INTRODUCTION
The adult stages of tapeworms of the cyclophyllidean family Anoplocephalidae Cholodkovsky, 1902, are intestinal parasites of mammals, birds and reptiles (Beveridge, Reference Beveridge, Khalil, Jones and Bray1994). In primates, several species of the subfamily Anoplocephalinae mostly from the genera Anoplocephala, Bertiella and Moniezia Blanchard, 1891, represent typical components of parasite communities. However, occasional infections by representatives of the subfamilies Linstowiinae (Mathevotaenia Akhumian, 1946) and Inermicapsiferinae (Thysanotaenia Beddard, 1911) have also been reported (Beveridge, Reference Beveridge, Khalil, Jones and Bray1994). Although the genus Anoplocephala mostly includes species from hyraxes, odd-toed ungulates and elephants, Anoplocephala gorillae Nybelin, Reference Nybelin1927, was described from eastern gorillas (Gorilla beringei) and is considered to be a common parasite in the Rwandan and Ugandan mountain gorilla populations (Sleeman et al. Reference Sleeman, Meader, Mudakikwa, Foster and Patton2000; Rothman et al. Reference Rothman, Pell and Bowman2008). In contrast, almost no cestodes have been found in western gorillas (Gorilla gorilla) of Gabon (Landsoud-Soukate et al. Reference Landsoud-Soukate, Tutin and Fernandez1995) and in G. gorilla from various localities in the Congo basin (author's unpublished data); only Lilly et al. (Reference Lilly, Mehlman and Doran2002) reported anoplocephalid eggs in two fecal samples of G. gorilla in the southern Dzanga-Ndoki National Park, Central African Republic, yet without specific identification. Other anoplocephalids, such as Moniezia rugosa Lühe, 1895, were encountered rather exceptionally in New World primates from the families Atelidae and Cebidae (Dunn, Reference Dunn1963; Fiennes, Reference Fiennes1967), with no recent records being available.
More common tapeworms in primates are members of the genus Bertiella, although other members of this genus parasitize a broad spectrum of mammalian hosts including marsupials, flying lemurs and rodents in Africa, Asia, Australia and South America (Denegri and Perez-Serrano, Reference Denegri and Perez-Serrano1997). The type species, Bertiella studeri, was described from a common chimpanzee Pan troglodytes by Blanchard (Reference Blanchard1891). Thereafter, more than 40 species have been assigned to the genus (Schmidt, Reference Schmidt1986; Caira et al. Reference Caira, Jensen and Barbeau2012), ten of them being known to infect primates (Schmidt, Reference Schmidt1986; Denegri and Perez-Serrano, Reference Denegri and Perez-Serrano1997; Galán-Puchades et al. Reference Galán-Puchades, Fuentes and Mas-Coma2000). Infections by three of the typical primate tapeworms, B. studeri (Blanchard, Reference Blanchard1891), Bertiella mucronata (Meyner, Reference Meyner1895) and Bertiella satyri (Blanchard, Reference Blanchard1891), are sporadically reported also from humans (Chandler, Reference Chandler1925; Cameron, Reference Cameron1929; Denegri and Perez-Serrano, Reference Denegri and Perez-Serrano1997; Bhagwant, Reference Bhagwant2004; Foitová et al. Reference Foitová, Mašová, Tenora, Koubková, Hodová, Vyskočilová, Baruš and Nurcahyo2011). Human infections with B. studeri were reported mostly from India, the Russian Far-East and Africa (Bhagwant, Reference Bhagwant2004; El-Dib et al. Reference El-Dib, Al-Rufaii, El-Badry, Al-Zoheiry and El-Aall2004), or from non-residents that visited these regions (Cameron, Reference Cameron1929; Galán-Puchades et al. Reference Galán-Puchades, Fuentes and Mas-Coma2000), while B. mucronata and B. satyri were confined to men and primates living in South America and South-East Asia, respectively (Gómez-Puerta et al. Reference Gómez-Puerta, López-Urbina and González2009; Foitová et al. Reference Foitová, Mašová, Tenora, Koubková, Hodová, Vyskočilová, Baruš and Nurcahyo2011). Diagnostics of the anoplocephaline infections are based mostly on the identification of gravid proglottids or eggs that possess a typical pyriform apparatus (Schmidt, Reference Schmidt1986; Denegri and Perez-Serrano, Reference Denegri and Perez-Serrano1997); yet identifying these parasites to the species level in clinical material from primates remains rather difficult.
In the last decade, molecular approaches helped to resolve the phylogenetic relationships of several genera of the Anoplocephalinae, including Bertiella (see e.g. Haukisalmi et al. Reference Haukisalmi, Hardman, Foronda, Feliu and Henttonen2010; Hardman et al. Reference Hardman, Haukisalmi and Beveridge2012; Taleb-Hossenkhan and Bhagwant, Reference Taleb-Hossenkhan and Bhagwant2012). Not surprisingly, molecular analyses showed that previous morphology-based classification may not reflect the actual relationships within the subfamily. In morphology-based phylogeny by Beveridge (Reference Beveridge, Khalil, Jones and Bray1994), Bertiella and Triplotaenia Boas, 1902, were sister genera, whereas the genus Anoplocephala was basal taxon of the subfamily (Fig. 1A). Molecular-based phylogeny by Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012), however, showed a sister relationship between Bertiella and Anoplocephala, with the genera Schizorchis Hansen, 1948, and Mosgovoyia Spasskii, 1951, emerging as basal taxa (Fig. 1B). The genus Bertiella deserves special attention, as previous studies did not include any isolates from primates (Hardman et al. Reference Hardman, Haukisalmi and Beveridge2012), or the used molecular marker was not sufficient for the establishment of intra-generic relationships within the cestodes (Taleb-Hossenkhan and Bhagwant, Reference Taleb-Hossenkhan and Bhagwant2012).
The phylogenetic trees of anoplocephaline cestodes: (A) position of Bertiella based on morphological-characters (Beveridge, Reference Beveridge, Khalil, Jones and Bray1994); (B) position of Bertiella based on analysis of the 28S rRNA gene (Hardman et al. Reference Hardman, Haukisalmi and Beveridge2012).
The present study provides the first molecular insight into the diversity of Bertiella and Anoplocephala from humans and non-human primates from Africa, Asia, Europe and South America, and describes their host specificity and phylogenetic relationships.
MATERIALS AND METHODS
Sample collection and preparation
The proglottids of anoplocephalid tapeworms were acquired from the following hosts and localities: (i) wild eastern chimpanzee P. troglodytes schweinfurthii from Uganda (Kalinzu Forest Reserve), (ii) semi-captive chimpanzees from Kenya (Sweetwaters Chimpanzee Sanctuary), (iii) wild western lowland gorilla Gorilla gorilla gorilla from the Central African Republic (Dzanga-Ndoki National Park), (iv) semi-captive Sumatran orangutan Pongo abelii from an Indonesian sanctuary (Gunung Leuser National Park), (v) captive Titi monkey Callicebus oenanthe from Peru (Moyobamba) and, (vi) humans from Brazil and Spain. The Spanish isolate from human and isolates from C. oenanthe and P. abelii were previously identified as B. studeri, B. mucronata and B. satyri, respectively, based on morphological characteristics of whole tapeworms or proglottids (Galán-Puchades et al. Reference Galán-Puchades, Fuentes and Mas-Coma2000; Gómez-Puerta et al. Reference Gómez-Puerta, López-Urbina and González2009; Foitová et al. Reference Foitová, Mašová, Tenora, Koubková, Hodová, Vyskočilová, Baruš and Nurcahyo2011). Furthermore, fecal samples containing cestode eggs with a characteristic pyriform apparatus (Beveridge, Reference Beveridge, Khalil, Jones and Bray1994) were collected from wild western chimpanzee P. troglodytes verus from Guinea-Bissau (Cantanhez National Park; Sá et al. Reference Sá, Petrášová, Pomajbíková, Profousová, Petrželková, Sousa, Cable, Bruford and Modrý2013), and also from wild mountain gorilla Gorilla beringei beringei from Rwanda (Volcanoes National Park). The morphology of the eggs of isolate from mountain gorilla corresponds to eggs of A. gorillae previously identified by Sleeman et al. (Reference Sleeman, Meader, Mudakikwa, Foster and Patton2000). In addition, eggs of Anoplocephala perfoliata isolated from domestic horse Equus cabalus (Brno, Czech Republic) were included into the study (summary of all isolates in Table 1).
The locality, source of DNA and GenBank accession numbers of particular genes of cestodes from non-human primates and humans. The extended dataset includes an isolate from horse
H.s., Homo sapiens; P.a., Pongo abelii; C.o., Callicebus oenanthe; P.t., Pan troglodytes; G.g., Gorilla gorilla; G.b., Gorilla beringei; E.c., Equus cabalus; ES, Spain, BR, Brazil; ID, Indonesia; PE, Peru; UG, Uganda; KE, Kenya; GW, Guinea-Bissau; CF, Central African Republic; RW, Rwanda; CZ, Czech Republic; c/s, captivity/sanctuary; w, wild.
a Galán-Puchades et al. (Reference Galán-Puchades, Fuentes and Mas-Coma2000).
b Foitová et al. (Reference Foitová, Mašová, Tenora, Koubková, Hodová, Vyskočilová, Baruš and Nurcahyo2011).
c Gómez-Puerta et al. (Reference Gómez-Puerta, López-Urbina and González2009).
Proglottids and fecal samples (~2 g) were preserved in 96% ethanol. The proglottids collected from feces were washed 5 times in phosphate buffered saline (PBS), disrupted with micro-pestles and washed 5 times in PBS followed by centrifugation at ~16 000 g for 5 min before DNA was extracted from the pellet. Feces were processed by flotation technique with modified Sheather's solution (Sheather, Reference Sheather1923). The eggs from surface of flotation solution were collected with a wire loop and transferred to a 10 mL vial tube containing 8 mL of PBS. Then, the vial was centrifuged at ~320 g for 2 min and the supernatant was carefully removed. The sediment was resuspended in 1 mL of PBS and 40 μL were transferred onto a microscopic slide with a dimple. Individual eggs were isolated with a thin glass micropipette, normally used for embryo transfers that had a short silicone hose with a 2 mm inner diameter and node on one end. The eggs (minimum 30) were transferred into 0·5 mL of PBS in a new 2 mL centrifuge tube. Prior to DNA extraction, the egg shells were disrupted with glass beads (0·5 and 1 mm) in a BeadBeater (Biospec, USA) by shaking at 2400 oscillations min−1 for 10 min.
DNA extraction
Before DNA extraction, 800 μL of NET buffer (4 m NaCl; 0·5 m Ethylenediaminetetraacetic acid; 1 m Tris, pH 8·0), 240 μL of N-lauroylsarcosine sodium salt solution (Sigma Aldrich, Germany) and 30 μL of proteinase K in concentration 100 μg mL−1 (Chemos, Czech Republic) were added to the sample. The homogenate was incubated in a dry bath at 56 °C for 15 h, and the lysate was subsequently extracted by phenol–chloroform. An equal volume of phenol (Sigma Aldrich, Germany) was added; gently vortexed for 10 min, spun for 10 min, and the procedure was repeated with equal volumes of phenol and subsequently chloroform (Sigma Aldrich, Germany). Following the extraction, DNA was precipitated with ethanol and sodium acetate, air-dried and resuspended in water.
Polymerase chain reaction (PCR) amplification and sequencing
For the purpose of identification of isolates as belonging to the genus Bertiella, we designed the primers amplifying 600 bp of 18S rRNA based on the published sequence of B. studeri obtained from a morphologically identified adult tapeworm from the crab-eating macaque Macaca fascicularis by Taleb-Hossenkhan and Bhagwant (Reference Taleb-Hossenkhan and Bhagwant2012). Using forward primer BF (5′-GGACACTATGAGGATTGACAGA-3′) and reverse primer BR (5′-CCTTTCGGGGCACCAAGATGG-3′), we amplified the fragment of 18S rRNA from our samples under following PCR conditions: an initial denaturation at 96 °C for 3 min and then 30 cycles of 1 min at 94 °C, 1 min at 58 °C and 1 min at 72 °C, followed by 10 min at 72 °C. These sequences obtained were later compared with the published sequence of B. studeri.
For further phylogenetic and taxonomic comparison, the cytochrome c oxidase subunit I (cox1) and partial nicotinamide adenine dinucleotide dehydrogenase dehydrogenase subunit 1 (nad1) of mitochondrial (mt) DNA, 5.8S-ITS2 and partial 28S rRNA of nuclear (n) DNA were sequenced, as published comparative sequences of these markers are available in larger abundance and broader taxonomic spectrum. The cox1 gene fragment was amplified using primers COX-F (5′-GATGTTTTCTTTACATTTATCTGGTG-3′) and COX-R (5′-GCCACCACAAATCAAGTATC-3′) following the protocol of Haukisalmi et al. (Reference Haukisalmi, Wickström, Henttonen, Hantula and Gubányi2004). The nad1 gene fragment was amplified by primers Cyclo-Nad1F (5′-GGNTATTSTCARTNTCGTAAGGG-3′) and Cyclo-trnNR (5′-TTCYTGAAGTTAACAGCATCA-3′) under conditions described elsewhere (Littlewood et al. Reference Littlewood, Waeschenbach and Nikolov2008). For 5.8S-ITS2 region primers Proteo1 (5′-CGGTGGATCACTCGGCTC-3′) and Proteo2 (5′-TCCTCCGCTTATTGATATGC-3′) designed by Škeříková et al. (Reference Škeříková, Hypša and Scholz2004) were used for 40 cycles (1 min at 94 °C, 1 min at 56 °C and 1 min at 72 °C) amplification. Additionally, the D1–D3 region of 28S rRNA gene was amplified using three alternative pairs of primers modified by Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012): (i) LSU5 (5′-TAGGTCGACCCGCTGAAYTTYAGCA-3′) and 1200R (5′-GCATAGTTCACCATCTTTCGG-3′) (c. 1400 bp); (ii) XZ-1 (5′-ACCCGCTGAATTTAAGCATAT-3′) and 1500R (5′-GCTATCCTGAGGGAAACTTCG-3′) (c. 1660 bp) and (iii) U178 (5′-GCACCCGCTGAAYTTAAG-3′) and L1642 (5′-CCAGCGCCATCCATTTTCA-3′) (c. 1500 bp). As used by Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012), the PCR conditions were following those of Lockyer et al. (Reference Lockyer, Olson and Littlewood2003), Waeschenbach et al. (Reference Waeschenbach, Webster, Bray and Littlewood2007) and Littlewood et al. (Reference Littlewood, Waeschenbach and Nikolov2008), respectively . Upon resolution in ethidium bromide-stained agarose gels, the amplicons were gel-purified using the QuickClean Gel Extraction Kit (GenScript, USA) and sequenced at Macrogen Inc. (Seoul, Korea). The assembled nucleotide sequences have been deposited in GenBank under the accession numbers JQ771093–JQ771118 and KJ888951–KJ888952 (Table 1).
Phylogenetic analyses
For phylogenetic analyses, we aligned seven 5.8.S-ITS2, 53 cox1, 17 nad1 and 24 28S rRNA GenBank sequences of anoplocephaline species (accession numbers of sequences are shown in the trees) with newly obtained tapeworm sequences from humans, non-human primates and horse, respectively. Based on availability in GenBank, sequences of Echinococcus spp. or Hymenolepis spp. were included as outgroup. Sequences of the respective genes were aligned in BioEdit (Hall, Reference Hall1999) using the Clustal W programme (Larkin et al. Reference Larkin, Blackshields, Brown, Chenna, McGettigan, McWilliam, Valentin, Wallace, Wilm, Lopez, Thompson, Gibson and Higgins2007), with a default scheme for introducing gaps into the alignment, and were further checked for inconsistencies and manually edited, when needed. Phylogenetic relationships among the Bertiella and Anoplocephala isolates and selected anoplocephaline species were estimated using Bayesian inference (BI) in the MrBayes 3.2 software (Ronquist and Huelsenbeck, Reference Ronquist and Huelsenbeck2003) under substitution models suggested by MrModeltest 2.3 (Nylander, Reference Nylander2004). Two independent runs were executed in each BI, lasting 1 million generations and sampled every 100 generations. The burn-in portion of the sampled trees was set to the default 25%. The mitochondrial and nuclear markers were analysed separately because of different included taxa available for each of them.
Mitochondrial cox1 (561 bp) and nad1 (747 bp) were analysed under the HKY+I+G and GTR+I+G substitution models, respectively. The large cox1 dataset containing multiple conspecific sequences of various anoplocephalid tapeworm genera was further used to assess the taxonomic significance of the genetic variability within our isolates based on the variation in percentage genetic distances within and between relevant clades representing distinct species. In addition to the 747 bp nad1 dataset, a portion thereof trimmed to 366 bp was alternatively analysed in order to include a shorter sequence of an isolate from G. beringei from Rwanda, which, despite repeated attempts of amplification and sequencing, could not be obtained in full length.
Due to short length of 5.8S rRNA, the nuclear 5.8S-ITS2 (896 bp after introduction of gaps) was analysed as an unpartitioned dataset under HKY+I+G substitution model.
Nuclear 28S rRNA sequences of Bertiella from primates were aligned together with the published data by Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012) in order to compare our data with published information on Bertiella from Australian marsupials and rodents. This alignment (1386 bp after introduction of gaps) was manually edited and ambiguously aligned positions were excluded prior to analysis, yielding a 1161 bp long alignment, which was analysed under GTR+I+G model using BI.
RESULTS
Low amount of available DNA was a limiting factor of our studies and in the case of P. troglodytes verus from Guinea-Bissau, we failed, despite multiple attempts, to obtain sufficient amount of DNA. In total, we managed to obtain eight sequences of 18S rRNA, six sequences of cox1, eight of nad1, eight of partial 5.8S-ITS2 rRNA and finally, two of 28S rRNA genes. The length of 18S amplicons varied from 680 to 710 bp, those of cox1 amplicons from 585 to 680 bp and nad1 from 730 to 900 bp, the size of partial 5.8S-ITS2 rRNA sequences obtained ranged from 620 to 820 bp and the size of 28S rRNA amplicons was 1460–1500 bp.
In 18S rDNA sequences, the query coverage of all our isolates with respect to the published sequence of B. studeri (GU323706) was 100% and the sequence identity reached as much as 97% for isolate from G. gorilla, 98% for isolates from Ugandan P. troglodytes, G. beringei, C. oenanthe and Brazilian isolate from human, and 99% for isolates from Kenyan P. troglodytes and P. abelii. An identity of 100% was revealed for Spanish isolate from human, which was previously determined as B. studeri by Galán-Puchades et al. (Reference Galán-Puchades, Fuentes and Mas-Coma2000).
The topology of the cox1 (Fig. 2), nad1 (Fig. 3A) and 5.8S-ITS2 (Fig. 4) trees showed that all isolates morphologically determined as Bertiella clustered together in a monophyletic group. This Bertiella cluster included also the isolate from G. beringei from Rwanda, morphologically determined as Anoplocephala cf. gorillae, from which only a partial fragment of nad1 was acquired (Fig. 3B). The internal topology of the Bertiella clade varied according to the respective markers and phylogenetic methods used, but in most analyses, no correlation with host or geography was observed. This means that cestodes from the same (humans) or congeneric hosts (two gorillas or three chimpanzees) were not closely related. In contrast, isolates from unrelated hosts and different locations were closely related, such as those of B. mucronata from C. oenanthe from Peru and Bertiella sp. from P. troglodytes from Uganda, or those from a Brazilian human and P. troglodytes from Kenya (Figs 2–4). Variability among Bertiella haplotypes expressed as pairwise genetic distance of the cox1 gene (Fig. 2) was 0·2–12·0% (Table 2).
Phylogenetic tree of 59 partial cox1 gene sequences of the subfamily Anoplocephalinae isolates. Sequences newly reported in the present study are in bold type and identified by name, host and locality of origin. Other sequences are identified by species name and GenBank accession number. The tree is rooted by Echinococcus granulosus. Posterior probabilities of BI are given above nodes; 1.00 support is indicated by an asterisk.
Bayesian inference phylogenetic tree based on 25 partial nad1 gene sequences (747 bp) from various tapeworm isolates of the subfamily Anoplocephalinae (A). Sequences newly reported in the present study are in bold type and identified by name, host and locality of origin. Other sequences are identified by species name and GenBank accession number. The tree is rooted by Echinococcus granulosus (Taeniidae). Posterior probabilities of BI are given above nodes; 1.00 support is indicated by an asterisk. (A, B). Topology of the Bertiella isolates that include the isolate from mountain gorilla after the trimming of nad1 sequences to 366 bp prior Bayesian analysis (B).
Phylogenetic tree of 15 partial 5.8S-ITS2 rRNA gene sequences of isolates from man and primates and selected species of subfamily Anoplocephalinae. Sequences newly obtained in the present study are in bold type and identified by name, host and locality of origin. Other sequences are identified by species name and their GenBank accession number. The tree is rooted by Echinococcus granulosus. Posterior probabilities of BI are given above nodes.
Pairwise genetic distance of cox1 sequences within (red colour) and among (green colour) chosen cestode species (whole analysis not shown). Newly obtained isolates from man, non-human primates and horse are bold typed
PK, Paranoplocephala kalelai; MB, Microticola blanchardi; EG, Eurotaenia gracilis; MV, Microcephaloides variabilis; AL, Anoplocephaloides lemmi; AD, Anoplocephaloides dentate; AK, Anoplocephaloides kontrimavichusi; AP, Anoplocephala perfoliata; ES, Spain; ID, Indonesia; PE, Peru; UG, Uganda; KE, Kenya; CZ, Czech Republic.
The analysis of partial 28S rRNA showed that the two Bertiella isolates from P. troglodytes from Uganda and G. gorilla do not cluster with other Bertiella isolates from marsupials and rodents, but form a separate clade, turning the genus into a paraphyletic assembly. As a consequence, the genus Anoplocephala represents a sister group of Bertiella spp. from marsupials and rodents (Fig. 5).
Phylogenetic tree of 26 partial 28S gene sequences of the subfamily Anoplocephalinae isolates. Sequences newly reported in the present study are in bold type and identified by name, host and locality of origin. Other sequences are identified by species name and GenBank accession number. The tree is rooted by Hymenolepis spp. Posterior probabilities of BI are given above nodes.
DISCUSSION
Despite the fact that anoplocephalid cestodes are known as parasites of humans, primates and other warm-blooded vertebrates for a long time, their taxonomy has far been unsatisfactorily addressed, especially in case of species parasitizing hosts other than rodents. Resolving the taxonomy of species from primates including man is complicated by the fact that the gravid proglottids and eggs obtained from the feces are mostly the only stages diagnosed, whereas the scolexes and larger parts of the strobila, including mature proglottids with taxonomically important structures, are not available (Beveridge, Reference Beveridge, Khalil, Jones and Bray1994; Sun et al. Reference Sun, Fang, Chen, Hu, Xia and Wang2006).
In present study, involving anoplocephalid cestodes from several primate hosts from different continents, the phylogenetic analyses based on nuclear (5.8S-ITS2) and mitochondrial (cox1, nad1) genes showed that all isolates acquired from humans and primates form a monophyletic group. Such a situation is in agreement with an 18S rRNA-based study of human Bertiella isolates and those from the crab-eating macaque M. fascicularis (Furtado et al. Reference Furtado, Batista Ede, Gonçalves, Silva, Melo, Giese and Santos2012; Taleb-Hossenkhan and Bhagwant, Reference Taleb-Hossenkhan and Bhagwant2012). However, clustering within this monophyletic group was inconsistent, probably as a consequence of a low number of, and/or high variability among, the Bertiella isolates. Thus, using the data of Haukisalmi et al. (Reference Haukisalmi, Wickström, Henttonen, Hantula and Gubányi2004, Reference Haukisalmi, Hardman, Hardman, Laakkonen, Niemimaa and Henttonen2007), we compared the genetic distances within and among the genera Paranoplocephala Lühe, 1910, Microcephaloides Haukisalmi, Hardman, Hardman, Rausch and Henttonen, 2008, Anoplocephaloides Baer, 1923 and Bertiella. The results revealed higher heterogeneity of Bertiella compared with that of the other genera, indicating the presence of several species within our dataset (Table 2). Although our collection of isolates from man and primates included also morphologically determined specimens, the isolates studied herein could not be referred to a particular Bertiella species. They are apparently congeneric; however, the lack of clear correlation between hosts or geography makes the identification to the species level impossible.
In morphology-based phylogenetic analyses of the Anoplocephalinae (Beveridge, Reference Beveridge, Khalil, Jones and Bray1994), the genus Anoplocephala appeared as a basal clade for all the other genera of the subfamily, with Bertiella forming a sister lineage to the Triplotaenia-Phascolotaenia-Progamotaenia subclade (Fig. 1A). In contrast to this arrangement, the recent 28S rRNA analyses by Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012) showed that the genus Anoplocephala forms a sister clade to the Bertiella isolates from marsupials and rodents (Fig. 1B). However, Bertiella species from primates, including the type species of the genus B. studeri originating from P. troglodytes (Blanchard, Reference Blanchard1891), have never been included in any previous analyses. Therefore, to amend the phylogenetic arrangement of the genus Bertiella; we added two new isolates from primates to the dataset of Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012). These isolates clearly represented genus Bertiella, as confirmed by comparison of our sequence data with morphologically identified samples from previous studies (Galán-Puchades et al. Reference Galán-Puchades, Fuentes and Mas-Coma2000; Gómez-Puerta et al. Reference Gómez-Puerta, López-Urbina and González2009; Foitová et al. Reference Foitová, Mašová, Tenora, Koubková, Hodová, Vyskočilová, Baruš and Nurcahyo2011). Surprisingly, position of these isolates (Fig. 5) confirmed neither the monophyly of Bertiella, nor its presumed close relationship to the genus Anoplocephala, to which Bertiella from marsupials and rodents remained as the sister group, as inferred by Hardman et al. (Reference Hardman, Haukisalmi and Beveridge2012). Such discrepancy could be explained by existence of paralogous forms of DNA in Bertiella formed due to genome duplications, as proposed to have happened in other cestodes, e.g. Ligula intestinalis Bloch, 1782 and Atractolytocestus huronensis Anthony, 1958 (see Bouzid et al. Reference Bouzid, Štefka, Hypša, Lek, Scholz, Legal, Ben Hassine and Loot2008; Králová-Hromadová et al. Reference Králová-Hromadová, Štefka, Špakulová, Orosová, Bombarová, Hanzelová, Bazsalovicsová and Scholz2010). On the other hand, there were none of the commonly encountered hints present in our sequence data, which are indicative of paralogs, such as double peaks in chromatograms or whole portions thereof unreadable due to confounded sequences (Griffin et al. Reference Griffin, Robin and Hoffmann2011; El-Sherry et al. Reference El-Sherry, Ogedengbe, Hafeez and Barta2013). Thus, the paraphyly between Bertiella from primates and Australian marsupials and rodents can be regarded as a genuine phylogenetic relationship. Supportive of such a hypothesis can be a fact mentioned by Beveridge (Reference Beveridge1985, Reference Beveridge1989) that almost all Bertiella species in marsupials and rodents differ morphologically from Bertiella in primates in mutual position of the uterus and osmoregulatory canals (medullary only in the latter group). This presumed difference between Bertiella species from primates and marsupials/rodents corroborated by our phylogenetic analysis, however, may not be so strict because this character was found also in the African species Bertiella douceti from West African scaly-tailed squirrels of the genus Anomalurus Waterhouse, 1843 (Beveridge Reference Beveridge1985, Reference Beveridge1989). The hypothesis of acquisition of this rodent parasite from primates sharing the same habitat (Baer, Reference Baer1953), as well as any presumptions on co-phylogeny, calls for a deeper analysis.
Although the egg morphology (data not shown) of our isolate from G. beringei corresponds to A. gorillae described by Nybelin (Reference Nybelin1927) and Eilenberger (Reference Eilenberger1998) and also is more similar to A. perfoliata from horses, in molecular phylogenies it invariably emerged within the Bertiella clade. This situation can be explained either as finding of Bertiella sp. newly in mountain gorillas, or, what is more probable, it is a discrepancy between phylogenic position of A. gorillae and its taxonomic placement. In latter case, a reclassification of A. gorillae to the genus Bertiella would solve the problem, however, additional molecular data resulting from well determined material are necessary for final solution.
Molecular markers are undoubtedly a powerful tool for distinguishing individual isolates, but the virtual absence of well-preserved adult tapeworms prevents unambiguous species identification. Although the analyses showed Bertiella spp. from primates monophyletic, the paraphyletic relationship of Bertiella isolates from primates and marsupials/rodents suggests that genus Bertiella, as traditionally acknowledged, and might actually be split into two genera. Then, species included in these genera and host specificity of individual members need to be evaluated using combined morphological and molecular data. The available data indicate that man and other primates can be parasitized by several Bertiella species, though a clear species concept has yet to be elaborated based on good-quality material suitable for morphological study. We conclude that humans and other primates may share different Bertiella species at any place where local ecological conditions allow the transmission of these anoplocephalid cestodes.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://dx.doi.org/10.1017/ S003118201500058.
ACKNOWLEDGEMENTS
The authors would like to thank the State Ministry of Research and Technology (RISTEK) and the Directorate General of Forest Protection and Nature Conservation (PHKA) for their cooperation and for their permission to conduct research in the Gunung Leuser National Park. Research in the Kalinzu Forest Reserve was approved by the Uganda Wildlife Authority and Uganda National Forest Authority. We thank Kateřina Pomajbíková and Martin Mulama for providing samples from the Sweetwaters Chimpanzee Sanctuary (Ol Pejeta, Kenya) and Rui Sá and the director Abubakar Serra of Guinea-Bissau Biodiversity and Protected Areas (IBAP – Instituto para a Biodiversidade e Áreas Protegidas da República da Guinea-Bissau) and the guards that helped with the collection of samples in Guinea Bissau. Special thank belongs to Michael R. Cranfield, other field veterinarians and trackers of Karisoke Research Center and the Rwanda Development board that aided with the collection of the samples. We also thank Ilona Pšenková and Angelique Todd for collecting and transporting material and the government of the Central African Republic and the World Wildlife Fund for providing research permits and logistic help in Dzanga-Sangha Protected Areas. The authors would like to thank Dr Galán-Puchades, Dr Gómez-Puerta and anonymous researcher from Brazil for their tapeworm donation and permission to include their isolates in this study. This publication is an outcome of the HPI-Lab (Laboratory for Infectious Diseases Common to Humans and Non-human Primates).
FINANCIAL SUPPORT
This work was supported by the project ‘CEITEC’ – Central European Institute of Technology (CZ.1.05/1.1.00/02.0068) (D. M., J. D.) from the European Regional Development Fund (D. M.), co-financed from the European Social Fund, from the state budget of the Czech Republic (project OPVK CZ.1.07/2.3.00/20.0300) (D. M., K. J. P., P. V.) and by the Czech Science Foundation (524/06/0264 and 206/09/0927 – D. M. and K. J. P.) and the Czech Grant Agency of the Academy of Sciences of the Czech Republic (M200961204 – J. L., GAP505/11/1163 – I. F.). This study was also financially supported by the Masaryk University (I. F.) and the Institute of Parasitology of the Biology Centre of the Czech Academy of Sciences (RVO: 60077344 – T. S., M. J., J. L.) and by institutional support of Institute of Vertbertae Biology of the Czech Academy of Sciences (RVO: 68081766 – K. J. P.). Field work at Sumatra was financially supported by Foundation UMI-Saving of Pongidaeâ (I. F, W. N.). J. L. and K. J. P. are supported by the Praemium Academiae award and is Fellow of the Canadian Institute for Advanced Research.
