Ctenocephalides felis and Ctenocephalides canis: introgressive hybridization?

1 Ctenocephalides felis and Ctenocephalides canis: Introgressive hybridization? 1 2 ANTONIO ZURITA1, ROCÍO CALLEJÓN1, MANUEL DE ROJAS1, ALI 3 HALAJIAN 2 and CRISTINA CUTILLAS1 4 5 1Department of Microbiology and Parasitology. Faculty of Pharmacy. University of 6 Sevilla. Profesor García González 2, 41012 Sevilla, Spain. 7 2 Department of Biodiversity (Zoology). University of Limpopo. Turfloop Campus, 8 Private Bag X1106, Sovenga, Polokwane, 0727 South Africa. 9 10 11 12 13 14 15 * Corresponding author: 16 Dr. Cristina Cutillas 17 Department of Microbiology and Parasitology. Faculty of Pharmacy. University of 18 Sevilla. Prof. García González 2, 41012 Sevilla, Spain. 19 Phone: +34954556773 20 e-mail: [email protected] 21 2 Abstract 22 In the present work, a comparative molecular study of Ctenocephalides felis and 23 Ctenocephalides canis isolated from dogs (Canis lupus familiaris) from different 24 geographical regions (Spain, Iran and South Africa) has been carried out. We have 25 found morphological variations in C. felis which do not correspond with molecular 26 differences. The Internal Transcribed Spacers 1 and 2 (ITS1, ITS2) and 18S rRNA 27 partial gene, and cytochrome c-oxidase 1 (cox1) mtDNA partial gene sequences were 28 determined to clarify the taxonomic status of these two species and to assess inter- 29 population variation and inter-specific sequence differences. In addition, a comparative 30 phylogenetic study with other species of fleas using Bayesian, Maximum Parsimony, 31 Maximum Likelihood and Neighbour-Joining analysis was performed. 18S rRNA 32 partial gene fragment is not useful to discriminate C. canis and C. felis and was useless 33 to infer phylogenetic relationships at this level while ITS1 and ITS2 assessed for 34 specific determination in the genus Ctenocephalides. Cox1 mtDNA sequences of C. 35 felis revealed three main haplotypes and we suggest that there has been introgression of 36 C. canis cox1 mtDNA into C. felis by Wolbachia pipientis. Based on cox1 sequences, 37 restriction mapping identified many endonucleases that could be used to delineate 38 different haplotypes of C. felis and to differentiate C. felis and C. canis. 39 3 Introduction 40 Fleas (Siphonaptera) constitute a highly distinct group of holometabolous bloodsucking 41 insects which currently involve about 2574 species-level taxa belonging to 16 families 42 and 238 genera (Bitam et al., 2010). 43 Some authors have argued that Siphonaptera is the most completely studied order of 44 insects (Medvedev, 1994), and although this is perhaps true from a morphological-level 45 classification point of view, from a phylogenetic standpoint they have been sorely 46 neglected as a group. Classically, the major obstacle in flea phylogeny has been their 47 extreme morphological specialization associated with ectoparasitism, and the inability 48 of systematics to homogenize characters adequately across flea and outgroup taxa 49 (Whiting et al., 2008). In the past 30 years, there have been over 3,000 publications 50 dealing with some aspects of fleas (Lewis & Lewis, 1985), but only a few instances of 51 formal cladistics analysis (Acosta, 2010; Gao et al., 2013 and Acosta & Morrone, 52 2013), so in depth and continuous studies based on molecular data are needed to clarify 53 the unknown phylogeny of this order. 54 The Order Siphonaptera has also been widely studied due to its clinical importance for 55 human health since species of this group may play a role as parasites by them causing 56 allergic dermatitis or other conditions as a result of their feeding activities. Furthermore, 57 fleas serve as intermediate hosts for parasitic worms, and transmitting important 58 pathogens such as Rickettsia typhi, Yersinia pestis, Bartonella henselae, or Francisella 59 tularensis (Eisen & Gage, 2012). 60 Within this order, the genus Ctenocephalides Stiles & Collins, 1930 is one of the most 61 studied lineages because they tend to parasitize domestic animals such as dogs, cats or 62 other pets which may play an important role as bridging hosts for fleas of different 63 animals (Dobler & Pfeffer, 2011). Thirteen species and four subspecies are recognized 64 4 within this genus (Beaucournu & Ménier, 1998, Lawrence et al., 2014) out of which 65 Ctenocephalides felis and Ctenocephalides canis have been the most studied species by 66 different authors (Gil Collado, 1949; Gil Collado, 1960; Beaucournu & Launay, 1990; 67 Lewis, 1993b; Beaucournu & Ménier, 1998; Ménier & Beaucournu, 1998; Linardi & 68 Guimarães, 2000, Durden & Traub, 2002; Linardi & Santos, 2012). From a 69 morphological point of view, four subspecies of C. felis have been distinguished: C. 70 felis felis Bouché, 1835, C. felis strongylus Jordan, 1925, C. felis orientis Jordan, 1925 71 and C. felis damarensis Jordan, 1936 (Hopkins & Rothschild, 1953). However, only a 72 few studies have been carried out based on molecular data (Vobis et al., 2004; Marrugal 73 et al., 2013; Lawrence et al., 2014 and Lawrence et al., 2015). Thus, some authors have 74 suggested that more molecular studies are needed in order to resolve and elucidate the 75 genetic diversity of C. felis (Lawrence et al., 2014). 76 The ribosomal DNA segments Internal Transcribed Spacer 1 and 2 have been shown to 77 be two of the best molecular markers for analyzing genetic relationships at families, 78 genus and species level in arthropods (Monje et al., 2013; Marcilla et al., 2002; De 79 Rojas et al., 2002, 2007). Recently, Marrugal et al. (2013) concluded that phylogenetic 80 analysis based on Internal Transcribed Spacer 1 (ITS1) region is a useful tool to 81 approach different taxonomic and phylogenetic issues in Ctenocephalides species and 82 they found clear molecular differences between C. felis and C. canis. In addition, 18S 83 rRNA gene has been widely used as a molecular marker in order to clarify molecular 84 relationships among families and subfamilies (Whiting et al., 2008; Acosta & Morrone, 85 2013; Díaz-Nieto et al., 2013). 86 Moreover, mtDNA markers have remained as the markers of choice in many 87 populations, biogeographic and phylogenetic studies. Though many species are still 88 described based on morphology, or morphometrics only, mtDNA markers has also been 89 5 used in taxonomic studies since all described species are given as mtDNA sequence tag 90 or bar code (Hebert et al., 2003). Recently, Lawrence et al. (2014) carried out a study 91 based on cytochrome c-oxidase subunits 1 and 2, to investigate evolutionary 92 relationships and the genetic diversity of C. felis and other flea species from the genus 93 Ctenocephalides from different geographical areas (Australia, Fiji, Thailand and 94 Seychelles), concluding that both markers can be used to identify species of the genus 95 Ctenocephalides. Nevertheless, some authors cited that it is not safe to assume a priori 96 that mtDNA evolves as a strictly neutral marker because both direct and indirect 97 selection influences mitochondria. Thus, they questioned its utility as a marker for 98 genomic history (Ballard & Whitlock, 2004). On the other hand, the presence of 99 symbionts like Wolbachia pipientis has shown cases of reduction and increases in the 100 mtDNA genetic diversity. Thus, some authors have concluded that mtDNA on its own 101 is an unsuitable marker for the study of recent historical events in arthropods, 102 suggesting the development and use of microsatellites for intraspecific study, and 103 nuclear coding genes for phylogenetic study as a requirement to reveal the history of 104 nuclear DNA (Hurst & Jiggins, 2005; Dean et al., 2003; Shoemaker et al., 2003; 105 Kodandaramaiah et al., 2013). 106 Here we present a comparative molecular study of C. felis and C. canis, from different 107 geographical regions (Spain, Iran and South Africa). The ITS1 and ITS2, 18S rRNA 108 partial gene and cytochrome c-oxidase 1 mtDNA partial gene were sequenced in order 109 to assess inter-population and inter-specific variations to clarify the taxonomic status of 110 these species. Furthermore, comparative phylogenetic and phylogeographic analyses 111 with other species of fleas by phylogenetic methods (Bayesian, Maximum Parsimony, 112 Maximum Likelihood and Neighbor-Joining inference) were done. Finally, the presence 113 6 of Wolbachia pipientis in C. canis and C. felis has been checked to assess their 114 influence in both species´ evolutionary histories. 115 7 Materials and methods 116 Collection of samples 117 Fleas were collected from dogs (C. l. familiaris) from different geographical regions of 118 Spain, South Africa and Iran (Tables 1 and 2). Single individuals were preserved in 70 119 % ethanol until required for subsequent identification and sequencing. 120 Morphological identification 121 For morphological studies fleas were cleared with 10 % KOH (Lewis, 1993) and 122 examined under stereomicroscope. Morphological differentiation between C. felis and 123 C. canis individuals was carried out according to the original descriptions (Gil Collado, 124 1949, 1960; Beaucournu & Launay, 1990; Lewis, 1993b; Beaucournu & Ménier, 1998; 125 Ménier & Beaucournu, 1998; Marrugal et al., 2013). 126 Criteria cited by different authors have been used for the specific determination of the 127 genus Ctenocephalides including: 128  Genal ctenidium formed of eight or nine spines oriented horizontally (Lewis, 129 1993b) and relative size of the first and second genal spines (Gil Collado, 1949; 130 Beaucournu & Ménier, 1998; Durden & Traub, 2002; Marrugal et al., 2013). 131  Shape of the front of head (Gil Collado, 1949; Lewis, 1993b; Beaucournu & 132 Launay, 1990; Linardi & Santos, 2012; Marrugal et al., 2013). Length/width ratio of the 133 head (Durden & Traub, 2002; Marrugal et al., 2013). 134  Male genitalia: manubrium and apex (degree of dilation) (Gil Collado 1949; 135 Lewis 1993b; Ménier & Beaucournu, 1998; Marrugal et al., 2013) and aedeagus 136 (Ménier & Beaucournu, 1998; Marrugal et al., 2013). 137  Female genitalia: spermatheca and hilla (degree of elongation of the apical part 138 of the spermatheca) (Gil Collado, 1949; Lewis, 1993b; Marrugal et al., 2013). 139 8  Presence of two to three bristles on the lateral metanotal area (LMA) (Gil 140 Collado, 1949; Beaucournu & Launay, 1990; Beaucournu & Ménier, 1998; Linardi & 141 Santos, 2012; Marrugal et al., 2013). 142  Hind tibia with a number of seta-bearing notches along dorsal margin (Lewis, 143 1993b; Beaucournu & Launay, 1990; Beaucournu & Ménier, 1998; Durden & Traub, 144 2002; Linardi & Santos, 2012; Marrugal et al., 2013). 145 Molecular study 146 All specimens were photographed and then were frozen in liquid nitrogen and 147 pulverized in a mortar. Genomic DNA was extracted using the DNeasy Blood and 148 Tissue Kit (Qiagen) following the manufacturer’s protocol and was checked in a 0.8 % 149 agarose gel electrophoresis using ethidium bromide. 150 All molecular markers sequenced in this study were amplified by PCR using a 151 thermocycler (Eppendorf AG). PCR mix, PCR conditions and PCR primers used to 152 sequence each marker have been summarized in Table S1. 153 All sequenced fleas were screened for Wolbachia sp. using specific 16S rRNA gene 154 primers (see Table S1). As positive control, DNA from Wolbachia pipientis (AN: 155 LN864488) derived from Ctenocephalides felis from Spain was available, whereas for 156 negative controls, we used DNA from Stenoponia tripectinata tripectinata negative for 157 Wolbachia pipientis. Negative and positive controls were tested after every PCR 158 reaction sets. Sequences obtained were compared with those in the GenBank DNA 159 database by using the program BLAST (version 2.0, National Center for Biotechnology 160 Information; available from: URL: http://www.ncbi.nlm.nih.gov). (Altschul et al., 161 1990). The ITS1, ITS2, 18S rRNA partial gene and cox1 partial gene sequences 162 obtained from C. felis and C. canis and 16S rRNA of Wolbachia pipientis gene were 163 deposited in GenBank database (Table 1). 164 9 The PCR products were checked on ethidium bromide stained 2 % Tris–Borate–EDTA 165 (TBE) agarose gels. Bands were eluted and purified from the agarose gel by using the 166 QWizard SV Gel and PCR Clean-Up System Kit (Promega). Once purified, the 167 products were sequenced by Stab Vida (Portugal). All the phylogenetic analyses were 168 performed on the rDNA and mtDNA datasets, and sequences were aligned using the 169 Clustal X program version 2.0 (Larkin et al., 2007). 170 The rDNA intra-individual variation was determined by sequencing four to five clones 171 of one individual per geographical population of C. felis and C. canis. The PCR 172 products were eluted from the agarose by using the WIZARD® SV Gel and PCR Clean- 173 Up System (Promega) and transformation was carried out as cited by Cutillas et al. 174 (2009). Plasmids were purified using a Wizard Plus SV (Promega) and sequenced by 175 Stab Vida (Portugal) with a universal primer (M13). 176 Restriction map of the cox1 sequences of C. felis and C. canis was performed using The 177 Sequence Manipulation Suite (Stothard, 2000; available at 178 http://www.bioinformatics.org/sms2/rest_map.html). 179 Phylogenetic trees were inferred using nucleotide data and produced using three 180 methods: Maximum Parsimony (MP) trees were generated using the MEGA 5 program 181 from Tamura et al. (2011), Maximum Likelihood (ML) using the PHYML package 182 from Guindon & Gascuel (2003) and Bayesian inferences (B) were performing from Mr 183 Bayes-3.2.6 (Ronquist & Huelsenbeck, 2003). JMODELTEST (Posada, 2008) program 184 was used to determinate the best fit substitution model for the parasite data (18S, ITS1, 185 ITS2 and cox1). Models of evolution were chosen for subsequent analyses according to 186 the Akaike Information Criterion (Huelsenbeck & Rannala, 1997; Posada & Buckley, 187 2004). For the study of the four concatenated datasets (cox1, ITS1, ITS2 and 18S), a 188 combined partitioned analysis was performed using the model-jumping option in Mr 189 16 (haplotypes B and C) together with C. canis (Iran) and C. f. felis (Australia). 317 Nevertheless, C. canis (Iran) seems to be close to C. f. felis (Haplotype C) with 1 318 mutational step. Clade 3 clustered C. felis (Haplotypes A, A1 and A2) together with C. 319 f. felis fron India, the Czech Republic and Australia. A majority haplotype (H1, 320 haplotype A) was observed including 14 individuals (C. felis from Spain and C. f. felis 321 from Australia and Czech Republic). Haplotype A1 (H2) and A2 (H3) appeared related 322 to the majority haplotype of C. felis (H1) separated with 1 mutational step in both cases 323 (Fig. 2). 324 In order to assess the influence of Wolbachia sp. on the relationship inferred from flea 325 cox1, the presence of Wolbachia pipientis was tested in all fleas studied. All the PCR 326 products obtained were sequenced and the length of the 16S rRNA gene region was 334 327 base pairs. BLAST analysis showed 99 – 100 % homology to Wolbachia pipientis 16S 328 rRNA (Genbank accession number AY026912). 329 17 4. Discussion 330 Genus and species determination of fleas is generally based on a variety of 331 morphological criteria (Lane & Crosskey, 1993; Kramer & Mencke 2001; Mehlhorn, 332 2001 and Linardi & Santos, 2012). However, a few studies have been carried out on 333 molecular differentiation of fleas. That means we have a great knowledge of flea 334 taxonomy at the species and subspecies level, and enough information to assess their 335 biology and role in diseases transmission, which has been studied worldwide in recent 336 years. In contrast, a rigorous exploration of phylogenetic relationships among fleas is 337 needed in order to clarify their complex systematics (Whiting et al., 2008). 338 The morphological study of fleas from dogs from different geographical regions 339 revealed the existence of two species: C. felis and C. canis. C. felis showed 340 morphological variations. This fact has been cited by different authors. Thus, in 341 Ctenocephalides spp. the most frequent morphological variations are observed in combs 342 and chaetotaxies of LMA and in hind tibia (Amin et al., 1974; Amin, 1976, Linardi & 343 Santos, 2012). These alterations in chaetotaxy on the LMA and metatibia have been 344 justified by different authors (Holland, 1949, Fox, 1952, Amin et al., 1974 and Amin, 345 1976) as the existence of hybridization between C. felis and C. canis. However, the 346 hypothesis of hybridization between both species must be rejected because commonly 347 species do not cross each other, as reinforced by Beaucournu & Guiller (2006). 348 Sometimes, specimens exhibiting variations have been improperly treated as hybrids, in 349 spite of the nonexistence of the two species in the same municipality or region. In our 350 case, the individuals exhibiting these “abnormalities” were collected from South Spain 351 and Mallorca, where only C. felis has been detected (Table 2). Furthermore, these 352 alterations in chaetotaxy on the LMA did not correspond with a determined cox1 353 haplotype (see asterisk, Table 2). 354 18 Our studies are in agreement with those of Linardi & Santos (2012), who concluded that 355 the separation of the two species of Ctenocephalides must be based on all 356 characteristics. Data on hosts, geographical distribution and prevalence of infestation 357 may support the identification of the species. In C. felis four subspecies have been 358 proposed on the basis of rather minor morphological differences (Hopkins & 359 Rothschild, 1953; Haeselbarth, 1966). Analysis of some sequences of the genome of 360 these species which show high variation can be helpful to assess the validity and 361 significance of such infraspecific taxa and in investigating evolutionary relationships 362 within and between species (Vobis et al., 2004). 363 In the present work, C. felis and C. canis isolated from C. l. familiaris from different 364 geographical areas were also analyzed by amplification and sequencing of ribosomal 365 (ITS1, ITS2, and 18S rRNA partial gene) and mitochondrial (cox1 mtDNA partial gene) 366 markers. 367 The comparative molecular study based on ITS1 sequences of C. felis within a 368 geographical region and from different geographical origins showed a 100 % similarity. 369 These results are in agreement with those of Vobis et al. (2004) and Marrugal et al. 370 (2013), who found that the ITS1 nucleotide sequences of different C. felis populations 371 from different geographical areas are practically identical. Nevertheless, other authors 372 such as Gamerschlag et al. (2008) reported different lengths in the ITS1 nucleotide 373 sequences in other species of fleas (Tunga penetrans from Africa and South America). 374 On the other hand, the comparison between C. canis and C. felis ITS1 sequences 375 revealed a clear inter-specific variability. These results are in agreement with Marrugal 376 et al. (2013), who concluded that the ITS1 region is a useful marker to approach 377 different taxonomic statements in the genus Ctenocephalides. 378 19 The ITS2 sequences of C. felis and C. canis were markedly shorter than ITS1. This fact 379 was reported by Vobis et al. (2004) in C. felis; however, these authors did not sequence 380 the ITS2 region of C. canis and we could not compare our data with theirs. The ITS2 381 sequence of C. canis has been reported for the first time in the present study. With 382 regard to our results, we can conclude that the analysis of ITS2 fragment constitutes a 383 useful tool for the differentiation of both Ctenocephalides species. There was no inter- 384 population variation and intra-population variation ranged from 99.6 % to 100 % in 385 both species. This result disagrees with that of Luchetti et al. (2007) who found two 386 genotypic groups (Atlantic and Pacific) in Tunga penetrans. Nevertheless, the peculiar 387 evolutionary dynamics of ITS2, which is a repeated sequence embodied of rDNA, is 388 well known. Therefore, this molecular marker evolves in a concerted way (Smith, 389 1976). This pattern is carried out through population dynamics processes and intra- 390 genomic unequal DNA exchanges (molecular drive; Dover, 2002). Our results confirm 391 this evolutionary process in fleas, where no marked differences were observed even 392 between different geographical populations. This phenomenon was recently reported by 393 Zurita et al. (2015) in Stenoponia tripectinata tripectinata (Siphonaptera) from the 394 Canary Islands and in Diptera species by Monje et al. (2013). In conclusion, ITS1 and 395 ITS2 sequences do not determine different populations within one species of the genus 396 Ctenocephalides; however, the analyses of both rDNA markers are a helpful tool for 397 differentiation at the species level. 398 The 18S rRNA partial gene sequences confirmed the low value of this marker for 399 phylogenetic studies at the species level. Thus, we could observe the existence of 400 politomy within the genus Ctenocephalides and other species of the family Pulicidae. 401 This fact is in agreement with Whiting et al. (2008). No differences were observed 402 between C. felis and C. canis. 403 20 Cox1 mtDNA sequences of C. felis revealed five different haplotypes: A, A1, A2, B 404 (individuals from Spain and Iran), and C (individuals from South Africa). On the other 405 hand, no different haplotypes were observed in cox1 mtDNA sequences of C. canis 406 from Iran. In addition, some specific recognition sites for endonucleases were detected 407 in order to differentiate both species (C. felis and C. canis) and to differentiate the C. 408 felis haplotypes. Thus, SacI, SstI, TaqI, DraI, BclI, HpaII and MspI sites have 409 diagnostic value for specific determination in C. felis and C. canis. 410 The phylogenetic tree of the cox1 sequences revealed three different clades within the 411 genus Ctenocephalides. The diversity showed even within individuals of C. felis from 412 the same geographical origins. For example, Fuentes de Andalucía (Haplotypes B, A 413 and A2) or Mairena (Haplotypes A, A1 and B) (Table 2) is in agreement with Lawrence 414 et al. (2014), who found different haplotypes in C. f. felis from Australia. In our case, 415 we found a dominant C. felis group including sequences from the Czech Republic, 416 Australia, India, Iran, Thailand, Fiji and Spain (haplotypes A, A1 and A2) clustering in 417 the same clade, while haplotypes B and C clustered with C. canis in another clade (Fig. 418 S3). Thus, we found individuals of Ctenocephalides defined as C. felis by ITS1 and 419 ITS2 sequences but showing cox1 partial gene sequences of C. canis. The phylogenetic 420 tree of the concatenated 18S, ITS1, ITS2 and cox1 sequences corroborated these results 421 (Fig. 1). This diversity could be explained due to the presence of the symbiont 422 Wolbachia pipientis. This bacterium is very common in arthropods, and in recent years 423 has been thoroughly studied because of its influence on the host´s mtDNA variability. 424 Some authors (Schulenburg et al., 2000; Shoemaker et al., 2003; Behura et al., 2001) 425 have reported the presence of two or more haplotypes associated with this bacterium in 426 some species, such as Adalia bipunctata (Coleoptera), Drosophila mauritania (Diptera) 427 or Solenopis invicta (Hymenoptera). If a population becomes infected with a symbiont 428 21 such as Wolbachia pipientis that has sufficient drive to spread, the mtDNA type 429 associated with the initial infection will hitchhike through the population and further the 430 ability of symbionts to spread between populations by occasional movements of hosts 431 setting up a process named ``indirect selection on the mtDNA´´ (Hurst & Jiggins, 2005). 432 In agreement with former studies in the Order Diptera (Monnerot et al., 1990 and 433 Rousset & Solignac, 1995), we suggest that there has been introgression of C. canis 434 cox1 mtDNA into C. felis by Wolbachia pipientis. According to Shaw (2002) it is 435 possible that symbiont-driven introgression may explain cases where mtDNA 436 phylogenetic conflicts with those obtained from nuclear DNA. The spread of the 437 introgressed symbiont would be associated with the spread of introgressed mtDNA, 438 homogenizing mtDNA variation across the species (Hurst & Jiggins, 2005). 439 Furthermore, C. orientis, often regarded as a subspecies of C. felis, should gain full 440 species status because it forms a sister clade to cox1 sequences of C. canis from the 441 Czech Republic. This is in agreement with Lawrence et al. (2015). Nevertheless, the 442 cox1 sequences of C. canis from Iran appeared separated from those sequences but 443 clustering with haplotypes B and C of C. felis. The network obtained for all the 444 haplotypes corroborated the phylogenetic results (Fig. S3 and 1). 445 In conclusion, we have found morphological variations in C. felis which do not 446 correspond with molecular differences. The analysis of 18S rRNA partial gene is not a 447 useful tool to discriminate C. canis and C. felis while ITS1 and ITS2 assessed for 448 specific determination in the genus Ctenocephalides. Cox1 mtDNA sequences of C. 449 felis revealed five different haplotypes and we suggest a possible introgression of C. 450 canis cox1 mtDNA into C. felis by Wolbachia pipientis. The presence of symbionts, 451 such as Wolbachia pipientis should be checked in species of arthropods showing 452 22 reduced or increased mtDNA diversity because these symbionts confound the inference 453 of an organism´s evolutionary history from mtDNA data (Hurst & Jiggins, 2005). 454 23 Acknowledgement 455 The present work was supported by a grant of the V Plan Propio de Investigación of the 456 University of Sevilla, Spain. Thanks to Biodiversity Research Chair (University of 457 Limpopo- Professor Wilmien J Luus-Powell) for funding the field works in South 458 Africa, to Firouz-Farideh for partly funding field works in Iran and Pietersburg 459 Veterinary Clinic (Polokwane, South Africa) for providing some of the samples. We 460 thank Mrs. Anne Kendall for the critical reading of the manuscript. 461 24 References 462 Acosta, R. (2010) Five New Mexican Species of the Flea Genus Strepsylla Traub, 1950 463 (Siphonaptera: Ctenophthalmidae: Neopsyllinae: Phalacropsyllini) with a Phylogenetic 464 Analysis. Journal of Parasitology, 96, 285-298. 465 Acosta, R. & Morrone, J.J. (2013) Phylogenetics of the tribe Phalacropsyllini 466 (Siphonaptera: Ctenophthalmidae: Neopsyllinae) based on molecular and morphological 467 evidence. Zootaxa, 3630, 333-346. 468 Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) Basic local 469 alignment search tool. Journal of Molecular Biology, 215, 403-410. 470 Amin O.M., Wells, T.R. & Gately, H.L. (1974) Comb variations in the cat flea, 471 Ctenocephalides f. felis (Bouché). Annals of the Entomological Society of America, 67, 472 831-834. 473 Amin, O.M. (1976) Host associations and seasonal occurrence of fleas from 474 Southeastern Wisconsin mammals with observations on morphologic variations. 475 Journal of Medical Entomology, 13, 179-192. 476 Ballard, J.W.O. & Whitlock, M.C. (2004) The incomplete natural history of 477 mitochondria. Molecular Ecology, 13, 729-744. 478 Bandelt, H.J., Forster, P. & Röhl, A. (1999) Median-joining networks for inferring 479 intraspecific phylogenies. Molecular Biology and Evolution, 16, 37-48. 480 Beaucournu, J.C. & Guiller, A. (2006) Réponse à l’article “Keine 481 molekularbiologischen Anzeichen für Unterarten beim Katzenfloh Ctenocephalides felis 482 (Siphonaptera: Ctenocephalidae)” par H Mehlorn, J D’Haese, M Vobis & N Mencke: 483 Entomologia Generalis 27: 295-301 [2005]. Entomologia Generalis, 28, 311-315. 484 25 Beaucournu, J.C. & Launay, H. (1990) Les Puces (Siphonaptera) de France et du Bassin 485 méditerranéen occidental. Faune de France, 76, Paris. Fedération Française des 486 Sociétés des Sciences Naturelles. 487 Beaucournu, J.C. & Ménier, K. (1998) Le genre Ctenocephalides Stiles et Collins, 1930 488 (Siphonaptera, Pulicidae). Parasite, 5, 3-16. 489 Behura, S.K., Sahu, S.C., Mohan, M., & Nair, S. (2001) Wolbachia in the Asian rice 490 gall midge, Orseolia oryzae (Wood-Mason): Correlation between host mitotypes and 491 infection status. Insect Molecular Biology, 10, 163-171. 492 Bitam, I., Dittmar, K., Parola, P., Whiting, M.F. & Raoult, D. (2010) Fleas and flea- 493 borne diseases. International Journal of Infectious Diseases, 14, 667-676. 494 Cassens, I., Van Waerebeek, K., Best, P.B., Crespo, E.A., Reyes, J. & Milinkovitch, 495 M.C. (2003) The phylogeography of dusky dolphins (Lagenorhynchus obscurus): a 496 critical examination of network methods and rooting procedures. Molecular Ecology, 497 12, 1781-1792. 498 Cutillas, C., Callejón, R., de Rojas, M., Tewes, B., Ubeda, J.M., Ariza, C. & Guevara, 499 D.C. (2009) Trichuris suis and Trichuris trichiura are different nematode species. Acta 500 Tropica, 111, 299-307. 501 De Rojas, M., Mora, M.D., Úbeda, J.M., Cutillas, C., Navajas, M. & Guevara, D.C. 502 (2002) Phylogenetic relationships in rhinonyssid mites (Acari: Rhinonyssidae) based on 503 ribosomal DNA sequences: insights for the discrimination of closely related species. 504 Parasitology Research, 88, 675-681. 505 De Rojas, M., Úbeda, J.M., Cutillas, C., Mora, D., Ariza, C. & Guevara, D.C. (2007) 506 Utility of ITS1-5.8S-ITS2 and 16S mitochondrial DNA sequences for species 507 32 Stothard, P. (2000) The sequence manipulation suite: JavaScript programs for analyzing 646 and formatting protein and DNA sequences. Biotechniques, 28, 1102-1104. 647 Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. & Kumar, S. (2011) 648 MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, 649 evolutionary distance, and maximum parsimony methods. Molecular Biology and 650 Evolution, 28, 2731-2739. 651 Vobis, M., D´Haese, J., Mehlhorn, H., Mencke, N., Blagburn, B.L., Bond, R., Denholm, 652 I., Dryden, M.W., Payne, P., Rust, M.K., Schroeder, I., Vaughn, M.B. & Bledsoe, D. 653 (2004) Molecular phylogeny of isolates of Ctenocephalides felis and related species 654 based on analysis of ITS1, ITS2 and mitochondrial 16S rDNA sequences and random 655 binding primers. Parasitology Research, 94, 219-226. 656 Whiting, M.F., Whiting, A.S., Hastriter, M.W. & Dittmar, K. (2008) A molecular 657 phylogeny of fleas (Insecta: Siphonaptera): origins and host associations. Cladistics, 24, 658 677-707. 659 Zurita, A., Callejón, R., De Rojas, M., Gómez-López, M.S. & Cutillas, C. (2015) 660 Molecular study of Stenoponia tripectinata tripectinata (Siphonaptera: 661 Ctenophthalmidae: Stenoponiinae) from the Canary Islands: taxonomy and phylogeny. 662 Bulletin of Entomological Research, doi: 10.1017/S0007485315000656. 663 33 Figure captions 664 Figure S1. Phylogenetic tree of Ctenocephalides felis and Ctenocephalides canis from 665 different geographical origins (see Table 1) based on 18S ribosomal RNA partial gene 666 of inferred using the Bayesian (B), Maximum Likelihood (ML) and Maximum 667 Parsimony (MP) methods and Bayesian topology. The percentage of replicate trees in 668 which the associated taxa clustered together in the bootstrap test (1,000 replicates) is 669 shown onto the branches (B/ML/MP). Bootstrap values lower than 60 % are not shown. 670 The Bayesian Posterior Probabilities (BPP) is percentage converted. 671 Figure S2. Alignment of the consensus nucleotide sequences of the Internal 672 Transcribed Spacer (ITS2) of Ctenocephalides felis and Ctenocephalides canis isolated 673 from Canis lupus familiaris (Gaps generated using Clustal W alignment program). 674 Figure S3. Phylogenetic tree of Ctenocephalides felis and Ctenocephalides canis from 675 different geographical origins (see Table 2) based on cytochrome c-oxidase 1 (cox1) 676 partial gene of mitochondrial DNA inferred using the Bayesian (B), Maximum 677 Likelihood (ML) and Maximum Parsimony (MP) methods and Bayesian topology. The 678 percentage of replicate trees in which the associated taxa clustered together in the 679 bootstrap test (1,000 replicates) is shown onto the branches (B/ML/MP). Bootstrap 680 values lower than 60 % are not shown. The Bayesian Posterior Probabilities (BPP) is 681 percentage converted. 682 Figure 1. Phylogenetic tree of Ctenocephalides felis and Ctenocephalides canis from 683 different geographical origins (see Table 1 and 2) based on concatenated 18S ribosomal 684 RNA partial gene, Internal Transcribed Spacers 1 and 2 (ITS1, ITS2) and cytochrome c- 685 oxidase 1 (cox1) partial gene of mitochondrial DNA inferred using the Bayesian (B) and 686 Maximum Likelihood (ML) methods and Bayesian topology. The percentage of 687 replicate trees in which the associated taxa clustered together in the bootstrap test (1,000 688 34 replicates) is shown onto the branches (B/ML). The Bayesian Posterior Probabilities 689 (BPP) is percentage converted. 690 Figure 2. A minimum spanning network constructed using 42 haplotypes of 691 mitochondrial cox1 partial gene sequences of Ctenocephalides spp. The geographical 692 origin for each haplotype is shown in Table 2. The size of the circle is proportional to 693 the number of haplotypes represented and the numbers correspond to the mutational 694 steps observed between haplotypes. H1 (14): C. felis from Spain and Iran (Haplo A), C. 695 felis felis from Australia and the Czech Republic; H2 (1): C. felis from Spain (Haplo 696 A1); H3 (1): C. felis from Spain (Haplo A2); H4 (2): C. felis from Spain (Haplo B); H5 697 (3): C. felis from South Africa (Haplo C); H6 (4): C. canis from Iran; H7 (2): C. felis 698 strongylus; H8 (2): C. orientis; H9 (2): C. felis felis from Australia; H10 (4): C. felis 699 felis from India, Thailand and Fiji; H11 (2): C. felis felis from the Czech Republic; H12 700 (2): C. felis felis from India; H13 (3): C. canis from the Czech Republic. 701 35 ITS1 Location/Country Number of fleas Number of base pairs (bp) Species Accession number Nashtarood, Mazandaran/Iran 1 668 C. felis LN827902 Pilas, Sevilla/Spain 4 668 C. felis Villamanrique de la Condesa, Sevilla/Spain 4 668 C. felis Lebrija, Sevilla/Spain 4 668 C. felis Dílar, Granada/Spain 3 668 C. felis Fuentes de Andalucia, Sevilla/Spain 4 668 C. felis La Luisiana, Sevilla/Spain 4 668 C. felis Dos Hermanas, Sevilla/Spain 1 668 C. felis ITS2 Location/Country Number of fleas Number of base pairs (bp) Species Accession number Sanlúcar de Barrameda, Cádiz/Spain 1 (No. 113) 327 C. felis LN827904 Sanlúcar de Barrameda, Cádiz/Spain (Clone 1) - 327 C. felis LN864484 Sanlúcar de Barrameda, Cádiz/Spain (Clone 2) - 327 C. felis Sanlúcar de Barrameda, Cádiz/Spain (Clone 3) - 327 C. felis Sanlúcar de Barrameda, Cádiz/Spain (Clone 4) - 327 C. felis Sanlúcar de Barrameda, Cádiz/Spain 2 327 C. felis LN827903 Mallorca/Spain 4 327 C. felis Pilas, Sevilla/Spain 3 327 C. felis Mairena, Sevilla/Spain 5 327 C. felis Villamanrique de la Condesa, Sevilla/Spain 3 327 C. felis Lebrija, Sevilla/Spain 4 327 C. felis Dílar, Granada/Spain 4 327 C. felis Fuentes de Andalucía, Sevilla/Spain 3 327 C. felis La Luisiana, Sevilla/Spain 3 327 C. felis Dos Hermanas, Sevilla/Spain 5 327 C. felis Nashtarood, Mazandaran/Iran 2 327 C. felis Polokwane, Limpopo/South Africa 8 327 C. felis Tonekabon, Mazandaran/Iran 1 327 C. canis LN827905 Kotra, Mazandaran/Iran 2 327 C. canis Kotra, Mazandaran/Iran (Clone 1) - 327 C. canis LN864485 Kotra, Mazandaran/Iran (Clone 2) - 327 C. canis Kotra, Mazandaran/Iran (Clone 5) - 327 C. canis Kotra, Mazandaran/Iran (Clone 3) - 327 C. canis LN864486 Kotra, Mazandaran/Iran (Clone 4) - 327 C. canis LN864487 18S rRNA gene Location/Country Number of fleas Number of base pairs (bp) Species Accession number Sanlúcar de Barrameda, Cádiz/Spain 2 989 C. felis LN651166 Mallorca/Spain 2 989 C. felis Pilas, Sevilla/Spain 2 989 C. felis Mairena, Sevilla/Spain 2 989 C. felis Villamanrique de la Condesa, Sevilla/Spain 1 989 C. felis Dílar, Granada/Spain 1 989 C. felis Fuentes de Andalucía, Sevilla/Spain 1 989 C. felis La Luisiana, Sevilla/Spain 1 989 C. felis Kotra, Mazandaran/Iran 1 989 C. felis Polokwane, Limpopo/South Africa 2 989 C. felis Kotra, Mazandaran/Iran 1 989 C. canis LN651167 16S rRNA gene Species Number of base pairs (bp) Accession number Wolbachia pipientis 334 LN864488 Table 1. GenBank accession numbers of ITS1, ITS2 and 18S rRNA partial gene sequences of individuals of Ctenocephalides felis and Ctenocephalides canis isolated from Canis lupus familiaris from different geographical regions. 36 702 703 Cox1 Ctenocephalides felis Number identifier Location/Country Wolbachia pipientis Sex Haplotype Accession Number 102 Sanlúcar de Barrameda, Cádiz/Spain - M A LN827896 31 Sanlúcar de Barrameda, Cádiz/Spain - M A 28 Sanlúcar de Barrameda, Cádiz/Spain + H A 223 Mairena, Sevilla/Spain + H A 228* Mairena, Sevilla/Spain - M A 221 Mairena, Sevilla/Spain + H A 219 Mairena, Sevilla/Spain + H A 218 Mairena, Sevilla/Spain - H A 338* Mallorca/Spain + H A 332 Mallorca/Spain + H A 330 Mallorca/Spain + H A 199 Kotra, Mazandaran/Iran + H A 200 Kotra, Mazandaran/Iran + H A 269 Nashtarood, Mazandaran/Iran - M A 271 Nashtarood, Mazandaran/Iran - M A 609 Villamanrique de la Condesa, Sevilla/Spain + H A 615 Villamanrique de la Condesa, Sevilla/Spain + H A 601 Villamanrique de la Condesa, Sevilla/Spain + H A 612 Villamanrique de la Condesa, Sevilla/Spain + H A 481 Pilas, Sevilla/Spain + M A 478 Pilas, Sevilla/Spain + H A 479 Pilas, Sevilla/Spain + H A 477 Pilas, Sevilla/Spain - M A 3 Dílar, Granada/Spain - M A 4 Dílar, Granada/Spain + H A 5 Dílar, Granada/Spain + H A 120 Lebrija, Sevilla/Spain - M A 118* Lebrija, Sevilla/Spain + H A 643 Fuentes de Andalucía, Sevilla/Spain + H A 647 Fuentes de Andalucía, Sevilla/Spain + H A 644 Fuentes de Andalucía, Sevilla/Spain + H A 617 La Luisiana, Sevilla/Spain - H A 619* La Luisiana, Sevilla/Spain + H A 620* La Luisiana, Sevilla/Spain - H A 222 Mairena, Sevilla/Spain + H A1 LN827897 642 Fuentes de Andalucía, Sevilla/Spain + H A2 LN827898 216* Mairena, Sevilla/Spain + H B LN827899 645* Fuentes de Andalucía, Sevilla/Spain + H B 595 Polokwane, Limpopo/South Africa - M C LN827900 591 Polokwane, Limpopo/South Africa + H C 589 Polokwane, Limpopo/South Africa + H C 586 Polokwane, Limpopo/South Africa + H C 548 Polokwane, Limpopo/South Africa + H C 491 Polokwane, Limpopo/South Africa + H C 487 Polokwane, Limpopo/South Africa + H C Cox1 Ctenocephalides canis Number identifier Location/Country Wolbachia pipientis Sex Haplotype Accession Number 197 Kotra, Mazandaran/Iran + H - LN827901 204 Nashtarood, Mazandaran/Iran + H - 214 Kotra, Mazandaran/Iran + H - 198 Kotra, Mazandaran/Iran + H - Table 2. GenBank accession numbers of cox1 partial gene sequences (600 base pairs) of individuals of Ctenocephalides felis and Ctenocephalides canis isolated from Canis lupus familiaris from different geographical regions. Asterisk represents individuals which showed two and three bristles in each side on the LMA of one individual. 37 Cox1 C. felis haplotype A and C. felis from New South Wales (Australia) C. felis haplotype A1 C. felis haplotype A2 C. felis haplotype B C. felis haplotype C (South Africa) C. canis from Iran C. felis strongylus C. orientis C. felis felis from Queensland, Cairns (Australia) C. felis felis from Fiji, Thailand and Mumbai (India) C. felis felis from Sikkim (India) C. felis felis from Jablonec nad Nisou (Czech Republic) C. felis felis from Pardubice (Czech Republic) C. canis from Czech Republic C. felis haplotype A and C. felis from New South Wales (Australia) - C. felis haplotype A1 99.8 % - C. felis haplotype A2 99.8 % 99.7 % - C. felis haplotype B 97.3 % 97.5 % 97.2 % - C. felis haplotype C (South Africa) 97.3 % 97.5 % 97.2 % 99 % - C. canis from Iran 97.7 % 97.8 % 97.5 % 99.3 % 99.7 % - C. felis strongylus 98 % 98.2 % 97.8 % 98 % 97.7 % 98 % - C. orientis 91.3 % 91.5 % 91.2 % 91.8 % 91.5 % 91.8 % 91.7 % - C. felis felis from Queensland, Cairns (Australia) 97.3 % 97.5 % 97.2 % 98.7 % 98.3 % 98.7 % 97.7 % 92.3 % - C. felis felis from Fiji, Thailand and Mumbai (India) 98.3 % 99 % 98.7 % 97.8 % 97.8 % 98.2 % 98.5 % 92 % 98.2 % - C. felis felis from Sikkim (India) 99.7 % 99.5 % 99.5 % 97.3 % 97.5 % 97.7 % 98 % 92.3 % 97.5 % 98.8 % - C. felis felis from Jablonec nad Nisou (Czech Republic) 100 % 99.8 % 99.8 % 97.3 % 97.3 % 97.7 % 98 % 91.3 % 97.3 % 98.8 % 99.7 % - Table 3. Percentage of similarity among all the cox1 mtDNA partial gene sequences of C. felis and C. canis from different geographical areas obtained in this work and from GenBank database. 38 704 C. felis felis from Pardubice (Czech Republic) 98.8 % 98.7 % 98.7 % 97.5 % 97.5 % 97.8 % 98.2 % 92.2 % 97.8 % 99 % 98.7 % 98.8 % - C. canis from Czech Republic 91.5 % 91.7 % 91.5 % 91.2 % 90.8 % 91.2 % 91.2 % 95.5 % 91.7 % 91.5 % 92.7 % 91.5 % 91.7 % - Pygiopsylla hilli JN008915 Acanthopsylla jordani JN008916 Pulex irritans KF479246 Pulex irritans KF479247 Echidnophaga gallinacea JN008921 Echidnophaga myrmecobii JN008919 Stenoponia tripectinata tripectinata KF479244 Stenoponia tripectinata tripectinata KF479243 Stenoponia tripectinata tripectinata LK937072 Stenoponia tripectinata tripectinata LK937071 Spilopsyllus cuniculi KF479237 Spilopsyllus cuniculi KF479236 Ctenocephalides orientis KF684868 Ctenocephalides orientis KF684867 Ctenocephalides canis from (Czech Republic) KP684208 Ctenocephalides canis from (Czech Republic) KP684209 Ctenocephalides canis from (Czech Republic) KP684210 Ctenocephalides felis from Mairena (Spain) LN827899 Ctenocephalides felis from Fuentes de Andalucía (Spain) LN827899 Ctenocephalides felis from Polokwane (South Africa) LN827900 Ctenocephalides felis from Polokwane (South Africa) LN827900 Ctenocephalides felis from Polokwane (South Africa) LN827900 Ctenocephalides canis from Kotra (Iran) LN827901 Ctenocephalides canis from Kotra (Iran) LN827901 Ctenocephalides canis from Nashtarood (Iran) LN827901 Ctenocephalides canis from Kotra (Iran) LN827901 Ctenocephalides felis felis from Queensland, Cairns (Australia) KF684905 Ctenocephalides felis felis from Queensland, Cairns (Australia) KF684904 Ctenocephalides felis strongylus KF684873 Ctenocephalides felis strongylus KF684872 Ctenocephalides felis felis from Thailand KF684866 Ctenocephalides felis felis from Fiji KF684877 Ctenocephalides felis felis from Mumbai (India) KP687810 Ctenocephalides felis felis from Mumbai (India) KP687811 Ctenocephalides felis felis from Pardubice (Czech Republic) KP684205 Ctenocephalides felis felis from Pardubice (Czech Republic) KP684204 Ctenocephalides felis felis from Jablonec nad Nisou (Czech Republic) KP684196 Ctenocephalides felis felis from Jablonec nad Nisou (Czech Republic) KP684194 Microchorista philpotti HQ696580 Boreus elegans HQ696579 Ctenocephalides felis from Dílar (Spain) LN827896 Ctenocephalides felis from Mairena (Spain) LN827896 Ctenocephalides felis from Mallorca (Spain) LN827896 Ctenocephalides felis from Lebrija (Spain) LN827896 Ctenocephalides felis from Pilas (Spain) LN827896 Ctenocephalides felis from Sanlúcar de Barrameda (Spain) LN827896 Ctenocephalides felis from Villamanrique de la Condesa (Spain) LN827896 Ctenocephalides felis from Fuentes de Andalucía (Spain) LN827896 Ctenocephalides felis from La Luisiana (Spain) LN827896 Ctenocephalides felis felis from Sikkim (India) KP229384 Ctenocephalides felis felis from Sikkim (India) KP229385 Ctenocephalides felis from Kotra (Iran) LN827896 Ctenocephalides felis from Fuentes de Andalucía (Spain) LN827898 Ctenocephalides felis from Mairena (Spain) LN827897 PYGIOPSYLLIDAE STENOPONIIDAE PULICIDAE PULICIDAE Haplo. A, A1 A2 Haplo. C NANNOCHORISTIDAE - BOREIDAE Haplo. B C. canis 100/99/- 99/62/- 81/-/- 100/100/99 100/99/98 99/89/99 100/98/99 100/-/99 97/97/71 86/77/94 74/-/76 73/-/84 74/87/69 100/82/69 100/95/96 Ctenocephalides felis felis from New South Wales (Australia) KF684894 Ctenocephalides felis felis from New South Wales (Australia) KF684893 FIGURE S3 FIGURE S2 Ctenocephalides canis AF423914 Ctenocephalides canis from Kotra (Iran) LN651167 Ctenocephalides felis from Dílar (Spain) LN651166 Panorpa striata AF423911 Neopanorpa harmandi AF423903 Ctenocephalides felis from Fuentes de Andalucía (Spain) LN651166 Ctenocephalides felis from Mairena (Spain) LN651166 Ctenocephalides felis from Villamanrique de la Condesa (Spain) LN651166 Ctenocephalides felis from Sanlúcar de Barrameda (Spain) LN651166 Ctenocephalides felis from Mallorca (Spain) LN651166 Ctenocephalides felis from Pilas (Spain) LN651166 Ctenocephalides felis KC177274 Ctenocephalides felis from Kotra (Iran) LN651166 Ctenocephalides felis from La Luisiana (Spain) LN651166 Ctenocephalides felis from Polokwane (South Africa) LN651166 Xenopsylla cunicularis EU336098 Xenopsylla cheopis EU336038 Parapulex chephrensis EU336052 Echidnophaga ibérica EU336099 Echidnophaga gallinacea EU336055 Pulex irritans AF423915 Spilopsyllus cuniculi EU336097 C. felis, C. canis PULICIDAE PANORPIDAE (MECOPTERA) Corrodopsylla curvata curvata EU336064 Corrodopsylla birulai EU336063 CTENOPHTHALMIDAE (Doratopsyllinae) Hystrichopsylla schefferi EU336084 Hystrichopsylla orophila EU336128 HYSTRICHOPSYLLIDAE Ophthalmopsylla volgensis palestinica AF423895 Ophthalmopsylla jettmari EU336101 Leptopsylla segnis DQ298442 LEPTOPSYLLIDAE Tunga libis EU336114 Tunga monositus EU336034 Tunga penetrans EU336115 TUNGIDAE Ceratophyllus petrochelidoni AF423888 Ceratophyllus gallinae EU336041 Citellophilus tesquorum mongolicus EU336103 CERATOPHYLLIDAE Ctenophthalmus pseudagyrtes AF423892 Ctenophthalmus formosanus EU336131 CTENOPHTHALMIDAE (Ctenophthalminae) Neopsylla bidentatiformis EU336074 Rhadinopsylla masculana EU336056 Rhadinopsylla difficilis EU336044 CTENOPHTHALMIDAE (Rhadinopsillinae, Neopsyllinae) Stephanocircus dasyuri EU336085 Stephanocircus pectinipes EU336088 STEPHANOCIRCIDAE Pygiopsylla tunneyi JN008933 Pygiopsylla hilli JN008926 Acanthopsylla rothschildi rothschildi AF286283 Acanthopsylla jordani JN008925 Lycopsylla nova EU336132 Pygiopsylla hoplia EU336086 PYGIOPSYLLIDAE Stenoponia tripectinata tripectinata LK937067 Stenoponia tripectinata tripectinata LK937066 Stenoponia tripectinata tripectinata LK937068 Stenoponia tripectinata tripectinata LK937069 Stenoponia tripectinata tripectinata LK937070 Stenoponia tripectinata medialis EU336050 Stenoponia americana AF423893 Stenoponia sidimi EU336078 STENOPONIIDAE 100/100/- 79/92/66 63/77/- 79/95/62 81/82/- 89/76/- 96/76/64 84/87/66 100/100/100 74/72/- 100/91/78 95/97/60 99/86/82 FIGURE S1