ABSTRACT Title of Disertation: EMERGING INFECTIOUS DISEASE: HOST AND PARASITE PERSPECTIVES Jon S. Beadel, Doctor of Philosophy, 2007 Directed By: Dr. Robert C. Fleischer, Head, Center for Conservation and Evolutionary Genetics, National Zoological Park Dr. Gerald S. Wilkinson, Profesor Behavior, Ecology, Evolution & Systematics Avian malaria and related haematozoa are nearly ubiquitous parasites that can impose fitnes costs of variable severity and may, in some cases, cause substantial mortality in their host populations. One example of the later, the emergence of avian malaria in the endemic avifauna of Hawai, has become a model for understanding the consequences of human-mediated disease introduction. The drastic declines of native Hawaian birds due to avian malaria provided the impetus for examining more closely several aspects of host-parasite interactions in this system. Host-specificity is an important character determining the extent to which a parasite may emerge. Traditional parasite clasification, however, has used host information as a character in taxonomical identification, potentialy obscuring the true host range of many parasites. To improve upon previous methods, I first developed molecular tools to identify parasites infecting a particular host. I then used these molecular techniques to characterize host-specificity of parasites in the genera Plasmodium and Haemoproteus. I show that parasites in the genus Plasmodium exhibit low specificity and are therefore most likely to emerge in new hosts in the future. Subsequently, I characterized the global distribution of the single lineage of P. relictum that has emerged in Hawai. I demonstrate that this parasite has a broad host distribution worldwide, that it is likely of Old World origin and that it has been introduced to numerous islands around the world, where it may have been overlooked as a cause of decline in native birds. I also demonstrate that morphological clasification of P. relictum does not capture diferences among groups of parasites that appear to be reproductively isolated based on molecular evidence. Finaly, I examined whether reduced imunological capacity, which has been proposed to explain the susceptibility of Hawaian endemics, is a general feature of an ?island syndrome? in isolated avifauna of the remote Pacific. I show that, over multiple time scales, changes in imune response are not uniform and that observed changes probably reflect diferences in genetic diversity, parasite exposure and life history that are unique to each species. EMERGING INFECTIOUS DISEASE: HOST AND PARASITE PERSPECTIVES By Jon Sumner Beadel Disertation submited to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfilment of the requirements for the degre of Doctor of Philosophy 2007 Advisory Commite: Profesor Gerald S. Wilkinson, Chair Profesor Robert C. Fleischer, Co-Advisor Profesor Dina M. Fonseca Profesor Mathew P. Hare Profesor David J. Hawthorne Profesor Sarah A. Tishkoff ? Copyright by Jon S. Beadel 2007 i Preface This disertation is comprised of an introduction and four chapters. Chapters I (Journal of Parasitology. 2005. 91, 683-685), I (Molecular Ecology. 2004. 13, 3829- 3844) and II (Procedings of the Royal Society London B. 2006. 273, 2935-2944) are presented in the formats in which they were originaly published. Chapter IV is formated for submision to The American Naturalist. A single bibliography provides information on references cited throughout the disertation. ii Dedication In memory of Far Far and Grandpa Chuck. iv Acknowledgements I am grateful to my advisor Dr. Robert Fleischer for opening the door to evolutionary genetics, for his generosity, for making available numerous opportunities to expand my horizons, and for his unfailingly positive view towards my work. Thanks, too, to my co-advisor Dr. Gerald Wilkinson and to Lois Reid for guiding an absente student through the mechanics of graduate life. And for making 5 years of life at the zoo a fun and productive time, I am especialy indebted to my pipete-wielding labmates at the Genetics Program. Thank you to Colm Atkins, Farah Ishtiaq, Carl McIntosh and Jen Red for their help in the lab and especialy to Erin Cashion and Michele Jonker for their valiant eforts in the field. Drs. Estele Russek-Cohen and Lary Douglas provided invaluable statistical advice. And many thanks to the long list of people in Hawai, Australia, French Polynesia, Christmas Island and even further afield who helped me coordinate fieldwork, provided tisues or data, or consulted on methodologies. My work was funded by the University of Maryland Biology Department (block grant), National Science Foundation (graduate felowship, doctoral disertation improvement grant, and grant to R. Fleischer), National Institutes of Health (grant to R. Flesicher), Smithsonian Institution (predoctoral felowship and Abbott Fund grant), Cosmos Club Foundation and the American Ornithologists? Union. v Table of Contents Preface...........................................................i Dedication........................................................ii Acknowledgements.................................................iv Table of Contents...................................................v List of Tables.....................................................vii List of Figures....................................................vii Introduction........................................................1 Chapter I A restriction enzyme-based asay to distinguish betwen avian hemosporidians....9 Abstract.......................................................9 Main Text.....................................................9 Table........................................................16 Figure Legends................................................17 Figures.......................................................18 Appendices...................................................20 Chapter I Prevalence and diferential host-specificity of two avian blood parasite genera in the Australo-Papuan region..............................................23 Abstract......................................................23 Introduction...................................................24 Materials and Methods...........................................26 Results.......................................................32 Discussion....................................................37 Table........................................................43 Figure Legends................................................44 Figures.......................................................45 Appendices...................................................49 Chapter II Global phylogeographic limits of Hawai?s avian malaria....................57 Abstract......................................................57 Introduction...................................................58 Material and Methods...........................................59 Results and Discussion..........................................63 Table........................................................73 Figure Legends................................................75 Figures......................................................... 77 Appendices...................................................82 vi CHAPTER IV Imunological change in island birds: no uniform evidence of an ?island syndrome? ................................................................95 Abstract......................................................95 Introduction...................................................96 Methods.....................................................101 Results......................................................106 Discussion...................................................112 Table.......................................................121 Figure Legends...............................................122 Figures......................................................123 Appendix....................................................125 Bibliography.....................................................127 vii List of Tables CHAPTER I Table 1. Host, geographical origin, and sources of parasite lineages for which full mitochondrial genomes were sequenced. CHAPTER I Table 1. Prevalence of Plasmodium and Haemoproteus asesed by PCR screning selected avian host families from the Australo-Papuan region. Estimates of prevalence are biased low because identification of genus was not possible for al samples (Genus Unknown). CHAPTER II Table 1. Sampling efort and frequency with which the Hawaian lineage of Plasmodium (GRW4, lineage 15) was recovered from regions shown in figure 2. Lineages defined as unique (difering by at least 1 bp) within a given region may be shared betwen regions. CHAPTER IV Table 1. Genetic diversity indices for introduced and endemic island taxa compared to their mainland counterpart. vii List of Figures CHAPTER I Figure 1. Frequency of insertions/deletions and nucleotide substitutions across five mitochondrial genomes from Plasmodium spp. and Haemoproteus spp. Upper limits of 200-nucleotide intervals are numbered with respect to positions in P. falciparum (AY283008, Joy et al., 2003). The locations of cytochrome c oxidase subunits I (CO1) and II (COII) and cytochrome b (CYb) are indicated at the top. Figure 2. Electrophoretic banding paterns indicative of infection with Plasmodium sp. (P), Haemoproteus sp. (H) or Leucocytozoon sp. (L) relative to a 100 base pair ladder. Multi-genus infections (M) may be resolved by electrophoresis of PCR products for 1.3 hours on a 4% agarose gel. CHAPTER I Figure 1. Location of sampling sites in tropical Australia and Papua New Guinea. Figure 2. Cladogram depiction of neighbor-joining tre based on LogDet distances betwen mitochondrial lineages of avian hematozoa. Region of origin (A for Australia, P for Papua New Guinea, B for both regions), lineage number, host species (color-coded for family), and frequency of detection (number in parenthesis when recovered more than once) are indicated at right. Red branches indicate bootstrap support greater than 70% (1000 replicates). Pale blue and dark blue dots indicate lineages used for binomial tests of host conservation. ix Figure 3. Relationships among hematozoan parasites based on maximum likelihood using the model GTR + I + G. Lineage number and host species (color-coded for family) are indicated at right. Red branches indicate bootstrap support greater than 70% (stepwise addition, 100 replicates). Figure 4. Logistic regresion curves relating the predicted probability of host relatednes to genetic diferentiation of parasites in the genus Plasmodium (P) and Haemoproteus clade A (HA) and clade B (HB). Dotted lines indicate the genetic distance at which parasite pairs from clades A and B were equaly likely to be found in hosts of the same or diferent families. CHAPTER II Figure 1. Phylogenetic relationships among morphologicaly-identified species of Plasmodium estimated using ML, MP and ME with cytochrome b sequences. Numbers above branches indicate bootstrap support based on 1000 replicates. Numbers before species names correspond to mitochondrial lineage numbers in figure 3. Sequences were obtained directly from Genbank (acesion number in italics) or from extracts of blood smears obtained from the International Reference Collection for Avian Haematozoa (IRCAH) and M. Peirce. Figure 2. Map depicting the global distribution of the single mitochondrial lineage of malaria parasite (GRW4) found in resident Hawaian paserines. Pie charts indicate x the proportion of al sequenced Plasmodium infections in a given region that were identical to GRW4 (red). Details concerning locations and sampling efort are in table 1. Red dots indicate additional locations in which GRW4 has been reported previously (Ishtiaq et al.; Mendes et al. 2005; Schrenzel et al. 2003; M. Kimura pers. comm.) or in which GRW4 was recovered from a relatively smal group of samples (Kazakhstan). Figure 3. Phylogenetic tres of parasite mitochondrial lineages (cyt b; left; numbered) and asociated nuclear haplotypes (DHFR-TS; right; letered), constructed using maximum likelihood (GTR+I+G for both markers). The distribution of mitochondrial lineages across global regions is indicated with squares, color-coded to help identify the asociated DHFR-TS sequence (when available, otherwise black). Background shading reflects the limits of two clades with good support in analysis of combined data (figure 4). Mitochondrial lineages which derived from at least one parasite identified as P. relictum by morphology are indicated in bold. Bootstrap support values (>60) are indicated above branches. Figure 4. Majority rule consensus tre of avian malaria parasite lineages generated by Bayesian analysis of combined mitochondrial (cytb b) and nuclear (DHFR-TS) sequence. Parasite genotypes are identified by their respective cytb b lineage (number) and DHFR-TS haplotype (leter), which are depicted separately in figure 3. Clade credibility values are indicated above branches. Background shading identifies two wel-supported clades (A and B) referenced in the text. Dots within Clade A xi indicate nodes for which New World (open circle) or non-New World (black) origin could be confidently asigned based on ancestral trait reconstruction performed with Mesquite. Figure 5. Statistical parsimony network of Plasmodium mitochondrial lineages related to the Hawaian strain (lineage 15). Sampled haplotypes are numbered as in figure 3 and infered haplotypes are indicated by black dots. Shading indicates whether the lineage was detected in the Old World (white), New World (black) or in both regions (grey). Lineages 9, 19, and 36 were not included due to mising sequence. Lineages 7, 8, and 29 were joined at the 90% connection limit (13 substitutions). CHAPTER IV Figure 1. Prevalence of the haematozoan parasites Plasmodium spp. (black), Haemoproteus spp. (light grey) and Leucocytozoon spp. (dark grey) across mainland and island populations of Z. lateralis, N. temporalis, and Acrocephalus spp. Asterisks indicate significantly lower parasite prevalences in recently introduced island populations (Haemoproteus spp. in Z. lateralis) or island endemic species (Plasmodium spp. in A. aequinoctialis, above, and A. rimitarae, below) relative to their mainland counterparts. Figure 2. Cel-mediated (PHA-induced patagial sweling) and innate (agglutination and lysis of rabbit red blood cels) measures of imune response in endemic and xii recently-introduced island taxa (circles) relative to a mainland control (centered at zero). Numbers indicate sample size and standard eror is depicted by vertical bars, or by grey shading in the case of the mainland control. Filed circles indicate a significant diference from the mainland control. 1 INTRODUCTION ?The role of disease in wildlife conservation has probably been radicaly underestimated? ? Aldo Leopold, 1933 Infectious disease, long recognized as a threat to human health and an economic problem in domestic animals, has increasingly been acknowledged as a major factor driving ecological and evolutionary proceses in wild animal populations (Dem et al. 2001). In the decades imediately following Leopold?s observation, the role of disease in wildlife population dynamics remained somewhat obscure due in part to the absence of proper tools to detect and identify pathogens and skepticism on the part of some researchers that disease could be important in regulating wild populations (reviewed in Price 1991, Toft 1991). The advent of molecular biology, however, has alowed for agents of disease to be identified and traced on an increasingly fine scale. In addition, theoretical models have increasingly revealed the role parasites can have in the regulation of host populations (May and Anderson 1983). And over the past several decades, globalization of travel and trade, increasing encroachment of a burgeoning human populace into previously isolated habitats, monotypic agricultural and animal husbandry practices, and changes in climate have precipitated the emergence of numerous infectious diseases with undeniably catastrophic consequences for both humans and wildlife (Daszak et al. 2000, Woolhouse and Gowtage-Sequeria 2005, Pearce-Duvet 2006). 2 Emerging disease may be broadly defined as any infectious agent that has been recently discovered, is newly evolved, or has recently increased in incidence, geographical distribution or host distribution (Cleaveland et al. 2001). Factors leading to emergence may be inherent to the host (e.g., imunosuppresion due to increased exposure to pollution or stres) or the parasite (e.g., mutation, selection for higher virulence in the presence of a competing parasite) as wel as extrinsic elements such as changes in the distribution, behavior or competency of vectors or alternative hosts. Although infectious disease may be extremely detrimental to wild populations, empirical data and theory both suggest that disease by itself is generaly not capable of driving populations to extinction. Smith et al. (2006) reported that disease was a contributing factor in fewer than 4% of al extinctions known to have occurred in the last 500 years and that fewer than 8% of criticaly endangered taxa are threatened by disease. And in fact, while hypothesized to be involved in several extinctions (de Castro and Bolker 2005), infectious disease has been demonstrated to be the ultimate cause of extinction only in the case of a captive remnant population of Polynesian tre snail (due to a microsporidian parasite; Cunningham and Daszak 1998) and one wild frog species (due to fungal infection; Schloegel et al. 2006). Theoretical models predict that host-specific diseases that cause high mortality and exhibit density- dependent transmision are likely to die out as the host population fals below some threshold density (Anderson and May 1992). Therefore, infectious disease may be more important in driving populations to smal sizes at which point stochastic fluctuations in demographics or environment, ale efects, or genetic complications of inbreding provide the ultimate force behind extinction. 3 Examples of emerging diseases that have detrimentaly impacted wildlife include chytridiomycosis (fungi) in amphibians, rinderpest in African ruminants, phocine distemper in North Sea seals, rabies in African wild dogs and Ethiopian wolves, and myxomatosis in European rabbits (summaries in Cleaveland et al. 2001, Dobson and Foufopoulos 2001, Daszak et al. 2003). In birds, several diseases have come to public awarenes not only because of their impact on their hosts, but also their potential to spilover into humans (e.g. HPAI H5N1, SARS, West Nile Virus). Other notable disease outbreaks (reviewed in Cooper 1993, Friend et al. 2001) have driven the decline of house finches on the eastern seaboard of the U.S. (mycoplasmosis; Hochachka and Dhondt 2000), pelicans in the Salton Sea (botulism), and the honeycrepers of Hawai (avian malaria; Warner 1968, van Riper et al. 1986). This last example, the human-mediated introduction of avian malaria into the native avifauna of Hawai, represents one of the best known examples of emerging disease and provides the motivation for the work described in this disertation. Avian malaria (family Plasmodidae) and related haematozoan parasites in the families Haemoproteidae and Leucocytozoidae have been detected on every continent except Antarctica. Excluding a few host taxa restricted to extreme arctic environments (Bennet et al. 1992) and perhaps several island taxa, the vast majority of bird species are host to avian haematozoa with estimated prevalences based on microscopy averaging about 10% (Janovy 1997). Recent application of sensitive molecular techniques to the detection of haematozoa have indicated that actual prevalences are probably substantialy higher (Jarvi et al. 2003, Kimura et al. 2006) though these values can vary dramaticaly with host age clas (Mendes et al. 2005), 4 season of sampling (Falon et al. 2004) and geographical location (Bensch and ?keson 2003). Molecular techniques have also revealed a previously unrecognized diversity of parasite lineages (Perkins 2000, Ricklefs and Falon 2002), caling into question previous species limits and raising the posibility that haematozoan species diversity is on the order of avian species diversity (Bensch et al. 2004). Supporting this hypothesis are surveys of single host species that have recovered betwen 5 and 26 distinct parasite mitochondrial lineages (Bensch et al. 2007, Durrant et al. unpublished, Falon et al. 2006, Ishtiaq et al. 2006, Kimura et al. 2006) and initial indications that mitochondrial lineages may represent good species (Bensch et al. 2000). Direct measures of haematozoon infection intensity (Bensch et al. 2007), as wel as the discrepancies typicaly encountered when detecting parasites by microscopy and PCR (Bentz et al. 2006), suggest that most infected birds cary low level chronic infections. While it is hypothesized that these infections are unlikely to have serious consequences for host fitnes (Valkiunas 2005), quantifying non-lethal efects of infection in wild populations is notoriously dificult and therefore, the true impact of haematozoans is likely to be blurred. At least several studies have demonstrated negative consequences of haematozoan infection for survival, clutch size, incubation period, fledging succes, motor activity and fat acumulation (Bennet et al. 1993, Gustafson et al. 1994, Nordling et al. 1998, Merino 2000, Valkiunas 2005) though numerous other studies have yielded more equivocal results. Bennet et al. (1993) suggested that severe mortality atributable to haematozoan parasites appears to be limited to cases involving domesticated birds or species 5 introduced to exotic ranges. If haematozoa are important in limiting the range of most bird species, as hypothesized by Bennet et al. (1993), then changes to the distribution of local haematozoan faunas could have disastrous consequences for bird populations worldwide. High mortality observed upon translocation of presumably na?ve species to foreign zoos (penguins, reviewed in Valkiunas 2005; New Zealand avifauna, Bennet et al. 1993, Tompkins and Gleson 2006) and upon introduction of malaria to Hawai provide experimental support for this possibility. Unlike most continental species, the native birds of Hawai were likely not exposed to malaria until some time in the early 19th century, when mosquitoes were first introduced to the Hawaian Islands (se Fonseca et al. 2006). Once established, the presence of a suitable vector (Culex quinquefasciatus) alowed for colonization of the islands by the parasites themselves, which may have arived via migratory ducks and shorebirds or via the importation of domestic fowl, gamebirds and caged song birds (Warner 1968). Preliminary screning of historical specimens suggests that avian malaria only arived as late as the middle of the twentieth century (Fleischer et al. unpublished data). Many native bird species, apparently unusualy susceptible to disease and already decimated by the introduction of mosquito-vectored pox virus, experienced further declines and possible extinction as a result of exposure to malaria. Disease-driven extinction in this case may have been made possible by the presence of competent alternative hosts (non-native birds) which are largely unafected by malaria (van Riper et al. 1986, Jarvi et al. 2003) yet provide a large reservoir of parasites which may then be transmited to native hosts. The presence of alternative host reservoirs provides an important exception to models which predict 6 the density-dependent decline of parasite impact along with the host (Smith et al. 2006). The extreme declines observed in Hawai raise the following questions: To what extent can avian malaria exploit various hosts and thus, to what extent does avian malaria present an emerging threat to bird populations elsewhere across the globe? What is the identity of the parasite introduced to Hawai? What is the geographical distribution of the Hawaian parasite, what is its prevalence, and can we use this information to infer anything about its virulence outside of Hawai? And finaly, are other remote avifaunas of the Pacific likely to exhibit susceptibility similar to that of the Hawaian honeycrepers? These questions are addresed in the chapters that follow. In Chapter I, I describe the development of a PCR asay that alows for the rapid and sensitive detection of haematozoan DNA within a sample extracted from bird tisue or blood. This method is an improvement upon previous asays based on microscopy, which can drasticaly underestimate the prevalence of chronic infections (Jarvi et al. 2003) and other PCR methods, which are ineficient because they employ nested amplifications (Valkiunas et al. 2006) or target relatively large fragments. The asay described in this chapter targets a smal conserved fragment of mitochondrial DNA identified by sequencing full mitochondrial genomes from five avian haematozoan parasites. The asay also employs restriction fragment length polymorphism to provide a genetic fingerprint that is diagnostic of the genus of parasite present in an infected sample. In Chapter I, I investigate the degre to which haematozoan parasites are constrained to a particular host species or family. Most emerging diseases infect a 7 broad range of hosts (Taylor et al. 2001, Woolhouse and Gowtage-Sequeria 2005), and therefore, host-specificity is likely a key factor determining both the extent to which a parasite may ?emerge? from its typical environment and also, whether density-dependent efects asociated with the decline of a primary host are likely to extinguish the pathogen threat. An initial application of molecular techniques to this question may have been misleading due to spartan sampling and the fact that the authors neglected to distinguish betwen haematozoan genera (Ricklefs and Falon 2002). Here, I present the results of a parasitological survey of closely related hosts from several avian families found in a single geographic region. I show that host- specificity of Haemoproteus spp. is high relative to Plasmodium spp. at multiple depths within the parasite phylogeny. This work highlights the potential for Plasmodium spp. to emerge as a problematic parasite in novel hosts. In addition, I show that host-specificity is not uniform even acros lineages of Haemoproteus, suggesting that parasite diferentiation has been the result of periodic host-switching events, followed by vicariance. In Chapter II, I use mitochondrial and nuclear markers to track the origins and spread of Plasmodium relictum, a species of avian malaria that emerged in the na?ve avifauna of Hawai. First I demonstrate that the Hawaian parasites exhibit a uniform genetic signature and therefore, may be defined much more narowly than the collection of parasites encompased by the morphological species P. relictum. Using this genetic signature, I show that Hawaian-like parasites are distributed across the globe, and currently occur on several other remote islands which could potentialy share a similar history of avian demise. Finaly, I use phylogenetic reconstructions of 8 additional parasite lineages recovered from a survey of over 13,000 specimens to show that the Hawaian parasite likely originated from an Old World source. This, in turn, may influence conclusions regarding the evolution of virulence. Finaly, in Chapter IV, I approach the topic of emerging disease from the perspective of a putatively na?ve island avifauna. Here I ask the questions a) Is loss of imune function a component of an ?island syndrome??, and b) On what time scale do changes in imunity occur? To answer these questions, I characterized parasite exposure, genetic diversity and imune response in recently-introduced and endemic island taxa and compared the results to related mainland counterparts. As expected, I found that parasite exposure on the islands was generaly lower than on the mainland and indices of genetic diversity also tended to be lower on the islands, particularly for long-term endemic residents. Interestingly, imune response in the island taxa was not necesarily lower than mainland taxa and the response was not uniform even among island taxa with similar residence times. 9 CHAPTER I A restriction enzyme-based asay to distinguish betwen avian hemosporidians ABSTRACT We describe a reliable and relatively inexpensive method for detecting and diferentiating betwen the commonly studied avian blood parasite genera Haemoproteus, Plasmodium, and Leucocytozoon. The asay takes advantage of a Haemoproteus-specific restriction site identified by sequencing full mitochondrial genomes from two Haemoproteus and thre Plasmodium lineages and an adjacent genus-specific restriction site identified in Leucocytozoon spp. The asay was sensitive to parasitemias of about 8 x 10 -6 and was 100% acurate in diferentiating betwen parasite genera isolated from a broad geographical and taxonomic sampling of infected hosts. MAIN TEXT Avian hemosporidia in the genera Haemoproteus, Plasmodium, and Leucocytozoon are widespread blood parasites occurring in diverse host species. The 3 genera are closely related evolutionarily (Perkins and Schal, 2002), but are characterized by considerable diferences in ecology and life history. Plasmodium spp. are transmited by culicine mosquitoes, whereas Haemoproteus spp. are transmited by hippoboscid 10 and ceratopogonid flies, and Leucocytozoon spp. are vectored by simulid flies. Species of Plasmodium generaly exhibit broader host specificity (Atkinson and van Riper II, 1991) while both Plasmodium spp. and Leucocytozoon spp. are considered to be more pathogenic (Bennet, Peirce, and Ashford, 1993) than Haemoproteus spp. Al 3 genera undergo alternating cycles of sexual and asexual reproduction, but only species of Plasmodium undergo schizogony in circulating erythrocytes (Atkinson and van Riper II, 1991), which leads to the symptoms of malaria (Garnham, 1966). Recently, the polymerase chain reaction (PCR) has become a valuable tool for detecting these parasites and sequencing has uncovered a wealth of genetic diversity (Bensch et al., 2000; Ricklefs and Falon, 2002; Schrenzel et al., 2003). PCR screns offer increased sensitivity over traditional analysis of blood smears (Feldman et al., 1995; Cann et al., 1996; Richard et al., 2002); however, the parasites (most often Haemoproteus spp. and Plasmodium spp.) are typicaly amplified indiscriminately, and identification to genus requires costly sequencing. Lineage-specific information may be useful when monitoring an epidemic or measuring migratory connectivity (Webster et al., 2002), but often, ecologists, wildlife managers, and zookeepers may benefit from genus-level knowledge of parasites present in an avian community. Here, we report the development of a relatively inexpensive restriction enzyme-based diagnostic for detecting and identifying Haemoproteus spp., Plasmodium spp. and Leucocytozoon spp. We have confirmed the utility of the test by correctly identifying the genus of geneticaly diverse parasite lineages from a broad taxonomic and geographic sampling of hosts. 11 We initialy searched for a method to distinguish Haemoproteus spp. and Plasmodium spp., the two most common genera, and the two most commonly co- amplified by existing PCR methods. To ensure that candidate nucleotide polymorphisms separating parasite genera fel in relatively conserved regions of DNA, we first sequenced the 6-kilobase mitochondrial genome of 5 avian parasite lineages. These included 2 lineages of Haemoproteus and 3 lineages of parasite putatively identified as P. relictum by morphology (Table I), but known to be geneticaly divergent (McConkey et al., 1996; J. Beadel unpubl. obs.). We extracted DNA using Qiaquick Dneasy kits (Qiagen, Valencia, California). We designed primers (available upon request) based on the sequences of relatively conserved regions of previously published mitochondrial genomes from mamalian malaria parasites. Primer pairs typicaly amplified betwen 500 and 600 base pairs (bp) and were spaced so that approximately 30 to 50 bp (excluding primed sequence) from one segment overlapped with the adjacent segment. Amplified DNA was sequenced bi- directionaly on an ABI 377 (Applied Biosystems, Inc., Foster City, California). Sequences were asembled, aligned and edited using the program SEQUENCHER version 4.1. Further alignment was performed by eye when necesary. Sequences of the five mitochondrial genomes have been deposited in GenBank (Acesion numbers AY733086 to AY733090, Table I). Comparison of the frequency of insertions/deletions and nucleotide substitutions across the 5 genomes with the location of mitochondrial gene sequences (COI, COII, CYb) identified in P. falciparum (AY283008, Joy et al., 2003) suggested that these regions were least conserved across hematozoan genera (Fig. 1). 12 Average uncorrected pairwise divergence among the 5 genomes was 2.5% outside gene-coding regions versus 7.7% within. Therefore, we targeted our search for polymorphisms to non-gene-coding regions where polymorphisms betwen genera were most likely to be fixed. We describe the resulting diagnostic below. We designed primers 213F (5?- GAG CTA TGA CGC TAT CGA -3?) and 372R (5?- GA TG AGA GT CAC GT TA -3?) to amplify a 160 bp fragment of DNA encoding an Xmn I restriction site unique to Haemoproteus. Both primers shared 100% identity with corresponding sequence of mitochondrial DNA from the parasites listed in Table I, and the final fragment size was minimized so as to improve the likelihood of recovering PCR product from old and potentialy degraded samples. We employed a PCR scheme typical of ?ancient DNA? amplifications: initial denaturing step at 94 C for 8 min, 45 cycles of 92 C for 30 sec, 52 C for 30 sec, 72 C for 30 sec, and a final extension at 72 C for 7 min. We amplified 1.8 ?l of DNA in a reaction volume of 25 ?l, using final concentrations of 1X PCR Gold Buffer (Applied Biosystems, Inc.), 0.8 mM dNTPs, 2.0 mM gCl 2 , 0.8 mg/ml BSA, 0.6 ?M each primer and 0.5 U of AmpliTaq Gold DNA Polymerase (Applied Biosystems, Inc.). We digested 5.0 ?l of the PCR product in a total volume of 10 ?l for 3 hr at 37 C, using 1 U of restriction enzyme Xmn I, 0.1 mg/ml BSA and 1X NEBuffer 2 (New England Biolabs, Beverly, MA). We electrophoresed the digested product at 80 volts on a 4% 3:1 agarose gel (Amresco, Solon, OH) with 0.2X Gel Star (Cambrex Bio Science Rockland, Inc., Rockland, ME). PCR products derived from Plasmodium spp. remained undigested while PCR products derived from Haemoproteus spp. were 13 cleaved into fragments of 121 and 39 base pairs (Fig. 2). The smalest band was typicaly faint or indistinguishable from primer artifacts. We tested the method described above on DNA extracted from 33 infected birds representing 21 families from 3 continents (Appendix I). The genus of the infecting parasite had been determined by prior sequencing and phylogenetic analysis of a 295 base pair segment of cytochrome b. Uncorected pairwise divergences ranged from 1.4 to 8.5% (average = 5.7%) among Plasmodium lineages tested (excluding samples from Schoeniophylax phyganophila and Gnorimopsar chopi for which the full 295 bp was not available) and from 0 to 9.5% (average = 5.2%) among Haemoproteus lineages. The diagnostic succesfully resolved the genus of infection in al cases (17 Plasmodium sp. and 16 Haemoproteus sp.). Because the diversity of lineages tested within each genus resembled that recovered in other surveys of hematozoa (Bensch et al., 2000; Ricklefs and Falon, 2002; Schrenzel et al., 2003), the test should be applicable on a broad scale. The test may also detect double infections by both genera when template DNA for each parasite occurs in sufficient concentration. We tested 4 samples for which prior sequencing and cloning had revealed infection with both Haemoproteus and Plasmodium. In each case, 2 bands were evident on the agarose gel (Fig. 2). Underdigestion of PCR products could also result in the appearance of a multi-genus infection, however, we did not observe double bands among the 33 samples tested above. The use of exces restriction enzyme and long incubation times should minimize these artifacts. We subsequently used primers 213F/372R to amplify DNA from diverse lineages of Leucocytozoon spp. extracted from blood smears (Appendix 2). 14 Sequencing revealed an absence of the Xmn I site, and digestion with Xmn I confirmed that the PCR products were indistinguishable from the products obtained from Plasmodium spp. However, comparison of the sequences with those from the lineages of Haemoproteus and Plasmodium tested above revealed an Xba I site unique to lineages of Leucocytozoon. Digestion of PCR products from each of the Leucocytozoon spp. with Xba I under conditions identical to those described above produced 2 fragments of about 109 and 54 bp (smal size diferences arose from single base indels at two sites within the larger fragment), while digestion of the products obtained from Haemoproteus spp. and Plasmodium spp. resulted in no cleavage. The use of Xmn I and Xba I together, therefore, generated a diagnostic molecular fingerprint for the thre hemosporidian genera (fragments of approx. 160 bp for Plasmodium spp., 121 bp for Haemoproteus spp., 109 and 54 bp for Leucocytozoon spp.) when PCR products were electrophoresed on a 4% agarose gel (Fig. 2). To determine the utility of these primers in detecting low-level infection, we tested the primers on extractions of serial dilutions of duck blood containing the Hawaian lineage of P. relictum. We estimated the parasitemia of the stock blood to be 3.23% by scanning 30 fields at 400x magnification. We consistently amplified extractions of stock blood diluted as much as 4,000-fold; however, amplification of more dilute DNA was sporadic or impossible. This level of sensitivity is similar to that for a protocol developed by Falon et al. (2003), which outperformed previously published methods. Serological asays may be even more sensitive (Jarvi et al., 2002); however, any single serological asay is specific for a limited set of lineages 15 and may detect both current and previous infections. PCR-based methods of detection may be imperfect (Fred and Cann, 2003), but using multiple primer sets can help reduce false negatives. We suggest that the asay described here wil complement previous methods by providing a relatively sensitive method for detecting current hematozoan infection and a robust and cost-efective means for identifying Haemoproteus spp., Plasmodium spp. and Leucocytozoon spp. We also note here that divergences (based on mean character diferences) betwen mitochondrial lineages from the thre parasites identified by morphology as P. relictum ranged from 3.4 to 3.9%. By comparison, P. reichenowi (AJ251941, Conway et al., 2000) and P. falciparum (AY283008, Joy et al., 2003), parasite species whose primary hosts are chimpanzes and humans, respectively, exhibited only 2.3% sequence divergence across their entire mitochondria. These data lend further support to authors who have questioned the phylogenetic significance of previous parasite clasifications based on morphology (Manwel, 1936; Bensch et al., 2000; Perkins and Schal, 2002). If mitochondrial lineages are indicative of species-level diferentiation, then extreme care should be taken in asesing evolutionary relationships or tracking the epidemiology of hematozoan parasites based solely on morphological and ecological phenotypes. This is particularly true for P. relictum, which has been implicated in the decline of Hawai?s native avifauna (van Riper et al., 1986; Atkinson et al., 1995), but has also been reported in blood smears from a broad distribution of avian hosts around the world (Bennet et al., 1993). 16 TABLE Table I. Host, geographical origin, and sources of parasite lineages for which full mitochondrial genomes were sequenced. Lineage Parasite name Host name Geographical Collector GenBank Scientific Common origin acesion no. 1 Haemoproteus sp. Meliphaga lewinii Lewin?s honeyeater Queensland, Australia J. Austin AY733086 2 Haemoproteus sp. Lichenostomus frenatus Bridled honeyeater Queensland, Australia J. Austin AY733087 3 Plasmodium relictum Spheniscus demersus African black- footed penguin Baltimore Zoo, Maryland, U.S.A. M. Cranfield AY733088 4 Plasmodium relictum Zenaida macroura Mourning dove Nebraska, U.S.A. ATC #30141 AY733089 5 Plasmodium relictum Hemignathus virens Amakihi Hawai, U.S.A. R. C. F. AY733090 17 FIGURE LEGENDS Figure 1. Frequency of insertions/deletions and nucleotide substitutions across five mitochondrial genomes from Plasmodium spp. and Haemoproteus spp. Upper limits of 200-nucleotide intervals are numbered with respect to positions in P. falciparum (AY283008, Joy et al., 2003). The locations of cytochrome c oxidase subunits 1 (CO1) and II (COII) and cytochrome b (CYb) are indicated at the top. Figure 2. Electrophoretic banding paterns indicative of infection with Plasmodium sp. (P), Haemoproteus sp. (H) or Leucocytozoon sp. (L) relative to a 100 base pair ladder. Multi-genus infections (M) may be resolved by electrophoresis of PCR products for 1.3 hours on a 4% agarose gel. 18 FIGURES 0 10 20 30 40 50 60 200600 1000140018002200260030003400380042004600500054005800 Position within mitochondria (nucleotides) F r e q u e n c y o f i n d e l o r n u c l e o t i d e s u b s t i u t i o n COIII COI CYb 19 20 APPENDICES Apendix I. Geographical origin and family designation of avian host species with known infections used to test ability of asay to distinguish betwen Plasmodium sp. (P) and Haemoproteus sp. (H). Host Host family Parasite Australia Acanthiza katherina Acanthizidae P Sericornis citreogularis Acanthizidae H Sericornis citreogularis Acanthizidae H Meliphaga notata Meliphagidae P Xanthotis macleayana Meliphagidae H Monarcha trivirgatus Monarchidae P Colluricincla megarhyncha Pachycephalidae H Eopsaltria australis Petroicidae H Eopsaltria australis Petroicidae H Pita versicolor Pitidae P Ailuroedus melanotis Ptilonorhynchidae P Gabon Ipsidina lecontei Alcedinidae H Merops breweri Meropidae H Trochocercus nigromitratus Monarchidae H Stiphrornis erythrothorax Muscicapidae H 21 Zoothera cameronensis Muscicapidae P Zoothera cameronensis Muscicapidae P Nectarinia fuliginosa Nectariniidae H Camaroptera brachyura Sylviidae P Iladopsis cleaveri Timalidae H Iladopsis cleaveri Timalidae P Iladopsis rufipennis Timalidae P Uruguay Basileuterus culicivorus Emberizidae P Gnorimopsar chopi Emberizidae P Icterus cayanensis Emberizidae H Molothrus badius Emberizidae H Schoeniophylax phyganophila Furnaridae P Mimus saturninus Mimidae P Colaptes campestris Picidae H Polioptila dumicola Polioptilidae P Aramides ypecaha Ralidae P Rallus sanguinolentus Ralidae H Troglodytes aedon Troglodytidae P 22 Apendix I. Host, geographical origin, source and number of Leucocytozoon spp. tested in asay. Parasite # Host Geographical origin Source L. mardouxi 2 Nesoenas mayeri Mauritius International Reference Centre for Avian Haematozoa, Brisbane, Australia L. toddi 2 Milvago chimango Chiloe Island, Chile U.S. National Parasite Collection, Maryland, U.S.A. Leucocytozoon sp. 1 Carduelis spinus Lithuania G. Valkiunas Leucocytozoon sp. 1 Sylvia borin Lithuania G. Valkiunas Leucocytozoon spp. 6 Junco hyemalis Alaska, U.S.A. P. Deviche 23 CHAPTER II Prevalence and diferential host-specificity of two avian blood parasite genera in the Australo-Papuan region ABSTRACT The degre to which widespread avian blood parasites in the genera Plasmodium and Haemoproteus pose a threat to novel hosts depends in part on the degre to which they are constrained to a particular host or host family. We examined the host distribution and host-specificity of these parasites in birds from two relatively understudied and isolated locations: Australia and Papua New Guinea. Using PCR, we detected infection in 69 of 105 species, representing 44% of individuals surveyed (n = 428). Across host families, prevalence of Haemoproteus ranged from 13% (Acanthizidae) to 56% (Petroicidae) while prevalence of Plasmodium ranged from 3% (Petroicidae) to 47% (Ptilonorhynchidae). We recovered 78 unique mitochondrial lineages from 155 sequences. Related lineages of Haemoproteus were more likely to derive from the same host family than predicted by chance at shalow (avg. LogDet genetic distance = 0, n = 12, P = 0.001) and greater depths (avg. distance = 0.014, n = 11, P < 0.001) within the parasite phylogeny. Within two major Haemoproteus subclades identified in a maximum likelihood phylogeny, host-specificity was evident up to parasite genetic distances of 0.029 and 0.007 based on logistic regresion. We found no significant host relationship among lineages of Plasmodium by any method of analysis. These results support previous evidence of strong host-family specificity 24 in Haemoproteus and suggest that lineages of Plasmodium are more likely to form evolutionarily-stable asociations with novel hosts. INTRODUCTION The application of molecular methods to the study of avian hematozoa has revealed surprising levels of genetic diversity. This diversity has been exploited to reveal phylogenetic relationships (Perkins and Schal 2002), ases disease linkage betwen breding and wintering grounds (Waldenstroem et al. 2002), and investigate host- parasite fidelity (Bensch et al. 2000, Ricklefs and Falon 2002, Falon et al. 2003). This last isue is of particular importance as human activities alter the ranges of vectors and avian hosts, thereby increasing exposure of potential hosts to novel parasites. In Hawai, the introduction of the malarial parasite Plasmodium relictum has been implicated in the decline of native honeycrepers (van Riper et al. 1986). The negative impact of hematozoa introduced to domesticated birds has also been wel documented (reviewed in Bennet et al. 1993a); however, discerning the fitnes consequences of infections in wild birds with long histories of parasite exposure has been more dificult (Hatchwel et al. 2001, Sikamaki et al. 1997). Predicting the consequences of introduced disease is dificult, but we can begin to ases the chances of an exotic parasite spreading to novel hosts by determining the extent to which that parasite is evolutionarily constrained to a particular host or host family. Two of the most common and best-studied genera of avian blood parasites are Plasmodium and Haemoproteus. Earlier studies have suggested that Haemoproteus exhibits greater host-specificity than Plasmodium (Bennet and Peirce 1988, Bennet et al. 1993b). Traditional means of clasifying parasites at the species level, however, 25 have often included host taxonomy as a character, thereby providing a biased estimate of host-parasite conservatism (Atkinson and van Riper 1991). In addition, reconstructions of parasite phylogenies based on DNA sequences have yielded evolutionary relationships that difer from those derived from traditional clasification methods (Escalante et al. 1998). A recent molecular study of Haemoproteus lineages in old world warblers and tits produced discordant host and parasite phylogenies, suggesting frequent host- switching (Bensch et al. 2000). A survey of parasites in African residents and European migrants revealed numerous cases of a single parasite lineage shared by multiple hosts; al Haemoproteus lineages were shared among hosts of the same family while at least one Plasmodium lineage occurred in multiple host families (Waldenstr?m et al. 2002). On a global scale, Ricklefs and Falon (2002) demonstrated relative conservatism of host-parasite evolution, but no distinction was made betwen the specificity of Plasmodium and Haemoproteus. Here, we atempt to merge the strengths of these studies by investigating host-parasite relationships at several evolutionary depths across multiple wel-diversified host families within a single region. As part of a global survey for the original host and geographical source of the Hawaian parasite, we examined malarial parasites from a subset of bird species from tropical Australia and Papua New Guinea. To our knowledge, this is the first molecular exploration of host-parasite relationships in this fauna. The avifauna of this region is relatively isolated, both taxonomicaly and geographicaly, potentialy reducing noise asociated with transient introduction of foreign parasites. Prior 26 surveys for hematozoan parasites have identified Haemoproteus and Plasmodium in many of the hosts included here, but relatively few parasites have been morphologicaly identified beyond the genus level (Ewers 1967, Bennet and Campbel 1973, Jones 1985). Our goals were to 1) characterize the prevalence of hematozoa across varied bird families in this region and 2) determine the extent to which Haemoproteus and Plasmodium difer in host-specificity. MATERIALS AND METHODS Sample Collection and Preparation JA collected blood samples from mist-neted birds in 2002 and 2003 at sites in the Wet Tropics of northeastern Queensland, Australia and at Eungela National Park, which encompases an isolated fragment of rainforest to the south (Figure 1). Blood smears for 40 samples were fixed with methanol and then stained with Giemsa for 30 min. For each slide, we searched 100 fields at 400x magnification to determine infection status. High resolution digital images of representative parasites were used for final identification. TP provided blood samples from birds captured in 2003 from the d?Entrecasteaux Islands, Papua New Guinea. JD provided blood and tisue samples of birds neted betwen 1991 and 2002 from forested sites across the main island of Papua New Guinea. We extracted host and parasite DNA from blood and tisue samples using the relevant protocols acompanying Qiagen DNeasy kits. Each extraction included a negative control, which was screned for contamination. Parasite Detection 27 In order to detect divergent and possibly degraded parasite DNA, we screned samples with two primer sets originaly designed to succesfully amplify Haemoproteus and Plasmodium DNA from dried blood smears up to 30 years old: 850F (5?-CT CA CTA TC TA TA AGT ATG T-3?) with 1024R (5?-AG TGA GTG TT TGC ATC AT-3?) and F2 (5?-AG TGA CC AC T AA AG-3?) with R2 (5?-GCT GTA TCA TAC CT AA G-3?). Prior use of these primers in a wide aray of avian hosts from varied geographical regions amplified no other hematozoa (e.g. Leucocytozoon, Trypanosoma, Hepatozoon). Primers 850F/1024R and F2/R2 amplify smal fragments (167 and 132 base pairs) with homology to portions of mitochondrial cytochrome oxidase II and cytochrome b genes (Feagin 1992), respectively. We used annealing temperatures of 50 C and 52 C, respectively, and typical PCR reactions employed conditions developed for amplification of ?ancient? DNA (Fleischer et al. 2000). For those samples that were positive based on the tests above, we amplified a larger fragment of cytochrome b (533 bp + primers) for use in phylogenetic analyses using primers 3760F (5?-GAG TG ATG TG TT TAG AT-3?) and 4292Rw2 (5?- TG AC AT ATG TAR AG AGT-3?). If this fragment did not amplify, we atempted to amplify smaler fragments of either 433 bp or 295 bp (+ primers) using either F1 (5?-CAT AT TAC T TAT CAT GA T-3?) or F3 (5?-CA GA CT GT TCA TG AT-3?) with 4292Rw2. The annealing temperature for these later reactions was 51 C. To ensure that DNA extractions were succesful for those samples in which we did not detect infection, we amplified a smal fragment (268 bp) of avian 28 cytochrome b DNA using primers cytb-2RC and cytb-wow following the methods described in Dumbacher et al. (2003). This amplification was succesful in al cases. Following purification of PCR products using Qiaquick kits (Qiagen), we bi- directionaly sequenced the largest fragment available for a given sample on an ABI 3100 Sequencer (Applied Biosystems, Inc.). Sequences were asembled, aligned and edited using the program SEQUENCHER version 4.1. Phylogenies based on cytochrome b sequence have consistently recovered two discrete clusters of lineages corresponding to Haemoproteus and Plasmodium (Bensch et al. 2000, Perkins and Schal 2002). Therefore, we asigned mitochondrial sequences (lineages) to each genus based on their asociations in a phylogenetic tre (se below). Inclusion of sequence data from prior studies and morphological asesments of parasites for which we had smears generaly alowed easy delineation of the two genera. In cases where limited sequence data did not provide sufficient resolution, we used a restriction enzyme test (JSB and RCF unpubl. data) to asign parasite lineages to genera. To ases whether prevalence of Haemoproteus and Plasmodium varied across host families, we performed an ANOVA (GLM in SAS v 8.2, SAS Institute, Inc., Cary, NC) on arcsine square-root transformed prevalences observed at the level of host species. We included only those species from families represented by greater than 10 individuals total (Table 1). We estimated the proportion of variance atributable to host family using the NESTED procedure in SAS. Cloning 29 In several cases, we detected multiple infections based on the occurrence of multiple peaks throughout the chromatogram. In these cases, we repeated the PCR and cloned the fragment using a TOPO-TA cloning kit (Invitrogen) following manufacturer guidelines. We picked 6 to 24 blue/white-selected colonies for each fragment cloned, boiled the colonies for 10 minutes, and amplified 2 ul of the resulting lysate for 30 cycles with the relevant primer set. Fragments from succesful amplifications were cleaned and sequenced as described above. Inspection of sequences obtained for a given clone, and comparison of those sequences with the original sequence, alowed for easy identification of PCR artifacts arising from polymerase eror or in vitro recombination (Thompson et al. 2002). Phylogenetic Analysis We estimated parasite phylogenetic relationships using al samples for which we had at least 295 base pairs of cytochrome b sequence, though 533 bp were available for most samples (se Appendix A). Following the phylogeny developed by Perkins and Schal (2002), al tres were rooted with mamalian Plasmodium sequences (GenBank acesion nos. AY069614, AF069624, AF055587, AY099051, AY283019, and AF069610). The program ModelTest v3.06 (Posada and Crandal 1998) indicated that the most likely model of base pair substitution was general time reversible (GTR), with the proportion of invariable sites = 0.3604 and gama shape parameter = 0.5372. We used maximum likelihood (ML) to reconstruct a phylogeny using these parameters. We used 100 replicates and the ?fast? heuristic in PAUP* (Swofford 1999) to estimate bootstrap support. We also performed a ful heuristic search for the shortest tre using tre-bisection reconnection (TBR) on both GTR and LogDet 30 (Lockhart et al. 1994) distances. We compared the resulting minimum evolution tre to 1000 tres generated by bootstrap resampling with a TBR heuristic search. Nodes with greater than 50% support were retained. Host-Specificity We followed the binomial probability approach of Ricklefs and Falon (2002) to ases the extent to which parasites of varying relatednes were likely to be found in host species from the same family. Host species were grouped into families as listed with the Handbook of the Birds of the World (2003), but we grouped al kingfishers in the Alcedinidae and included Rhipidura fantails within the Monarchidae (Sibley and Ahlquist 1985). First, we tested for a significant diference betwen the observed and expected probability that a shared parasite lineage (i.e., mitochondrial haplotypes indicated by light blue dots in Figure 2) derived from two host species of the same family. We calculated this separately for shared Haemoproteus and Plasmodium lineages. In cases where a single parasite lineage was found in more than two host species, we randomly paired hosts to represent that lineage. For example, if a lineage occurred in six diferent hosts, we randomly paired those hosts to form thre observations. Subsequently, we repeated the analysis using pairs of parasite lineages joined by 1 st -step nodes with greater than 70% bootstrap support (dark blue dots in Figure 2). When a 1 st -step node joined more than two host taxa, we randomly chose just a single independent pair. To quantify the phylogenetic depth being analyzed, we calculated average pairwise LogDet distances among parasites compared at each level. For al comparisons we used only lineages with greater than 470bp of sequence. 31 In order to extend the analysis beyond 1 st -step nodes and to ases the parasite genetic distance at which host family conservatism was lost, we performed a logistic regresion of host family (same or diferent) versus LogDet parasite distance (Ricklefs and Falon 2002). We tested for a significant influence of region (same or diferent) on host family similarity before using the full data set for each application of the model. Logistic regresion employs the model ln(P/(1-P) = a + b*d where ?P? is the probability that two parasites derive from hosts of the same family, ?d? is genetic distance, and ?a? and ?b? are coeficients estimated by the model. We performed this regresion on al pairwise comparisons of parasite lineages and their hosts at several levels of evolutionary organization. Because multiple pairwise distance comparisons violate asumptions of independence, we determined significance of the coeficients using a permutation of the original data. We randomly reasigned host families to the parasite phylogeny 999 times and performed logistic regresion upon each iteration. Coeficients based on the original data were compared to those generated by randomization in order to estimate the probability of recovering the original estimates by chance alone. 32 RESULTS Parasite Prevalence We used PCR to scren 428 individuals in total. Of 209 individuals from Papua New Guinea, 64 (31%) tested positive for Haemoproteus and 20 (10%) tested positive for Plasmodium. Of 77 species tested, 46 were positive for one or both genera and we detected infection in al 12 species for which we tested 5 or more individuals. Of 219 individuals tested from Australia, 62 (28%) were positive for Haemoproteus and 30 (14%) were positive for Plasmodium. We recovered Haemoproteus or Plasmodium from 27 of 32 species tested, and we found infection in 17 of 19 species for which 5 or more individuals were screned. Chi-squared tests revealed no significant diference in prevalence of either parasite betwen regions. Low PCR amplification, poor-quality sequence, or unresolved multiple infections reduced the number of samples for which we could identify parasites to genus, and therefore, estimates of prevalence (Table 1) were biased low. Prevalence of Haemoproteus, which ranged from 13% in the Acanthizidae to 56% in the Petroicidae was not uniform across diferent host families (F = 3.71, df = 7, p = 0.002), however, host family grouping explained only 22% of the total variance in prevalence among diferent host species. Except in the Ptilonorhynchidae, prevalence of Plasmodium was relatively low, and no significant diference was evident among families (F = 1.39, df = 7, p = 0.223). Only about 4% of the variance in prevalence betwen species could be atributed to host family. Family asignment and frequency of parasite detection for al host species examined is listed in Appendix B. 33 We detected mixed infections in 29 individuals. Among those with enough sequence data to identify parasite genera present, one individual harbored two Plasmodium lineages (66 and 72), 11 harbored two Haemoproteus lineages (se below), and four harbored mixed Plasmodium/Haemoproteus infection (11 and 61, 16 and 72, 21 and 72, 18 and 70). Of the lineages involved in mixed Haemoproteus infections, four pairs derived from within wel-supported clades composed of non- paserines (3 and 4), Meliphagidae (28 and 29), or Petroicidae (35 and 36, 35 and 37). The remaining pairs (10 and 22, 13 and 37, 14 and 38, 14 and 39 repeatedly) were composed of parasites from each of the two main subclades (se phylogenetic results below). The average LogDet genetic distance betwen parasite combinations was 0.0623 (n = 1) for mixed Plasmodium, 0.0414 (n = 11) for mixed Haemoproteus, and 0.1352 (n = 4) for mixed Haemoproteus/Plasmodium. Reliability of Methods Failure to detect infection by PCR may have been due to low-quality or insufficient template, smal daily variation in PCR conditions and reaction composition, and mismatches betwen the primer and parasite DNA template. To generate a minimum estimate of our detection eror, we divided the number of false negatives produced by a given primer set by the total number of samples that were known to be positive by either primer set. By this method, the primer set F2/R2 had an eror rate of 30%, while primer set 850F/1024R mised infections at a rate of 17%. Therefore, even under favorable PCR conditions, the chance that both primer sets failed to detect an infection was about 5%. 34 Estimation of hematozoa presence/absence was identical for 35 of 40 samples analyzed by both PCR and visual inspection of blood smears. PCR screning detected infection in thre samples that went undetected by examination of blood smears. Conversely, an initial inspection of blood smears suggested that PCR had mised infections in two samples. Subsequent scanning of the slides by an unbiased second observer (MP), however, suggested that artifacts in these two slides had been misidentified as parasites. In samples where both methods identified a parasite to genus, 7 of 8 matched. The single disparity in genus identification was atributed to a poorly prepared slide and a second appraisal of the slide suggested that the parasite was representative of either Haemoproteus or Plasmodium. No other hematozoa were observed in blood smears. Phylogenetics Among the 165 samples for which we had at least 295 bp of sequence, we found 78 unique mitochondrial lineages: 60 Haemoproteus and 18 Plasmodium (GenBank acesion numbers listed in Appendix A). Lineage 60, isolated from Macropygia amboienensis, was included with Haemoproteus based on evidence from the restriction asay and morphological asesment of a parasite with a closely related mtDNA sequence (EG unpubl. data). Related lineages have also been found in Columbina passerina from North America (unpubl. data) and other doves (S Falon pers. comm.). Phylogenies developed using ML and LogDet and GTR distances were similar. Because each of these methods yielded similar topologies and for consistency with previous work by Ricklefs and Falon (2000), we used a tre derived from LogDet distances for tests of host-parasite specificity (Figure 2). 35 Within Haemoproteus, our data could not resolve deep hierarchical relationships, which resulted in a large basal polytomy. Parasites from two non- paserine host families occurred in a unique, wel-supported clade (top of Figure 2). Other clades descending from the genus-level polytomy included several which were largely derived from a single host family (Meliphagidae, Petroicidae, Pachycephalidae) and one wel-supported clade with diverse host family representation (clade A). Several wel-supported host-family-specific clades (Petroicidae, Pachycephalidae, Monarchidae) were nested within clade A. An ML estimate of the phylogeny (Figure 3) identified thre major clades within Haemoproteus: two lineages derived from paserine hosts (clades A and B) and a third composed of lineages from the two non-paserine families studied. Bootstrap support was relatively low for al but the non-paserine clade. The ML phylogeny also indicated monophyly of al unshared parasites recovered from Meliphagidae. Deper level relationships among Plasmodium lineages were similarly unresolved in a distance-based phylogeny (Figure 2). Beneath the genus-level polytomy, only a pair of lineages (64 and 65) from Meliphagidae fel into a smal wel-supported host-specific clade. Host-Specificity We found 12 Haemoproteus lineages that were each shared by two diferent host species and we found 6 Plasmodium lineages in more than one host species. Thre of these Plasmodium lineages were each found in thre to six host species. Related lineages of Haemoproteus were more likely to be found in related hosts than predicted by chance. At average parasite genetic distances of 0 (shared identical 36 lineages) and 0.014 (1 st -step nodes), the probability of related parasites deriving from the same host family was 0.58 (n = 12, P = 0.001) and 0.73 (n = 11, P < 0.001), respectively. Sample sizes for comparisons within the Plasmodium genus were smaler. The probability that a shared Plasmodium lineage derived from the same host was 0.13 (average distance = 0, n = 8, P = 0.65). This value was not significant even if pairs of hosts were chosen so as to maximize the probability (probability = 0.38, n = 8, P = 0.11). Similarly, sister lineages joined by 1 st -step nodes were not significantly likely to have derived from the same host family (average distance = 0.008, probability = 0.33, n = 3, P = 0.61). We applied logistic regresion to four groups of parasites: al Plasmodium lineages, al Haemoproteus lineages, Haemoproteus clade A, and Haemoproteus clade B. Because ?region? did not contribute significantly to the regresion of host family on distance, we considered Australia/Papua New Guinea to be one region for al logistic regresion analyses. Regresion coeficients for the genus Haemoproteus were significant (a = 0.1909, b = -38.57, P < 0.001) as were coeficients for clade A (a = 1.5701, b = -214.6, P < 0.001) and clade B (a = 1.5513, b = -53.7664, P <0.001). Coeficients for the genus Plasmodium were not significant (a = -1.7248, b = -1.8373, P = 0.63). By evaluating the regresion equation at P = 0.5, we could estimate the genetic distance at which pairs of parasite lineages were equaly likely to have derived from the same or diferent host family. This distance at which host-family signal was lost was 0.005 for al Haemoproteus lineages, 0.007 for clade A, and 0.029 for clade B. Evaluating the regresion equation at a distance of zero, the predicted 37 probability of finding identical parasites in hosts of the same family was 0.55 evaluated over al Haemoproteus lineages, 0.83 for clade A, and 0.83 for clade B. Figure 4 depicts the predicted regresion curves for Plasmodium and Haemoproteus clades A and B. DISCUSSION Epizootiology Blood parasites in the genera Haemoproteus and Plasmodium appear to be nearly ubiquitous in avian communities. We detected one or both of these genera in almost 66% of species and this number would likely rise substantialy with deeper sampling of individual species. In the Australo-Papuan region studied, we estimated an overal prevalence of about 44% with no significant diferences betwen northeast Australia and New Guinea lowlands. Estimates of prevalence in tropical regions have ranged from about 10% in Costa Rica and the Neotropics (White et al. 1978, Young et al. 1993; by blood smear) to 28% in the Leser Antiles (S Falon pers. comm.), 40% in Central Africa (Richard et al. 2002) and 59% in American Samoa (Plasmodium only, Jarvi et al. 2003). Comparison of prevalence across surveys is confounded by diferences in sensitivity of the diagnostics employed (Richard et al. 2002) and our PCR technique underestimated infection by at least 5%. Serological tests may provide the most acurate estimate of infection by detecting low-level chronic infections (Jarvi et al. 2002), but interpretation of the asays can be dificult (Jarvi et al. 2003), lineage identification is impossible, and the methods may not be applicable across varied hosts and parasite lineages. 38 Comparison of prevalence among regions is also likely to be confounded by the host families sampled. Except in the Ptilonorhynchidae, which were sampled only sparsely, prevalence of Plasmodium was low and fairly uniform among wel- represented host families. On the other hand, prevalence of Haemoproteus varied significantly among host families, and this could bias regional comparisons in cases where families are not represented equaly. Although certain host families such as the Columbidae repeatedly exhibit relatively high prevalence of infection across studies (Atkinson and van Riper 1991), estimates of prevalence, even if acurate, should be considered snapshots in time and host space (Bensch and Akeson 2003, Scheuerlein and Ricklefs 2004). Infection rates can vary dramaticaly betwen years and may be more representative of diferences in vector abundance and their distribution within diferent habitats than family-level diferences in host imune response or other evolved characters (Bennet and Cameron 1974). Within the Pachycephalidae, we expected that parasite prevalence might have been lower among pitohuis, a group which produces varying amounts of a toxic alkaloid potentialy active in invertebrate vectors (Dumbacher et al. 1992). The overal infection rate in this group (40%), however, was close to both the value for the entire family (35%) and the average prevalence within Papua New Guinea (46%). This suggests litle role for the toxin in vector deterence, however, collection of Papuan birds occurred over several years and the caveats mentioned above may apply. We uncovered multiple infections from a wide range of hosts. Given that at least 40% of individuals were infected by either Haemoproteus or Plasmodium, the 39 prevalence of mixed infections should have been fairly high if not constrained by parasite-parasite interaction (Hatchwel et al. 2000). The 29 cases of multiple infection that we uncovered fel below the expected number of about 60 (based on overal prevalences of about 30% and 12% for Haemoproteus and Plasmodium, respectively). While this may be indicative of competitive exclusion, the cases of multiple infection observed represent a minimum since we did not recover sequence data from every infected individual and even succesful PCR was likely to mis some multiple infections due to primer bias or unequal quantities of parasite DNA. Hematozoan genera may have evolved distinctive antigenic signatures that avoid cross-generic imunity in a common host (Atkinson and van Riper 1991), but the extent to which the evolutionary relatednes of parasites within genera influences inter-lineage competition and thus, the distribution of parasites, should be addresed more carefully in the future. Host-Parasite Evolution Parasite lineages found in more than one host have often been cited as evidence of host-switching. While the introduction of parasites into novel hosts is a prerequisite for host-switching, the current distribution of parasites may not reflect long-term co- evolution betwen the parasite and its vertebrate host, but may be more indicative of the cosmopolitan feding of its invertebrate vector. Generalist feders such as mosquitoes or ceratopogonid flies may drive the continuous introduction of varied Plasmodium and Haemoproteus lineages into diverse hosts. Not al of these interactions wil necesarily be stable throughout time. For example, Atkinson (1986) demonstrated that Haemoproteus meleagridis, a parasite commonly found in turkeys, 40 was capable of developing in other Galiformes, but infections were transient and rapidly cleared from these secondary hosts to which the parasite may have been poorly adapted. We found several lineages of both Plasmodium and Haemoproteus in multiple host families, however, identical Haemoproteus lineages were more likely to derive from related hosts than Plasmodium. Even if we asume that these cases represent evolutionarily stable changes in host afinity, recent host-switching by Haemoproteus lineages has been relatively constrained to related hosts. The significant signal of host family specificity observed in Haemoproteus at greater depths within the phylogeny, however, suggests that not al of the apparent asociations betwen a single parasite lineage and multiple host families represent stable interactions. Given the host-family conservatism at 1 st -step nodes and the strong signals from logistic regresion, evolutionarily stable jumps betwen host families are likely to be rare in the genus Haemoproteus. Across the genus Haemoproteus, the signal for host family specificity was lost at a parasite divergence of about 0.005. The atenuation in the signal, measured across the entire genus, was probably due to the structure of relationships betwen lineages within the genus. Analyzed separately, the two large subclades of Haemoproteus lineages derived from paserine hosts both exhibited strong host specificity. For clade B, in which the average pairwise divergence among parasites was about 0.075, the host signal extended to a parasite divergence of about 0.029. Within clade A, average pairwise parasite divergence was only about 0.021, and host-specificity was evident up to a parasite divergence of only 0.007. 41 Lineages within clades A and B may have diversified via periodic host- switching following an early vicariance event in an ancient Haemoproteus lineage. In both clades, however, we were largely unable to recover wel-supported hierarchical relationships among groups of parasites derived from diferent host families, suggesting that the comon ancestor to each clade spread rapidly across host families. Asuming that rates of nucleotide substitution are similar across various lineages of Haemoproteus, the short branch lengths in clade A suggest a relatively recent radiation of parasites acros host taxa. Without further sampling, it wil remain unclear how frequently lineages have escaped otherwise strong host constraint. If younger parasite radiations have spread broadly across avian hosts in the past, this phenomenon of escape and radiation would continualy reset the parasite molecular clock relative to the avian clock. This in turn could help to explain the apparent slow divergence of parasite DNA relative to host DNA noted by Ricklefs and Falon (2002). Parasites in the genus Plasmodium appeared to be les constrained by the phylogenetic relationships of their hosts and showed no evidence of host-specificity at any depth within the parasite phylogeny. Our relatively smal sample of Plasmodium may have limited our power to detect a signal, however, we detected host-specificity within the equaly smal Haemoproteus clade A. Interestingly, Ricklefs and Falon (2002) detected host conservatism across both Haemoproteus and Plasmodium up to a parasite divergence of 0.026. Because they applied logistic regresion across lineages from both genera, however, it is unclear how that value partitioned betwen genera or betwen distinct radiations within genera. The evidence 42 here supports a broad host range for at least some Plasmodium parasites and indicates a tendency for a high level of evolutionarily stable host-switching. Of the two parasite genera studied, Plasmodium likely presents the greatest threat of colonizing novel hosts and may warant the most atention when managing the welfare of isolated and na?ve hosts. Most of the avian lineages sampled for parasites derived from a radiation of songbirds unique to the Australo-Papuan region (Sibley and Ahlquist 1985). In addition, birds from tropical Australia and New Guinea may be more isolated than their continental counterparts such that interactions betwen hosts, vectors, and parasites that would otherwise confound estimates of host-specificity are minimized. Nonetheles, trends in host-specificity observed in the Australo-Papuan region appear to be in line with the picture emerging from many other regional studies (partial summary in Schrenzel et al. 2003). Additional molecular surveys of parasites at the regional level wil add further insight into paterns of host-parasite interaction. 43 TABLE Table 1. Prevalence of Plasmodium and Haemoproteus asesed by PCR screning selected avian host families from the Australo- Papuan region. Estimates of prevalence are biased low because identification of genus was not possible for al samples (Genus Unknown). Host Family # Species # Samples Plasmodium Haemoproteus Genus Unknown Screned Screned # Positive % of Total # Positive % of Total # Positive % of Total Acanthizidae 12 69 9 13 9 13 1 2 Alcedinidae 8 23 1 4 5 22 1 4 Columbidae 8 17 1 6 9 53 1 6 Meliphagidae 15 70 7 10 34 49 1 1 onarchidae 13 54 13 24 14 26 0 0 Pachycephalidae 17 94 9 10 20 21 7 7 Petroicidae 6 34 1 3 19 56 1 3 Ptilonorhynchidae 2 15 7 47 6 40 0 0 Al Families 80 376 48 13 116 31 12 3 44 FIGURE LEGENDS Figure 1. Location of sampling sites in tropical Australia and Papua New Guinea. Figure 2. Cladogram depiction of neighbor-joining tre based on LogDet distances betwen mitochondrial lineages of avian hematozoa. Region of origin (A for Australia, P for Papua New Guinea, B for both regions), lineage number, host species (color-coded for family), and frequency of detection (number in parenthesis when recovered more than once) are indicated at right. Red branches indicate bootstrap support greater than 70% (1000 replicates). Pale blue and dark blue dots indicate lineages used for binomial tests of host conservation. Figure 3. Relationships among hematozoan parasites based on maximum likelihood using the model GTR + I + G. Lineage number and host species (color-coded for family) are indicated at right. Red branches indicate bootstrap support greater than 70% (stepwise addition, 100 replicates). Figure 4. Logistic regresion curves relating the predicted probability of host relatednes to genetic diferentiation of parasites in the genus Plasmodium (P) and Haemoproteus clade A (HA) and clade B (HB). Dotted lines indicate the genetic distance at which parasite pairs from clades A and B were equaly likely to be found in hosts of the same or diferent families. 45 FIGURES 46 P 1. Common paradise kingfisher (CPK) P 2. CPK (3), Yellow-billed kingfisher P 3. Superb fruit-dove P 4. Superb fruit-dove (3), Wompoo fruit-dove P 5. Yellow-billed kingfisher P 6. White-breasted fruit-dove 7. Northern fantail A 8. Yellow-throated scrubwren, Large-billed scrubwren (LBSW, 4) P 9. Dwarf honeyeater (6), Rusty mouse-warbler P 10. Western mountain white-eye P 11. Black sunbird P 12. Crested pitohui (2) A 13. Pale-ylow robin 14. Grheade i (GHR, 6) P 15. Papuan black myzomela, Varied triller A 16. Yellow-throated scrub wren, Large-billed scrubwren (2) 17. Bassian thrush A 18. Little shrike-thrush (4), Bower!s shrike-thrush (BST) P 19. Little shrike thrush (3) P 20. White-bellied pitohui, Rusty pitohui A 21. Tooth-billed bowerbird (TBBB, 3) P 22. Western mountain white-eye (2) P 23. Northern fantail (2) P 24. Shining flycatcher P 25. Spot-winged monarch A 26. Yellow-breasted boatbill P 27. Papuan black myzomela A 28. Dusky honeyeater (3) 29. Dusky honeyeater P 30. Helmeted friarbird P 31. Papuan black myzomela, Little shrike-thrush P 32. Long-billed honeyeater, Rufous babbler A 3. Eastern spinebil A 34. Pale-yellow robin (2) 35. A 36. Pale-yellow robin (2) 37. A 38. Grey-heade robin (2) 9. i 7 A 40. Pale-yellow robin (2) 41. Eastern yellow robin A 42. P 43. Black berrypecker A 44. Lewin!s honeyeater (2) 45. Bridled honeyeater (BHON) A 46. Bridled honeyeater (5) A 47. Chowchila A 48. Macleay!s honeyeater (3) P 49. Tawny-breasted honeyeater P 50. P 51. Smoky honeyeater (2) P 52. Spot-winged monarch, Frilled monarch P 53. Grey whistler P 54. Variable pitohui B 55. , Eungella honeyeater P 56. Hooded pitohui (2) P 57. Magnificent riflebird P 58. Golden monarch (3) P 59. Black butcherbird (2) P 60. Brown cuckoo-dove P 61. Black sunbird A 62. Golden whistler, Grey fantail 63. Rufous fantail A 64. Yellow-spotted honeyeater B 65. Mimic honeyeater, Lewin!s honeyeater A 66. Spectacled monarch (SMON, 3), Pied monarch (3), GHR P 67. Shining flycatcher (2) P 68. Wattled ploughbill A 69. Noisy pitta 70. Mountain thornbill, Little shrike-thrush, CPK P 71. White-bellied pitohui A 72. TBBB (3), SPC (2), BHON, LBSW (2), BST, SMON P 73. Hooded pitohui, Large scrubwren A 74. Yellow-spotted honeyeater 75. Spotted catbird (SPC) P 76. Emerald dove P 77. Mountain mouse-warbler P 78. Rusty mouse-warbler Mammalian Plasmodium Outgroup Haemoproteus Plasmodium Alcedinidae Columbidae Monarchidae Acanthizidae Meliphagidae Zosteropidae Nctarini Phycephalide etroiid Ptilonorhynchidae Family represented one time Host Family Key Clade A 47 1. Common paradise kingfisher 2. , Yellow-billed kingfisher 3. Superb fruit-dove 4. , Wompoo fruit-dove 5. Yelow-biled kingfisher 6. White-breasted fruit-dove 7. Northern fantail 8. Yelow-throated scrubwren, Large-billed scrubwren 9. Dwarf honeyeater, Rusty mouse-warbler 10. Western mountain white-eye 52. Spot-winged monarch, Frilled monarch 53. Grey whistler 27. Papuan black myzomela 28. Dusky honeyeater 29. 30. Helmeted friarbird 33. Eastern spinebill 31. Papuan black myzomela, Little shrike-thrush 32. Long-billed honeyeater, Rufous babbler 44. Lewin!s honeyeater 45. Bridled honeyeater (BHON) 46. Bridled honeyeater 48. Macleay!s honeyeater 49. Tawny-breasted honeyeater 50. 51. Smoky honeyeater 43. Black berrypecker 59. Black butcherbird 54. Variable pitohui 55. Variable pitohui, Eungella honeyeater 56. Hooded pitohui 47. Chowchilla 58. Golden monarch 57. Magnificent riflebird 34. Pale-yellow robin 35. 36. Pale-yellow robin 37. 40. Pale-yellow robin 41. Eastern yellow robin 42. 38. Grey-headed robin 39. 11. Black sunbird 23. Northern fantail 24. Shining flycatcher 25. Spot-winged monarch 22. Western-mountain white-eye 12. Crested pitohui 15. Papuan black myzomela, Varied triller 16. Yellow-throated scrubwren, Large-billed scrubwren (LBSW) 17. Bassian thrush 21. Tooth-billed bowerbird (TBBB) 13. Pale-yellow robin 14. Grey-headed robin 18. Little shrike-thrush, Bower!s shrike-thrush (BST) 19. 20. White-bellied pitohui, Rusty pitohui 26. Yellow-breasted boatbill 60. Brown cuckoo-dove 61. Black sunbird 62. Golden whistler, Grey fantail 70. Mountain thornbill, Little shrike-thrush, Common paradise kingfisher 71. White-bellied pitohui 63. Rufous fantail 64. Yellow-spotted honeyeater 65. Mimic honeyeater, Lewin!s honeyeater 66. Spectacled monarch (SMON), Pied monarch, Grey-headed robin 72. TBBB, SPC, BHON, LBSW, BST, SMON 68. Wattled ploughbill 67. Shining flycatcher 73. Hooded pitohui, Large scrubwren 74. Yellow-spotted honeyeater 75. Spotted catbird (SPC) 77. Mountain mouse-warbler 76. Emerald dove 78. Rusty mouse-warbler 69. Noisy pitta P. reichenowi P. falciparum P. vivax P. simiovale P. yoelli P. malriae 0.1 substiutions/ite Alcedinidae Columbidae Monarchidae Acanthizidae Meliphagidae Zosteropidae Nctarini Phycephalidae etroiid Ptilonorhynchidae Family represented one time Host Family Key Haemoproteus Plasmodium Clade B Clade A 48 0 0.2 0.4 0.6 0.8 1 00.010.020.030.040.050.060.070.080.090.1 LogDet pairwise parasite genetic distance Probability of same host family (predicted) HB P HA 49 APPENDICES Apendix A. Avian host names, geographical origin, sequence length and GenBank acesion numbers for parasite lineages listed in Figure 2. Host Information Lineage Family Genus Species Comon Name Locality bp GenBank 1 Alcedinidae Tanysiptera galatea Comon paradise kingfisher PNG 487 AY714134 2 Alcedinidae Tanysiptera galatea Comon paradise kingfisher PNG 53 AY714135 2 Alcedinidae Halcyon torotoro Yelow-biled kingfisher PNG 53 AY714135 3 Columbidae Ptilinopus superbus Superb fruit-dove PNG 53 AY714136 4 Columbidae Ptilinopus superbus Superb fruit-dove PNG 53 AY714137 4 Columbidae Ptilinopus magnificus Wompo fruit-dove PNG 53 AY714137 5 Alcedinidae Halcyon torotoro Yelow-biled kingfisher PNG 53 AY714138 6 Columbidae Ptilinopus rivoli White-breasted fruit-dove PNG 53 AY714139 7 Monarchidae Rhipidura rufiventris Northern fantail PNG 295 AY714140 8 Acanthizidae Sericornis citreogularis Yelow-throated scrubwren AUS 53 AY714141 8 Acanthizidae Sericornis magnirostris Large-biled scrubwren AUS 53 AY714141 9 Meliphagidae Oedistoma iliolophus Dwarf honeyeater PNG 53 AY714142 9 Acanthizidae Crateroscelis murina Rusty mouse-warbler PNG 53 AY714142 10 Zosteropidae Zosterops fuscicapilus Western mountain white-eye PNG 53 AY714143 1 Nectarinidae Nectarinia aspasia Black sunbird PNG 295 AY71414 12 Pachycephalidae Pitohui cristatus Crested pitohui PNG 53 AY714145 13 Petroicidae Tregelasia capito Pale-yelow robin AUS 53 AY714146 14 Petroicidae Heteromyias albispecularis Grey-headed robin AUS 53 AY714147 15 Campephagidae Lalage leucomela Varied triler PNG 53 AY714148 15 Meliphagidae Myzomela nigrita Papuan black myzomela PNG 53 AY714148 16 Acanthizidae Sericornis citreogularis Yelow-throated scrubwren AUS 53 AY714149 16 Acanthizidae Sericornis magnirostris Large-biled scrubwren AUS 53 AY714149 17 Muscicapidae Zothera lunulata Basian thrush AUS 53 AY714150 50 18 Pachycephalidae Coluricincla megarhyncha Litle shrike-thrush AUS 53 AY714151 18 Pachycephalidae Coluricincla boweri Bower?s shrike-thrush AUS 53 AY714151 19 Pachycephalidae Coluricincla megarhyncha Litle shrike-thrush AUS 53 AY714152 20 Pachycephalidae Pitohui incertus White-belied pitohui PNG 53 AY714153 20 Pachycephalidae Pitohui ferugineus Rusty pitohui PNG 53 AY714153 21 Ptilonorhynchidae Scenopoetes dentirostris Toth-biled bowerbird AUS 53 AY714154 2 Zosteropidae Zosterops fuscicapilus Western mountain white-eye PNG 53 AY71415 23 Monarchidae Rhipidura rufiventris Northern fantail PNG 53 AY714156 24 onarchidae Myiagra alecto Shining flycatcher PNG 53 AY714157 25 Monarchidae Monarcha gutula Spot-winged monarch PNG 53 AY714158 26 onarchidae Machaerirhynchus flaviventer Yelow-breasted boatbil AUS 271 AY714159 27 Meliphagidae Myzomela nigrita Papuan black myzomela PNG 53 AY714160 28 eliphagidae Myzomela obscura Dusky honeyeater AUS 53 AY714161 29 Meliphagidae Myzomela obscura Dusky honeyeater AUS 53 AY714162 30 eliphagidae Philemon buceroides Helmeted friarbird PNG 53 AY714163 31 Meliphagidae Myzomela nigrita Papuan black myzomela PNG 53 AY714164 31 Pachycephalidae Coluricincla megarhyncha Litle shrike-thrush PNG 53 AY714164 32 Meliphagidae Melilestes megarhynchus Long-biled honeyeater PNG 53 AY714165 32 Pomatostomidae Pomatostomus isodorei Rufous babler PNG 53 AY714165 3 Meliphagidae Acanthorhynchus tenuirostris Eastern spinebil AUS 53 AY71416 34 Petroicidae Tregelasia capito Pale-yelow robin AUS 53 AY714167 35 Petroicidae Tregelasia capito Pale-yelow robin AUS 53 AY714168 36 Petroicidae Tregelasia capito Pale-yelow robin AUS 53 AY714169 37 Petroicidae Tregelasia capito Pale-yelow robin AUS 53 AY714170 38 Petroicidae Heteromyias albispecularis Grey-headed robin AUS 53 AY714171 39 Petroicidae Heteromyias albispecularis Grey-headed robin AUS 53 AY714172 40 Petroicidae Tregelasia capito Pale-yelow robin AUS 53 AY714173 41 Petroicidae Eopsaltria australis Eastern yelow robin AUS 53 AY714174 42 Petroicidae Eopsaltria australis Eastern yelow robin AUS 53 AY714175 43 Melanocharitidae Melanocharis nigra Black berypecker PNG 53 AY714176 4 eliphagidae Meliphaga lewini Lewin?s honeyeater AUS 53 AY71417 45 Meliphagidae Lichenostomus frenatus Bridled honeyeater AUS 53 AY714178 46 eliphagidae Lichenostomus frenatus Bridled honeyeater AUS 53 AY714179 47 Orthonychidae Orthonyx spaldingi Chowchila AUS 53 AY714180 48 Meliphagidae Xanthotis macleayana Macleay?s honeyeater AUS 53 AY714181 49 eliphagidae Xanthotis flaviventer Tawny-breasted honeyeater PNG 53 AY714182 51 50 Meliphagidae Xanthotis flaviventer Tawny-breasted honeyeater PNG 53 AY714183 51 eliphagidae Melipotes fumigatus Smoky honeyeater PNG 53 AY714184 52 Monarchidae Monarcha gutula Spot-winged monarch PNG 53 AY714185 52 onarchidae Arses telescophthalmus Friled monarch PNG 53 AY714185 53 Pachycephalidae Pachycephala simplex Grey whistler PNG 53 AY714186 54 Pachycephalidae Pitohui kirhocephalus Variable pitohui PNG 53 AY714187 5 Pachycephalidae Pitohui kirhocephalus Variable pitohui PNG 53 AY71418 5 Meliphagidae Lichenostomus hindwodi Eungela honeyeater AUS 53 AY71418 56 Pachycephalidae Pitohui dichrous Hoded pitohui PNG 53 AY714189 57 Paradisaeidae Ptiloris magnificus Magnificent riflebird PNG 53 AY714190 58 Monarchidae Monarcha chrysomela Golden monarch PNG 53 AY714191 59 Cracticidae Cracticus quoyi Black butcherbird PNG 53 AY714192 60 Columbidae Macropygia amboinensis Brown cucko-dove PNG 53 AY714193 61 Nectarinidae Nectarinia aspasia Black sunbird PNG 295 AY714194 62 Pachycephalidae Pachycephala pectoralis Golden whistler AUS 53 AY714195 62 Monarchidae Rhipidura fuliginosa Grey fantail AUS 53 AY714195 63 onarchidae Rhipidura rufifrons Rufous fantail AUS 53 AY714196 64 Meliphagidae Meliphaga notata Yelow-spoted honeyeater AUS 53 AY714197 65 eliphagidae eliphaga analoga Mimic honeyeater PNG 53 AY714198 65 Meliphagidae Meliphaga lewini Lewin?s honeyeater AUS 53 AY714198 6 onarchidae Monarcha trivirgatus Spectacled monarch AUS 53 AY71419 6 Monarchidae Monarcha kaupi Pied monarch AUS 53 AY71419 6 Petroicidae Heteromyias albispecularis Grey-headed robin AUS 53 AY71419 67 Monarchidae Myiagra alecto Shining flycatcher PNG 53 AY71420 68 Pachycephalidae Eulecestoma nigripectus Watled ploughbil PNG 53 AY714201 69 Pitidae Pita versicolor Noisy pita AUS 53 AY714202 70 Acanthizidae Acanthiza katherina Mountain thornbil AUS 53 AY714203 70 Pachycephalidae Coluricincla megarhyncha Litle shrike-thrush AUS 53 AY714203 70 Alcedinidae Tanysiptera galatea Comon paradise kingfisher PNG 53 AY714203 71 Pachycephalidae Pitohui incertus White-belied pitohui PNG 53 AY714204 72 Ptilonorhynchidae Scenopoetes dentirostris Toth-biled bowerbird AUS 53 AY714205 72 Ptilonorhynchidae Ailuroedes melanotis Spoted catbird AUS 53 AY714205 72 Meliphagidae Lichenostomus frenatus Bridled honeyeater AUS 53 AY714205 72 Acanthizidae Sericornis magnirostris Large-biled scrubwren AUS 53 AY714205 72 Pachycephalidae Coluricincla boweri Bower?s shrike-thrush AUS 53 AY714205 72 Monarchidae Monarcha trivirgatus Spectacled monarch AUS 53 AY714205 52 73 Pachycephalidae Pitohui dichrous Hoded pitohui PNG 53 AY714206 73 Acanthizidae Sericornis nouhuysi Large scrubwren PNG 469 AY714206 74 Meliphagidae Meliphaga notata Yelow-spoted honeyeater AUS 53 AY714207 75 Ptilonorhynchidae Ailuroedes melanotis Spoted catbird AUS 485 AY714208 76 Columbidae Chalcophaps indica Emerald dove PNG 295 AY714209 7 Acanthizidae Crateroscelis robusta Mountain mouse-warbler PNG 295 AY714210 78 Acanthizidae Crateroscelis murina Rusty mouse-warbler PNG 295 AY71421 53 Apendix B. Frequency of detection of Haemoproteus (H), Plasmodium (P), unknown genus (U) or mixed infection (M) across host families from Australia (AUS) and Papua New Guinea (PNG). Composition of mixed infections is indicated at right. Host Location Total H P U M INF Mixed Acipitridae Acipiter poliocephalus PNG 1 0 Megapodidae egapodius reinwardt PNG 1 0 Columbidae Chalcophaps indica PNG 3 1 1 1 3 H Chalcophaps stephani PNG 1 1 1 Ducula pinon PNG 1 0 Macropygia amboinensis PNG 3 1 1 Ptilinopus magnificus PNG 2 1 1 Ptilinopus pulchelus PNG 1 0 Ptilinopus rivoli PNG 2 1 1 Ptilinopus superbus PNG 4 2 1 1 4 H Podargidae Podargus ocelatus PNG 2 1 1 Aegothelidae Aegotheles benneti PNG 2 0 Alcedinidae Alcedo azurea PNG 1 0 Alcedo pusila PNG 1 0 Halcyon chloris PNG 1 0 Halcyon sancta PNG 1 0 Halcyon torotoro PNG 4 2 1 3 Melidora macrorhina PNG 1 0 Tanysiptera danae PNG 4 0 Tanysiptera galatea PNG 10 3 1 4 Pitidae Pita versicolor AUS 3 1 1 Climacteridae Cormobates leucophaeus AUS 3 0 Ptilonorhynchidae Ailuroedus melanotis AUS 8 3 2 1 6 P Scenopoeetes dentirostris AUS 7 2 2 2 6 P,PH Acanthizidae Acanthiza katherina AUS 8 1 1 54 Crateroscelis murina PNG 5 1 3 4 Crateroscelis robusta PNG 6 3 3 Gerygone mouki AUS 2 0 Oreoscopus gutturalis AUS 6 1 1 Sericornis citreogularis AUS 12 2 2 Sericornis frontalis AUS 7 0 Sericornis keri AUS 4 0 Sericornis magnirostris AUS 14 5 1 6 PH Sericornis nouhuysi PNG 1 1 1 Sericornis papuensis PNG 2 0 Sericornis perspicilatus PNG 2 0 Meliphagidae Acanthorhynchus tenuirostris AUS 7 1 1 2 Lichenostomus frenatus AUS 8 6 1 7 Lichenostomus hindwoodi AUS 4 1 1 Melilestes megarhynchus PNG 3 1 1 eliphaga analoga PNG 1 1 1 Meliphaga aruensis PNG 4 1 1 eliphaga lewinii AUS 16 5 2 7 Meliphaga notata AUS 3 2 2 elipotes fumigatus PNG 5 2 2 Myzomela nigrita PNG 2 1 1 2 H yzomela obscura AUS 3 2 1 3 H Oedistoma iliolophus PNG 6 6 6 Philemon buceroides PNG 1 1 1 Xanthotis flaviventer PNG 2 2 2 Xanthotis macleayana AUS 5 4 4 Petroicidae Amalocichla incerta PNG 2 0 Eopsaltria australis AUS 3 2 1 3 Heteromyias albispecularis AUS 17 5 1 5 11 H Melanodryas cucullata PNG 1 0 Petroica rosea PNG 1 0 Tregelasia capito AUS 10 3 4 7 H Orthonychidae Orthonyx spaldingii AUS 4 1 1 Pomatostomidae Pomatostomus isidorei PNG 5 1 1 Cinclosomatidae Cinclosoma ajax PNG 2 0 Psophodes olivaceus AUS 7 0 Psophodes olivaceus PNG 1 0 Pachycephalidae Colluricincla boweri AUS 8 1 1 2 Colluricincla harmonica PNG 1 0 55 Colluricincla megarhyncha AUS 11 4 1 5 PH Colluricincla megarhyncha PNG 16 4 1 5 Colluricincla woodwardi PNG 1 1 1 ? Eulecestoma nigripectus PNG 1 1 1 Falcunculus frontatus PNG 1 1 1 Pachycephala melanura PNG 2 0 Pachycephala olivacea PNG 1 1 1 Pachycephala pectoralis AUS 11 1 1 Pachycephala pectoralis PNG 1 0 Pachycephala schlegeli PNG 1 0 Pachycephala simplex PNG 3 1 1 2 Pitohui cristatus PNG 3 2 2 Pitohui dichrous PNG 10 2 1 3 Pitohui ferugineus PNG 14 1 1 1 3 ? Pitohui incertus PNG 3 1 1 2 Pitohui kirhocephalus PNG 5 1 1 2 4 PH Rhagologus leucostigma PNG 1 0 Paradisaeidae Cicinnurus magnificus PNG 1 0 Paradiseae raggiana PNG 1 1 1 Ptiloris magnificus PNG 1 1 1 Cracticidae Cracticus quoyi PNG 2 2 2 Campephagidae Lalage leucomela PNG 1 1 1 Dicruridae Chaetorynchus papuensis PNG 1 0 Dicrurus hottentottus PNG 2 0 Monarchidae Arses telescopthalmus PNG 3 1 1 2 H Machaerirynchus flaviventer AUS 3 1 1 2 Monarcha chrysomela PNG 3 3 3 onarcha guttula PNG 5 3 3 Monarcha kaupi AUS 3 3 3 onarcha trivirgatus AUS 12 3 1 4 P Myiagra alecto PNG 3 1 2 3 Rhipidura albolimbata PNG 2 0 Rhipidura atra PNG 3 0 Rhipidura brachyrhyncha PNG 1 0 Rhipidura fulginosa AUS 7 2 2 Rhipidura rufifrons PNG 2 0 Rhipidura rufifrons AUS 1 1 1 Rhipidura rufiventris PNG 6 3 1 5 H Sylviidae Phyloscopus trivirgatus PNG 1 0 56 Muscicapidae Zoothera lunulata AUS 2 1 1 Melanocharitidae elanocharis nigra PNG 1 1 1 Nectariniidae Nectarinia aspasia PNG 2 1 1 PH Zosteropidae Zosterops fuscicapilus PNG 3 1 1 2 H Zosterops griseotinctus PNG 2 0 Paseridae Erythrura trichroa PNG 1 0 57 CHAPTER III Global phylogeographic limits of Hawai?s avian malaria ABSTRACT The introduction of avian malaria (Plasmodium relictum) to Hawai has provided a model system for studying the influence of exotic disease on na?ve host populations. Litle is known, however, about the origin or genetic variation of Hawai?s malaria and traditional clasification methods have confounded atempts to place the parasite within a global ecological and evolutionary context. Using fragments of the parasite mitochondrial gene cytochrome b and the nuclear gene DHFR-TS obtained from a global survey of >13,000 avian samples, we show that Hawai?s avian malaria, which can cause high mortality and is a major limiting factor for many species of native paserines, represents just one of numerous lineages composing the morphological parasite species. The single parasite lineage detected in Hawai exhibits a broad host distribution worldwide and is dominant on several other remote oceanic islands, including Bermuda and Moorea, French Polynesia. Rarity of this lineage in the continental New World and the restriction of closely-related lineages to the Old World suggest limitations to the transmision of reproductively-isolated parasite groups within the morphological species. 58 INTRODUCTION The introduction of avian malaria (Plasmodium relictum) to the remote Hawaian Islands has been implicated in the widespread decline and possible extinction of many species within the endemic avian radiation of honeycrepers (Warner 1968; van Riper et al. 1986). While mortality in introduced bird species is negligible, mortality in many endemic species can range from 50-90% (Jarvi et al. 2001), possibly reflecting their long isolation (ca. 4 milion years; Fleischer & McIntosh 2001) from malarial parasites. Although the epidemiology of malaria in Hawaian birds has been wel studied, litle is known about the diversity of parasite strains in Hawai or their origin. Transmision of malaria was impossible until the human-mediated introduction of a competent vector (Culex quinquefasciatus) to Hawai in 1826 (Warner 1968). Since then, resident Hawaian birds may have been exposed to reservoirs of parasites harbored by the hundreds of exotic birds released in the late 19 th and early 20 th centuries (Long 1981) and by the thousands of ducks and shorebirds which annualy migrate to Hawai from their breding grounds in the arctic. Understanding the host range of P. relictum in Hawai and elsewhere across the globe is esential for the purpose of identifying its original host, for understanding limits to its transmision, and for eventualy understanding its extreme virulence in native honeycrepers relative to its efects in other hosts. Unfortunately, clasical techniques for identifying parasites may be confounding this understanding. P. relictum has been reported to occur in a broad spectrum of hosts from al continents except Antarctica (Bennet et al. 1993), however, to date, identification has been based both on morphology, which can vary within strains of the same parasite species 59 (Peirce 1979; van Riper 1991), and on biological characteristics such as vector, endogenous development and host range (Garnham 1966). These characters may not acurately reflect phylogenetic relationships among parasites (Escalante et al. 1998). The recent detection of extensive genetic diversity across avian malaria parasites (Ricklefs & Falon 2002; Waldenstr?m et al. 2002; Beadel et al. 2004) suggests that cryptic structure in parasite populations may underlie diferences in host susceptibility, vector competence and parasite virulence. Here, we use molecular markers to characterize Hawai?s avian malaria and to place this genotype in a global context in order to beter understand its origin, its current and historic impact, and limits to its transmision. MATERIAL AND METHODS Lineage Identification Although phylogenetic species limits have not been wel defined in avian malaria parasites, a previous study indicated that mitochondrial lineages appear to represent reproductively-isolated units (Bensch et al. 2004), and therefore, we characterized the Hawaian strain and its global distribution using cytochrome b (cyt b). Determination of parasite haplotype frequencies in Europe and Nigeria (table 1) folowed PCR methods described in Waldenstr?m et al. (2004). Data for the Leser Antiles and Venezuela were provided by S. Falon. Detection and identification of al other parasite mitochondrial lineages generaly followed the methods described in Beadel et al. (2004). Briefly, we screned DNA extracted from blood or tisue for parasites using primers F2/R2, 850F/1024R or 213F/372R (Beadel & Fleischer 2005). For positively infected samples, we then sequenced a larger fragment generated with 60 primers 3760F/4292rw2 (533 bp) or with primers Fifi/4292rw2 (351 bp; Ishtiaq et al. 2006) or F2/4292rw2 (295 bp) in cases where degraded template precluded the amplification of the larger piece (appendix 1). For lineages used in detailed phylogenetic analysis, we also obtained an additional 220 bp of sequence using primers L15368/H15730 (Falon et al. 2003), which generated a 338 bp fragment overlapping previously-generated fragments by 118 bp. Sequences were asembled, aligned and edited using SEQUENCHER v. 4.1. For the purpose of describing the worldwide distribution of Hawai?s parasite lineage (GRW4; Bensch et al. 2000), we identified as GRW4 any sequences that identicaly matched the largest fragment of parasite cyt b isolated from a Hawaian host (AY733090). In Korea, we detected the presence of GRW4 based on matching a single 91 bp fragment alone, but in al other locations, detection of GRW4 was based on recovery of identical sequences of betwen 256 and 753 bp (appendix 1). When defining al other lineages, we grouped together only identical sequences exhibiting matching sequence of 256 bp or more. Therefore, within the limits of the sequence examined, parasite lineages are defined by unique mitochondrial haplotypes. Phylogeny of Plasmodium sp. To examine the concordance betwen parasite clasification based on morphology and on DNA, we asembled cyt b sequence from avian Plasmodium spp. previously identified by morphology and recognized as valid by Bennet et al. (1993). We obtained sequences from Genbank and from DNA extracted from blood smears, which were obtained from the International Reference Collection for Avian Haematozoa (IRCAH; Brisbane, Australia) or from M. Peirce. Clasification of 61 parasites in IRCAH smears were checked by C. Atkinson and M. Peirce and smears exhibiting multiple infections (upon visual inspection or after molecular analysis) were not used. We constructed phylogenetic tres using taxa for which we had recovered betwen 335 and 753 bp of sequence and rooted tres with sequences from Haemoproteus spp., parasites in the sister genus to avian Plasmodium (Perkins & Schal 2002). We estimated phylogenies using minimum evolution (ME; on K2P and GTR distances), maximum likelihood (ML), and maximum parsimony (MP) as implemented by PAUP* (Swofford 1999). For ML analysis, we chose the most likely model of base pair substitution (GTR+I+G) and parameters (pinv = 0.5509, shape = 0.6505) based on a likelihood ratio test employed by Modeltest version 3.07 (Posada & Crandal 1998). Bootstrap support was estimated for each method using 1000 replicates. Phylogenetic Analysis of Lineages Related to GRW4 We asembled a total of 166 unique Plasmodium cyt b sequences gathered from our global survey and from GenBank. Due to the large number of mitochondrial lineages, we initialy constructed an ME tre using K2P distances and PAUP* in order to identify those lineages which shared most recent common ancestry with GRW4 and which were most relevant to tracing the origin of this lineage. The resulting tre (data not shown) exhibited a clade containing GRW4 (lineages 15 through 37), a sister clade (lineages 1 through 14) and two clades imediately ancestral (lineages 38 through 51); combined, these 51 lineages formed a monophyletic clade nested within the other 115 lineages. For subsequent analyses, therefore, we focused only on these 51 lineages plus several lineages with known morphological clasification to help 62 polarize the tre. Lineage 56 was included because a representative (AF254962) was originaly clasified as P. nucleophilum, however, its status is undergoing revision (G. Valkiunas pers. comm.). Employing the longest fragment of cyt b available for each lineage (appendix 1), we used ML to re-estimate a phylogeny using a model (GTR+I+G, pinv = 0.5445, shape = 0.7154) chosen by Modeltest. We estimated support for nodes based on 100 replicates. Due to uncertainty about parasite species limits and because a dichotomously-branching tre may not appropriately capture relationships among mitochondrial lineages within a species, we also generated a haplotype network using statistical parsimony as implemented in TCS1.21 (Clement et al. 2000). Lineages were joined at the 95% confidence criterion unles noted. Parasite lineages which could not be joined to GRW4 at the 90% level were excluded. In order to generate a second, independent estimate of relationships among lineages, we followed the protocols in Bensch et al. (2004) to amplify and sequence a portion of the nuclear gene dihydrofolate reductase-thymidylate synthase (DHFR-TS; 236 bp) from samples for which we had already recovered mitochondrial sequence (appendix 2). Because nuclear DNA occurs in much lower copy number than mitochondrial DNA, we recovered DHFR-TS from only a fraction of the samples for which we recovered mitochondrial lineages. In addition, we did not include DHFR- TS sequences from samples for which nuclear or mitochondrial sequence provided evidence of multiple infections (e.g., double peaks in the chromatogram or diferent sequences from diferent primer sets). Nuclear haplotype Q (AY033582) was derived from P. gallinaceum but not necesarily the same strain of P. gallinaceum from which mitochondrial lineage 50 (AY099029) was derived. We used the methods 63 described above to estimate an ML tre using the model GTR+I+G (pinv = 0.5620, shape = 2.0791). Bootstrap support was based on 1000 replicates. To improve resolution of hierarchical relationships among lineages, we used Bayesian analysis as implemented in MrBayes v3.1.1 (Ronquist & Huelsenbeck 2003) to estimate phylogenetic relationships among lineages for which we could combine both mitochondrial and nuclear markers. Parameters of the GTR+I+G model of DNA substitution were alowed to vary independently for each marker within the concatenated dataset. We performed two runs of 25 milion generations, each with one cold and thre heated chains and sampled the resulting tres every 1000 generations. Graphical plotting of ML scores suggested that stationarity was reached after approximately 100,000 generations, however, we discarded the first milion generations as burn-in. Posterior probabilities of nodes were estimated from the remaining 24,000 tres. Flat priors were asumed for al parameters. Using a wel-supported group of lineages within the Bayesian tre (Clade A) and the program Mesquite v1.05 (Maddison & Maddison 2004), we calculated the likelihood that ancestral parasites were found in the New World. We employed the Mk1 model and considered ancestral state reconstruction to be signficant when raw likelihood scores for the two possible states (in New World or not-in-New World) difered by greater than 2 and the proportional likelihood of the best state was > 0.95. RESULTS AND DISCUSSION From 245 introduced and endemic resident forest birds captured during various seasons betwen 1971 and 1998 on Hawai, Maui, Molokai, Oahu and Kauai, we recovered only a single mitochondrial lineage of Plasmodium (lineage 15; n = 75 64 sequenced infections; previously identified as GRW4, Bensch et al. 2000) and only one nuclear haplotype (DHFR haplotype G). We detected a second lineage of parasite (lineage 43) in a single migratory golden plover (Pluvialis fulva) from the Northwestern Hawaian Islands, however we found no evidence for transmision of this parasite to Hawaian forest birds. This result corroborates reports of just a single morphological subspecies of parasite in Hawai (P. relictum capistranoae) (Laird & van Riper 1981; van Riper et al. 1986) and suggests that the recent expansion of native host populations into low-elevation forests over the last decade (Woodworth et al. 2005) has not been facilitated by the cryptic introduction of diferent parasite lineages of lower virulence. While Hawaian mitochondrial lineages of P. relictum were monotypic, pairwise divergence of other parasites identified as P. relictum averaged 4.0% (range: 0 to 7.6%), substantialy greater than the intraspecific divergence observed in the human parasite P. falciparum across the entire mitochondrial genome (0.2%; Joy et al. 2003) and the divergence observed betwen sympatric haematozoan parasites restricted to diferent avian hosts (0.6%; Ricklefs et al. 2004). In addition, phylogenetic analysis of avian malaria parasites clasified by morphology indicated that P. relictum does not form a monophyletic clade (figure 1). Dep nodes were generaly not wel supported, however, the close relationship of one isolate of P. cathemerium and two isolates of P. elongatum to P. relictum indicated that either morphology does not reflect evolutionary relationships or that previous clasification has been in eror. This conflict, as wel as the disparity betwen the broad genetic diversity of parasites identified as P. relictum by morphology and the single type 65 found in Hawai, suggests that ecological data, such as host and geographical ranges, which have been compiled for parasites clasicaly identified as P. relictum are not necesarily applicable to the Hawaian parasite. Pinpointing the original host is dificult given the broad host range of Plasmodium spp. in general (Bennet et al. 1993; Beadel et al. 2004; Falon et al. 2005) and of the GRW4 lineage in particular. Worldwide, we recovered GRW4 from 39 species of birds, representing 13 families (appendix I). We found the lineage frequently in continental populations of common mynas (Acridotheres tristis) and house sparows (Passer domesticus), both of which were introduced to Hawai, and also in great red warblers (Acrocephalus arundinaceus). We did not detect GRW4 in a survey of 75 migratory shorebirds sampled from Hawai, the French Frigate Shoals and Laysan Island, however, the lineage has been detected in a shorebird from Mauritania (Mendes et al. 2005). The apparent lack of host-specificity of GRW4 is reflected by its broad geographic distribution. In addition to Hawai, we detected GRW4 throughout the Old World, where it was particularly common relative to other malaria lineages in Europe (but only in adults of migratory species), India, and on several Indian Ocean islands (figure 2; table 1). Evidence from our surveys and extensive sampling of thousands of North American birds by R. Ricklefs and coworkers (R. E. Ricklefs, personal communication), however, suggests that the lineage is rare in birds from mainland North and South America. To our knowledge, the only two mainland hosts in which GRW4 has been detected are a house sparow from California (Schrenzel et al. 2003) and a house finch from Arizona (M. Kimura pers. comm.). The only other 66 New World records of this lineage were derived from two individuals in the Leser Antiles (Falon et al. 2005). Given the wide host range of GRW4 and its presence in at least some New World host families (Emberizidae and Mimidae), our failure to detect GRW4 more widely in the New World likely reflects its rarity there and not simply an artifact atributable to diferences in the composition of hosts sampled in diferent regions. In contrast to its rarity relative to other lineages elsewhere in the New World, GRW4 was the only lineage detected in resident paserines of Bermuda, an oceanic island of volcanic origin lying 1000 km off the coast of North America. Colonization of Bermuda by GRW4 is likely to have occurred only recently since mosquitoes were reported as absent from Bermuda by the Spanish sailor Diego Ramirez, who was shipwrecked on the then-uninhabited island in 1603 (acount published in Wilkinson 1950). Given subfossil evidence of unique endemic resident paserines existing in Bermuda prior to human colonization (Olson et al. 2005), it is possible that, as in Hawai, the arival of a competent vector (C. quinquefasciatus is currently present) and an Old World lineage of Plasmodium may have contributed to the extinction of another island avifauna. In the Pacific, the Hawaian form of malaria was also the only lineage of malaria parasite detected in paserines of French Polynesia, though we detected a second lineage exclusively in junglefowl (Gallus gallus). In a survey of birds from Moorea, we found GRW4 at low frequency in several introduced species including red-browed firetails (Neochmia temporalis; 2 of 34 individuals), silvereyes (Zosterops lateralis; 1 of 60 individuals) and common mynas (Acridotheres tristis; 2 of 10 67 individuals). We did not sample any of the native paserines on Moorea because populations of those species, namely the Tahiti red warbler (Acrocephalus caffer longirostris), Pacific swalow (Hirundo tahitica) and Polynesian swiftlet (Aerodramus leucophaeus), if extant, were extremely smal. However, among a smal sample of endemic Marquesan red warblers (Acrocephalus mendanae) collected in 1987 on Nuku Hiva, we detected GRW4 in 9 of 11 individuals. Because populations of Marquesan red warblers remain fairly robust (Holyoak & Thibault 1984; J.-C. Thibault personal communication), this finding presents the possibility that, unlike its efect on Hawaian honeycrepers, GRW4 may not pose a threat to these endemic French Polynesian paserines, which are relatively recently diverged from a mainland ancestor (ca. 1-2 Ma; Fleischer et al. unpublished data) and of Old World descent. Conversely, for older Polynesian endemics such as the Pomarea flycatchers, which have likely evolved for a longer time in isolation (ca. 3.6 Ma; Cibois et al. 2004), the introduction of GRW4 may represent a previously unrecognized factor driving the decline of these species, most of which are threatened or endangered (BirdLife International 2000). The detection of GRW4 in French Polynesia and the Cook Islands (Ishtiaq et al. 2006), and of a closely related parasite in the Marianas (lineage 32), warants further investigation into the efects of avian malaria on isolated avifaunas outside of Hawai. While GRW4 itself exhibited a broad geographical distribution, the distribution of related parasite lineages provided evidence of an Old World ancestry for GRW4. Phylogenetic reconstructions of mitochondrial parasite lineages and asociated nuclear haplotypes yielded broadly concordant topologies (figure 3). 68 Although there was litle support for deep nodes when loci were analyzed separately, Bayesian analysis of data from both mitochondrial and nuclear loci combined recovered similar clustering of parasite lineages and provided support for the monophyly of parasite genotypes 1B through 38P (Clade A, figure 4). Within this group, but with the exception of 15G (GRW4), al of the parasite genotypes detected in the New World fel within a wel-supported clade (Clade B, genotypes 1B through 14F) that was either sister to or derived from the remainder of genotypes in Clade A. Among the remainder of Clade A, al genotypes except 15G (GRW4) were recovered exclusively from the Old World and likelihood estimation of ancestral origins confirmed that the imediate ancestors of GRW4 likely derived from the Old World (figure 4). Among the 24 parasite mitochondrial lineages composing this group, 18 were recovered from hosts in Africa (figure 3). A haplotype network (figure 5), which may more appropriately describe non-bifurcating relationships among mitochondrial lineages derived from a single species, similarly indicated broad geographical substructure within Clade A and close asociation of GRW4 (lineage 15) with Old World lineages recovered from Africa (16) and New Guinea (17). The derived position of GRW4 relative to other lineages from the Old World further suggests that its range has only recently expanded to include parts of the New World. The mitochondrial lineage GRW4 was asociated with nuclear haplotype G everywhere except on several Indian Ocean islands where GRW4 was instead asociated with haplotype H (figure 3). We identified several additional cases of a single mitochondrial lineage (e.g., 1, 30 and 31) asociated with multiple nuclear haplotypes as wel as cases of a single DHFR-TS sequence asociated with multiple 69 divergent mitochondrial lineages (e.g., haplotype G and lineages 15, 16, 17, 18, 27, 28, 32, 33; figure 3). If every mitochondrial lineage of Plasmodium represents a sexualy-isolated unit (Bensch et al. 2004), then these results may reflect either incomplete lineage sorting amongst genes in otherwise reproductively-isolated species or insufficient variation in our markers. On the other hand, the sharing of mitochondrial or nuclear haplotypes among diferent parasite lineages may simply represent intraspecific genetic variation. The resolution of our data precludes investigation of this on a fine scale, however, several cases in which nuclear and mitochondrial haplotypes exhibit similar clustering (e.g., lineages 11, 12, 13 and nuclear haplotypes D, E), combined with an apparent lack of genetic exchange with closely related parasites, provide an indication that species limits may be very narow. Within the resolution of our data, the complete linkage disequilibrium of mitochondrial and nuclear markers found among parasites in Clade B (average cyt b p-dist: 1.5%) relative to other parasites in Clade A (avg. p-dist: 1.9%), and among parasites in Clade A (avg. p-dist: 2.2%) relative to the next most related lineages, suggests that these groups, at least, are reproductively isolated. Except for two parasites, which were described as P. elongatum and P. cathemerium, al other morphologicaly-described parasites with asociated mitochondrial sequences faling within Clade A were identified as P. relictum (lineages 2, 5, 15, 16 and 22; figure 3). Given the results above, the morphological taxon P. relictum appears to be composed of at least two, and probably several more, reproductively-isolated groups. The geographical structuring of parasite lineages within Clade A is surprising in light of the masive commercial and migratory movement of birds worldwide. 70 Parasites are often lost when their hosts are introduced to novel regions (Torchin et al. 2003; Colautti et al. 2004) and competence of novel hosts, host migration paterns, and competing strains of parasite may retard the exchange of parasites betwen hemispheres. Nonetheles, the prominence of GRW4 on several remote oceanic islands and its wide host distribution suggest that these are not primary factors limiting the range of GRW4. Instead, diferential vector-parasite compatibility may be limiting transmision of GRW4 and driving genetic isolation betwen populations of P. relictum. Vector incompatibility may be preventing the transmision of GRW4 in northern Europe (Waldenstr?m et al. 2002) and appears to be responsible for the isolation of New World and Old World forms of P. vivax (Li et al. 2001), the dominant form of malaria in humans. We found further evidence for transmision limits in Bermuda, where we recovered thre lineages of Plasmodium, but GRW4 was the only lineage of Plasmodium detected in blood from resident Bermuda paserines (n = 42 sequenced infections) sampled betwen 2002 and 2004. Among resident birds, we detected GRW4 in both introduced Old World hosts [house sparows (P. domesticus) and European starlings (Sturnus vulgaris)] and New World hosts [grey catbirds (Dumetela carolinensis) and Eastern bluebirds (Sialia sialis), but never in white-eyed vireos (Vireo griseus; n=16) or great kiskadees (Pitangus sulphuratus; n= 33)]. As in Hawai, Bermuda provides a wintering ground for numerous North American migrants, some of which may exhibit transmisible erythrocytic-stage malaria infections. Asuming that winter parasitemias are not low enough to prevent transmision, the absence of al lineages except GRW4 in Bermuda residents suggests 71 that either resident Bermuda birds are not competent hosts for most North American Plasmodium lineages or that the local vector is refractory to these lineages. The former is unlikely to be true for al resident paserines since many of these species colonized or were introduced to the island from the New World only within the last several hundred years (C. E. McIntosh et al., unpublished work). In addition, we found one of the non-GRW4 lineages (lineage 1) in a migratory ovenbird in Bermuda and in house sparows from continental North America, but never in resident house sparows from Bermuda (n = 15 sequenced infections). The other non-GRW4 lineage in Bermuda was recovered from two migratory yelow-throated warblers (Dendroica dominica). This sequence matched a parasite lineage recovered in several other North American species and was only distantly related to lineages in Clades A. Given the distinct parasite lineages in resident and migratory species, it appears that refractorines of the local strain of C. quinquefasciatus to parasites caried by migrants may be important in structuring the parasite community in Bermuda. If transmision of GRW4 was initialy limited to the Old World, as sems possible given the distribution of lineages most closely related to it, the spread and admixture of Old World populations of C. quinquefasciatus with geneticaly- diferentiated New World populations (Fonseca et al. 2006) may be facilitating the expansion of GRW4 into new locations. Future experimental infections of New World mosquitoes with isolates of P. relictum from diferent regions could shed light on the mechanism underlying the current rarity of GRW4 in the New World. Given the diversity of lineages encompased by the morphological taxon P. relictum, future asesment of the ecological and evolutionary impacts of GRW4 and other avian 72 malarias wil require a molecular characterization of the pathogen in question. This wil be particularly valuable, for example, when identifying an independent source of GRW4 with which to ases co-evolutionary models of virulence change in Hawai. Previous hypotheses of virulence change (van Riper 1991; Atkinson et al. 1995) have been based in part on comparisons of pathogenicity betwen the Hawaian parasite and a North American strain that was presumed to be its closest counterpart (van Riper 1991). Our results, which provide evidence of cryptic population structure within Plasmodium relictum and an Old World origin for the Hawaian parasite, should provide a more robust foundation for understanding the evolution of virulence and the dynamics of host-parasite-vector interactions in Hawai?s model system. 73 TABLE Table 1. Sampling efort and frequency with which the Hawaian lineage of Plasmodium (GRW4, lineage 15) was recovered from regions shown in figure 2. Lineages defined as unique (difering by at least 1 bp) within a given region may be shared betwen regions. # region total host individuals sampled (n) total species sampled (n) host species with Plasmodium (n) Plasmodium sequences recovered (n) sequences matching GRW4 (n) minimum unique lineages (n) 1. Hawaian Archipelago 320 17 8 79 78 2 2. French Polynesia a 161 8 4 14 14 1 3. USA 161 21 10 61 0 12 4. Bermuda 142 14 7 42 39 3 5. Antiles/Venezuela 5553 169 47 303 2 17 6. Guyana 195 53 22 42 0 23 7. Uruguay 322 111 33 57 0 13 8. Northern Europe b 2835 26 19 305 131 36 9. Southern Europe c 1151 9 8 206 4 16 10. Nigeria 827 71 33 101 7 30 11. Western Africa d 656 105 62 174 0 44 12. South Africa 171 15 8 60 1 15 13. Indian Ocean Islands e 150 20 15 48 23 8 14. India 259 43 23 71 18 23 15. Burma 344 133 42 60 0 28 16. Japan/Korea 209 58 26 48 1 15 74 17. Australia/Papua New Guinea 454 106 30 56 3 22 a French Polynesia: Moorea (Society Islands), Nuku Hiva (Marquesas). b Northern Europe: Belarus, Belgium, England, Germany, Lithuania, Sweden. c Southern Europe: France, Israel, Italy, Spain, Ukraine. d Western Africa: Annabon, Bioko, Cameroon, Gabon, Principe, Sao Tome. e Indian Ocean Islands: Anjouan, Fregate, Grand Comore, Madagascar, Mauritius, Mayotte, Moheli, Praslin, Reunion, Rodrigues. 75 FIGURE LEGENDS Figure 1. Phylogenetic relationships among morphologicaly-identified species of Plasmodium estimated using ML, MP and ME with cytochrome b sequences. Numbers above branches indicate bootstrap support based on 1000 replicates. Numbers before species names correspond to mitochondrial lineage numbers in figure 3. Sequences were obtained directly from Genbank (acesion number in italics) or from extracts of blood smears obtained from the International Reference Collection for Avian Haematozoa (IRCAH) and M. Peirce. Figure 2. Map depicting the global distribution of the single mitochondrial lineage of malaria parasite (GRW4) found in resident Hawaian paserines. Pie charts indicate the proportion of al sequenced Plasmodium infections in a given region that were identical to GRW4 (red). Details concerning locations and sampling efort are in table 1. Red dots indicate additional locations in which GRW4 has been reported previously (Ishtiaq et al.; Mendes et al. 2005; Schrenzel et al. 2003; M. Kimura pers. comm.) or in which GRW4 was recovered from a relatively smal group of samples (Kazakhstan). Figure 3. Phylogenetic tres of parasite mitochondrial lineages (cyt b; left; numbered) and asociated nuclear haplotypes (DHFR-TS; right; letered), constructed using maximum likelihood (GTR+I+G for both markers). The distribution of mitochondrial lineages across global regions is indicated with squares, color-coded to help identify the asociated DHFR-TS sequence (when available, otherwise black). 76 Background shading reflects the limits of two clades with good support in analysis of combined data (figure 4). Mitochondrial lineages which derived from at least one parasite identified as P. relictum by morphology are indicated in bold. Bootstrap support values (>60) are indicated above branches. Figure 4. Majority rule consensus tre of avian malaria parasite lineages generated by Bayesian analysis of combined mitochondrial (cytb b) and nuclear (DHFR-TS) sequence. Parasite genotypes are identified by their respective cytb b lineage (number) and DHFR-TS haplotype (leter), which are depicted separately in figure 3. Clade credibility values are indicated above branches. Background shading identifies two wel-supported clades (A and B) referenced in the text. Dots within Clade A indicate nodes for which New World (open circle) or non-New World (black) origin could be confidently asigned based on ancestral trait reconstruction performed with Mesquite. Figure 5. Statistical parsimony network of Plasmodium mitochondrial lineages related to the Hawaian strain (lineage 15). Sampled haplotypes are numbered as in figure 3 and infered haplotypes are indicated by black dots. Shading indicates whether the lineage was detected in the Old World (white), New World (black) or in both regions (grey). Lineages 9, 19, and 36 were not included due to mising sequence. Lineages 7, 8, and 29 were joined at the 90% connection limit (13 substitutions). 77 78 INWHITECOLOR 1. 3. 4. 5. 6. 7. 8. 9. 13. 12. 10. 11. 14. 15. 16. 17. 2. 79 80 81 82 APPENDICES Apendix 1. Avian hosts, geographical origin, frequency of detection and GenBank acesion numbers for parasite mitochondrial (cyt b) lineages. GenBank numbers for sequences obtained from previously published data are in italics. lineage host information n length GenBank family genus species common name locality a (bp) no. 1 Fringilidae Carpodacus mexicanus House finch USA 1 753 DQ659538 1 Parulidae Seiurus aurocapilus Ovenbird BER 1 256 DQ838987 1 Paseridae Paser domesticus House sparow USA 4 533 DQ838988 1 Troglodytidae Thryomanes bewicki Bewick?s wren USA 1 533 DQ838989 1 Cardinalidae Cardinalis cardinalis Northern cardinal USA 28 533 DQ838990 1 Emberizidae Melospiza georgiana Swamp sparow USA - 551 AY640130 1 Hirundinidae Tachycineta bicolor Tre swalow USA - 551 AY640130 2 Icteridae Molothrus ater Brown-headed cowbird USA 1 753 DQ659539 2 Fringilidae Carpodacus mexicanus House finch USA 2 653 DQ659540 3 Fringilidae Carpodacus mexicanus House finch USA 1 753 DQ659541 4 Parulidae Geothlypis trichas Common yelowthroat USA 1 753 DQ659542 5 Muscicapidae Luscinia svecica Bluethroat NOR 1 753 DQ659543 5 Phasianidae Phasianus colchicus Ring-necked pheasant KOR 1 533 DQ838991 5 Motacilidae Anthus hodgsoni Olive-backed pipit KOR 2 753 DQ838992 5 Corvidae Corvus corone Carion crow JAP 1 433 DQ659544 6 Icteridae Icterus cayanensis Epaulet oriole URU 1 753 DQ659545 83 6 Emberizidae Poospiza lateralis Red-rumped warbling finch URU 1 256 DQ838993 6 Emberizidae Loxigila noctis Leser Antilean bullfinch DOM - 312 AF465558 6 Troglodytidae Troglodytes aedon House wren URU 1 515 DQ838994 7 Parulidae Icteria virens Yelow-breasted chat USA 2 753 DQ659546 7 Pachycephalidae Eulacestoma nigropectus Watled ploughbil PNG - 533 AY714201 8 Parulidae Icteria virens Yelow-breasted chat USA 2 753 DQ659547 9 Muscicapidae Cercotrichas podobe Black scrub-robin NIG - 478 AF495549 10 Tyrannidae Myiarchus tyrannulus Brown-crested flycatcher USA 1 753 DQ659548 11 Parulidae Geothlypis trichas Common yelowthroat USA 1 753 DQ659549 11 Paseridae Passer domesticus House sparow USA - 533 AF069611 11 Icteridae Gnorimopsar chopi Chopi blackbird URU 2 322 DQ838995 11 Turdidae Turdus rufiventris Rufous-belied thrush URU 1 313 DQ838996 11 Troglodytidae Troglodytes aedon House wren URU 3 753 DQ838997 11 Icteridae Pseudoleistes guirahuro Yelow-rumped marshbird URU 1 533 DQ838998 11 Thraupidae Tangara preciosa Chestnut-backed tanager URU 1 533 DQ838999 11 Thraupidae Stephanophorus diadematus Diademed tanager URU 1 335 DQ839000 12 Emberizidae Emberizoides herbicola Wedge-tailed gras- finch GUY 1 753 DQ659550 12 Icteridae Sturnela militaris Red-breasted blackbird GUY 1 256 DQ839001 13 Icteridae Sturnela superciliaris White-browed URU 2 753 DQ659551 84 blackbird 14 Turdidae Alethe diademata Fire-crested alethe GAB 4 753 DQ659552 15 Drepanididae Hemignathus virens Hawai amakihi HI 12 753 DQ659553 15 Drepanididae Hemignathus flavus Oahu amakihi HI 6 533 DQ839002 15 Drepanididae Himatione sanguinea Apapane HI 4 315 DQ839003 15 Drepanididae Vestiaria coccinea I?iwi HI 2 338 DQ839004 15 Emberizidae Carpodacus mexicanus House finch HI 2 351 DQ839005 15 Emberizidae Loxigila violacea Greater Antilean bullfinch DOM 1 533 DQ839006 15 Estrildidae Lonchura punctulata Nutmeg mannikin HI 1 351 DQ839007 15 Estrildidae Lonchura malacca Black-headed munia IND 1 256 DQ839008 15 Estrildidae Neochmia temporalis Red-browed firetail FP 2 753 DQ839009 15 Mimidae Dumetela carolinensis Grey catbird BER 7 533 DQ839010 15 imidae Mimus gilvus Tropical mockingbird GRE 1 533 DQ839011 15 Muscicapidae Cercomela fusca Indian chat IND 1 351 DQ839012 15 uscicapidae Humblotia flavirostris Grand Comoro flycatcher GCO 1 295 DQ839013 15 Muscicapidae Luscinia svecica Bluethroat SWE 5 478 DQ839014 15 Nectariniidae Nectarinia humbloti Humblot?s sunbird GCO 1 295 DQ839015 15 Nectariniidae Nectarinia notata Long-biled gren sunbird GCO 1 533 DQ839016 15 Nectariniidae Cinnyris sovimanga Souimanga sunbird MAD 1 91 DQ839017 15 Paradoxornithidae Paradoxornis webbianus Vinous-throated parotbil KOR 1 91 DQ839018 15 Paseridae Passer domesticus House sparow BER, HI, IND 67 533 DQ839019 15 Ploceidae Foudia madagascarensis Red fody GCO, MAD, PRA 5 533 DQ839020 15 Ploceidae Foudia eminentisima Red-headed fody GCO 2 533 DQ839021 15 Ploceidae Foudia flavicans Rodrigues fody ROD 1 533 DQ839022 85 15 Ploceidae Quelea quelea Red-biled quelea SAF 1 295 DQ839023 15 Sturnidae Acridotheres tristis Common myna AUS, HI, IND, FP 19 533 DQ839024 15 Sturnidae Sturnus vulgaris European starling BER 13 533 DQ839025 15 Sylviidae Acrocephalus arundinaceus Gret red-warbler BEL, ISR, NIG, SWE,UKR 133 478 DQ839026 15 Sylviidae Acrocephalus baeticatus African red- warbler NIG 1 478 DQ839027 15 Sylviidae Acrocephalus gracilirostris Leser swamp- warbler NIG 1 478 DQ839028 15 Sylviidae Acrocephalus schoenobaenus Sedge warbler NIG 1 478 DQ839029 15 Sylviidae Acrocephalus mendanae Marquesan red- warbler FP 9 753 DQ839030 15 Sylviidae Hippolais pallida Olivaceous warbler NIG 1 478 DQ839031 15 Sylviidae Megalurus palustris Striated grasbird IND 1 351 DQ839032 15 Sylviidae Orthotomus cuculatus Common tailorbird IND 1 351 DQ839033 15 Timalidae Stachyris pyrhops Black-chinned babbler IND 1 351 DQ839034 15 Zosteropidae Zosterops borbonicus Mascarene white- eye MAU, REU 3 533 DQ839035 15 Zosteropidae Zosterops mouroniensis Comoro white-eye GCO 2 295 DQ839036 15 Zosteropidae Zosterops maderaspatanus Madagascar white- eye GCO, MAD, AY 5 533 DQ839037 15 Zosteropidae Zosterops choloronothos Mauritius white-eye MAU 1 121 DQ839038 15 Zosteropidae Zosterops lateralis Silvereye FP 1 351 DQ839039 16 Fringilidae Linurgus olivaceus Oriole finch BIO 1 753 DQ659554 16 Fringilidae Serinus atrogularis Black-throated canary SAF 2 753 DQ839040 16 Ploceidae Ploceus velatus African masked- BOT, SAF 3 653 DQ659555 86 weaver 16 Ploceidae Quelea quelea Red-biled quelea SAF, ZIM 19 533 DQ839041 16 Ploceidae Euplectes orix Red bishop SAF 1 295 DQ839042 16 Cisticolidae Cisticola fulvicapilus Piping cisticola SAF 1 295 DQ839043 16 Sulidae Sula capensis Cape gannet SAF 1 533 DQ659556 17 Monarchidae Myiagra alecto Shining flycatcher PNG 2 753 DQ659557 18 Ploceidae Ploceus velatus African masked weaver SAF 1 753 DQ659558 18 Ploceidae Quelea quelea Red-biled quelea BOT, SAF, ZIM 15 533 DQ839044 19 Ploceidae Quelea quelea Red-biled quelea BOT 1 351 DQ659559 20 Nectariniidae Cinnyris coquereli Mayotte sunbird MAY 3 753 DQ659560 20 Nectariniidae Cinnyris sovimanga Souimanga sunbird AD 1 533 DQ839045 21 Ploceidae Foudia seychelarum Seycheles Fody FRE 1 753 DQ659561 21 Nectariniidae Cinnyris dussumieri Seycheles sunbird FRE 1 533 DQ839046 22 Sylviidae Acrocephalus arundinaceus Great red warbler SWE 1 753 DQ659562 22 Sylviidae Sylvia atricapila Blackcap SPA - 478 AF495571 22 Paseridae Passer luteus Sudan golden- sparow NIG - 478 AF495571 22 Fringilidae Carduelis chloris European grenfinch UKR 1 456 DQ659563 22 Paridae Parus major Great tit KOR 1 351 DQ839047 22 Paridae Sitiparus varius Varied tit KOR 1 256 DQ839048 22 Corvidae Garrulus glandarius Eurasian jay KOR 1 753 DQ839049 22 Sturnidae Acridotheres tristis Common myna IND, NZ 3 256 DQ839050 23 Paseridae Passer melanurus Mossie SAF 1 753 DQ659564 24 Motacilidae Motacila alba White wagtail KOR 1 753 DQ659565 24 Sylviidae Acrocephalus orientalis Oriental red warbler KOR 1 256 DQ839051 25 Ploceidae Quelea quelea Red-biled quelea SAF 1 753 DQ659566 26 Zosteropidae Zosterops senegalensis African yelow SAF 1 753 DQ659567 87 white-eye 27 Nectariniidae Cyanomitra oritis Cameroon sunbird BIO 1 753 DQ659568 27 Ploceidae Ploceus melanogaster Black-biled weaver BIO 1 517 DQ839052 28 Alaudidae Alauda arvensis Sky lark KOR 1 753 DQ659569 28 Paridae Parus major Great tit KOR 2 256 DQ839053 29 Nectariniidae Cyanomitra olivacea Olive sunbird GAB, CAM, PRI 6 753 DQ659570 29 Nectariniidae Cinnyris chloropygius Olive-belied sunbird GAB 3 533 DQ839054 30 Ploceidae Ploceus nigerimus Vieilot?s weaver GAB 1 753 DQ659571 30 Ploceidae Ploceus nigricollis Black-necked weaver GAB 1 533 DQ839055 30 Muscicapidae Copsychus malabaricus White-rumped shama BUR 1 533 DQ839056 30 Muscicapidae Copsychus saularis Oriental magpie- robin BUR 1 533 DQ839057 30 Estrildidae Lonchura punctulata Nutmeg manikin IND 1 351 DQ839058 30 Monarchidae Hypothymis azurrea Black-naped monarch BUR 1 295 DQ839059 31 Ploceidae Foudia madagascariensis Red fody MAD, MAY, MAU 5 753 DQ659572 31 Ploceidae Ploceus grandis Giant weaver SAO 2 533 DQ839060 31 Ploceidae Ploceus princeps Principe golden- weaver PRI 1 491 DQ839061 32 Rhipiduridae Rhipdiura rufifrons Rufous fantail MAR 2 753 DQ659573 33 Monarchidae Monarcha kaupi Pied monarch AUS 3 753 DQ659574 33 onarchidae Monarcha trivirgatus Spectacled monarch AUS 3 533 AY714199 33 Petroicidae Heteromyias albispecularis Grey-headed robin AUS 1 533 AY714199 34 Ploceidae Ploceus princeps Principe golden- weaver PRI 1 753 DQ659575 88 34 Nectariniidae Hedydipna collaris Collared sunbird GAB 1 533 DQ839062 34 Nectariniidae Hedydipna platura Pygmy sunbird NIG - 478 AF495566 34 Sturnidae Lamprotornis splendidus Splendid glossy- starling PRI 1 518 DQ839063 35 Muscicapidae Fraseria cinerascens White-browed forest flycatcher GAB 1 753 DQ659576 36 Nectariniidae Cinnyris chloropygius Olive-belied sunbird CAM 1 533 DQ659577 37 Sylviidae Phyloscopus trochilus Wilow warbler SWE 1 753 DQ659578 38 Nectariniidae Cyanomitra olivacea Olive sunbird GAB 1 753 DQ659579 39 Ptilonorhynchidae Scenopoeetes dentirostris Tooth-biled bowerbird AUS 3 753 DQ659580 39 Ptilonorhynchidae Ailuroedes melanotis Spotted catbird AUS 2 533 AY714205 39 Acanthizidae Sericornis magnirostris Large-biled scrubwren AUS 2 533 AY714205 39 Meliphagidae Lichenostomus frenatus Bridled honeyeater AUS 1 533 AY714205 39 onarchidae Monarcha trivirgatus Spectacled monarch AUS 1 533 AY714205 39 Pachycephalidae Colluricincla boweri Bower?s shrike- thrush AUS 1 533 AY714205 40 Aegithinidae Aegithina tiphia Common iora BUR 2 753 DQ659581 41 Turdidae Turdus migratorius American robin USA - 753 AY099033 42 Emberizidae Emberiza rutila Chestnut bunting KOR 1 753 DQ659582 42 Emberizidae Emberiza spodocephala Black-faced bunting KOR 2 533 DQ839064 43 Charadridae Pluvialis fulva Pacific golden plover HI 1 753 DQ659583 43 Fringilidae Carpodacus erythrinus Common rosefinch KOR 1 753 DQ839065 43 Paridae Parus major Great tit SWE - 413 AF254978 44 Paridae Parus major Great tit SE - 447 AF495564 45 Turdidae Turdus philomelos Song thrush SWE - 472 AF495576 46 Cisticolidae Camaroptera brachyura Gren-backed GAB 1 753 DQ659584 89 camaroptera 46 Dicruridae Dicrurus adsimilis Fork-tailed drongo GAB 1 334 DQ839066 46 Estrildidae Pyrenestes ostrinus Black-belied sedcracker GAB 1 335 DQ839067 46 Estrildidae Estrilda astrild Common waxbil TAN 1 533 DQ839068 46 Estrildidae Spermophaga haematina Western bluebil GAB 1 334 DQ839069 46 Muscicapidae Cossypha niveicapila Snowy-crowned robin-chat GAB 1 333 DQ839070 46 Muscicapidae Muscicapa olivascens Olivaceous flycatcher GAB 1 533 DQ839071 46 Muscicapidae Stiphrornis erythrothorax Forest robin GAB 2 533 DQ839072 46 Nectariniidae Chalcomitra rubescens Gren-throated sunbird GAB 1 334 DQ839073 46 Nectariniidae Cyanomitra olivacea Olive sunbird GAB 1 334 DQ839074 46 Nectariniidae Hedydipna collaris Collared sunbird GAB 1 533 DQ839075 46 Ploceidae Malimbus nitens Grey?s malimbe GAB 1 334 DQ839076 46 Ploceidae Ploceus cucullatus Vilage weaver GAB 1 503 DQ839077 46 Pycnonotidae Andropadus gracilis Grey grenbul GAB 1 533 DQ839078 46 Pycnonotidae Andropadus virens Litle grenbul CAM 2 533 DQ839079 46 Pycnonotidae Bleda notata Leser bristlebil GAB 1 334 DQ839080 46 Pycnonotidae Bleda syndactyla Common bristlebil GAB 2 334 DQ839081 46 Pycnonotidae Criniger calurus Red-tailed grenbul GAB 1 533 DQ839082 46 Pycnonotidae Criniger chloronotus Eastern bearded grenbul GAB 1 334 DQ839083 46 Pycnonotidae Nicator chloris Yelow-spotted nicator GAB 1 315 DQ839084 46 Rhipiduridae Rhipidura rufifrons Rufous fantail MAR 2 753 DQ839085 46 Turdidae Alethe poliocephala Brown-chested alethe GAB 4 533 DQ839086 46 Turdidae Neocossyphus fraseri Rufous thrush GAB 4 533 DQ839087 90 46 Turdidae Neocossyphus poensis White-tailed ant- thrush GAB 2 334 DQ839088 46 Turdidae Neocossyphus rufus Red-tailed ant- thrush GAB 1 533 DQ839089 46 Turdidae Zoothera cameronensis Black-eared ground- thrush GAB 1 533 DQ839090 46 Turdidae Cyanomitra olivacea Olive sunbird GAB 1 533 DQ839091 47 Timalidae Pomatorhinus feruginosus Coral-biled scimitar-babbler BUR 1 753 DQ659585 47 Timalidae Alcippe morrisonia Grey-cheeked fulveta BUR 1 295 DQ839092 47 Muscicapidae Cinclidium leucurum White-tailed robin BUR 1 533 DQ839093 48 uscicapidae Niltava sundara Rufous-belied niltava BUR 1 533 DQ659586 49 Nectariniidae Cyanomitra olivacea Olive sunbird GAB 1 533 DQ659587 50 Icteridae Quiscalus quiscula Common grackle USA - 753 AY099031 50 n/a n/a n/a n/a VIE - 753 AY099029 51 Strigidae Ninox scutulata Brown hawk-owl SIN - 753 AY099035 52 Ardeidae Ardea herodias Great blue heron USA 1 753 DQ659588 53 Columbidae Zenaida macroura Mourning dove USA - 753 AY099032 53 Spheniscidae Spheniscus demursus African penguin SAF 1 335 DQ659589 54 Paridae Parus major Great tit SWE 1 753 DQ659590 55 n/a n/a n/a n/a n/a - 478 AY178904 56 Sylviidae Acrocephalus arundinaceus Great red warbler SWE 1 753 DQ659591 57 Sylviidae Acrocephalus arundinaceus Great red warbler SE - 753 AY099040 58 Paridae Cyanistes caeruleus Blue tit SWE - 753 AY099045 59 Alcedinidae Alcedo leucogaster White-belied kingfisher GAB 1 753 DQ659592 91 a AUS, Australia; BER, Bermuda; BEL, Belarus; BIO, Bioko; BOT, Botswana; BUR, Burma; CAM, Cameroon; DOM, Dominican Republic; FRE, Fregate; FP, French Polynesia; GAB, Gabon; GCO, Grand Comore; GRE, Grenada; GUY, Guyana; HI, Hawai; IND, India; ISR, Israel; KOR, Korea; MAD, Madagascar; MAU, Mauritius; MAR, Marianas; MAY, Mayotte; NIG, Nigeria; NOR, Norway; NZ, New Zealand; PNG, Papua New Guinea; PRA, Praslin; PRI, Principe; REU, Reunion; ROD, Rodrigues Island; SAF, South Africa; SAO, Sao Tome; SIN, Singapore; SWE, Sweden; TAN, Tanzania; UKR, Ukraine; URU, Uruguay; USA, United States; VIE, Vietnam; ZIM, Zimbabwe Apendix 2. Avian hosts, geographical origin, frequency of detection, asociated mitochondrial lineage and GenBank acesion numbers for DHFR-TS haplotypes shown in figure 3. DHFR-TS did not amplify from al samples for which a mitochondrial lineage was recovered. lineage host location a n asociated mtDNA lineage GenBank no. A Yelow-breasted chat USA 2 7 DQ659597 A Yelow-breasted chat USA 3 8 DQ659597 B House finch USA 1 1 DQ659598 B Brown-headed cowbird USA 1 2 DQ839094 B Bluethroat NOR 1 5 DQ839095 B Carion crow JAP 1 5 DQ839096 B Olive-backed pipit KOR 2 5 DQ839097 92 B Ring-necked pheasant KOR 1 5 DQ839098 B Epaulet oriole URU 1 6 DQ839099 B Red-rumped warbling finch URU 1 6 DQ839100 B House wren URU 1 6 DQ839101 C Northern cardinal USA 2 1 DQ659599 D Chestnut-backed tanager URU 1 11 DQ659600 D Wedge-tailed gras-finch GUY 1 12 DQ839102 E hite-browed blackbird URU 1 13 DQ659601 F Fire-crested alethe GAB 2 14 DQ659602 G Hawai amakihi HI 4 15 DQ659603 G Gray catbird BER 4 15 DQ839103 G Greater Antilean bullfinch DOM 1 15 DQ839104 G Red-browed firetail FP 2 15 DQ839105 G Marquesan red-warbler FP 2 15 DQ839106 G European starling BER 4 15 DQ839107 G House sparow BER, HI 7 15 DQ839108 G Iiwi HI 2 15 DQ839109 G Oahu amakihi HI 1 15 DQ839110 G Shining flycatcher PNG 1 17 DQ839111 G African masked weaver SAF 1 18 DQ839112 G Red-biled quelea SAF 1 18 DQ839113 G African masked weaver SAF 1 16 DQ839112 G Oriole finch BIO 1 16 DQ839114 G Black-biled weaver BIO 1 27 DQ839115 G Sky lark KOR 1 28 DQ839116 G Rufous fantail MAR 2 32 DQ839117 G Spectacled monarch AUS 2 33 DQ839118 H Madagascar white-eye GCO, MAD 4 15 DQ659604 I Black-naped monarch BUR 1 30 DQ659605 J Red fody MAD, MAY 2 31 DQ659606 93 K White-rumped shama BUR 1 30 DQ659607 L Collared sunbird GAB 1 34 DQ659608 M Vieilot?s weaver GAB 1 30 DQ659609 Giant weaver SAO 1 31 DQ839119 M Principe golden-weaver PRI 1 31 DQ839120 N Eurasian jay KOR 1 22 DQ659610 N Great tit KOR 1 22 DQ839121 N White wagtail KOR 1 24 DQ839122 N Blackcap SPA 1 22 AY560372 O Mossie SAF 1 23 DQ659611 P Olive sunbird GAB 1 38 DQ659612 Q n/a ? Plasmodium gallinaceum n/a - - AY033582 R Bower?s shrike thrush AUS 1 39 DQ659613 R Bridled honeyeater AUS 1 39 DQ839123 R Spotted catbird AUS 1 39 DQ839124 S Common iora BUR 1 40 DQ659614 T Black-eared ground thrush GAB 1 46 DQ659615 T Litle grenbul CAM 1 46 DQ839125 T Rufous fantail MAR 2 46 DQ839126 U Common rosefinch KOR 1 43 DQ659616 U Black-faced bunting KOR 2 42 DQ839127 U Chestnut bunting KOR 1 42 DQ839128 V Mourning dove USA 1 53 DQ659617 W Great red-warbler SWE - 56 AY560373 X Blue tit SE - 58 AY560369 94 a AUS, Australia; BER, Bermuda; BIO, Bioko; BUR, Burma; CAM, Cameroon; DOM, Dominican Republic; FP, French Polynesia; GAB, Gabon; GCO, Grand Comore; GUY, Guyana; HI, Hawai; JAP, Japan; KEN, Kenya; KOR, Korea; MAD, Madagascar; MAR, Marianas; MAY, Mayotte; NOR, Norway; PNG, Papua New Guinea; PRI, Principe; SAF, South Africa; SAO, Sao Tome; URU, Uruguay; USA, United States 95 CHAPTER IV Immunological change in island birds: no uniform evidence of an ?island syndrome? ABSTRACT Dramatic declines of native Hawaian avifauna due to the human-mediated emergence of avian malaria and pox prompted an examination of whether reduced imune response is a generalized component of an island syndrome, potentialy driven by increased inbreding and reduced exposure to parasites. We tested this hypothesis by characterizing parasite exposure, genetic diversity and several measures of imune response in both recently-introduced and endemic island taxa and by comparing the results to those observed in closely-related mainland counterparts. We show that significant reorganization of the inflamatory cel- mediated response as wel as levels of innate imune compounds such as natural antibodies and complement may occur following island colonization. However, we did not find evidence of uniformly reduced imune response in island taxa, even amongst those taxa with the longest residence times. Our results point to the potential importance of smal diferences in the pathogenic landscape and the stochastic history of mutation and genetic drift in shaping the imunological profiles of smal isolated populations. Consequently, predicting the impact of introduced disease on the many other endemic faunas of the remote Pacific wil remain a chalenge. 96 INTRODUCTION Emerging disease in wildlife is an important force driving the decline and extinction of threatened populations (Cooper 1993, Lyles and Dobson 1993, Wikelski et al. 2004) and may pose a threat to worldwide biodiversity (Daszak et al. 2000). Human- mediated environmental changes are often the root cause of disease emergence (Friend et al. 2001) and may have particularly dire consequences in island ecosystems. In Hawai, for example, the introduction of non-native songbirds and the mosquito vector Culex quinquefasciatus has led to the emergence of avian malaria and avian poxvirus in endemic honeycrepers (Drepanididae), contributing to the dramatic declines and contracting range limits of several species (Warner 1968, Atkinson et al. 1995, 2000, Yorinks and Atkinson 2000). While host species that have been introduced to Hawai from continental sources over the last several centuries are largely unafected by avian malaria, endemic species may exhibit mortalities ranging as high as 100% (Atkinson et al. 2001, summary in Jarvi et al. 2001). This suggests that the introduced strain of malaria is not unusualy virulent; instead, it appears that at least some long-term island residents are unusualy susceptible to this parasite. High susceptibility of island endemics to infectious disease has been proposed as a component of an ?island syndrome? (Hochberg and Moller 2001, Matson 2006), which seks to codify typical changes observed in body size (Milien 2006), life history traits such as survival and fecundity (Wiggins et al. 1998, Goltsman et al. 2005) and other features asociated with insular organisms (Blondel 2000). Several factors common to insular life could be driving susceptibility in island endemics. First, the strength of selection exerted by infectious agents on the imune systems of 97 remote island taxa is likely to be lower than that experienced by mainland birds over their evolutionary history. Because the parasites that are succesfully transported to an island by avian colonists are only a subsample of those present in the source host population and because even those parasites may go extinct due to reduced transmision probabilities while the smal island host population becomes established (Colautti et al. 2004), parasite richnes is typicaly low on islands compared to the mainland (Fromont et al. 2001, Beadel et al. 2006), though prevalence may vary depending on relative transmision eficiency and host densities (Dobson 1988). On the remote islands of the Pacific in particular, even acounting for recent extinctions (Steadman 1997), bird communities are extremely depauperate and unlikely to sustain the diversity or abundance of parasites observed in large and diverse mainland host communities. Given the physiological costs asociated with developing, maintaining, and using an imune system (Klasing and Barnes 1988, Scrimshaw 1991, Lochmiler and Derenberg 2000, Zuk and Stoehr 2002, Martin et al. 2003), in a parasite- impoverished environment, selection should favor birds that maximize fitnes by alocating resources away from the imune system and perhaps towards other fitnes-related traits such as reproductive efort (Gustafson et al. 1994, Derenberg et al. 1997, Nordling et al. 1998), survival (Gonzalez et al. 1999) or the expresion of sexual ornaments (Hilgarth and Wingfield 1997, Peters et al. 2004). A second factor which may contribute to susceptibility of island fauna is the low genetic diversity typicaly asociated with smal population sizes (Frankham 1997). Theory (Wright 1931, Nei et al. 1975) and observations on natural avian systems (Baker and Moeed 1987, Tar et al. 1998) suggest that bottlenecks, such as 98 those experienced upon colonization of an island, are most likely to decrease alelic diversity (due to the loss of rare aleles) while heterozygosity wil decline only if the bottleneck is severe and the growth rate of the population is low. Additional diversity may be lost due to serial bottlenecks (Clegg et al. 2002) if island populations, already constrained to be smal by island size, are repeatedly reduced due to demographic stochasticity. This later efect may be important in driving the diferences in disease susceptibility observed in recently introduced versus endemic species. Observations in wild populations have confirmed the deleterious impacts of bottlenecks and inbreding on imunological parameters (Reid et al. 2003, Hawley et al. 2005, Hale and Briskie 2007) and parasite susceptibility (Acevedo-Whitehouse et al. 2006, Pearman and Garner 2005, Whiteman et al. 2006), however, drift is unlikely to afect al populations similarly and thus, the impact of inbreding on disease susceptibility is not likely to be uniform (Spielman et al. 2004). The Hawaian honeycrepers have become a model for understanding the susceptibility of a na?ve fauna to exotic disease, but given a relative paucity of data on disease prevalence and consequences in island taxa (but se Goltsman et al. 1996, Wikelski et al. 2004, Gottdenker et al. 2005, Smits et al. 2005, Cliford et al. 2006), the extent to which this model applies elsewhere across the globe is not obvious. For example, in contrast to Hawai, the avifauna of American Samoa is characterized by stable native communities exhibiting relatively high prevalence of chronic infection with possibly indigenous blood parasites (Jarvi et al. 2003, Atkinson et al. 2007). Lack of clear paralels to the Hawaian model may reflect Hawai?s unique position as the most remote archipelago in the world or the unique susceptibility of the 99 Drepanidine radiation to exotic disease. Alternatively, introduced pathogens may have decimated similarly susceptible species so quickly that paralel declines have gone unrecorded elsewhere in the world. In the Pacific region especialy, which harbors 24% of al threatened birds species (BirdLife International 2006), Hawai, the Galapagos (Wikelski et al. 2004, Gottdenker et al. 2005, Parker et al. 2006) and New Zealand (Tompkins and Gleson 2006) have received the vast majority of atention, to the neglect of numerous other archipelagoes, many of which are extremely isolated and home to smal populations of endemics. To investigate the evolution of imunity in island taxa and to test for a common imunological signature of an island syndrome, we characterized imunological responses in endemic and recently-introduced bird populations on remote islands of the Pacific and compared the results to closely-related taxa from mainland Australia. Because vertebrate imunity depends on a diversity of defenses of variable specificity and inducibility (Schmid-Hempel and Ebert 2003) and because succesful imune defense may emphasize just a single component of those defenses (Zuk and Stoehr 2002), we characterized multiple components of imunity using techniques that were applicable to wild and, in some cases, vulnerable populations. As a measure of constitutive innate imunity, we asayed levels of natural antibodies and complement in plasma (Matson et al. 2005). Natural antibodies are germ-line encoded molecules that are important in initial recognition of pathogens (Ochsenbein et al.1999) and may be linked to activation of the B-cel mediated production of specific antibodies (Parmentier et al. 2004). They are also integral for initiating the action of complement, a suite of enzymes that function together to lyse foreign cels 100 (Janeway et al. 2005). As an index of cel-mediated imunity, we measured the delayed-type hypersensitivity response to injection with the plant-derived mitogen PHA. The sweling that results reflects the action of T-lymphocytes, which secrete cytokines and direct the recruitment of macrophages, basophils, heterophils and B- lymphocytes to the site of injection (Janeway et al. 2005). This response is potentialy important in the defense against intracelular parasites such as viruses and haemosporidia (Gonzalez et al. 1999). A strong response has been linked to increased probability of survival (Gonzalez et al. 1999, Merino et al. 2000, Moller and Saino 2004) and may be indicative of high exposure to parasites over evolutionary time (Martin et al. 2001). If evolution on remote, parasite-impoverished islands necesarily leads to increased susceptibility to exotic parasites, then we would expect imune responses to be lower in island populations relative to their mainland counterparts. In addition, we would expect this patern to be most evident in island endemics relative to species that have been introduced to an island only recently. Alternatively, island colonization may lead to variable upregulation or downregulation of imune components depending on the costs of those components in a new environment, their lability in the face of genetic changes, and the particular parasites with which they are chalenged. While a previous study pointed to defined paterns of imunological reorganization characteristic of an island syndrome (Matson 2006), here, we find no evidence for a uniformly reduced imune response or for other fixed paterns of change in island taxa. We explore this outcome in light of the genetic and parasitological context in which imunity has evolved and, through the consideration 101 of both recently-introduced and endemic island species, we addres the timescale on which imunological changes have occurred in an extremely isolated avifauna. METHODS Avian System We characterized imune response, genetic variability, and parasite exposure in mainland populations of thre species of songbirds and compared the results to closely related island populations or species representing isolation at two diferent time scales. As a model of short-term isolation on islands, we sampled populations of Red-browed firetails (Neochmia temporalis) and Silvereyes (Zosterops lateralis) from their native range in Australia (Blakers et al. 1985) and also from French Polynesia. N. temporalis was introduced to French Polynesia in the late 19 th century and may have been reintroduced in 1938, while Z. lateralis was most likely introduced in 1938 (Long 1981). As a model of long-term evolution in an island environment, we compared two island endemic species of Acrocephalus red warbler to their most closely related mainland form (A. australis, Fleischer et al., unpublished manuscript). We sampled the Rimitara red warbler (A. rimitarae) on Rimitara, Austral Islands, French Polynesia and the Bokikokiko (A. aequinoctialis) on Kiritimati, Line Islands, Kiribati as these species likely represent two distinct lineages of Pacific warblers (Fleischer et al., unpublished manuscript) and their populations were sufficiently large and acesible to acommodate sampling. Australian populations of N. temporalis, Z. lateralis and A. australis were sampled using mistnets betwen June and July, 2005 at several sites in the region from Brisbane south to the Clarence River. Island populations of N. temporalis and Z. 102 lateralis were sampled at two sites on Moorea, French Polynesia in July and August, 2005. Rimitara red warblers were sampled in August 2005 and Bokikokiko were sampled in March 2006. The timing of sampling insured that birds were not breding, except in the case of the Bokikokiko which may bred opportunisticaly throughout the year given evidence of territoriality and nesting by some individuals in both March (pers. obs.) and mid-July (Milder and Schreiber 1982). In this case, we did not perform imune asays on females that were obviously atending a nest (i.e. females carying an egg or possesing a wel-developed brood patch). Protocols for handling birds were approved by Animal Care and Use Commites at the University of Maryland (R-05-19) and the Smithsonian National Zoological Park (05-10). Parasite Screning Al captured birds were visualy inspected for evidence of exposure to Avipoxvirus spp. (wartlike lesions on exposed skin). In addition, we screned blood smears for trypanosomes and microfilaria. For each slide, we scanned 30 fields at 100x and 50 fields at 500x magnification. Finaly, we screned DNA, extracted from blood samples using DNeasy kits (Qiagen), for evidence of haematozoa in the genera Haemoproteus, Leucocytozoon, and Plasmodium using molecular methods described previously (Beadel et al. 2004, Beadel and Fleischer 2005). PCR methods have proven more likely to detect haematozoon infection than microscopy alone (Richard et al. 2002). Briefly, we used primers F2/R2 and 213F/372R to detect parasite infections. The later includes restriction sites that are diagnostic for the thre diferent parasite genera. In order to evaluate the diversity of parasite lineages present in any population, we used forward primers F2, Fifi, or 3760F with reverse primer 103 4292rw2 to amplify a 295 to 533 bp fragment of cytochrome b, which was then sequenced and compared to sequences on GenBank to confirm parasite identification. We tested for significant diferences in prevalence betwen island and mainland populations using Fisher?s exact test, since al comparisons involved cels with low values. Genetic Variability We quantified levels of genetic diversity using microsatelites designed by previous authors for use with taxa related to N. temporalis (6 loci; Sefc et al. 2001), Z. lateralis (12 loci; Degnan et al. 1999, Frentiu et al. 2003), and Acrocephalus spp. (12 loci; Hansson et al. 2000, Richardson et al. 2000). Loci chosen for use across species of Pacific Acrocephalus were originaly isolated from distantly related taxa (A. arundinaceus or A. seychelensis) and therefore, ascertainment bias should not contribute to any diferences in diversity observed betwen species. Generaly, PCR reactions were caried out in a total volume of 10 uL with 1x PCR buffer, 1 U of AmpliTaq DNA polymerase (Applied Biosystems), 0.2 mM each dNTP (NEB), 0.5 uM each primer, and concentrations of MgCl and/or betaine and other conditions as indicated in the online appendix. Products were separated on an ABI 3100 Genetic Analyzer (Applied Biosystems). Aleles were aligned and scored using Genotyper 2.5 (Perkin-Elmer) and manualy binned. Al loci were tested for significant deviations from Hardy-Weinberg equilibrium (HWE) in Genepop. Samples that yielded homozygotes at any locus showing significant departures from HWE were rerun at les stringent conditions to reduce the likelihood of alelic dropout. 104 Values for observed heterozygosity were obtained from Genepop while values for alelic richnes, genic diversity and the coeficient of inbreding (F IS ) were calculated using FSTAT v 2.9.3. Diferences betwen mainland and island values were tested for significance across loci within each species (i.e., N. temporalis, Z. lateralis) or species group (i.e., Acrocephalus spp.) using a Wilcoxon signed-ranks test implemented with SAS v 9.1. In addition, we calculated internal relatednes (IR) for each individual using microsatelite alele data and the Excel spreadsheet IR_macroN3 (Amos et al. 2001). IR provides a measure of inbreding (i.e. parental relatednes), which is similar to standardized heterozygosity but weights aleles by their frequency in the population. Thus, an individual homozygous for a rare alele wil be scored as more inbred than an individual homozygous for a common alele. To test for a significant efect of inbreding on imune measures, we reran the analysis of imune diferences (described below), but added IR, as wel as the interaction of IR with population, as fixed efects in the ANCOVA. Imunological Tests We characterized multiple components of the avian imune system. As a measure of investment in the cel-mediated imune response, we chalenged a subsample of captured birds with the plant-derived mitogen phytohaemagglutinin (PHA). Following the basic protocol of Smits et al. (1999), we measured the patagium of captured birds to the nearest 0.01 m with a digital smal-face spline micrometer (Fowler) prior to and 24 hours after injection with PHA. Each measurement was repeated thre times and averaged. Resultant sweling, which is widely used as a measure of the cel-mediated response (reviewed in Martin et al. 2001), reflects the 105 recruitment and proliferation of T-lymphocytes and macrophages at the site of injection, as wel as the action of other leucocytes (Martin et al. 2006). For N. temporalis and Z. lateralis, we injected 20uL of a 1.5mg/mL solution of PHA (Sigma L9017) in PBS buffer (Sigma P4417) into the patagium. For al Acrocephalus spp., which were larger, we injected 25ul of a 3mg/mL solution. Birds were housed in portable cages in the shade and provided with ad libitum water and food (seds, fruit, or larval-stage invertebrates depending on the species). We also characterized two components of constitutive innate imunity in plasma sampled from island and mainland populations. We measured levels of natural antibodies, as indicated by the agglutination of foreign red blood cels, and we measured levels of complement, as indicated by the lysis of these foreign cels (Matson et al. 2005). In the case of the island endemics A. aequinoctialis and A. rimitarae, as wel as island and mainland populations of N. temporalis and Z. lateralis, plasma was obtained from blood samples taken imediately after capture of individuals that did not undergo further imunological testing. In the case of A. australis, several blood samples (n = 15) were collected after completion of the PHA asay (se results). Plasma was stored in a minus 20?C frezer in the field and then transfered to a -80?C frezer until asayed (June 2006). We performed serial dilutions (in PBS, Sigma P-4417) of plasma samples across 12 wels of a 96-wel plate (Corning 3798) and then added an equal volume (25 uL) of 1% washed rabbit red blood cels (Hemostat R83546). Each plate included a dilution series of 7 randomly chosen plasma samples plus a row of negative controls (PBS only) to aid in scoring agglutination and lysis endpoints. Plates were covered with parafilm, mixed 106 for 2 minutes on a rotary shaker, then placed in a 37?C incubator for 90 minutes. Following tilting of the plate at 45 degres for 20 minutes, we created a computerized image for future scoring of agglutination, and then incubated the plate at 37?C for an additional 70 minutes. At this point, plates were again scanned to record the extent of lysis. Agglutination titer, scored blindly with respect to species and location, was determined by the last plasma dilution at which aggregated blood cels showed ?peaking? relative to the negative control. Lysis titer was determined by the lowest concentration of plasma at which >50% of rabbit red blood cels had ruptured. Asays were repeated on samples for which sufficient plasma was available and scores for these samples were averaged. Least squares means for mainland and island imune responses were generated and tested separately for each taxa using contrasts in PROC MIXED (SAS v 9.1). We employed an ANCOVA framework with imune response as the dependent variable, population (or species in the case of Acrocephalus) as the main efect, and body condition (the residuals of the regresion of mas on tarsus length) as a covariate. In al cases, we tested for a significant interaction of condition with population before proceding with a model that did not include the interaction term. RESULTS Parasite Prevalence and Diversity We did not detect trypanosomes, microfilarae or pox lesions in any individuals and therefore, parasite analyses are limited to haematozoan infections (fig. 1). Across al individuals of the thre species surveyed in Australia (n = 165), we detected 8 unique lineages of blood parasite. In contrast, we detected only a single lineage of blood 107 parasite among the four island species (n = 174) surveyed on Kiritimati, Moorea and Rimitara. Within Z. lateralis, prevalence of Haemoproteus spp. in the introduced population on Moorea, French Polynesia (0%, n = 59) was significantly lower than that observed in Australia (39.1%, n = 64; p < 0.001), where only a single lineage was detected. Prevalence of Plasmodium spp. did not difer significantly betwen island (1.7%) and mainland (6.3%) populations of Z. lateralis, however, mainland populations harbored at least 2 lineages of Plasmodium, both of which were divergent from the single lineage detected in just one individual from the French Polynesian population (se below). No significant diferences were observed in the prevalence of any blood parasites betwen island (n = 34) and mainland (n = 67) populations of N. temporalis. We detected a single individual infected with Leucocytozoon spp. in Australia and this parasite was not detected in the introduced population on Moorea, French Polynesia. In addition, Plasmodium spp. was detected in individuals from both populations at low prevalence (1.5 to 5.9%), however the lineage in Australia was diferent from that found in N. temporalis from French Polynesia. Interestingly, the only lineage of blood parasite recovered from any forest bird sampled on Moorea (including introduced species N. temporalis, Z. lateralis, Pycnonotus cafer (n = 10), Lonchura castaneothorax (n = 24), Acridotheres tristis (n = 10), Estrilda astrild (n = 10), Geopelia striata (n = 8) and the endemic dove Ptilinopus purpuratus (n = 5)) identicaly matched the strain of avian malaria introduced to Hawai (Beadel et al. 2006), suggesting that it is a relatively recent introduction. 108 Plasmodium spp. was the only blood parasite detected in the mainland taxon A. australis. The prevalence of Plasmodium spp. in A. australis was 17.7% (n = 34) and we detected thre distinct lineages of parasite. In contrast, we did not detect any blood parasites in the endemic red warblers A. aequinoctialis (n = 25) or A. rimitarae (n = 56). The diference in the prevalence of Plasmodium spp. betwen island and mainland taxa was significant (p = 0.03 for both comparisons). Genetic Diversity No locus exhibited a significant departure from HWE within any island or mainland population when p-values were Bonferoni corrected for multiple comparisons (p > 0.004 for Acrocephalus spp. and Z. lateralis; p > 0.008 for N. temporalis). Ase13 and Ase58 exhibited significant linkage disequilibrium, but only within A. aequinoctialis (p < 0.0001). This is likely an artifact of low diversity given that these loci have been mapped to distinct linkage groups in the related warbler A. arundinaceus (B. Hansson pers. comm.) and that we did not detect linkage disequilibrium betwen these loci in either A. australis or A. rimitarae. Al other pairs of loci appeared to segregate independently within each population when p-values were corrected for multiple comparisons (p > 0.0001 for Acrocephalus spp. and Z. lateralis; p > 0.003 for N. temporalis) and therefore, we treated locus-specific indices of diversity as independent samples when comparing genetic diversity betwen populations. Al loci examined were polymorphic in mainland populations of Z. lateralis and N. temporalis as wel as recently introduced island populations of these species. In contrast, while 100% of loci were polymorphic in the mainland species A. australis, 3 of 12 loci were fixed in the island endemic A. aequinoctialis and 7 of 12 109 loci were fixed in A. rimitarae (genetic diversity indices summarized in table 1). Alelic richnes tended to be lower in island populations of al species (or species groups) and the diference was significant in the case of the recently introduced island population of Z. lateralis (minimum n = 59, diference = -4.3 aleles, S = 33, p = 0.001) as wel as the endemic warblers A. aequinoctialis (minimum n = 25, diference = -5.8 aleles, S = 39, p = 0.001) and A. rimitarae (minimum n = 25, diference = -6.6 aleles, S = 39, p = 0.001) relative to mainland counterparts. For Z. lateralis and N. temporalis, which were recently introduced to French Polynesia and in which alele frequency changes were unlikely to have been altered by mutation events, we tested whether low alelic diversity in French Polynesian populations could be atributed to the loss of rare aleles. For each species, we divided aleles recovered from the Australian source population into two clases depending on whether they had been retained or lost upon founding of the French Polynesian population. We excluded loci in which al aleles had been retained. For the remaining loci, we calculated the average frequency of aleles in each clas and compared the diference across loci using a Wilcoxon signed-ranks test. As expected, the average frequency of aleles (in the Australian population) that were lost in French Polynesia tended to be lower than the frequency of aleles that were retained folowing colonization in N. temporalis (n = 5, ? frequency = -0.14, S = 7.5, p = 0.063) and this diference was signficant in Z. lateralis (n = 11, ? frequency = -0.226, S = 33, p = 0.001). No significant diferences were detected in either gene diversity (H E ) or observed heterozygosity (H o ) betwen recently introduced taxa and their mainland counterparts. However, both measures of heterozygosity were significantly lower in 110 the island endemics A. aequinoctialis (?H E = -0.34, S = 36, p = 0.002, ?H o = -0.42, S = 39, p = 0.001) and A. rimitarae (?H E = -0.50, S = 39, p = 0.001, ?H o = -0.51, S = 39, p = 0.001) compared to A. australis. F IS , a measure of the overal level of inbreding in a population, tended to be slightly, but not significantly, lower in recently introduced populations of Z. lateralis and N. temporalis relative to their mainland source (table 1). Conversely, F IS was higher in both island endemic warblers compared to the mainland taxon however this diference was not significant. Our power to detect a significant diference among warblers was impaired by high levels of fixation across microsatelite loci, which alowed for the comparison of F IS at just 4 polymorphic loci across al thre species. Imune Response Across al imune tests, we did not observe consistent changes in island birds relative to their mainland counterparts, nor did we observe consistent trends even when island birds were grouped by island residence time (fig. 2). With regard to recently-introduced island residents, Z. lateralis in French Polynesia exhibited a significant decrease in cel-mediated response relative to the Australian population (diference = -0.34 m, t = -4.37, df = 60, p < 0.001), however, no diference was evident betwen the innate response of island and mainland populations as measured by agglutination. The French Polynesian population of N. temporalis, on the other hand, showed no significant diferences in either cel-mediated or innate imune response when compared to a mainland population. Neither species demonstrated a measurable lysis response in either island or mainland populations. 111 The two Acrocephalus species representing long-term island evolution exhibited strikingly divergent paterns of imune response across al thre tests. A Wilcoxon-Mann-Whitney test for diferences in innate imune responses betwen A. australis that had been treated with PHA (n = 15) and those that were untreated (n = 4) revealed no significant diferences (agglutination, Z = -1.18, p = 0.24; lysis, Z = - 1.52, p = 0.13), and therefore data were combined to provide a baseline for comparison to responses in endemic warblers. A. aequinoctialis exhibited litle change in cel-mediated imune response relative to A. australis, however both measures of innate imunity were significantly lower (agglutination diference = -1.1 titers, t = -2.42, df = 40, p = 0.020; lysis diference = -1.1 titers, t = -3.15, df = 41, p = 0.003). In contrast, A. rimitarae exhibited significantly increased imune responses relative to A. australis across al thre asays (patagial sweling diference = 0.26 m, t = 3.66, df = 62, p < 0.001; agglutination diference = 1.3 titers, t = 3.91, df = 40, p < 0.001; lysis diference = 0.6 titers, t = 2.38, df = 41, p = 0.022). Across Acrocephalus spp., agglutination titers and lysis titers were significantly correlated (n = 46, r = 0.69, p < 0.001). Among the 15 individuals of A. australis that were asayed for both cel-mediated and innate components of imunity, no correlation was evident betwen degre of patagial sweling and either agglutination or lysis titers. The efect of body condition on imune response was significant in the models of cel-mediated (F = 8.47, df = 60, p = 0.005) and innate (F = 4.96, df = 21, p = 0.037) responses observed in Z. lateralis. Condition did not contribute significantly to any other model of imune response in either N. temporalis or Acrocephalus spp. 112 Individual measures of breding (IR) did not acount for a significant proportion of variance in any imune response in any of the taxa asayed, nor did we recover a significant interaction betwen IR and population (or species in the case of Acrocephalus spp.). We examined the efect of infection status on imune response in A. australis and Z. lateralis from Australia, two populations in which infection rates were high enough to warant tests. In A. australis, cel-mediated response tended to be lower in individuals infected with Plasmodium spp. (mean patagial sweling = 0.22, n = 5) compared to those that were uninfected (mean = 0.38, n = 19) but this diference was not signficant based on a t-test (t = 1.36, df = 22, p = 0.19). Similarly, no significant diferences were observed in innate imune responses betwen infected and uninfected A. australis. Cel-mediated responses in Z. lateralis that were infected with Haemoproteus spp. (mean patagial sweling = 0.471, n = 9) also showed a decline relative to uninfected individuals (mean = 0.7088, n = 22) but this diference was not significant (t = 1.79, df = 29, p = 0.084), and the relationship became les obvious when comparing cel-mediated responses of mainland Z. lateralis infected (mean = 0.55, n = 11) or uninfected (mean = 0.69, n = 20) with any blood parasite (t = 1.10, df = 29, p = 0.28). Again, no significant diferences were observed in innate imune response betwen infected and uninfected individuals. DISCUSSION Our results do not uniformly support the hypothesis that island taxa exhibit reduced imunological capacity compared to mainland relatives, despite the fact that our system incorporated two key components that could theoreticaly drive an island 113 syndrome: reduced exposure to pathogens and reduced genetic diversity. Those taxa with the longest history of evolution on islands were no more likely to exhibit reduced imune responses than taxa with shorter island residence times. Furthermore, changes in the response of island taxa relative to their mainland counterparts were not consistent among taxa with similar island residency times. These results, along with similar reports of significant population-level variation in imune response (Lindstr?m et al. 2004, Matson 2006, Whiteman et al. 2006), suggest that this response may be more appropriately considered in light of the particular parasite communities to which a population is exposed and the particular genetic background in which imunity is evolving. Recently Introduced Island Populations In keeping with the prediction of an island syndrome, the island population of Z. lateralis showed a significant decrease in cel-mediated imunity relative to its mainland source population, although innate imunity was unchanged. In contrast to Z. lateralis, however, the imunological profile of the other recently-introduced island resident N. temporalis was largely similar to that found in mainland individuals, though our sample size for innate imunity provided low power with which to detect a diference. The diferential efects of island life on the cel-mediated imune response in these two species may be atributable to the degre to which the island forms have escaped mainland parasites. While the island population of Z. lateralis appeared to have lost a common mainland parasite (Haemoproteus spp.), prevalence of haemosporidian parasites in N. temporalis was largely unchanged in the island population relative to the mainland. N. temporalis exhibited a low prevalence 114 of infection with avian malaria (Plasmodium spp.) both on the mainland and in French Polynesia and the one parasite that may have been lost by the island population occurred only infrequently on the mainland (Leucocytozoon spp.). Haemosporidia represent only one clas of parasite to which island birds may be diferentialy exposed. Given their potentialy deleterious efects on reproduction and survival (Bennet et al. 1993, Merino et al. 2000, Valkiunas 2005), however, they are likely to be important factors structuring host imunity. The absence of Haemoproteus spp. in the island population of Z. lateralis may represent a relaxation of selection on the imune system, alowing for a shift in resources away from a costly nonspecific inflamatory cel-mediated response. Given that we did not detect any diference in response to PHA betwen mainland individuals that were infected or uninfected with Haemoproteus spp, it is unlikely that the population-level reduction in cel-mediated response observed in French Polynesia can be atributed to direct efects of parasitism on the response of mainland individuals. Parasitized individuals on the mainland tended to have lower cel-mediated response and thus, if these individuals were removed from consideration, the diference in response betwen island and mainland populations would have been even larger. Among recently-introduced taxa, the efects of reduced parasite exposure on imune response cannot be fully decoupled from the potential efects of reduced genetic diversity. As with blood parasite exposure, genetic composition was similar betwen the imunologicaly indistinguishable island and mainland populations of N. temporalis. On the other hand, the island population of Z. lateralis, which exhibited lower prevalence of parasites, also exhibited reduced alelic richnes. Conforming to 115 theoretical expectation, the aleles that were lost were rare in the mainland source population. If this reduction in alelic richnes at neutral loci was acompanied by a paralel loss of functional aleles (e.g., antigen-binding motifs encompased by the major histocompatibility locus and expresed on T-cels, Hansson and Richardson 2005; but se Aguilar et al. 2004), then this loss might be reflected by reduced sensitivity to a novel antigen such as PHA. Increased inbreding has also been demonstrated to be negatively corelated to cel-mediated imune response within bottlenecked populations of song sparows (Reid et al. 2003) and house finches (Hawley et al. 2005). In Z. lateralis, however, we observed similar levels of heterozygosity in island and mainland populations and found no support for a relationship betwen imune response and individual measures of inbreding. Therefore, while we cannot exclude the possibility that the loss of a particular alele has contributed to a population-level efect of reduced cel-mediated imune response, inbreding is unlikely to have afected this response Island Endemics While the change observed in cel-mediated response across populations of Z. lateralis could simply reflect plasticity in the imune system, the changes observed in the two island endemics relative to their mainland counterpart may represent long- term evolutionary responses to their isolated environment. A. aequinoctialis and A. rimitarae exhibit average pairwise cytochrome b divergences of 1.7% and 2.3%, respectively, from their closest mainland relative A. australis (based on 551 bp of cytb, K2P distances, data not shown). Using a molecular clock calibrated for paserine cytb (1.6% per milion years; Fleischer et al. 1998) and correcting for 116 ancestral polymorphism (Nei and Li 1979, Avise et al. 1998) by subtracting the mean intraspecific divergence observed in the continental form (0.4%), these divergences equate to approximate separation times of 0.8 and 1.2 milion years. Given this period of isolation, we expected that the strongest signal of an island syndrome would be found in the endemic red warblers. Interestingly, the imunological profiles observed in A. aequinoctialis and A. rimitarae were extremely divergent and did not support this hypothesis, even though both species exhibited significant reductions in genetic diversity and reduced exposure to parasites. While A. aequinoctialis exhibited only reduced innate imune response relative to the baseline provided by A. australis, A. rimitarae exhibited significantly higher innate and cel-mediated responses than the mainland control. If investment in imunity is costly, then the later result is particularly surprising given the degre to which A. rimitarae has likely been exposed to pathogens. A. rimitarae, like A. aequinoctialis, appears to have escaped the avian malaria parasites present in its mainland congener, though the presence of a recently introduced lineage of parasite on adjacent archipelagoes suggests that this may soon change. Considering the isolation of Rimitara (ca. 3200 km east of Fiji, 6400 km east of Australia), the smal island size (ca. 9km 2 ), and the paucity of alternative terestrial reservoirs for pathogens (four species including junglefowl and one introduced finch), our failure to detect haematozoa in A. rimitarae (as wel as in the co-occurring finch L. castaneothorax, n = 10) may reflect an impoverished pathogen community in general. The same is likely true for Kiritimati, which is substantialy larger, but similarly isolated (ca. 6400 km east-northeast of Australia) and host to just two 117 additional terestrial bird species (a native lorikeet and junglefowl). On the other hand, native paserines may be exposed to at least some of the pathogens known to be present in semi-domesticated junglefowl in both French Polynesia and Kiribati (Pacific Animal Health Information System http:/ww.spc.int/rahs, Gottdenker et al. 2005). We surveyed only a smal fraction of the total pathogen community that may occur on these islands and therefore, it is possible that diferential exposure to just a handful of unmeasured pathogens could be driving diferential imune response. Futhermore, even if parasite communities on both islands are currently impoverished, slightly diferent histories of pathogen colonization and extinction on Rimitara and Kiritimati could be sufficient to drive diferential investment in the imune system as wel as diferential partitioning of resources betwen arms of imunity. Another possible explanation for the generaly high response observed in A. rimitarae may be that the particular imune reponses that we measured are not extremely costly to maintain and use. For example, Matson (2006) has proposed that insular birds may actualy favor innate imune responses over adaptive humoral imunity. Likewise, while the inflamatory cel-mediated response can be costly in terms of both nutrients required (Lochmiler and Derenberg 2000) and potential damage inflicted to the organism itself (Janeway et al. 2005), it is not wel understood how these costs compare to those required to support the adaptive antibody mediated response, which we did not measure. Adaptive responses are generaly cheap to use but can incur substantial developmental costs asociated with the time and resources required to produce a diverse B-cel repertoire (Humphrey et al. 2002). Imunocompetence in the face of a particular chalenge may manifest itself by the 118 absence of a response (Boots and Bowers 2004), or at least by varied emphasis on any particular arm of the imune system (Zuk and Stoehr 2002). Therefore, our results for A. rimitarae would support the hypothesis of an island syndrome if the high responses observed were coupled to downregulation of a much more costly adaptive imune response. The increased cel-mediated response observed in A. rimitarae is consistent with the gradient of increasing inflamatory response observed in Galapagos finches exposed to fewer pathogens on increasingly smaler islands (Lindstr?m et al. 2004). However, we did not observe a concomitant reduction in natural antibody titers, which was observed in the Galapagos finches and which may be indicative of adaptive antibody imune capacity (Parmentier et al. 2004). In addition, the contrasting imunological profile of A. aequinoctialis highlights the fact that imunological reorganization in island taxa, if it occurs at al, is not uniform. One factor underlying this lack of uniformity in imune response may be the stochastic nature of genetic drift acting on regulatory regions or functional genes asociated with imunity. As observed in Z. lateralis, an initial bottleneck asociated with island colonization may result in a loss of alelic diversity across the genome, which if asociated with a concomitant loss of MHC diversity (Richardson and Hansson 2005) could lead to reduced surveilance for foreign antigens. Over the longer-term, smal island populations may exhibit further erosion of alelic diversity, as wel as reduced heterozygosity and higher levels of inbreding, as evidenced in A. aequinoctialis and A. rimitarae. Changes in imune response observed in bottlenecked or inbred populations (Reid et al. 2003, Hawley et al. 2005, Hale and Briskie 2007) may be linked to indirect efects of inbreding, the random loss of 119 resistance aleles, and also the loss of any advantages that may be asociated with overdominance. In addition, deleterious mutations in imunologicaly-important regions of DNA may become fixed given the increasing strength of drift over selection in smal populations and the fact that selection may be reduced if the pathogenicity of the island environment is indeed reduced. Fixation of a mutation afecting regulation of the cel-mediated imune pathway could explain the unusualy strong response observed in A. rimitarae. An optimal imune response is not necesarily a maximal response (Viney et al. 2005) and therefore, the strong imune response that we observed may be more indicative of a damaging alergic reaction than increased investment in that particular arm of imunity or increased ability to fight of disease. Conclusions Our results indicate that 1) significant reorganization of the inflamatory cel- mediated response as wel as levels of innate imune compounds such as natural antibodies and complement may occur following island colonization and 2) imunological reorganization in island taxa is not uniform, even among taxa with similar residence times. Variation in imunological profiles likely reflect smal diferences in the pathogenic landscape and the stochastic history of mutation and genetic drift in smal populations. Unfortunately, it is not clear to what extent high or low imune response can be linked to the phenotypes in which we are most interested: disease resistance or susceptibility (Adamo 2004). Similarly, while numerous studies have equated reduced genetic diversity to increased disease susceptibility (O?brien et al. 1985, Roelke et al. 1993, Acevedo-Whitehouse 2006, 120 Reid et al. 2003, Hawley et al. 2005, Pearman and Garner 2005), this relationship is not universaly applicable (Reid et al. 2003, Spieleman 2004). Therefore, predicting the susceptibility of the many smal and threatened populations of birds residing on islands of the remote Pacific may not be feasible. In the case of the genus Acrocephalus, the detection of a Hawaian-like strain of avian malaria in populations of A. mendanae from the Marquesas, French Polynesia (Beadel et al. 2006) has provided the basis for a natural experiment. This species, like A. aequinoctialis and A. rimitarae, is a long-term island resident and exhibits reduced genetic diversity similar to its congeners studied here (unpubl. data). Interestingly, A. mendanae is apparently robust to infection with the introduced parasite and provides some hope that related members of the widely-distributed and highly endemic Pacific genus Acrocephalus wil not be decimated by this parasite. We should be careful, however, in extrapolating even this far, considering that even the Hawaian honeycrepers exhibit substantial variation in susceptibility to malaria (Jarvi et al. 2001) although they share a common evolutionary background. Imunity integrates not only energetic investments, which may be constrained by variable physiological demands of alternate life histories and diferential parasite exposure, but also geneticaly- determined molecular recognition and regulation systems which are subject to random, population-specific efects of drift. Therefore, imunity in isolated fauna is unlikely to follow the simple heuristic of an island syndrome and may be best asayed with experimental chalenges employing the particular pathogen of interest. 121 TABLE Table 1. Genetic diversity indices for introduced and endemic island taxa compared to their mainland counterpart. Population n % Loci polymorphic (total surveyed) Alelic richnes H E H o F IS a Introduced Z. lateralis Mainland 64 100 (12) 8.1 0.57 0.54 0.048 Island 59 100 (12) 3.8 0.52 0.53 -0.027 N. temporalis Mainland 68 100 (6) 10.1 0.69 0.67 0.032 Island 34 100 (6) 8.0 0.70 0.69 0.007 Endemic Acrocephalus spp. Mainland (A. australis) 34 100 (12) 8.3 0.69 0.68 -0.005 Island (A. aequinoctialis) 25 75 (12) 2.5 0.35 0.28 0.124 Island (A. rimitarae) 56 42 (12) 1.7 0.19 0.17 0.145 Bold face indicates significant diferences in the island index relative to the mainland (p < 0.05). a For comparative purposes, F IS for Acrocephalus is based on only 4 loci for which gene diversity (i.e., expected heterozygosity) was non-zero in al thre species. 122 FIGURE LEGENDS Figure 1. Prevalence of the haematozoan parasites Plasmodium spp. (black), Haemoproteus spp. (light grey) and Leucocytozoon spp. (dark grey) across mainland and island populations of Z. lateralis, N. temporalis, and Acrocephalus spp. Asterisks indicate significantly lower parasite prevalences in recently introduced island populations (Haemoproteus spp. in Z. lateralis) or island endemic species (Plasmodium spp. in A. aequinoctialis, above, and A. rimitarae, below) relative to their mainland counterparts. Figure 2. Cel-mediated (PHA-induced patagial sweling) and innate (agglutination and lysis of rabbit red blood cels) measures of imune response in endemic and recently-introduced island taxa (circles) relative to a mainland control (centered at zero). Numbers indicate sample size and standard eror is depicted by vertical bars, or by grey shading in the case of the mainland control. Filed circles indicate a significant diference from the mainland control. 123 FIGURES Z. lateralis N. temporalis Acrocephalus spp. Mainland Island n = 64 n = 59 n = 67 n = 34 n = 56 n = 34 n = 25 * * * 124 125 APPENDIX Apendix 1. PCR conditions for microsatelite loci. Locus Tm (?C) Mg (mM) Betaine (%) BSA (mg/mL) Cycles Source N. temporalis IND7 54 1.5 - - 35 Sefc et al. 2001 IND8 56 1.0 - - 35 Sefc et al. 2001 IND28 56 1.5 - - 35 Sefc et al. 2001 IND29 56 1.5 - - 35 Sefc et al. 2001 IND38 54 1.5 - - 35 Sefc et al. 2001 IND41 54 1.5 - - 35 Sefc et al. 2001 Z. lateralis ZL12 58 1.5 10 - 35 Degnan et al. 1999 ZL14 58 2.0 - 1.0 35 Degnan et al. 1999 ZL18 58 1.5 10 1.0 35 Degnan et al. 1999 ZL22 57 1.5 10 - 35 Degnan et al. 1999 ZL35 60 1.5 10 - 35 Degnan et al. 1999 ZL38 56 2.0 - - 35 Degnan et al. 1999 ZL41 53 2.0 - 1.0 35 Frentiu et al. 2003 ZL44 53 1.0 10 - 30 Frentiu et al. 2003 ZL45 58 1.5 - - 39 Frentiu et al. 2003 ZL46 54 2.0 10 1.0 35 Frentiu et al. 2003 ZL50 59 1.5 - - 35 Frentiu et al. 2003 ZL54 58 1.5 10 - 35 Frentiu et al. 2003 Acrocephalus spp. Aar2 60 1.5 - - 35 Hansson et al. 2000 Ase7 60 1.0 - - 35 Richardson et al. 2000 126 Ase9 60 1.0 - - 35 Richardson et al. 2000 Ase11 62 1.5 - - 35 Richardson et al. 2000 Ase12 60 1.5 - - 35 Richardson et al. 2000 Ase13 62 1.5 - - 35 Richardson et al. 2000 Ase34 60 1.5 - - 35 Richardson et al. 2000 Ase48 60 2.0 - - 35 Richardson et al. 2000 Ase51 60 1.5 - - 35 Richardson et al. 2000 Ase56 60 1.5 - - 35 Richardson et al. 2000 Ase57 TD a 1.5 - - 35 Richardson et al. 2000 Ase58 60 2.0 - - 35 Richardson et al. 2000 a Touch-down cycle as in Richardson et al. 2000. 127 Bibliography Acevedo-Whitehouse, K., Gulland, F., Greig, D. & Amos, W. 2006. Inbreding: disease susceptibility in California sea lions. Nature, 422, 35. Adamo, S. A. 2004. How should behavioural ecologists interpret measurements of imunity? Animal Behaviour, 68, 1443-1449. Aguilar, A., Roemer, G., Debenham, S., Binns, M., Garcelon & Wayne, R. K. 2004. High MHC diversity maintaine by balancing selection in an otherwise geneticaly monomorphic mamal. Procedings of the National Academy of Sciences U. S. A., 101, 3490-3494. Amos, W., Wilmer, J. W., Fullard, K., Burg, T. M., Croxal, J. P., Block, D. & Coulson, T. 2001. The influence of parental relatednes on reproductive succes. Procedings of the Royal Society London B, 268, 2021-2027. Anderson, R. M. & May, R. M. 1992. Infectious diseases of humans, dynamics and control. Oxford University Pres, New York. Atkinson, C. T. 1986. Host specificity and morphometric variation of Haemoproteus meleagridis Levine, 1961 (Protozoa: Haemosporina) in galinaceous birds. Canadian Journal of Zoology, 64, 2634-2638. Atkinson, C. T., Dusek, R. J., Woods, K. L. & Iko, W. M. 2000. Pathogenecity of avian malaria in experimentaly-infected Hawai Amakihi. Journal of Wildlife Diseases, 36, 197-204. Atkinson, C. T., Lease, J. K., Drake, B. M. & Shema, N. P. 2001. Pathogenicity, serological responses, and diagnosis of experimental and natural malarial infections in native Hawaian thrushes. The Condor, 103, 209-218. Atkinson, C. T., Utzurrum, R. C., Seamon, J. O., Savage, A. F. & LaPointe, D. A. 2007. Hematozoa of forest birds in American Samoa ? evidence for a diverse, indigenous parasite fauna from the South Pacific. Pacific Conservation Biology, 13, 229-237. Atkinson, C. T., & van Riper II, C. 1991. Pathogenicity and epizootiology of avian haematozoa: Plasmodium, Leucocytozoan, and Haemoproteus. In Bird- parasite interactions, ecology, evolution and behavior, J. E. Loye & M. Zuk (eds.). Oxford University Pres, Oxford, U.K., p. 19-48. Atkinson, C. T., Woods, K. L., Dusek, R. J., Sileo, L. S. & Iko, W. M. 1995. Wildlife 128 disease and conservation in Hawai: pathogenecity of avian malaria (Plasmodium relictum) in experimentaly infected Iiwi (Vestiaria coccinea). Parasitology, 111, S59-S69. Avise, J. C., Walker, D. & Johns, G. C. 1998. Speciation durations and Pleistocene efects on vertebrate phylogeography. Procedings of the Royal Society London B, 265, 1707-1712. Baker, A. J. & Moeed, A. 1987. Rapid genetic diferentiation and founder efect in colonizing populations of common mynas (Acridotheres tristis). Evolution, 41, 525-538. Beadel, J. S. & Fleischer, R. C. 2005. A restriction enzyme-based asay for distinguishing betwen avian haematozoa. Journal of Parasitology, 91, 683- 685. Beadel, J. S., Gering, E., Austin, J., Dumbacher, J. P., Peirce, M., Prat, T. K., Atkinson, C. T. & Fleischer, R. C. 2004. Prevalence and diferential host- specificity of two avian blood parasite genera in the Australo-Papuan region. Molecular Ecology, 13, 3829-3844. Beadel, J. S., Ishtiaq, F., Covas, R., Melo, M., Waren, B. H., Atkinson, C. T., Bensch, S., Graves, G. R., Jhala, Y. V., Peirce, M. A., Rahmani, A. R., Fonseca, D. M. & Fleischer, R. C. 2006. Global phylogeographic limits of Hawai?s avian malaria. Procedings of the Royal Society London B, 273, 2935-2944. Bennet, G. F., Bishop, M. A. & Peirce, M. A. 1993. Checklist of the avian species of Plasmodium Marchiafava & Celi, 1885 (Apicomplexa) and their distribution by avian family and Walacean life zones. Systematic Parasitology, 26, 171- 179. Bennet, G. F. & Cameron, M. 1974. Seasonal prevalence of avian hematozoa in paseriform birds of Atlantic Canada. Canadian Journal of Zoology, 52, 1259- 1264. Bennet, G. F. & Campbel, A. G. 1973 Avian Haemoproteidae. II. Description of Haemoproteus fusca sp. N. and re-description of Haemoproteus halcyonis De Melo, 1935, parasites of the kingfisher family Alcedinidae. Journal of Parasitology, 59, 337-343. Bennet, G. F., Montgomerie, R. & Seutin, G. 1992. Scarcity of haematozoa in birds breding on the Arctic tundra of North America. Condor, 94, 289?292. Bennet, G. F. & Peirce, M. A. 1988 Morphological form in the avian 129 Haemoproteidae and an annotated checklist of the genus Haemoproteus Kruse, 1890. Journal of Natural History, 22, 1683-1696. Bennet, G. F., Peirce, M. A. & Ashford, R. W. 1993. Avian haematozoa: mortality and pathogenicity. Journal of Natural History, 27, 993-1001. Bensch, S. & Akeson, S. 2003 Temporal and spatial variation of hematozoans in Scandinavian wilow warblers. Journal of Parasitology, 89, 388-391. Bensch, S., P?rez-Tris, J., Waldenstr?m, J. & Helgren, O. 2004. Linkage betwen nuclear and mitochondrial DNA sequences in avian malaria parasites: multiple cases of cryptic speciation? Evolution, 58, 1617-1621. Bensch, S., Stjernman, M., Haselquist, D., Ostman, O., Hansson, B., Westerdahl, H. & Torres Pinheiro, R. 2000. Host specificity in avian blood parasites: a study of Plasmodium and Haemoproteus mitochondrial DNA amplified from birds. Procedings of the Royal Society London B, 267, 1583-1589. Bensch, S., Waldenstr?m, J., Jonz?n, N., Westerdahl, H., Hansson, B., Sejberg, D. & Haselquist, D. 2007. Temporal dynamics and diversity of avian malaria parasites in a single host species. Journal of Animal Ecology, 76, 112-122. Bentz, S., Rigaud, T., Baroca, M., Martin-Laurent, F., Bru. D., Moreau, J. & Faivre, B. 2006. Sensitive measure of prevalence and parasitaemia of haemosporidia from European blackbird (Turdus merula) populations: value of PCR-RFLP and quantitative PCR. Parasitology, 133, 685-692. BirdLife International. 2000. Threatened birds of the world. Lynx Edicions, Barcelona, and BirdLife International, Cambridge, UK. Blakers, M., Davies, S. J. J. F. & Reily, P. N. 1985. The atlas of Australian birds. International Scholarly Book Service, Inc., Beaverton. 738pp. Blondel, J. 2000. Evolution and ecology of birds on islands: trends and prospects. Vie et Mileu, 50, 205-220. Boots, M. & Bowers, R. G. 2004. The evolution of resistance through costly acquired imunity. Procedings of the Royal Society London B, 271, 715-723. Cann, R. L., Feldman, R. A., Agullana, L. & Fred, L. A. 1996. A PCR approach to detection of malaria in Hawaian birds. In Molecular genetic approaches in conservation, T. B. Smith & R. K. Wayne (eds.). Oxford University Pres, New York, New York, p. 202-213. Cibois, A., Thibault, J.-C. & Pasquet, E. 2004. Biogeography of eastern Polynesian 130 monarchs (Pomarea): an endemic genus close to extinction. Condor, 106, 837-851. Cleaveland, S., Laurenson, M. K. & Taylor, L. H. 2001. Diseases of humans and their domestic mamals: pathogen characteristics, host range and the risk of emergence. Philosophical Transaction of the Royal Society London B, 356, 991-999. Clegg, S. M., Degnan, S. M., Kikkawa, J., Moritz, C., Estoup, A. & Owens, I. P. F. 2002. Genetic consequences of sequential founder events by an island- colonizing bird. Procedings of the National Academy of Sciences, 99, 8127- 8132. Clement, M., Posada, D. & Crandal, K. A. 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology, 9, 1657-1660. Cliford, D. L., Mazet, J. A. K., Dubovi, E. J., Garcelon, D. K., Coonan, T. J., Conrad, P. A. & Munson, L. 2006. Pathogen exposure in endangered island fox (Urocyon litoralis) populations: implications for conservation management. Biological Conservation, 131, 230-243. Colautti, R. I., Riciardi, A., Grigorovich, I. A. & MacIsac, H. J. 2004. Is invasion succes explained by the enemy release hypothesis? Ecology Leters, 7, 721- 733. Conway, D. J., Fanelo, C., Lloyd, J. M., Al-Joubori, B. M. A., Baloch, A. H., Somanath, S. D., Roper, C., Oduola, A. M. J., ulder, B., Povoa, M. M., Singh, B. & Thomas, A. W. 2000. Origin of Plasmodium falciparum malaria is traced by mitochondrial DNA. Molecular and Biochemical Parasitology, 111, 163-171. Cooper, J. E. 1993. Historical survey of disease in birds. Journal of Zoo and Wildlife Medicine, 24, 256-264. Cunningham, A. A. & Daszak, P. 1998. Extinction of a species of land snail due to infection with a microsporidian parasite. Conservation Biology, 12, 1139- 1141. Daszak, P., Cunningham, A. A. & Hyat, A. D. 2000. Emerging infectious diseases of wildlife ? threats to biodiversity and human health. Science, 287, 443-449. Daszak, P., Cunningham, A. A. & Hyat, A. D. 2003. Infectious disease and amphibian population declines. Diversity and Distributions, 9, 141-150. de Castro, F. & Bolker, B. 2005. Mechanisms of disease-induced extinction. Ecology Leters, 8, 117-126. 131 Dem, S. L., Karesh, W. B. & Weisman, W. 2001. Putting theory into practice: wildlife health in conservation. Conservation Biology, 15, 1224-1233. Derenberg, C., Arpanius, V., Dan, S. & Bos, N. 1997. Reproductive efort decreases antibody responsivenes. Procedings of the Royal Society London B, 264, 1021-1029. Degnan, S. M., Robertson, B. C., Clegg, S. M. & Moritz, C. 1999. Microsatelite primers for the study of gene flow and mating systems in white-eyes (Zosterops). Molecular Ecology, 8, 157-158. Dobson, A. P. 1988. Restoring Island Ecosystems: the potential of parasites to control introduced mamals. Conservation Biology, 2, 31-39. Dobson, A. & Foufopoulos, J. 2001. Emerging infectious pathogens of wildlife. Philosophical Transactions of the Royal Society London B, 356, 1001-1012. Dumbacher, J. P., Behler, B. M., Spande, T. F., Garafo, H. M. & Daly, J. W. 1992. Homobatrachotoxin in the genus Pitohui: chemical defense in birds? Science, 258, 799-801. Dumbacher, J. P., Prat, T. K. & Fleischer, R. C. 2003. Phylogeny of the owlet- nightjars (Aves: Aegothelidae) based on mitochondrial DNA sequence. Molecular Phylogenetics and Evolution, 29, 540-549. Escalante, A. A., Freland, D. E., Collins, W.E. & Lal, A. A. 1998 The evolution of primate malaria parasites based on the gene encoding cytochrome b from the linear mitochondrial genome. Procedings of the National Academy of Science, 95, 8124-8129. Ewers, W. H. 1967. The blood parasites of some New Guinea birds. Nova Guinea, 38, 427-432. Falon, S. M., Bermingham, E. & Ricklefs, R. E. 2003. Island and taxon efects in parasitism revisited: avian malaria in the Leser Antiles. Evolution, 57, 606- 615. Falon, S. M., Bermingham, E. & Ricklefs, R. E. 2005. Host specialization and geographic localization of avian malaria parasites: a regional analysis in the Leser Antiles. American Naturalist, 165, 466-480. Falon, S. M., Fleischer, R. C. & Graves, G. R. 2006. Malarial parasites as geographical markers in migratory birds? Biology Leters, 2, 213-216. Falon, S. M., Ricklefs, R. E., Lata, S. C. & Bermingham, E. 2004. Temporal 132 stability of insular avian malarial parasite communities. Procedings of the Royal Society London B, 271, 493-500. Falon, S. M., Ricklefs, R. E., Swanson, B. L. & Bermingham, E. 2003. Detecting avian malaria: an improved polymerase chain reaction diagnostic. Journal of Parasitology, 89, 1044-1047. Feagin, J. E. 1992. The 6-kb element of Plasmodium falciparum encodes mitochondrial cytochrome genes. Molecular and Biochemical Parasitology, 52, 145-148. Feldman, R. A., Fred, L. A. & Cann, R. L. 1995. A PCR test for avian malaria in Hawaian birds. Molecular Ecology, 4, 663-673. Fleischer, R. C. & McIntosh, C. E. 2001. Molecular systematics and biogeography of the Hawaian avifauna. Studies in Avian Biology, 2, 51-60. Fleischer, R. C., McIntosh, C. E. & Tar, C. L. 1998. Evolution on a volcanic conveyor belt: Using phylogenetic reconstructions and K-Ar-based ages of the Hawaian Islands to estimate molecular evolutionary rates. Molecular Ecology, 7, 533-545. Fleischer, R. C., Olson, S., James, H. F. & Cooper, A. C. 2000. The identity of the extinct Hawaian eagle (Haliaeetus) as determined by mitochondrial DNA sequence. Auk, 117, 1051-1056. Fonseca, D., Smith, J., Wilkerson, R. & Fleischer, R. 2006. Pathways of expansion and multiple introductions ilustrated by large genetic diferentiation among worldwide populations of the southern house mosquito. American Journal of Tropical Medicine and Hygiene, 74, 284-289. Frankham, R. 1997. Do island populations have les genetic variation than mainland populations? Heredity, 78, 311-327. Fred, L. A. & Cann, R. L. 2003. On polymerase chain reaction tests for estimating prevalence of malaria in birds. Journal of Parasitology, 89, 1261-1264. Frentiu, F. D., Lange, C. L., Burke, T. & Owens, I. P. F. 2003. Isolation of microsatelite loci in the Capricorn silvereye, Zosterops lateralis chlorcephalus (Aves: Zosteropidae). Molecular Ecology Notes, 3, 462-464. Friend, M., McLean, R. G. & Dein, F. J. 2001. Disease emergene in birds: chalenges for the twenty-first century. The Auk, 118, 290-303. Fromont, E., Morviliers, L., Artois, M. & Pontier, D. 2001. Parasite richnes and 133 abundance in insular and mainland feral cats: insularity or density? Parasitology, 123, 143-151. Garnham, P. C. C. 1966. Malaria parasites and other Haemosporidia. Blackwel Scientific Publishers, Oxford, U.K., 1114 p. Goltsman, M., Kruchenkova, E. P. & Macdonald, D. W. 1996. The Mednyi Arctic foxes: treating a population imperiled by disease. Oryx, 30, 251-258. Goltsman, M., Kruchenkova, E. P., Sergeev, S., Volodin, I. & Macdonald, D. W. 2005. ?Island syndrom? in a population of Arctic foxes (Alopex lagopus) from Mednyi Island. Journal of Zoology, 267, 405-418. Gonzalez, G., Sorci, G., Moller, A. P., Ninni, P., Haussy, C. & de Lope, F. 1999. Imunocompetence and condition-dependent sexual advertisement in male house sparows (Passer domesticus). Journal of Animal Ecology, 68, 1225- 1234. Gottdenker, N. L., Walsh, T., Vargas, H., Merkel, J., Jimenez, G. U., Miler, R. E., Dailey, M. & Parker, P. G. 2005. Asesing the risks of introduced chickens and their pathogens to native birds in the Galapagos archipelago. Biological Conservation, 126, 429-439. Gustafson, L., Nordling, D., Anderson, M. S., Sheldon, B. C. & Qvarnstr?m, A. 1994. Infectious diseases, reproductive efort and the cost of reproduction in birds. Philosophical Transactions of the Royal Society London B, 346, 323- 331. Hale, K. A. & Briskie, J. V. 2007. Decreased imunocompetence in a severely bottlenecked population of an endemic New Zealand bird. Animal Conservation, 10, 2-10. Handbook of the Birds of the World Volume 8 Broadbils to Tapaculos. 2003. J. del Hoyo, A. Eliot & D. Christie (eds.). Lynx Edicions, Barcelona, Spain. Hansson, B., Bensch, S., Haselquist, D., Lilandt, B.-G., Wennerberg, L. & von Schantz, T. 2000. Increase of genetic variation over time in a recently founded population of great red warblers (Acrocephalus arundinaceus) revealed by microsatelites and DNA fingerprinting. Molecular Ecology, 9, 1529-1538. Hansson, B. & Richardson, D. S. 2005. Genetic variation in two endangered Acrocephalus species compared to a widespread congener: estimates based on functional and random loci. Animal Conservation, 8, 83-90. Hatchwel, B. J., Wood, M. J., Anwar, M. A., Chamberlain, D. E. & Perins, C. M. 134 2001. The haematozoan parasites of Common Blackbirds Turdus merula: asociations with host condition. Ibis, 143, 420-426. Hatchwel, B. J., Wood, M. J., Anwar, M. A. & Perins, C. M. 2000. The prevalence and ecology of the haematozoan parasites of European blackbirds, Turdus merula. Canadian Journal of Zoology, 78, 684-687. Hawley, D. M., Sydenstricker, K. V., Kollias, G. V. & Dhondt, A. A. 2005. Genetic diversity predicts pathogen reisistance and cel-mediated imunocompetence in house finches. Biology Leters, 1, 326-329. Hilgarth, N. & Wingfield, J. C. 1997. Parasite-mediated sexual selection: endocrine aspects. In Host-parasite evolution, D. H. Clayton & J. Moore (eds.). Oxford University Pres, Oxford, p. 78-104 Hochachka, W. M. & Dhondt, A. A. 2000. Density-dependent decline of host abundance resulting from a new infectious disease. Procedings of the National Academy of Sciences U. S. A., 97, 5303-5306. Hochberg, M. E. & Moller, A. P. 2001. Insularity and adaptation in coupled victim- enemy asociation. Journal of Evolutionary Biology, 14, 539-551. Holyoak, D. T. & Thibault, J.-C. 1984. Contribution ? l??tude des oiseaux de Polyn?sie orientale. M?moires Museum National Histoire Naturele, Paris (s?r. A). Zoologie, 127, 1-209. Humphrey, B. D., Koutsos, E. A. & Klasing, K. C. 2002. Requirements and priorities of the imune system for nutrients. In Nutrition biotechnology in the fed and food industries: Procedings of Altech?s 18th annual symposium, T. P. Lyons & K. A. Jasques (eds.). Nottingham University Pres, Nottingham, p. 69-77. Ishtiaq, F., Beadel, J., Baker, A. J., Rahmani, A. R., Jhala, Y. V. & Fleischer, R. 2006. Prevalence and evolutionary relationships of haematozoan parasites in native versus introduced populations of common myna Acridotheres tristis. Procedings of the Royal Society London B, 273, 587-594. Janeway, C. A., Travers, P., Walport, M. & Sclomchik, M. J. 2005. Imunobiology, 6th ed. Garland Science Publishing, New York. Janovy, J. 1997 Protozoa, helminths, and arthropods of birds. In D. Clayton & J. Moore (eds.). Host-parasite evolution. New York: Oxford University Pres. Jarvi, S. I., Atkinson, C. T. & Fleischer, R. C. 2001. Imunogenetics and resistance to avian malaria in Hawaian honeycrepers (Drepanidinae). Studies in Avian Biology, 22, 254-263. 135 Jarvi, S. I., Farias, M. E. M., Baker, H., Freifeld, H. B., Baker, P., van Gelder, E., Masey, J. G. & Atkinson, C. T. 2003. Detection of avian malaria (Plasmodium spp.) in native land birds of American Samoa. Conservation Genetics, 4, 629-637. Jarvi, S. I., Schultz, J. J. & Atkinson, C. T. 2002. PCR diagnostics underestimate the prevalence of avian malaria (Plasmodium relictum) in experimentaly-infected paserines. Journal of Parasitology, 88, 153-158. Jones, H. I. 1985. Hematozoa from montane forest birds in Papua New Guinea. Journal of Wildlife Diseases, 21, 7-10. Joy, D. A., Feng, X., Mu, J., Furuya, T., Chotivanich, K., Kretli, A. U., Ho, M., Wang, A., Suh, E., Berli, P. & Su, X.-Z. 2003. Early origin and recent expansion of Plasmodium falciparum. Science, 300, 318-321. Kimura, M., Dhondt, A. A. & Lovete, I. J. 2006. Phylogeographic structuring of Plasmodium lineages across the North American range of the house finch (Carpodacus mexicanus). Journal of Parisitology, 92, 1043-1049. Klasing, K. C. & Barnes, D. M. 1988. Decreased amino acid requirements of growing chicks due to imunologic stres. Journal of Nutrition, 118, 1158-1164. Laird, M. & van Riper II, C. 1981. Questionable reports of Plasmodium from birds in Hawai, with the recognition of P. relictum sp. capistranoae (Russel, 1932) as the avian malaria parasite there. In Parasitological Topics, E. V. Canning (ed.). Alen Pres, Lawrence, pp. 159-165. Li, J., Collins, W. E., Wirtz, R. A., Rathore, D., Lal, A. & McCutchan, T. R. 2001. Geographic subdivision of the range of the malaria parasite, Plasmodium vivax. Emerging Infectious Disease, 7, 35-42. Lindstr?m, K. M., Foufopoulos, J., P?rn, H. & Wikelski, M. 2004. Imunological investments reflect parasite abundance in island populations of Darwin?s finches. Procedings of the Royal Society London B, 271, 1513-1519. Lochmiler, R. L. & Derenberg, C. 2000. Trade-ofs in evolutionary imunology: just what is the cost of imunity? Oikos, 88, 87-98. Lockhart, P. J., Stel, M. A., Hendy, M. D. & Penny, D. 1994. Recovering evolutionary tres under a more realistic model of sequence evolution. Molecular Biology and Evolution, 11, 605-612. Long, J. L. 1981. Introduced birds of the world. Universe Books, New York. 528 p. Lyles, A. M. & Dobson, A. P. 1993. Infectious disease and intensive management: 136 population dynamics, threatened hosts, and their parasites. Journal of Zoo and Wildlife Medicine, 24, 315-326. Maddison, W. P. & Maddison, D. R. 2004. Mesquite: a modular system for evolutionary analysis. Version 1.05 http:/mesquiteproject.org. Manwel, R. D. 1936. The problem of species, with special reference to the malaria parasites. Annals of Tropical Medicine, 30, 435-439. Martin I, L. B., Han, P., Lewites, J., Kuhlman, J. R., Klasing, K.C. & Wikelski, M. 2006. Phytohemagglutinin-induced skin sweling in birds: histological support for a clasic imunoecological technique. Functional Ecology, 20, 290-299. Martin, L. B., Scheuerlein, A. & Wikelski, M. 2003. Imune activity elevates energy expenditure of house sparows: a link betwen direct and indirect costs? Procedings of the Royal Society of London B, 270, 153-158. Martin, T. E., Moller, A. P., Merino, S. & Clobert, J. 2001. Does clutch size evolve in response to parasites and imunocompetence? Procedings of the National Academy of Sciences U. S. A., 98, 2071-2076. Matson, K. D. 2006. Are there diferences in imune function betwen continental and insular birds? Procedings of the Royal Society London B, 273, 2267- 2274. Matson, K. D., Ricklefs, R. E. & Klasing, K. C. 2005. A hemolysis-hemagglutination asay for characterizing constitutive innate humoral imunity in wild and domestic birds. Developmental and Comparative Imunology, 29, 275-286. May, R. M. & Anderson, R. M. 1983. Epidemiology and genetics in the coevolution of parasites and hosts. Procedings of the Royal Society London B, 219, 281- 313. McConkey, G. A., Li, J., Rogers, M. J., Seley Jr., D. C., Graczyk, T. K., Cranfield, M. R. & McCutchan, T. F. 1996. Parasite diversity in an endemic region for avian malaria and identification of a parasite causing penguin mortality. Journal of Eukaryotic Microbiology, 43, 393-399. Mendes, L., Piersma, T., Lecoq, M., Spaans, B. & Ricklefs, R. E. 2005. Disease- limited distributions? Contrasts in the prevalence of avian malaria in shorebird species using marine and freshwater habitats. Oikos, 109, 396-404. Merino, S., Moreno, J., Sanz, J. J. & Ariero, E. 2000. Are avian blood parasites pathogenic in the wild? A medication experiment in blue tits (Parus caeruleus). Procedings of the Royal Society London B, 267, 2507-2510. 137 Milder, S. L. & Schreiber, R. W. 1982. Notes on the nesting behavior of Acrocephalus aequinoctialis. Bulletin of the British Ornithological Council,102, 20-22. Milien, V. 2006. Morphological evolution is acelerated among island mamals. PLoS Biology, 10, 1863-1868. Moller, A. P. & Saino, N. 2004. Imune response and survival. Oikos, 104, 299-304. Nei, M., Maruyama, T. & Chakraborty, R. 1975. The bottleneck efect and genetic variability in populations. Evolution, 29, 1-10. Nei, M. & Li, W.-H. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Procedings of the National Academy of Sciences U. S. A., 76, 5269-5273. Nordling, D., Anderson, M., Zohari, S. & Gustafson, L. 1998. Reproductive efort reduces specific imune respone and parasite resistance. Procedings of the Royal Society London B, 265, 1291-1298. O?Brien, S. J., Roelke, M. E., Marker, L., Newman, A., Winkler, C. A., Meltzer, D., Colly, L., Evermann, J. F., Bush, M. & Wildt, D. E. 1985. Genetic basis for species vulnerability in the cheetah. Science, 227, 1428-1434. Ochsenbein, A. F., Fehr, T., Lutz, C., Suter, M., Brombacher, F., Hengartner, H. & Zinkernagel, R. M. 1999. Control of early viral and bacterial distribution and disease by natural antibodies. Science, 286, 2156-2159. Olson, S. L., Wingate, D. B., Hearty, P. J. & Grady, F. V. 2005. Prodromus of vertebrate paleontology and geochronology of Bermuda. In Insular Vertebrate Evolution: the Palaeontological Approach, J. Alcover & P. Bover (eds.). Procedings of the International Symposium ?Insular Vertebrate Evolution: the Paleontological Approach.? Monografies de la Societat d?Historia Natural de les Balears, 12, 219-232. Parker, P. G., Whiteman, N. K. & Miler, R. E. 2006. Conservation medicine on the Galapagos Islands: partnerships among behavioral, population, and veterinary scientists. Auk, 123, 625-638. Parmentier, H. K., Lamers, A., Hoekman, J. J., Reilingh, G. de V., Zanen, I. T. A. & Savelkoul, H. F. J. 2004. Diferent levels of natural antibodies in chickens divergently selected for specific antibody responses. Developmental and Comparative Imunology, 28, 39-49. Pearce-Duvet, J. M. C. 2006. The origin of human pathogens: evaluating the role of 138 agriculture and domestic animals in the evolution of human disease. Biological Reviews, 81, 369-382. Pearman, P. B. & Garner, T. W. J. 2005. Susceptibility of Italian agile frog populations to an emerging strain of Ranavirus paralels population genetic diversity. Ecology Leters, 8, 401-408. Peirce, M. A. 1979. Some additional observations on haematozoa of birds in the ascarene Islands. Bulletin of the British Ornithological Council, 99, 68-71. Perkins, S. L. 2000. Species concepts and malaria parasites: detecting a cryptic species of Plasmodium. Procedings of the Royal Society London B, 267, 2345-2350. Perkins, S. L. & Schal, J. J. 2002. A molecular phylogeny of malarial parasites recovered from cytochrome b gene sequences. Journal of Parasitology, 88, 972-978. Peters, A., Delhey, K., Denk, A. G. & Kempenaers, B. 2004. Trade-ofs betwen imune investment and sexual signaling in male malards. American Naturalist, 164, 51-59. Posada, D. & Crandal, K. A. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics, 14, 817-818. Price, P. 1991. In Host-parasite interactions, J. E. Loye & M. Zuk (eds.). Oxford University Pres, New York. Reid, J. M, Arcese, P. & Keler, L. F. 2003. Inbreding depreses imune response in song sparows (Melospiza melodia): direct and inter-generational efects. Procedings of the Royal Society London B, 270, 2151-2157. Richard, F. A., Sehgal, R. N. M., Jones, H. I. & Smith, T. B. 2002. A comparative analysis of PCR-based detection methods for avian malaria. Journal of Parasitology, 88, 819-822. Richardson, D. S., Jury, F. L., Dawson, D. A., Salgueiro, P., Komdeur, J. & Burke, T. 2000. Fifty Seycheles warbler (Acrocephalus sechelensis) microsatelite loci polymorphic in Sylviidae species and their cross-species amplification in other paserine birds. Molecular Ecology, 9, 2155-2234. Richardson, D. S. & Westerdahl, H. 2003. MHC diversity in two Acrocephalus species: the outbred Great red warbler and the inbred Seycheles warbler. Molecular Ecology, 12, 3523-3529. Ricklefs, R. E. & Falon, S. M. 2002. Diversification and host switching in avian 139 malaria parasites. Procedings of the Royal Society London B, 269: 885-892. Ricklefs, R. E., Falon, S. M. & Bermingham, E. 2004 Evolutionary relationships, cospeciation, and host switching in avian malaria parasites. Systematic Biology, 53, 111-119. Roelke, M. E., Martenson, J. S. & O?Brien, S. J. 1993. The consequences of demographic reduction and genetic depletion in the endangered Florida panther. Current Biology, 3, 340-350. Ronquist, F. & Huelsenbeck, J. P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572-1574. Scheuerlein, A. & Ricklefs, R. E. 2004. Prevalence of blood parasites in European paseriform birds. Procedings of the Royal Society, London B, 271, 1363- 1370. Schloegel, L. M., Hero, J.-M., Berger, L. Speare, R., McDonald, K. & Daszak, P. 2006. The decline of the sharp-snouted day frog (Taudactylus acutirostris): the first documented case of extinction by infection in a fre-ranging wildlife species? Ecohealth, 3, 35-40. Schmid-Hempel, P. & Ebert, D. 2003. On the evolutionary ecology of specific imune defence. Trends in Ecology and Evolution, 18, 27-32. Schrenzel, M. D., Malouf, G. A., Kener, L. L. & Gafney, P. M. 2003. Molecular characterization of malarial parasites in captive paserine birds. Journal of Parasitology, 89, 1025-1033. Scrimshaw, N. S. 1991. Efect of infection on nutrient requirements. Journal of Parenteral and Enteral Nutrition, 15, 589-600. Sefc, K. M., Payne, R. B. & Sorenson, M. D. 2001. Characterization of microsatelite loci in vilage indigobirds Vidua chalybeata and cross-species amplification in estrildid and ploceid finches. Molecular Ecology Notes, 1, 252-254. Sibley, C.G. & Ahlquist, J. E. 1985. The phylogeny and clasification of the Australo- Papuan paserine birds. Emu, 85, 1-14. Sikamaki, P., Rati, O., Hovi, M. & Bennet, G. F. 1997 Asociation betwen haematozoan infections and reproduction in the Pied Flycatcher. Functional Ecology, 11, 176-183. Smith, K. F., Sax, D. F. & Laferty, K. D. 2006. Evidence for the role of infectious disease in species extinction and endangerment. Conservation Biology, 20, 1349-1357. 140 Smits, J. E., Bortolotti, G. R. & Tela, J. L. 1999. Simplifying the phytohaemagglutinin skin-testing technique in studies of avian imunocompetence. Functional Ecology, 13, 567-572. Smits, J. E., Tela, J. L., Carete, M., Serano, D. & Lopez, G. 2005. An epizootic of avian pox in endemic short-toed larks (Calandrela rufescens) and Berthelot?s pipits (Anthus berthelotti) in the Canary Islands, Spain. Veterinary Pathology, 42, 59-65. Speilman, D., Brook, B. W., Briscoe, D. A. & Frankham, R. 2004. Does inbreding and loss of genetic diversity decrease disease resistance? Conservation Genetics, 5, 439-448. Steadman, D. 1997. The historic biogeography and community ecology of Polynesian pigeons and doves. Journal of Biogeography, 24, 737-753. Swofford, D. L. 1999 PAUP*: Phylogenetic analysis using parsimony (*and other methods). Sinauer, Sunderland, MA. Tar, C. L., Conant, S. & Fleischer, R. C. 1998. Founder events and variation at microsatelite loci in an insular paserine bird, the Laysan finch (Telespiza cantans). Molecular Ecology, 7, 719:731. Taylor, L. H., Latham, S. M. & Woolhouse, M. E. 2001. Risk factors for human disease emergence. Philosophical Transactions of the Royal Society London B, 356, 983-989. Thompson, J. R., Marcelino, L. A. & Polz, M. F. 2002. Heteroduplexes in mixed- template amplifications: formation, consequence and elimination by ?reconditioning PCR?. Nucleic Acids Research, 30, 2083-2088. Toft, C. A. 1991 Current theory of host-parasite interactions. In Host-parasite interactions, J. E. Loye & M. Zuk (eds.). Oxford University Pres, New York. Tompkins, D. M. & Gleson, D. M. 2006. Relationship betwen avian malaria distribution and an exotic invasive mosquito in New Zealand. Journal of the Royal Society of New Zealand, 36, 51-62. Torchin, M. E., Laferty, K. D., Dobson, A. P., McKenzie, V. J. & Kuris, A. M. 2003 Introduced species and their mising parasites. Nature, 421, 628-630. Valkiunas, G. 2005. Avian malaria parasites and other haemosporidia. CRC Pres, Boca Raton. Valkiunas, G., Bensch, S., Iezhova, T. A., Krizanauskiene, A., Helgren, O. & 141 Bolshakov, C. V. 2006. Nested cytochrome B polymerase chain reaction diagnostics underestimate mixed infections of avian blood haemosporidian parasites: Microscopy is stil esential. Journal of Parasitology, 92, 418-422. van Riper II, C. 1991. The impact of introduced vectors and avian malaria on insular paseriform bird populations in Hawai. Bul. Soc. Vect. Ecol., 16, 59-83. van Riper, C., S. G. van Riper, M. L. Goff & Laird, M. 1986. The epizootiology and ecological significance of malaria on the birds of Hawai. Ecological Monographs, 56, 327-344. Viney, M. E., Riley, E. M. & Buchanan, K. L. 2005. Optimal imune responses: imunocompetence revisited. Trends in Ecology and Evolution, 20, 665-669. Waldenstr?m, J., Bensch, S., Haselquist, D. & Ostman, O. 2004. A new nested polymerase chain reaction method very eficient in detecting Plasmodium and Haemoproteus infections from avian blood. Journal of Parasitology, 90, 191- 194. Waldenstr?m, J., Bensch, S., Kiboi, S., Haselquist, D. & Otosson, U. 2002. Cross- species infection of blood parasites betwen resident and migratory songbirds in Africa. Molecular Ecology, 11, 1545-1554. Warner, R. E. 1968. The role of introduced diseases in the extinction of the endemic Hawaian avifauna. The Condor, 70, 101-120. Webster, M. S., Mara, P. P., Haig, S. M., Bensch, S. & Holmes, R. T. 2002. Links betwen worlds: unraveling migratory connectivity. Trends in Ecology and Evolution, 17, 76-83. White, E. M., Greiner, E. C., Bennet, G. F. & Herman, C. M. 1978. Distribution of hematozoa of Neotropical birds. Revista de Biologia Tropical, 26, 43-102. Whiteman, N. K., Matson, K. D., Bollmer, J. L. & Parker, P. G. 2006. Disease ecology in the Galapagos Hawk (Buteo galapagoensis): host genetic diversity, parasite load and natural antibodies. Procedings of the Royal Society London B, 273, 797-804. Wiggins, D. A., Moller, A. P., Sorensen, M. F. L. & Brand, L. A. 1998. Island biogeography and the reproductive ecology of great tits Parus major. Oecologia, 115, 478-482. Wikelski, M., Foufopoulos, J., Vargas, H. & Snel, H. 2004. Galapagos birds and diseases: invasive pathogens as threats for island species. Ecology and Society, 9, [online]. Available at ww.ecologyandsociety.org/vol19/is1/art5. 142 Wilkinson, H. 1950. Spanish intentions for Bermuda 1603-1615. Bermuda Historical Quarterly, 7, 50-89. Woodworth, B., Atkinson, C. T., LaPointe, D. A., Hart, P. J., Spiegel, C. S., Twed, E. J., Henneman, C., LeBrun, J., Denete, T., DeMots, R., Kozar, K. L., Triglia, D., Lease, D., Gregor, A., Smith, T. & Duffy, D. 2005. Host population persistence in the face of introduced vector-borne diseases: Hawai amakihi and avian malaria. Procedings of the National Academy of Sciences U. S. A., 102, 1531-1536. Woolhouse, M. E. J. & Gowtage-Sequeria, S. 2005. Host range and emerging and remerging pathogens. Emerging Infectious Disease, 11, 1842-1847. Wright, S. 1931. Evolution in Mendelian populations. Genetics, 16, 97-159. Yorinks, N. & Atkinson, C. T. 2000. Efects of malaria (Plasmodium relictum) on activity budgets of experimentaly infected juvenile Apapane (Himatione sanguinea). Auk, 117, 731-738. Young, B. E., Garvin, M. C. & McDonald, D. B. 1993. Blood parasites in birds from Monteverde, Costa Rica. Journal of Wildlife Diseases, 29, 555-560. Zuk, M. & Stoehr, A. M. 2002. Imune defense and host life history. The American Naturalist, 160, S9-S22.