates for North America (NAM, dark grey, n=79), Central and South America (SAM, lime green, n=13), Eurasia (EUR, pink, n=49) and Africa and Oceania (AFOC, blue, n=7).

North America dataset ran for 50,000,000 generations, Eurasia 20,000,000 generations, Central/South America and Africa/Oceania ran for 5,000,000 generations. B. Estimates by FWECs sampled: Rio Grande (RIO, red, n=16),

Coastal (COA, blue, n=13), Colorado (COL, purple, n=11), Atlantic (ATL, orange, n=9), Mississippi (MISS, green, n=26), the Great Basin and St. Lawrence FWECs were excluded because of low sample size. All FWEC datasets were run for

5,000,000 generations, with the exception of Atlantic (3,000,000). C. Estimates of recovered Clades, Clade A (A, aqua, n=116) and B (B, dark green, n=33) compared to total (TOT, light grey, n=149). The total dataset and Clade A ran for 80,000,000 generations, Clade B ran for 10,000,000 generations. Bayesian skyline plots were constructed in TRACER, the data was exported and visualized using Excel.

188

APPENDIX B

CHAPTER III ADDITIONAL FILES

Additional File 1. Locality information for TQ, TP and TA. Migratory flyways (MWF). Voucher material housed at the Museum of Southwestern Biology, Parasite Division (MSB:Para).

189

Latitude: MSB:Para: Specimen ID Locality MFW Longitude Date Host Species # Source Trichobilharzia querquedulae Lousiana: Brant and Loker Rockefeller Wildlife 29.67304, - 2009; Ebbs et al. W135 Refuge M 92.689902 11/15/2003 Anas clypeata 183 2016 Lousiana: Rockefeller Wildlife 29.67304, - W135.2 Refuge M 92.689902 11/15/2003 Anas clypeata 183 This Study Lousiana: Brant and Loker Rockefeller Wildlife 29.67304, - 2009; This Study W137 Refuge M 92.689902 11/15/2003 Anas discors 19497 Lousiana: Rockefeller Wildlife 29.67304, - W137.2 Refuge M 92.689902 11/15/2003 Anas discors 19497 This Study Lousiana: Rockefeller Wildlife 29.67304, - W138 Refuge M 92.689902 11/15/2003 Anas clypeata 19498 This Study Lousiana: Rockefeller Wildlife 29.67304, - W138.3 Refuge M 92.689902 11/15/2003 Anas clypeata 19498 This Study Lousiana: Timberton Hunt 30.1705, - W140 Club M 90.868 11/13/2003 Anas discors 19500 This Study Lousiana: Ebbs et al. 2016; Timberton Hunt 30.1705, - This Study W142 Club M 90.868 11/13/2003 Anas discors 19502 Brant and Loker New Mexico: San 33.7131, - Anas 2009; Ebbs et al. W148.1 Marcial C 106.9579 3/25/2004 cyanoptera 18584 2016

190

New Mexico: San 33.7131, - Anas Brant and Loker W148.2 Marcial C 106.9579 3/25/2004 cyanoptera 18584 2009; This Study New Mexico: San 33.7131, - W152.1 Marcial C 106.9579 4/6/2004 Anas clypeata 18587 This Study New Mexico: San 33.7131, - W152.2 Marcial C 106.9579 4/6/2004 Anas clypeata 18587 This Study New Mexico: San 33.7131, - W152.4 Marcial C 106.9579 4/6/2004 Anas clypeata 18587 This Study New Mexico: San 33.7131, - Ebbs et al. 2016; W153 Marcial C 106.9579 4/6/2004 Anas clypeata 18587 This Study New Mexico: San 33.7131, - Anas W154.1 Marcial C 106.9579 4/6/2004 cyanoptera 18588 This Study New Mexico: San 33.7131, - Anas W154.2 Marcial C 106.9579 4/6/2004 cyanoptera 18588 This Study New Mexico: San 33.7131, - Anas Brant and Loker W155.1 Marcial C 106.9579 4/6/2004 cyanoptera 18589 2009; This Study New Mexico: San 33.7131, - Anas W155.2 Marcial C 106.9579 4/6/2004 cyanoptera 18589 Ebbs et al. 2016 Brant and Loker New Mexico: San 33.7131, - Anas 2009; Ebbs et al. W155.3 Marcial C 106.9579 4/6/2004 cyanoptera 18589 2016; This Study Brant and Loker New Mexico: Bitter 2009; Ebbs et al. W156 Lake C 33.45, -104.4 4/11/2004 Anas discors 18590 2016 New Mexico: Bitter W157 Lake C 33.45, -104.4 4/11/2004 Anas clypeata 18591 This Study New Mexico: Bitter Brant and Loker W158 Lake C 33.45, -104.4 4/12/2004 Anas clypeata 18592 2009 New Mexico: Bitter W159 Lake C 33.45, -104.4 4/12/2004 Anas discors 18593 This Study

191

New Mexico: Bitter Ebbs et al. 2016; W160 Lake C 33.45, -104.4 4/12/2004 Anas discors 18594 This Study New Mexico: Bitter Anas W161 Lake C 33.45, -104.4 4/12/2004 cyanoptera 18605 This Study Brant and Loker New Mexico: Bitter 2009; Ebbs et al. W162 Lake C 33.45, -104.4 4/12/2004 Anas clypeata 18602 2016 New Mexico: San 33.7131, - W166.2 Marcial C 106.9579 9/9/2004 Anas discors 181 This Study New Mexico: San 33.7131, - Ebbs et al. 2016; W167 Marcial C 106.9579 9/9/2004 Anas discors 26231 This Study California: Wister Brant and Loker Unit Waterfowl 33.284266, - Anas 2009; This Study W180 Area P 115.57921 11/24/2004 cyanoptera 18573 California: Imperial 33.284266, - Anas W181 County P 115.579216 2004 americana 18572 This Study California: Wister Brant and Loker Unit Waterfowl 33.284266, - 2009; Ebbs et al. W183 Area P 115.57921 11/24/2004 Anas clypeata 18575 2016 California: Wister Unit Waterfowl 33.284266, - Anas W188 Area P 115.57921 11/24/2004 cyanoptera 18580 This Study California: Wister Unit Waterfowl 33.284266, - Brant and Loker W190 Area P 115.57921 11/24/2004 Anas discors 18582 2009 California: Wister Unit Waterfowl 33.284266, - Anas W191 Area P 115.57921 11/24/2004 cyanoptera 18583 This Study Alaska: Yukon- 65.665, - 5/29/2005 Brant and Loker W203 Koyukuk Borough P 149.098 Anas clypeata 18636 2009

192

Alaska: Yukon- 67.13369, - 5/30/2005 W204 Koyukuk Borough P 150.35594 Anas clypeata 18637 This Study Alaska: Yukon- 67.492186, - 5/30/2005 W207 Koyukuk Borough P 149.86944 Anas clypeata 18640 This Study Alaska: North 68.64636, - W2135 Slope Borough P 149.44891 6/1/2005 Anas clypeata 18647 This Study New Mexico: Bitter W259 Lake C 33.45, -104.4 3/17/2006 Anas discors 19515 This Study Manitoba: Brant and Loker Churchill, Goose 58.67583, - 2009; Ebbs et al. W345 Creek M 94.1677 8/3/2007 Anas clypeata 18626 2016 Manitoba: Churchill, Goose 58.67583, - W345.2 Creek M 94.1677 8/3/2007 Anas clypeata 18626 This Study Manitoba: Churchill, Goose 58.67583, - W345.3 Creek M 94.1677 8/3/2007 Anas clypeata 18626 This Study South Africa: Free -29.208716, W650 State Province SH 27.18945 1/13/2012 Anas smithii 18990 Ebbs et al. 2016 South Africa: Free -29.14433, W664 State Province SH 27.02565 1/21/2012 Anas smithii 19000 Ebbs et al. 2016 W698 North Dakota C - - Anas clypeata - This Study New Zealand, South Island, Lake -44.35124, Anas W703 Benmore SH 170.2091 12/31/2012 rhynchotis - Ebbs et al. 2016 New Zealand, South Island, Lake -44.35124, Anas W704 Benmore SH 170.2091 12/31/2012 rhynchotis 20793 Ebbs et al. 2016 W743 New Mexico C - - Anas clypeata - This Study Florida: Hamilton 30.42307, - W746 County A 82.7936 1/12/2013 Anas clypeata - This Study

193

Florida: Hamilton 30.42307, - W747 County A 82.7936 1/12/2013 Anas clypeata - This Study Florida: Hamilton 30.42307, - W748 County A 82.7936 1/12/2013 Anas clypeata - This Study Florida: Hamilton 30.42307, - W749 County A 82.7936 1/12/2013 Anas clypeata - This Study Florida: Hamilton 30.42307, - W750 County A 82.7936 1/12/2013 Anas discors - This Study Florida: Hamilton 30.42307, - W750.2 County A 82.7936 1/12/2013 Anas discors - This Study Florida: Hamilton 30.42307, - W752 County A 82.7936 1/12/2013 Anas clypeata 19534 This Study Florida: Hamilton 30.42307, - W753 County A 82.7936 1/12/2013 Anas clypeata - This Study Florida: Hamilton 30.42307, - W754 County A 82.7936 1/12/2013 Anas clypeata - This Study Florida: Hamilton 30.42307, - W755.2 County A 82.7936 1/12/2013 Anas discors - This Study Florida: Hamilton 30.42307, - W756 County A 82.7936 1/12/2013 Anas discors - This Study Florida: Hamilton 30.42307, - W756.2 County A 82.7936 1/12/2013 Anas discors - This Study Florida: Hamilton 30.42307, - W757 County A 82.7936 1/12/2013 Anas clypeata - This Study New Mexico: Eddy 32.4122, - W758 County C 103.938 2/2/2013 Anas clypeata - This Study Florida: Palm 26.42068, - W762 Beach County A 80.5853 2/21/2013 Anas discors 19283 This Study New Mexico: San 36.44, - W763 Juan County C 108.27213 3/24/2013 Anas clypeata - This Study

194

Florida: Palm 26.42068, - W766 Beach County A 80.5853 2/21/2013 Anas discors 19284 This Study Florida: Palm 26.42068, - W767 Beach County A 80.5853 2/21/2013 Anas discors 19285 This Study Florida: Palm 26.42068, - W767.2 Beach County A 80.5853 2/21/2013 Anas discors - This Study Florida: Palm 26.42068, - W767.3 Beach County A 80.5853 2/21/2013 Anas discors - This Study Florida: Palm 26.42068, - W767.4 Beach County A 80.5853 2/21/2013 Anas discors - This Study Alaska: Fairbanks 64.6655, - W771 North Star Borough P 147.101388 10/8/2013 Anas clypeata 19474 This Study Minnesota: Polk 47.69405, - W774 County M 96.31216 9/21/2013 Anas discors 19475 This Study Minnesota: Polk 34.27864, - W775 County C 106.8527 1/9/2014 Anas discors 19576 This Study Minnesota: Polk 47.69405, - W776 County M 96.31216 9/21/2013 Anas discors 19481 This Study Minnesota: Polk 47.69405, - W777 County M 96.31216 9/21/2013 Anas discors 19482 This Study Minnesota: Polk 47.69405, - W778 County M 96.31216 9/21/2013 Anas discors 19483 This Study Minnesota: Polk 47.69405, - W779 County M 96.31216 9/21/2013 Anas discors 19484 This Study Minnesota: Polk 47.69405, - W814 County M 96.31216 9/21/2013 Anas discors 19617 This Study Minnesota: Polk 47.69405, - W815 County M 96.31216 9/21/2013 Anas discors 19625 This Study Minnesota: Polk 47.69405, - W816 County M 96.31216 9/21/2013 Anas discors 19626 This Study

195

Minnesota: Polk 47.69405, - W816.2 County M 96.31216 9/21/2013 Anas discors - This Study Argentina: -30.0999, - W831 Corrientes SH 59.49915 6/1/2013 Anas versicolor 23178 This Study Argentina: -30.0999, - Ebbs et al. 2016; W833 Corrientes SH 59.49915 6/1/2013 Anas versicolor 23180 This Study New Zealand: -44.568863, Anas W845 South Island SH 170.1872 5/24/2015 rhynchotis 24887 This Study New Zealand: -44.568863, Anas W846 South Island SH 170.1872 5/24/2015 rhynchotis 24888 This Study North Carolina: W857 Hyde County A 1/15/2016 Anas clypeata - This Study W925 Puerto Rico A - - Anas discors - This Study Brant and Loker 2009; Ebbs et al. E64 Florida A - 10/1/2002 Anas discors 24777 2016 Brant and Loker E45 Florida A - 10/1/2002 Anas discors 24778 2009 E45.2 Florida A - 10/1/2002 Anas discors - Ebbs et al. 2016 Brant and Loker SDS1006 Nebraska C - 11/1/2004 Anas clypeata 24776 2009; This Study New Zealand: Anas TshovNZ South Island SH rhynchotis 20794 Ebbs et al. 2016 Minnesota: 45.0684361, W413 Ramsey County M -93.1257 6/17/2009 Physa gyrina 186 Brant et al. 2011 Trichobilharzia physellae New Mexico: 33.7131, - Brant and Loker W146 Soccoro County C 106.9579 3/25/2004 Physa gyrina - 2009; This Study New Mexico: 33.7131, - Brant and Loker W147 Soccoro County C 106.9579 3/25/2004 Physa gyrina - 2009; This Study

196

New Mexico: 33.7131, - Brant and Loker W150 Soccoro County C 106.9579 3/25/2004 Anas strepera 18586 2009; This Study Pennsylvania: Erie 41.170333, - Brant and Loker W171 County A 80.086883 11/10/2004 Aythya affinis 18565 2009 Pennsylvania: 41.59215, - Anas Ebbs et al. 2016; W175 Crawford County A 80.4093666 11/10/2004 platyrhynchos 18567 This Study Pennsylvania: 41.59215, - Anas W176 Crawford County A 80.4093666 11/10/2004 platyrhynchos 18568 This Study New Mexico: Brant and Loker W193 Chaves County C 33.45, -104.4 2005 Aythya affinis 18610 2009; This study New Mexico: Brant and Loker W194 Chaves County C 33.45, -104.4 2005 Aythya collaris 18611 2009; This study New Mexico: Ebbs et al. 2016; W196 Chaves County C 33.45, -104.4 2005 Aythya affinis 18613 This Study Alaska: North 68.64636, - Clangula W211 Slope Borough P 149.44891 2005 hyemalis 18644 This Study Alaska: North 68.98204, - Brant and Loker W212 Slope Borough P 148.83182 2005 Aythya affinis 18645 2009 Mergus merganser 45.581, - Mergus Brant and Loker W230I1I2 eggs MI M 84.697 2005 merganser - 2009; This study Michigan: Emmet 45.581, - Brant and Loker W234 County M 84.697 2005 Physa gyrina 18656 2009 Michigan: Emmet 45.581, - Brant and Loker W235 County M 84.697 2005 Physa gyrina 18657 2009; This study Michigan: Emmet 45.581, - Brant and Loker W236 County M 84.697 2005 Physa gyrina 18658 2009; This study Nevada: Churchill 39.9, - Aythya Brant and Loker W249 County P 118.8166 2005 valisineria 18554 2009; This study New Mexico: Bucephala Brant and Loker W255 Chaves County C 33.45, -104.4 2006 albeola 19159 2009; This study

197

New Mexico: Brant and Loker W256 Chaves County C 33.45, -104.4 2006 Aythya affinis 177 2009 New Mexico: 35.8485, - Brant and Loker W263 Sandoval County C 106.4907 2006 Physa gyrina 178 2009; This study New Mexico: 33.79336, - W408 Soccoro County C 106.8912 2008 Physa gyrina 18686 This Study New Mexico: 35.2166138, W497 Bernalillo County C -106.59698 2009 Physa acuta 18700 This Study Bucephala W602 Alaska: Fairbanks P - 2011 albeola 25267 This Study Bucephala W604 Alaska: Fairbanks P - 2011 clangula 25269 This Study W608 Alaska: Fairbanks P - 2011 Aythya collaris 25260 This Study Florida: Hamilton 30.42307, - W745 County A 82.7936 1/12/2013 Aythya collaris - This Study New Mexico: 33.79336, - W765 Soccoro County C 106.8912 2014 Aythya affinis - This Study New Mexico: 33.79336, - W765.3 Soccoro County C 106.8912 2014 Aythya affinis - This Study New Mexico: 33.79336, - W765.4 Soccoro County C 106.8912 2014 Aythya affinis - This Study New Mexico: 33.79336, - W765.5 Soccoro County C 106.8912 2014 Aythya affinis - This Study New Mexico: 33.79336, - W765.6 Soccoro County C 106.8912 2014 Aythya affinis - This Study MGC804 Alberta C - - Physa gyrina - Gordy et al. 2016 Aythya SAL92-09 Canada: Alberta C - - americana - This Study Trichobilharzia sp. A

198

New Mexico: 33.7131, - Anas Brant and Loker W149.1 Socorro County C 106.9579 2004 americana 18585 2009; This Study New Mexico: 33.7131, - Anas W149.2 Socorro County C 106.9579 2004 americana 18585 California: Imperial 33.284266, - Anas Brant and Loker W182.1 County P 115.579216 2004 americana 18574 2009 California: Imperial 33.284266, - Anas W182.2 County P 115.579216 2004 americana - New Mexico: 32.9071, - Anas Brant and Loker W192 Sierra county C 107.3116 2004 americana 18609 2009; This Study Alaska: Yukon- 67.4093611, Anas W206 Koyukuk Borogh P -150.104883 2005 americana 18639 This Study Alaska: Yukon- 67.4093611, Anas W206.2 Koyukuk Borogh P -150.104883 2005 americana 18639 Ebbs et al. 2016 Alaska: North 68.98204, - Anas Brant and Loker W213 Slope Borough P 148.8318 2005 americana 18646 2009 Alaska: North 68.98204, - Anas W213.2 Slope Borough P 148.8318 2005 americana 18646 This Study New Mexico: 36.46363, - Stagnicola W510 Colfax county C 105.2733 2011 elodes 18705 This Study New Mexico: 36.46363, - Stagnicola W510.2 Colfax county C 105.2733 2011 elodes 18705 This Study New Mexico: 36.46363, - Stagnicola W511 Colfax county C 105.2733 2011 elodes 18706 This Study New Mexico: 36.46363, - Stagnicola W511_2 Colfax county C 105.2733 2011 elodes 18706 This Study New Mexico: 36.46363, - Stagnicola W512 Colfax county C 105.2733 2011 elodes 18707 This Study Anas W600 Alaska: Fairbanks P - 2011 americana 25265 This Study

199

Anas Brant and Loker W601 Alaska: Fairbanks P - 2011 americana 25266 2009; This Study Anas Brant and Loker W607 Alaska: Fairbanks P - 2011 americana 25258 2009; This Study Anas Brant and Loker W609 Alaska: Fairbanks P - 2011 americana 25259 2009; This Study Anas Brant and Loker W611 Alaska: Fairbanks P - 2011 americana 25262 2009; This Study Anas Brant and Loker VVT3043 North Dakota C - - americana - 2009; This Study

200

p- distance Taxa CO1 ND4 ITS1 Trichobilharzia querquedulae 0.01 0.014 0.002 vs. T. querquedulae like 0.047 0.035 - vs. T. physellae 0.095 0.133 0.012 vs. Trichobilharzia sp. A 0.112 0.1 0.009 vs. T. franki 0.075 - 0.016 vs. Trichobilharzia sp. C 0.089 - 0.034 vs. W701 0.098 - 0.014 vs. T. regenti 0.125 0.129 0.337 vs. T. mergi 0.116 - - vs. T. anseri 0.13 - - vs. T. stagnicolae 0.121 - - vs. T. szidati 0.106 - - vs. Anserobilharzia brantae 0.141 - - Trichobilharzia querquedulae - like - 0.036 - Trichobilharzia phsyellae 0.009 0.034 0.006 vs. T. querquedulae like 0.083 0.136 - vs. Trichobilharzia sp. A 0.107 0.136 0.006 vs. T. franki 0.081 - 0.014 vs. Trichobilharzia sp. C 0.095 - 0.027 vs. W701 0.066 - 0.016 vs. T. regenti 0.1 0.163 0.333 vs. T. mergi 0.131 - - vs. T. stagnicolae 0.143 - vs. T. anseri 0.126 - - vs. T. szidati 0.124 - - vs. Anserobilharzia brantae 0.14 - - Trichobilharzia sp. A 0.014 0.007 0.003 vs. T. querquedulae like 0.1 0.098 - vs. T. franki 0.096 - 0.011 vs. Trichobilharzia sp. C 0.093 - 0.029 vs. W701 0.097 - 0.008 vs. T. regenti 0.116 0.127 0.333 vs. T. mergi 0.124 - - vs. T. anseri 0.135 - - vs. T. stagnicolae 0.14 - - vs. T. szidati 0.116 - - vs. Anserobilharzia brantae 0.137 - - Trichobilharzia sp. B 0.01 0.16 0.002 vs. Trichobilharzia sp. A 0.075 0.094 0.007 vs. Trichobilharzia sp. C 0.092 - 0.033 vs. T. physellae 0.103 0.159 0.01 vs. T. querquedulae 0.094 0.139 0.013 vs. T. fanki 0.081 - 0.008 Trichobilharzia franki 0.004 - 0.001 vs. Trichobilharzia sp. C 0.072 - 0.037 vs. W701 0.091 - 0.026 vs. T. regenti 0.093 - 0.34 vs. T. mergi 0.114 - - vs. T. anseri 0.129 - - vs. T. stagnicolae 0.135 - - vs. T. szidati 0.121 - -

201

vs. Anserobilharzia brantae 0.145 - - Trichobilharzia regenti 0.002 - 0.011 vs. T. mergi 0.111 - - vs. T. anseri 0.116 - - vs. T. stagnicolae 0.127 - - vs. T. szidati 0.124 - - vs. Anserobilharzia brantae 0.126 - - Trichobilharzia mergi 0 - - Trichobilharzia anseri 0.003 - - Trichobilharzia stagnicolae 0.007 - - Trichobilharzia szidati 0.008 - - Anserobilhariza brantae 0.013 - - vs. Allobilhariza visceralis 0.143 - - Allobilharzia visceralis 0.01 - -

Additional File 2. Pairwise p-distances within and between Trichobilharzia species.

202

Additional File 3. nad4 Phylogeny. BI inference of TQ, TP and TA, only posterior probabilities ≥.95 are presented. Branches are colored by species; TQ= orange,

TP=green and TA=blue. Individual worms are labeled by the migratory flyway they were collected from, denoted by a colored box. Average pairwise p-distances between TQ subclades are indicated.

203

Additional File 4. ITS1 Phylogeny. BI inference of TQ, TP and TA, only posterior probabilities ≥.95 are presented. Branches are colored by species; TQ= orange,

TP=green and TA=blue.

204

W137 W752 W756.2 W162 W767.4 W743 W925 W748 W750.2 W203 W190 W183 SDS1006 W156 W159 W154.2 W750 W160 W180 W753 W777 W166.2 W664 W213 W831 W259 W833 W747 W778 .99 E45 W774 W148.2 W154 W191 W152.4 W704 W153 W158 W746 W703 W749 W161 W181 W754 W756 W148 W767.3 W155.3 Tshov W755.2 W135 W142 Trichobilharzia sp. C Trichobilharzia franki W701 W604 W194 W497 .97 W234 W235 W193 W602 W249 W510 W608 W230 W211 W213 W146 W147 W236 W745 W196 W765.3 W765.6 W765 W765.5 W765.4 W256 W171 W175 W255 W263 Trichobilharzia sp. B W149 .97 W212 Pacific W601 W609 Central W182 W192 Mississippi W600 W511 Atlantic W607 S. Hemisphere Trichobilharzia regenti 0.02

205

Additional File 5. nad4 + cox1+ ITS1 phylogeny. BI inference of TQ, TP and TA, only posterior probabilities ≥.95 are presented. Branches are colored by species;

TQ= orange, TP=green and TA=blue. Individual worms are labeled by the migratory flyway they were collected from, denoted by a colored box.

206

Additional File 6. Minimum spanning network, showing relationships of TP haplotypes by host genera.

207

Additional File 7. Minimum spanning network, showing relationships of TQ haplotypes by host species.

208

Additional File 8. Bayesian skyline plot of cox1 datasets, all mutation rates. Estimates of Ne (y-axis) over time (kya, x-axis), using a 2%(dashed line) and

4% (solid line) rate of cox1 change. TA populations were further subdivided by flyway as significant among flyway divergence was suspected. Lines are colored by species; TQ= orange, TP=green and TA=blue.

209

Trichobilharzia querquedulae Pacific Central Mississippi Atlantic S. Hemisphere AK CA NM NE ND MB MN LA FL PR AR ZA NZ AK / CA 0.03448 / NM 0.10577 -0.02813 / NE 1 -0.4 -0.34135 / ND 0.71429 0.06728 -0.02875 0.5 / MB 0.18519 0.39558** 0.38637* -0.04762 0.35 / MN 0.2381 -0.00378 -0.02657 -0.23077 -0.09615 0.3543 / LA -0.02703 -0.04893 0.00183 -0.31034 0.08042 0.39144 0.0293 / FL -0.10238 -0.0388 -0.0066 -0.43789 -0.08456 0.29923** -0.04908 -0.02889 / PR 1 0.5942 0.55854 1 0.85714 0.46341 0.64444 0.54217 0.3987 / AR 0.5 0.0884 0.02967 0.2 0 0.34783 -0.03125 0.1289 -0.03954 0.73333 / ZA 1 -0.47368 -0.56157 1 -1 -0.15789 -1 -0.40741 -0.5938 1 -1 / NZ 0.37931 0.21437* 0.14928 0 0.04545 0.40517* 0.10526 0.17263* 0.10268* 0.64706 0.07692 -0.63636 /

Trichobilharzia physellae Pacific Central Miss Atlantic AK CA NM AB MI PA AK / CA -0.7193 / NM 0.29524** 0.11111 / AB -0.30667 1 -0.85507 / MI 0.0582 -0.66667 0.24159** -0.33333 / PA 0.12695 1 -0.25703 0 0.11111 /

Additional File 9. Pairwise ΦST values by flyway for TQ and TP. (*) indicates a p-value ≤ 0.05, (**) indicates a p-value ≤ 0.01.

210

MSB: Para: Specimen ID Locality Host Species # NCBI Source

Trichobilharzia sp. B W205 Alaska Anas americana 18638 ─ This study W210 Alaska Anas americana 18643 ─ This study Trichobilharzia stagnicolae W164 New Mexico Stagnicola elodes 18659 FJ174461 Brant and Loker, 2009 Brant and Loker 2009; This W221 Michigan Mergus merganser 18649 ─ study W222 Michigan Mergus merganser 18650 KP734304 Brant and Loker 2009 Brant and Loker 2009; This W223 Michigan Mergus merganser 18651 ─ study Brant and Loker 2009; This W224 Michigan Stagnicola emarginata 18652 ─ study Brant and Loker 2009; This W225 Michigan Stagnicola emarginata 19505 ─ study W229 Michigan Stagnicola emarginata 179 FJ1744543 Brant and Loker 2009 Brant and Loker 2009; This W230 Michigan Mergus merganser 18652 ─ study Brant and Loker 2009; This W231 Michigan Mergus merganser 18654 ─ study W240 Michigan Stagnicola emarginata 18673 FJ174462 Brant and Loker, 2009 Trichobilharzia franki BERS1 France Radix auricularia ─ HM131199 Jouet et al. 2010 EAN77 France Radix auricularia ─ HM131201 Jouet et al. 2010 BERS67 France Radix auricularia ─ HM131200 Jouet et al. 2010 STRS2 France Radix auricularia ─ HM131202 Jouet et al. 2010 FORS3 France Radix auricularia ─ HM131198 Jouet et al. 2010 FORS4 France Radix auricularia ─ HM131197 Jouet et al. 2010 Trichobilharzia regenti ACI25 Iceland Anas platyrhynchos ─ HM439504 Jouet et al. 2010 BERS58 France Anas platyrhynchos ─ HM439502 Jouet et al. 2010

211

HAR1 France Mergus merganser ─ HM439501 Jouet et al. 2010 CYA18 France Cygnus olor ─ HM439500 Jouet et al. 2010 AC122 Iceland Anas platyrhynchos ─ HM439503 Jouet et al. 2010 EAN9 France Radix peregra ─ HM439499 Jouet et al. 2010 AY157190 Czech Republic Radix peregra ─ AY157190 Lockyer et al. 2003 Trichobilharzia szdati Lsk2 Siberia Lymnaea stagnalis ─ JF838203 Korsunenko et al. 2012 Lsm20 Siberia Lymnaea stagnalis ─ JF838202 Korsunenko et al. 2012 Lsm7 Siberia Lymnaea stagnalis ─ JF838201 Korsunenko et al. 2012 Lsm4 Siberia Lymnaea stagnalis ─ JF838200 Korsunenko et al. 2012 Lsm3 Siberia Lymnaea stagnalis ─ JF838199 Korsunenko et al. 2012 Lsm2 Siberia Lymnaea stagnalis ─ JF838198 Korsunenko et al. 2012 Lsm1 Siberia Lymnaea stagnalis ─ JF838197 Korsunenko et al. 2012 AY157191 Czech Republic Lymnaea stagnalis ─ AY157191 Lockyer et al. 2003 Trichobilharzia anseri ANS41 France Anser anser ─ KP901385 Jouet et al 2015 ANS40 France Anser anser ─ KP901384 Jouet et al 2015 ANS36 France Anser anser ─ KP901383 Jouet et al 2015 ANS32 France Anser anser ─ KP901382 Jouet et al 2015 FCI1 Iceland Radix balthica ─ KP901381 Jouet et al 2015 ICR9 Iceland Radix balthica ─ KP901380 Jouet et al 2015

Additional File 10. Trichobilharzia samples included in estimates of species wide diversity

212

APPENDIX C CHAPTER IV ADDITIONAL FILES

MSB: Taxa Sample ID Host Locality Lat:Long PARA: # Anserobilharzia brantae W340/352 Branta canadensis Manitoba, CAN 58.7419, -93.9747 176/18628 Austrobilharzia variglandis W697 Larus argentatus Quebec, CAN 48.674, -66.237 - Bilharziella polonica W930 Anas platyrhynchos Ukraine - - Dendritobilharzia pulverulenta W926 Anas crecca New Mexico, USA - - Dendritobilharzia pulverulenta W813 Anas discors Minnesota, USA 47.69405, -96.31216 19616 Gigantobilharzia huronensis W414/513 Physa sp. New Mexico, USA 35.21166, -106.5969 18687/25488 Gigantobilharzia huronensis W678 Physa sp. New Mexico, USA 35.21166, -106.5969 - Heterobilharzia americana W805 Procyon lotor Louisiana, USA 32.3258, -91.3855 19286 Macrobilharzia macrobilharzia W931 Anhinga anhinga Mississippi, USA - - Ornithobilharzia canaliculata W393 Larus occidentalis California, USA - 18542 Trichobilharzia physellae W604 Bucephala clangula Alaska, USA - 25269 Trichobilharzia querquedulae W203 Anas clypeata Alaska, USA 65.665, -149.098 18636 Trichobilharzia querquedulae W929 Anas discors Puerto Rico, USA - - Trichobilharzia stagnicolae W233 Stagnicola emarginata Michigan, USA 45.569,-84.71215 19509 Trichobilharzia sp. A W510/511 Stagnicola elodes New Mexico, USA 36.4636, -105.2733 18705 Trichobilharzia sp. E W606 Anas crecca Alaska, USA - 25257 Avian Schistosome C11 W607 Anas americana Alaska, USA - 25258 Avian Schistosome D W342 Gyraulus parvus Manitoba, CAN 58.7369, -93.7982 18619 Schistosomatidae W216 W216 Haminoea japonica California, USA 37.7678, -122.2776 18660 Schistosomatidae W688 W688 Indoplanorbis exustus Nepal 27.616575, 83.097119 18710

Additional File 1. Locality and accession information for schistosome specimens.

213

UCE Sample ID Taxa Loci Schistosomatidae SAMEA2201407 Allobilharzia visceralis 1254 PRJEB519 Schistosoma curassoni 2421 PRJNA273970 Schistosoma haematobium 2423 PRJEB2425 Schistosoma haematobium 2422 PRJNA354903 Schistosoma japonicum 2384 PRJNA286685 Schistosoma japonicum 2382 PRJEB13625 Schistosoma mansoni 2417 PRJNA302212 Schistosoma mansoni 2411 PRJEB523 Schistosoma mattheei 2422 PRJEB522 Schistosoma margrebowiei 2421 PRJEB526 Schistosoma rhodani 2424 SAMEA1920831 Schistosomatium douthitti 115 SAMEA2422295 Trichobilharzia regenti 2400 PRJEB4661 Trichobilharzia szidati 1219 Out-groups PRJEB3954 Dicrocoelium dendriticum 58 PRJEB1207 Echinostoma caproni 296 PRJNA438721 Fasciola gigantica 260 PRJNA392273 Opisthorchis viverrini 214 PRJNA248332 Paragonimus westermani 246

Additional File 2. Published genomes mined for UCE loci.