Accepted Manuscript Sexual differences in prevalence of a new species of trypanosome infecting túngara frogs Ximena E. Bernal, C.Miguel Pinto PII: S2213-2244(16)30005-0 DOI: 10.1016/j.ijppaw.2016.01.005 Reference: IJPPAW 147 To appear in: International Journal for Parasitology: Parasites and Wildlife Received Date: 24 August 2015 Revised Date: 25 December 2015 Accepted Date: 17 January 2016 Please cite this article as: Bernal, X.E, Pinto, C.M., Sexual differences in prevalence of a new species of trypanosome infecting túngara frogs, International Journal for Parasitology: Parasites and Wildlife (2016), doi: 10.1016/j.ijppaw.2016.01.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Graphical abstract M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Sexual differences in prevalence of a new species of trypanosome infecting túngara frogs 1 2 Ximena E Bernal1,2, C Miguel Pinto3,4,5 3 1 Department of Biological Sciences, Purdue University, West Lafayette, Indiana, 47907, USA 4 2Smithsonian Tropical Research Institute, Apartado 2072, Balboa, Republic of Panama 5 3Sackler Institute for Comparative Genomics and Department of Mammalogy, American 6 Museum of Natural History, New York, NY 10024, USA 7 4Centro de Investigación en Enfermedades Infecciosas, Pontificia Universidad Católica del 8 Ecuador, Quito, Ecuador 9 5Division of Mammals, National Museum of Natural History, Smithsonian Institution, 10 Washington, DC 20004, USA 11 12 Email addresses: 13 XEB: xbernal@purdue.edu 14 CMP: pintom@si.edu 15 16 17 18 19 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Abstract 20 Trypanosomes are a diverse group of protozoan parasites of vertebrates transmitted by a variety 21 of hematophagous invertebrate vectors. Anuran trypanosomes and their vectors have received 22 relatively little attention even though these parasites have been reported from frog and toad 23 species worldwide. Blood samples collected from túngara frogs (Engystomops pustulosus), a 24 Neotropical anuran species heavily preyed upon by eavesdropping frog-biting midges 25 (Corethrella spp.), were examined for trypanosomes. Our results revealed sexual differences in 26 trypanosome prevalence with female frogs being rarely infected (<1%). This finding suggests 27 this protozoan parasite may be transmitted by frog-biting midges that find their host using the 28 mating calls produced by male frogs. Following previous anuran trypanosome studies, we 29 examined 18S ribosomal RNA gene to characterize and establish the phylogenetic relationship of 30 the trypanosome species found in túngara frogs. A new species of giant trypanosome, 31 Trypanosoma tungarae n. sp., is described in this study. Overall the morphometric data revealed 32 that the trypomastigotes of T. tungarae n. sp. are similar to other giant trypanosomes such as T. 33 rotatorium and T. ranarum. Despite its slender and long cell shape, however, 18S rRNA gene 34 sequences revealed that T. tungarae n. sp. is sister to the rounded-bodied giant trypanosome, T. 35 chattoni. Therefore, morphological convergence explains similar morphology among members 36 of two non-closely related groups of trypanosomes infecting frogs. The results from this study 37 underscore the value of coupling morphological identification with molecular characterization of 38 anuran trypanosomes. 39 40 41 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Keywords: Engystomops pustulosus, Corethrella, frog-biting midges, Panamá, Physalaemus, 42 species delimitation, Trypanosome phylogeny 43 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT 1. Introduction 44 Trypanosomes are protozoan parasites that are ubiquitous across invertebrate and vertebrate 45 species. Indeed, trypanosomes infect species across all vertebrate classes. Anuran trypanosomes, 46 however, have received considerably less attention than those in other vertebrates even though 47 they infect frog and toad species worldwide (Bardsley and Harmsen, 1973; Desser and Yekutiel, 48 1986; Werner, 1993; Desser, 2001; Žičkus, 2002; Lemos et al. 2008). Since many anurans spend 49 at least their early developmental stages in aquatic environments and return to breed as adults, 50 leeches have long been considered the main vectors of trypanosomes in this group (Reilly and 51 Woo, 1982). As adults, however, many species of frogs are preyed upon by a variety of 52 opportunistic and specialized hematophagous insects that may act as possible vectors of blood 53 parasites. Phlebotomine sandflies (Phlebotomus squamirostris), for instance, transmit 54 Trypanosoma bocagei França 1911 to European toads, Bufo bufo (Feng and Chung, 1940). 55 Similarly, trypanosomes may be mosquito-borne parasites for anurans. Mosquitoes, such as 56 Culex territans, that feed mainly on anuran hosts have been implicated in the transmission of T. 57 ranarum Lankester 1871 (Desser et al., 1973) but their role as trypanosome vectors is still 58 controversial (Ferguson and Smith, 2012). Other mosquito species such as Aedes aegypti and 59 Culex pipiens can transmit trypanosomes (T. rotatorium Mayer 1843 complex) to some frogs 60 even though they do not usually feed on anurans (Ramos and Urdaneta-Morales 1977). Closely 61 related to mosquitoes, frog-biting midges (Corethrellidae) are small hematophagous flies 62 specialized at feeding on anurans (Borkent, 2008). These midges are thus potentially important 63 vectors of blood parasites of this vertebrate clade. In fact, in the Southeastern United States, one 64 species of frog-biting midge (Corethrella wirthi) transmits trypanosomes to green treefrogs, 65 Hyla cinerea (Johnson et al., 1993). The family Corethrellidae contains 107 species of frog-66 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT biting midges, in which females are specialized in using the mating call of frogs to localize them 67 and obtain a blood meal (Borkent, 2014). The frog’s mating call is the main cue used by the 68 midges for long-distance host detection (Bernal and de Silva, 2015). Further studies that examine 69 the role of other species of frog-biting midges at transmitting trypanosomes are necessary to 70 understand the evolutionary ecology of these interactions. In this study we investigate 71 trypanosome infection in a Neotropical anuran species, the túngara frog (Engystomops 72 pustulosus), which is heavily preyed upon by frog-biting midges. 73 Túngara frogs are small anurans that occur from southern Mexico to northern South 74 America (Colombia, Venezuela, and Belize) and Trinidad and Tobago. Males aggregate during 75 the rainy season at temporary puddles from where they produce mating calls (Ryan, 1985). 76 While calling to attract a mate, túngara frog males also attract frog-biting midges (Corethrella 77 spp). These eavesdroppers prey upon túngara frogs in great numbers (Figure 1a). A speaker 78 broadcasting calls equivalent to those produced by a motivate túngara frog male, attracts up to 79 511 midges in 30min (average=142 midges/30min; Bernal et al., 2006). Túngara frogs represent 80 an ideal opportunity to investigate trypanosome infection potentially transmitted by frog-biting 81 midges. 82 The goals of this study were twofold: firstly, to determine the presence of trypanosomes 83 in túngara frogs along with the characterizing of these parasites, and secondly, to examine 84 whether the prevalence differs between females and males. Since as in most anuran species, 85 túngara frog females do not produce mating calls (Ryan, 1985), eavesdropping frog-biting 86 midges most likely only feed on male frogs. We thus expected differences in trypanosome 87 prevalence between male and female túngara frogs reflecting the feeding habits of the frog-biting 88 midges. As predicted, we found trypanosome infected male túngara frogs and thus implemented 89 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT morphological and molecular methods to characterize and infer the phylogenetic relationship of 90 this Trypanosoma species to other trypanosomes that parasitize other vertebrates that inhabit 91 aquatic and marine environments. The characterization and phylogenetic relationships of this 92 new Trypanosoma species provide new information on anuran trypanosomes, a group with 93 poorly known taxonomic relationships (Martin et al., 2002). In addition, we provide insights 94 about the prevalence of this trypanosome species on its type host. 95 2. Materials and methods 96 2.1. Study site and sample collection 97 Túngara frogs were captured at their breeding areas during the rainy season around the 98 Smithsonian Tropical Research Facilities in Gamboa (9º 79’ N, 79 º 42.9’ W), Panama. 99 Individuals were brought to the laboratory where they were measured and blood samples were 100 collected by toe-clipping as well as via the orbital sinus following Lynch et al. (2006). After 101 collecting blood samples, the frogs were placed in individual containers with sufficient amounts 102 of water and released within 24hrs at the exact location where they were captured. This 103 procedure was approved by the Smithsonian Tropical Research Institute IACUC (#2009-0616-104 2012-11). To examine the presence of trypanosomes in túngara frogs and test our prediction of 105 sexual differences in infection, we collected 25 calling males and 15 females approaching the 106 puddle or in amplexus. We performed 2-5 blood smears per individual to include both thin and 107 thick smears for each frog, for a total of 112 blood smears (2.8 blood smears/individual). Given 108 that some trypanosomes in anurans are known to have nocturnal peripheral parasitemia, bleeding 109 of all túngara frogs was performed between 2000-0100hrs when trypanosome parasitemia is 110 higher in other anuran species (Johnson et al., 1993). Túngara frogs are not preyed upon by other 111 biting insects and liver-baited traps at the small temporary pools in which they breed revealed 112 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT that leeches are absent (N= 5 nights, two traps a night). In addition, usually when leeches feed on 113 amphibian hosts they leave distinct hematomas on their skin. Careful inspection of túngara frogs 114 did not revealed signs of skin lesions such as those that result from leeches (McCallum et al., 115 2011; Rhoden and Bolek, 2012). 116 To characterize the trypanosome species using molecular techniques, additional blood 117 samples were collected from individuals that had been confirmed to be infected with the 118 trypanosome species described here using microscopy. Those samples were stored in lysis buffer 119 and preserved at 4°C for molecular analysis (Innis et al., 1990; Longmire et al., 1997). Some 120 frogs were kept in captivity for longer periods to conduct behavioral experiments as part of an 121 additional study. 122 2.2. Morphological characterization 123 After performing the blood smears, the slides were air dried, fixed with absolute methanol and 124 later stained using Giemsa stain following Mohr (1981). Blood smears were thoroughly 125 screened, covering the entire smear at 400X magnification (1-3 hrs per slide) using a Nikon 126 Eclipse E 200 (Nikon, Chiyoda, Tokyo, Japan) microscope. Once trypomastigotes were found, 127 they were photographed at 400X and 1000X magnification using a Nikon high-definition color 128 camera head DS-Fi2 and the images were transferred onto a computer screen via a Nikon 129 Camera Control unit DS–L2. We measured trypomastigote morphology (total body length and 130 maximum width, N = 39) with Nikon's NIS-Elements D research application. Given the dark and 131 uniform coloration of the stained trypomastigotes, other morphological characters could not be 132 measured in a reliable way for any of the specimens. Additional blood samples from ten 133 individuals were collected and blood smears prepared and stained using Hemacolor® Giemsa 134 stain kit (Merck KGaA, Darmstadt, Hesse, Germany) in an attempt to obtain images revealing 135 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT kinetoplastic morphology. Both stain techniques, however, had limited success revealing the 136 kinetoplast and nucleus in the stained trypomastigote. Therefore, we could only make 137 morphological measures of the internal structure in a subset of the specimens (N = 14). 138 Measurements are given as the mean ± standard deviation in micrometers. 139 All blood smears were labeled and arranged in such a way to prevent biased screening of 140 the slides. Statistical analysis was performed on the proportion of individuals infected across 141 each sex, using a two-tailed Z-test for population proportion implemented through STATA 10 142 (StataCorp, College Station, TX, USA) (StataCorp, 2007). 143 2.3. Phylogenetic relationships 144 We extracted DNA directly from blood samples using DNeasy kits (Qiagen, Valencia, CA, 145 USA) following the manufacturer’s recommendations. Following Martin et al. (2002), we 146 examined 18S ribosomal RNA gene (18S rRNA). We amplified by PCR two overlapping 147 fragment of 18S rRNA with newly designed primers from an alignment of frog trypanosomes. 148 The first fragment—955 bp—was amplified with primers SSU1_F 149 (TCTGGTTGATTCTGCCAGTAG) and SSU1_R (AAAACCAACAAAAGCCGAAA); the 150 second fragment—980 bp—was amplified with primers SSU2_F 151 (CCAAAGCAGTCATCCGACTT) and SSU2_R (AGGAGCATCACAGACCTGCT). These 152 primers were designed from a large alignment of trypanosome species (Hamilton et al., 2007); 153 these primer sequences are highly conserved among trypanosomes, likely are able to amplify 154 multiple species of anuran trypanosomes. Both PCR amplifications were conducted with a 155 touchdown PCR profile (Murphy and O’Brien, 2007). After cleaning the PCR product with 156 ExoSAP-IT (Affymetrix, Santa Clara, CA, USA), we completed sequencing reactions in both 157 directions with the ABI BigDye chemistry (Applied Biosystems, Inc., Foster City, CA, USA), 158 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT and sequenced the fragments on an ABI 3730xl DNA Analyzer automatic sequencer (Applied 159 Biosystems, Inc., Foster City, CA, USA). We assembled contigs with the obtained sequence 160 chromatograms in Geneious 6.1.6 (Biomatters, Auckland, New Zealand), resulting in sequences 161 of 1688 bp for male #165504 (GenBank accession number: KM406915) and 1689 bp for male 162 #165507 (GenBank accession number: KM406916). 163 We built an 18S rRNA gene matrix with the newly generated data and previously 164 published sequences of members of the aquatic clade of Trypanosoma, using T. avium 165 Danilewsky 1885, T. lewisi Kent 1880 and T. theileri Laveran 1902 as outgroups (Martin et al., 166 2002; Ferreira et al., 2007; Ferreira et al., 2008; Hayes et al., 2014). We aligned the sequences 167 using the MUSCLE (Edgar, 2004) plugin within Geneious 6.1.6, and edited manually obvious 168 misplacements and removed suspicious ends of sequences (i.e., ends with abundant substitutions 169 while the remaining of the alignment is conserved). The aligned matrix comprised 67 terminals 170 and a length of 2,364 bp. We ran Bayesian and maximum likelihood analyses with a single 171 partition with the model GTR+Γ in MrBayes 3.1.2 (Ronquist and Huelsenbeck, 2003) and 172 RAxML 8.0.12 (Stamatakis, 2014) respectively. For the Bayesian analysis we did two 173 independent runs with 1 cold and 3 heated chains, with sampling the chains every 100 174 generations. The analysis was allowed to run until reaching estationarity—stopval set at 0.01, 175 and later confirmed by the potential scale reduction factor values close to 1—which occurred at 176 1,185,000 generations, and 10% of generated trees were discarded as burn in. Nodal support was 177 estimated with posterior probabilities. For the Maximum Likelihood we estimated the nodal 178 support with 1,000 bootstrap pseudo replicates. 179 As an additional confirmation of the species status of this new trypanosome we ran a 180 coalescent-based species delimitation analysis using Poisson tree processes (PTP) model (Zhang 181 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT et al., 2013). New probabilistic approaches for species delimitation provide alternatives to using 182 arbitrary genetic thresholds and arbitrary monophyletic groupings. In particular, the PTP analysis 183 is a fast species delimitation approach that attempts to identify putative species using a single 184 input phylogenetic tree—usually built with a single locus by modeling speciation rates directly 185 from the number of substitutions. We run the analysis in the bPTP web server with our 186 maximum likelihood tree, using 500,000 Markov chain Monte Carlo generations, a thinning of 187 100 and a burn-in of 0.25. 188 3. Results 189 3.1. Species description 190 The trypanosomes observed in the blood smears have a unique set of morphological characters 191 that differentiate them from previously described species. Morphology, however, often does not 192 allow researchers to distinguish trypanosomes species and is problematic for determining species 193 relationships. We obtained DNA sequences that revealed this lineage constitute a new species of 194 trypanosome that we describe below. 195 Taxonomic summary: Phylum Euglenozoa, Cavalier-Smith, 1981; class Kinetoplastea, 196 Honigberg, 1963; order Trypanosomatida, (Kent, 1880) Hollande, 1952; family 197 Trypanosomatidae, Doflein, 1951. Trypanosoma tungarae n. sp. Bernal and Pinto (201x) 198 Type material: type blood smears of three infected frogs are deposited in the Smithsonian 199 National Museum of Natural History (USNM Numbers TBD). Type Host: Vertebrate host is the 200 túngara frogs Engystomops pustulosus (Amphibia: Leuperidae); putative vectors are Corethrella 201 spp. midges (Diptera: Corethrellidae). Type Locality: Panamá, Colon Province, Gamboa (30 202 m.a.s.l., 9º 79’ N, 79 º 42.9’ W) (Figure 2). Location on hosts: In the vertebrate hosts peripheral 203 blood . The location in their putative vector frog-biting midges is unknown (possibilities include 204 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT the digestive tract, the hemocele and the salivary glands). Distribution: Currently known only 205 from the type locality, Gamboa, Panama. Diagnosis: Monomorphic trypanosome with an 206 elongated body (52.13 ± 12.94 µm) and thin soma (5.41 ± 3.62 µm). Free flagellar length (FF), 207 13.20 ± 5.11 µm; midnucleus to anterior end (MA), 42.71 ± 13.77µm; midnucleus to posterior 208 end (MP), 29.67 ± 10.59 µm; midnucleus to kinetoplast, 20.31 ± 7.41 µm; posterior end to 209 kinetoplast, 9.71 ± 3.50 µm; relative size of flagellum (FF/MA), 0.34 ± 0.14 µm; length of 210 nucleus, 3.63 ± 1.67 µm; nuclear index (MP/MA), 0.97 ± 0.60 µm. In general, this species 211 resembles other anuran trypanosomes from Central and South America (Desser, 2001; Ferreira et 212 al., 2007; McKenzie and Starks 2008). This species is longer and thinner that T. rotatorium – like 213 species found in other leptodactilyd anuran host in South America (Lemos et al. 2008). In 214 particular, this species corresponds to the morphology of anuran trypanosomes with elongated 215 trypomastigotes with pointed ends observed in Bufonidae, Leiuperidae and Leptodactylidae from 216 Brazil (Group I, Ferreira et al., 2007). The morphology of this species, however, is most similar 217 in general to Trypanosoma sp. (e) and Trypanosoma sp. (f) described from Lithobates vaillanti 218 syn. Rana vaillanti by Desser (2001). Although the measurements of the species described here 219 match closely some characteristics of rypanosoma sp. (e) such as the relative length of the free 220 flagellum, other features, including total body length and the distance from the posterior end to 221 the kinetoplast, are closer to the morphology of Trypanosoma sp. (f). Some other features, 222 however, are distinct from both Trypanosoma sp. (e) and (f) (e.g. distance from the center of the 223 nucleus to the anterior end). A T. montrealis-like species was found to be transmitted by North 224 American frog-biting midges (C. wirthi) in Florida (Johnson et al. 1993). Although the body 225 length and width of Trypanosoma montrealis (Fantham et al. 1942) fall within the dimensions of 226 the species described here, that previously described species has a much shorter free flagellum 227 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT than T. tungarae n. sp (3-5.5 µm vs 13.20 ± 5.11 µm). The validity of T. montrealis, however, 228 has been questioned (Werner et al. 1988). More detailed morphological comparisons with 229 previously described species of anuran trypanosomes from the same geographical area are 230 unfeasible given that detailed morphological measurements are not often reported and recent, 231 updated species descriptions frequently focus on the species genotypes (e.g Ferreira et al. 2007). 232 Intraspecific morphological variation of amphibian trypanosomes, however, is so high that 233 precludes its use for species identification. For example, amphibian trypanosomes can 234 significantly change their morphotype when infecting different hosts (Hysek 1976). 235 This species does not resemble in morphology T. chattoni, the closest related species 236 known to date (see under Phylogenetic relationships below), that has a characteristic round to 237 oval body (Lemos et al 2008). Trypomastigotes of both species, however, have large size and 238 this new species thus becomes a new member of the giant trypanosomes that includes species 239 such as T. mega, T. ranarum and T. rotatorium (Martin et al., 2002). Despite the widespread 240 distribution of T. chattoni including Asia (China, Werner, 1993; Kyushu and Ryukyu Islands, 241 Miyata, 1978; Thailand, Sailasuta et al., 2011), North America (United Sates, Diamond, 1965; 242 Canada, Jones and Woo, 1986) and South America (Brazil, Lemos et al., 2008), this species is 243 monomorphic with little geographic variation. Both T. chattoni and T. tungarae n. sp. have 244 heavily stained cytoplasms that often obscure the nucleus and kinetoplast. When visible, the 245 kinetoplast lays towards the anterior end at about a fourth of the total length of the cell. Glass 246 slides of Giemsa-stained smears from túngara frog blood samples and DNA samples are kept at 247 the Smithsonian National Museum of Natural History, Washington, DC. To comply with the 248 regulations of the International Code of Zoological Nomenclature (ICZN), details of this species 249 have been submitted to ZooBank with the Life Science Identifier (LSID) zoobank.org:pub:TBD. 250 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Etymology: Túngara (English pronunciation: toon-gah-rra) is the common name of the frog 251 Engystomops pustulosus, the vertebrate host of this new species of trypanosome. Túngara is a 252 feminine Spanish onomatopoeic word resembling part of the singing repertoire of the 253 Engystomops pustulosus males. We treat tungarae as a feminine noun in the genitive case. 254 3.2. Host prevalence: Consistent with our prediction, we found sexual differences in 255 trypanosomes infection in túngara frogs (Z-test, Z=2.28, p=0.022).While 40% of male túngara 256 frogs sampled were infected with this blood parasite, only 6.6% of the females were infected 257 (males: 10/25; females: 1/15). We were, however, expecting that no females would be infected 258 since female túngara frogs do not vocalize. Frog-biting midges are attracted to the mating calls 259 produced by males (Bernal et al. 2006; Borkent, 2008; McKeever and Hartberg, 1980), so our 260 results beg the question, if frog- biting midges are the vectors, how did a female become infected 261 with this new species of trypanosomes? Careful inspections of our records confirmed this result 262 and field observations revealed a potential path of transmission for female frogs to be infected. 263 When túngara frog are in amplexus, frog-biting midges attempting to feed on the calling male 264 have an opportunity to move directly from their original victim, the male, to the female and 265 obtain a blood meal (Figure 1b,c). 266 3.3. Phylogenetic relationships: The maximum likelihood and the Bayesian inference 267 phylogenies of the 18S rRNA gene are highly concordant, and show strong support for the 268 placement of the new species, Trypanosoma tungarae, in the clade with aquatic trypanosomes; 269 however, several internal branches are poorly supported for both methods. Trypanosoma 270 tungarae n. sp. is sister to T. chattoni, and both form a highly supported clade sister to other 271 trypanosomes of South American frogs (Fig. 4). 272 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Both, the maximum likelihood and Bayesian solutions of the putative species delimitation 273 analysis in PTP indicate that T. tungarae n. sp. is a different species from other trypanosomes for 274 which molecular data is available. Also, the PTP analyses indicate that it might be some over 275 splitting of species in fish trypanosomes, and several unrecognized species of frog trypanosomes 276 (Fig. 4). The two sequences of T. tungarae n. sp. diverge in eight nucleotides, and it is likely that 277 additional genetic variation can be found within the study area. Despite that the 18S rRNA gene 278 is a slowly evolving marker, the variation that we found is not surprising given the complex 279 patterns of intra and inter specific trypanosome diversity found in this geographic region (Pinto 280 et al. 2012; Cottontail et al. 2014). 281 4. Discussion 282 Our results revealed that while male túngara frogs are frequently infected with trypanosomes, 283 females rarely carry these parasites. Since females do not vocalize, they do not attract frog-biting 284 midges (Bernal et al., 2006) and are thus rarely in contact with this putative vector. Similarly, 285 sexual differences in prevalence of trypanosomes in green treefrogs, Hyla cinerea, were reported 286 in the Southeastern United States where frog-biting midges (Corethrella wirthi) were implicated 287 as vectors of this parasite (Johnson et al., 1993). Transmission of T. tungarae n. sp. by vectors 288 other than frog-biting midges seems unlikely. Leeches, common vectors of trypanosomes of the 289 aquatic clade (Hamilton et al., 2007), are absent from the breeding puddles of túngara frogs in 290 the study population. Although we collected leeches at our study site in larger ponds where other 291 anurans breed, no leeches were found using the same traps in the puddles where túngara frogs 292 breed. During the time we have spent observing túngara frogs in the field and collecting insects 293 biting them (>100 hrs), no other blood-sucking insects or lesions potentially inflicted by leeches 294 have been detected. The high numbers of frog-biting midges that bite túngara frogs (Bernal et al., 295 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT 2006), combined with their ability to transmit this parasite to other frogs (Johnson et al., 1993), 296 strongly suggest frog-biting midges may be the main vectors of T. tungarae n. sp. The 297 advertisement call of túngara frogs attracts at least seven species of frog-biting midges (Bernal et 298 al., 2006) and it is unclear if all, or only some, of those species may act as vectors of T. tungarae 299 n. sp. Further studies that confirm the presence of T. tungarae n. sp. in the midgut or salivary 300 glands of frog-biting midges and examine species differences in transmission of trypanosomes 301 among frog-biting midges are necessary to confirm that the midges are indeed the vectors of T. 302 tungarae n. sp. These studies would also provide valuable insights by clarify the degree of 303 species specificity of trypanosomes and the midges. 304 In addition to frog-biting midges, there are other dipterans that are potential vectors of 305 blood parasites that in general should be considered when investigating the transmission patterns 306 of amphibian trypanosomes. There are, for instance, at least two species of mosquitos that use 307 the mating calls of frogs to find their victim and feed exclusively on anurans (Uranotaenia lowii, 308 Borkent and Belton, 2006; Culex territans, Bartlett-Healy et al., 2008). Other frog-biting insects 309 such as Forcipomyia species specialize on amphibians (Thompson 1969) and could also act as 310 vectors of anuran trypanosomes. Although at our study site túngara frogs are only preyed upon 311 by frog-biting midges, frogs and toads are often bitten by a wide range of insects. Considering all 312 potential vectors of anuran trypanosomes is essential to understand the dynamics of these 313 protozoan parasites. 314 This description of a new species of Trypanosoma here highlights an interesting pattern 315 of convergence in morphology among members of two non-closely related groups of 316 trypanosomes infecting frogs. The morphometric data revealed that the trypomastigotes of T. 317 tungarae n. sp. have overall similarity to other giant trypanosomes such as T. rotatorium and T. 318 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT ranarum. Despite its slender and long cell shape, however, T. tungarae n. sp. is sister to T. 319 chattoni—a highly derived trypanosome with a large rounded body, lacks a free flagellum, and 320 lack of undulating membrane (Martin et al., 2002; Lemos et al. 2008). This convergence in 321 morphology, however, could be explained by functionality; the sizes of the host’s erythrocytes 322 are correlated with the morphology of trypanosomes suggesting adaptations of the trypanosomes 323 to the host environment (Wheeler et al., 2013). We limit our discussion to comparisons between 324 T. tungarae n. sp. and species of trypanosomes from the phylogenetic tree used here because (i) 325 we are confident they represent separate lineages, and (ii) it is difficult to rely on morphology to 326 discern between blood trypanosomes (Lima et al. 2012; Fermino et al. in press). Sequences, 327 however, are not available for all anuran trypanosomes described to date. Therefore, it is possible 328 that T. tungarae n. sp. may be equivalent to a previously described, unnamed trypanosome for 329 which no molecular data is yet available. Further studies of trypanosome diversity in anurans that 330 include a combination of morphological and molecular work would provide an opportunity to 331 identify further cases of morphological convergence and overall patterns of evolution within 332 members of the aquatic clade. 333 Despite significant efforts to revise the phylogenetic relations and taxonomic status of 334 anuran trypanosomes (Diamond, 1965; Ayala, 1970, 1971; Desser and Yekutiel, 1986; Desser, 335 2001; Martin et al., 2002; Ferreira et al., 2007, 2008; Lemos et al., 2008), there is still an urgent 336 need for an extensive revision of this group of parasites. The phylogeny of anuran trypanosomes 337 needs in particular the advancement of the development of tools to include additional genes. 338 Traditionally only the 18S rRNA and gGAPDH genes have been used for trypanosome 339 phylogenetics (e.g. Hamilton et al., 2007), and most of the work conducted on the aquatic clade 340 has relied only on data from one gene (this study, Martin et al., 2002; Ferreira et al., 2007; 341 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Ferreira et al., 2008; Hayes et al., 2014). Perhaps the difficulties associated with amplifying 342 markers different to the 18S rRNA directly from DNA extracted from blood and tissues have 343 hampered the efforts to build stronger phylogenies of this clade. In this study, we failed to 344 amplify the gGAPDH gene using published and new primers. As a consequence, our analysis 345 only includes one gene and several relationships are thus not well supported in our phylogeny. 346 For example, there is little support for the relationships among the subclades that we identified. 347 Despite this challenge, however, fragments of the 18S rRNA gene have been successfully used to 348 characterize trypanosome species in other systems (e.g., Hayes et al. 2014) and the DNA 349 sequences found in this study indicate the trypanosome examined here represents a single, new 350 lineage. 351 The PTP species delimitation approach we used here is a reliable method to tentatively 352 identify trypanosome species using phylogenetic data. Another study explored the usefulness of 353 this method in uncovering several species of trypanosomes in a single location providing 354 convincing evidence of its reliability (Cottontail et al., 2014). Multiple loci and multiple 355 delimitation approaches, however, are necessary to confirm these inferences (Carstens et al., 356 2013). Nonetheless, for organisms as poorly studied as the trypanosomes of wildlife, the PTP 357 method is promising—at least until generating data from multiple genes is a common practice. 358 Trypanosoma tungarae n. sp. is currently only known from its type host, túngara frogs. 359 Although a trypanosome was previously reported to be transmitted by frog-biting midges to 360 another anuran in the Southeastern US (Johnson et al., 1993), its relationship to T. tungarae n. 361 sp. is unknown because it was not characterized. Few studies have investigated host specificity 362 of anuran trypanosomes. Studies have described the presence of the same trypanosome species 363 across a broad range of frogs and toads (Ray and Choudhury, 1983) and, given that vectors are 364 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT often associated with several species of vertebrate hosts (Ferreira et al., 2008), their association 365 to only one or few anuran species seems unlikely. The diversity of hosts used by T. tungarae n. 366 sp. requires further examination. Other potential anuran hosts include, for instance, include the 367 hourglass frog (Dendrosophus ebbracatus) and yellow cricket treefrog (D. microcephalus) that 368 are also victims of frog-biting midges (de Silva et al., 2014), the putative vector of T. tungarae n. 369 sp. in this area. Investigating the diversity of hosts of T. tungarae in further studies will be 370 important to understand the patterns of this blood parasite's dynamics in this anuran community. 371 Acknowledgements 372 We thank the Smithsonian Tropical Research Institute (STRI) for logistical support. Mabelle 373 Chong, Johanna Goyes, Simone Loss, Rachel Page and Miryam Venegas provided invaluable 374 help during the field and laboratory work. Daniel Castillo from the Universidad de Panama 375 provided valuable advice about staining techniques. We are also grateful to Cameron Smith and 376 Erik Galindo who helped inspect blood smears. Susan Perkins, Matthew Bolek, and Agustín 377 Jiménez read different versions of this manuscript and provided insightful comments. Two 378 anonymous reviewers provided valuable comments that helped improve this study. This research 379 was approved by the Panamanian Authorities (Autoridad Nacional del Medio Ambiente, ANAM 380 permit #SE/A-29-09, SEX/A-60-09). XEB was supported by a grant from the National Science 381 Foundation (IOS#1433990). 382 383 References 384 Ayala, S.C.1970. Two new trypanosomes from California toads and lizards. J. Protozool. 17, 385 370-373. 386 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Ayala, S.C. 1971. Trypanosomes in wild California sandflies, and extrinsic stages of 387 Trypanosoma bufophlebotomi. J. Protozool. 18, 433-438. 388 Bardsley, J.E., Harmsen, R. 1973. The Trypanosomes of Anura. Adv. Parasitol. 11, 1-73. 389 Bartlett-Healy, K., Crans, W., Gauger, E. 2008. Phonotaxis to amphibian vocalizations in Culex 390 territans (Diptera: Culicidae). Ann Entomol Soc Am 101, 95-103. 391 Bernal, X.E., Rand, A.S., Ryan, M.J. 2006. Acoustic preferences and localization performance of 392 blood-sucking flies (Corethrella Coquillett) to túngara frog calls. Behav. Ecol. 17, 709-715. 393 Bernal, X.E., de Silva. P. 2015. Cues used in host-seeking behavior by frog-biting midges 394 (Corethrella spp Coquillet). J. Vect. Ecol. 40, 122-128. 395 Borkent, A. 2008. The frog-biting midges of the world (Corethrellidae: Diptera). Zootaxa 1804, 396 1-456. 397 Borkent, A. 2008. The frog-biting midges of the world (Corethrellidae: Diptera). Zootaxa 1804, 398 1–456. 399 Borkent, A. 2014. World Catalog of extant and fossil Corethrellidae (Diptera). Zootaxa 3796, 400 453-468. 401 Borkent, A., Belton, P. 2006. Attraction of female Uranotaenia lowii (Diptera: Culicidae) to 402 frogs in Costa Rica. Can Entomol 138, 91-94. 403 Carstens, B.C., Pelletier, T.A., Reid, N.M., Satler, J.D. 2013. How to fail at species delimitation. 404 Mol. Ecol. 22, 4369-4383. 405 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Cottontail, V.M., Kalko, E.V., Cottontail, I., Wellinghausen, N., Tschapka, M., Perkins, S., 406 Pinto, C.M. 2014. High local diversity of Trypanosoma in a common bat species, and 407 implications for the biogeography and taxonomy of the T. cruzi clade. PLoS One 9, e108603. 408 de Silva, P., Jaramillo, C.A., Bernal, X.E. 2014. Selection of biting sites on anuran hosts by 409 Corethrella Coquillett species. J. Insect Behav. 27, 302-316. 410 Desser, S.S., Yekutiel, D. 1986. Blood parasites of amphibians and reptiles in Israel. Isr. J. Zool. 411 34, 77-90. 412 Desser, S.S. 2001. The blood parasites of anurans from Costa Rica with reflections on the 413 taxonomy of their trypanosomes. J. Parasitol. 87, 152-160. 414 Desser, S.S., Mciver , S.B., Ryckman, A. 1973. Culex territans as a potential vector of 415 Trypanosoma rotatorium. I. Development of the flagellate in the mosquito. J. Parasitol. 59, 353-416 358. 417 Diamond, L.S. 1965. Studies on the morphology, biology and taxonomy of the trypanosomes of 418 Anura. Wildl. Dis. 44, 1-77. 419 Edgar, R.C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high 420 throughput. Nucleic Acids Res. 32, 1792-1797. 421 Fantham, H.B., A. Porter, Richardson, L.R. 1942. Some Haematozoa observed in vertebrates in 422 eastern Canada. Parasitology 34, 199-226. 423 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Feng, L.C., Chung, H.L. 1940. Phlebotomus squamirostris Newstead, transmitter of 424 Trypanosoma bocagei França in the toad, Bufo bufo gargarizans (Cantor). China Med. J. 3, 198-425 211. 426 Ferguson, L.V., Smith, T.G. 2012. Reciprocal trophic interactions and transmission of blood 427 parasites between mosquitoes and frogs. Insects 3, 410-423. 428 Ferreira, R.C., Campaner, M., Viola, L.B., Takata, C.S.A., Takeda, G.F., Teixeira, M.M.G. 2007. 429 Morphological and molecular diversity and phylogenetic relationships among anuran 430 trypanosomes from the Amazonia, Atlantic Forest and Pantanal biomes in Brazil. Parasitology 431 134,1623-1638. 432 Ferreira, R.C., De Souza, A.A., Freitas, R.A., Campaner, M., Takata, C.S.A., Barrett, T.V., 433 Shaw, J.J., Teixeira, M.M.G. 2008. A Phylogenetic Lineage of Closely Related Trypanosomes 434 (Trypanosomatidae, Kinetoplastida) of Anurans and Sand Flies (Psychodidae, Diptera) Sharing 435 the Same Ecotopes in Brazilian Amazonia. J. Eukaryot. Microbiol. 55, 427-435. 436 Hamilton, P.B., Gibson, W.C., Stevens, J.R. 2007. Patterns of co-evolution between 437 trypanosomes and their hosts deduced from ribosomal RNA and protein-coding gene 438 phylogenies. Mol. Phylogenet. Evol. 44, 15-25. 439 Hayes, P.M., Lawton, S.P., Smith, N.J., Gibson, W.C., Davies, A.J. 2014. Morphological and 440 molecular characterization of a marine fish trypanosome from South Africa, including its 441 development in a leech vector. Parasit. Vectors 7, 1-11. 442 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Hysek, J, Zizka, Z. 1976. Transmission of Trypanosoma rotatorium from frogs to white mice. 443 Nature 260, 608-609 444 Innis, M.A, Gelfand, D.H., Sninsky, J.J., White, T.J. 1990. PCR Protocols: A Guide to Methods 445 and Applications. Academic Press, London. 446 Johnson, R.N., Young, D.G., Butler, J.F. 1993. Trypanosome transmission by Corethrella wirthi 447 (Diptera, Chaoboridae) to the green treefrog, Hyla cinerea (Anura, Hylidae). J. Med. Entomol. 448 30, 918-921. 449 Jones, S.R.M., Woo, P.T. 1986. Trypanosoma chattoni Mathis and Leger, 1911 in Rana pipiens 450 of southern Ontario: morphometrics and a description of the division process. Syst. Parasitol. 9, 451 57–62. 452 Lemos, M., Morais, D.H., Carvalho, V.T., D’Agosto, M. 2008. First record of Trypanosoma 453 chattoni in Brazil and occurrence of other Trypanosoma Species in Brazilian frogs (Anura, 454 Leptodactylidae). J. Parasitol. 94, 148-151. 455 Lynch, K.S., Crews, D., Ryan, M.J., Wilczynski, W. 2006. Hormonal state influences aspects of 456 female mate choice in the túngara frog (Physalaemus pustulosus). Horm Behav 49, 450-457. 457 Longmire, J.L., Maltbie, M., Baker, R.J. 1997. Use of “Lysis Buffer” in DNA Isolation and Its 458 Implication for Museum Collections. Occas. Pap. Museum Texas Tech Univ. 1997, 163. 459 McCallum, M.L., Moser, W.E., Wheeler, B.A., Trauth, S.E. 2011. Amphibian infestation and 460 host size preference by the leech Placobdella picta (Verrill, 1872) (Hirudinida: Rhynchobdellida: 461 Glossiphoniidae) from the Eastern Ozarks, USA. Herpetol. Notes 4, 147-151. 462 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT McKenzie, V.J., Starks, H.A. 2008. Blood parasites of two Costa Rican amphibians with 463 comments on detection and microfiliaria density associated with adult filarial worm intensity. J. 464 Parasitol. 94, 824-829. 465 Martin, D.S., Wright, A-D.G., Barta, J.R., Desser, S.S. 2002. Phylogenetic position of the giant 466 anuran trypanosomes, Trypanosoma chattoni, Trypanosoma fallisi, Trypanosoma mega, 467 Trypanosoma neveulemairei, and Trypanosoma ranarum inferred from 18S rRNA gene 468 sequences. J. Parasitol. 88, 566-571. 469 McKeever, S., Hartberg, W.K. 1980. An effective method for trapping adult female Corethrella 470 (Diptera, Chaoboridae). Mosq. News 40, 111-112. 471 McKeever, S., French, F.E. 2000. Corethrellidae (Diptera), vectors of present and perhaps some 472 of the earliest Anuran trypanosomes. Int. Virtual Conf. Vet. Med. Proc. Dis. Reptil. Amphib. 473 October 1-31, 2000. 474 Miyata, A. 1978. Anuran trypanosomes in Kyushu and Ryukyu Islands, with descriptions of six 475 new species. Trop. Med. 20, 51-80. 476 Mohr, J.L. 1981. Methods for staining protozoans in blood, macrophages or body fluids. In: 477 Baltimore, C.G. (Ed.), Staining Procedures. Fourth ed. Williams & Wilkins, Philadelphia. p. 303. 478 Murphy, W.J., O’Brien, S.J. 2007. Designing and optimizing comparative anchor primers for 479 comparative gene mapping and phylogenetic inference. Nat. Protoc. 2, 3022-3030. 480 Pinto, C.M., Kalko, E.K.V., Cottontail, I., Wellinghausen, N., Cottontail, V.M. 2012. TcBat, a 481 bat-exclusive lineage of Trypanosoma cruzi, in the Panama Canal Zone, with comments on its 482 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT classification and the use of the 18S rRNA gene for lineage identification. Infect. Genet. Evol. 483 12, 1328-1332. 484 Ray, R., Choudhury, A. 1983. Trypanosomes of Indian Anurans. Volume 8. Zoological Survey 485 of India, Calcutta. 486 Ramos, B., Urdaneta-Morales, S. 1977. Hernatophagous insects as vectors for frog 487 trypanosomes. Rev. Biol. Trop. 25, 209-217. 488 Reilly, B.O., Woo, P.T. 1982. Susceptibility of the leech Batracobdella picta (Verrill) to 489 Trypanosoma andersoni Reilly and Woo and Trypanosoma gryl. Can. J. Zool. 60, 1441-1445. 490 Rhoden, H.R., Bolek, M.G. 2012. Helminth and leech community structure in tadpoles and 491 caudatan larvae of two amphibian species from western Nebraska. J. Parasitol. 98, 236-244. 492 Ronquist, F., Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed 493 models. Bioinformatics 19, 1572-1574. 494 Ryan, M.J. 1985. The Túngara Frog: A Study in Sexual Selection and Communication. 495 University of Chicago Press, Chicago. 496 Sailasuta, A., Satetasit, J., Chutmongkonkul, M. 2011. Pathological study of blood parasites in 497 rice field frogs, Hoplobatrachus rugulosus (Wiegmann, 1834). Vet. Med. Int, Article ID 850568. 498 Stamatakis, A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of 499 large phylogenies. Bioinformatics, btu033. 500 StataCorp. 2007. Stata Statistical Software: Release 10. College Station, TX: StataCorp LP. 501 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Thompson, P. H. 1969. Feeding of Forcipomyia fairfaxensis on grass frogs (Rana spp.) in New 502 Jersey. Ann. Ent. Soc. Am. 62, 451-452. 503 Werner, J. 1993. Blood parasites of amphibians from Sichuan Province, People’s Republic of 504 China. J. Parasitol. 79, 356-363. 505 Werner. J.K., Davis, J.F., Slaght, K.S. 1988. Trypanosomes of Bufo americanus from Northern 506 Michigan. J. Wild. Dis. 24, 647-649. 507 Wheeler, R.J., Gluenz, E., Gull, K. 2013. The limits on trypanosomatid morphological diversity. 508 PLoS ONE 8, e79581. 509 Zhang, J., Kapli, P., Pavlidis, P., Stamatakis, A. 2013. A general species delimitation method 510 with applications to phylogenetic placements. Bioinformatics 29, 2869-2876. 511 Žičkus, T. 2002. The first data on the fauna and distribution of blood parasites of amphibians in 512 Lithuania. Acta Zool. Litu. 12, 197-202. 513 514 Figure Legends 515 Fig. 1. Photographs of túngara frogs (Engystomops pustulosus) and frog-biting midges 516 (Corethrella spp). (a) Calling male túngara frog preyed upon by frog-biting midges; (b) female 517 (bottom) in amplexus with a male (top) covered with biting midges; (c) female (bottom) with a 518 biting midge on her nostril that was passed from the male during amplexus. Túngara frogs are 519 about 30mm long while the frog-biting midges are only about 1.5mm. Photos taken by 520 Alexander Baugh (a) and Ximena E Bernal (b,c). 521 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Fig. 2. Map of the Republic of Panamá indicating with a star the location of Gamboa, the type 522 locality of Trypanosoma tungarae n. sp. Insert shows the location of Panamá in the New World. 523 Fig. 3. Light microscopy of Trypanosoma tungarae n. sp. (Giemasa-staining). (a-e) 524 Trypomastigotes stained using Hemacolor® Giemsa stain kit (Voigt Global Distribution Inc, 525 USA); (f-i) Trypomastigotes stained using Giemsa stain following Mohr (1981). Scale bars: 10 526 µm. 527 Fig. 4. Phylogeny of the aquatic clade, and PTP species delimitation results. Best maximum 528 likelihood tree of the18S rRNA gene of member of the aquatic clade and selected outgroups. 529 Numbers on the branches represent support values corresponding to ≥70% bootstrap replicates 530 (left) and ≥0.9 Bayesian posterior probabilities (right). Subclades are highlighted with colored 531 boxes to indicate host associations. Color of the branches indicate the PTP species delimitation 532 results; monophyletic groups in red indicate members of a single species, blue terminal branches 533 indicate that only one sample is included in such species. Names of the terminals indicate the 534 GenBank accession numbers, scientific name, and sample or isolate code. Star indicates the 535 position of T. tungarae n. sp. 536 M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT M AN US CR IP T AC CE PT ED ACCEPTED MANUSCRIPT Highlights • There is higher prevalence of trypanosome in male than female túngara frogs • Sexual differences in infection suggest potential transmission by frog-biting midges • Trypanosoma tungarae n. sp. is a new species infecting túngara frogs • This parasite resembles other giant frog trypanosomes from the Aquatic clade.