The cestode genus GoezeellaFuhrmann, 1916 is reviewed on the basis of taxonomic evaluation of type and newly collected material from pimelodid and heptapterid catfishes (Siluriformes) in the Amazon River basin, South America, and its generic diagnosis is amended. The genus is typified by the exclusively ventral position of the cortical vitelline follicles, the inner longitudinal musculature formed by dense individual muscle fibres, rather than forming bundles as in other proteocephalids, a well-developed metascolex and biloculate suckers. The type species, G. siluriFuhrmann, 1916, is redescribed based on its syntype from Cetopsis coecutiens (type-host) (Cetopsidae) and specimens from Pinirampus pirinampu (Pimelodidae). The validity of G. danbrooksi de Chambrier, Rego & Mariaux, 2004 from Ageneiosus pardalis (Auchenipteridae) is confirmed and some erroneous morphological traits are corrected based on examination of its holotype. A new species, Goezeella mariae sp. nov., is described from the heptapterid catfish Pimelodella cristata (Heptapteridae). The new species differs from congeners by its overall size (much smaller compared to G. siluri), morphology of the scolex (strongly wrinkled metascolex and weakly developed interlocular septum of suckers) and number of the testes. This is the third proteocephalid cestode described from a heptapterid catfish in South America and the first helminth parasite reported from P. cristata. Goezeella is unusual among other Neotropical proteocephalids by its occurrence in catfishes of as many as four families; all species of the genus are known only from the Amazon and Orinoco River basins. Molecular data on two of the three valid species and a key to their identification are provided.
INTRODUCTION
South American catfishes (Siluriformes) are hosts of a vast diversity of tapeworms (Cestoda) of the recently erected order Onchoproteocephalidea Caira, Jensen, Waeschenbach, Olson & Littlewood, 2014 (see Alves et al., 2017a; Scholz & Kuchta, 2017). This order includes the previously recognized order Proteocephalidea Mola, 1928 (all in one family, Proteocephalidae La Rue, 1911) and some members of the polyphyletic order ‘Tetraphyllidea’ Carus, 1863 (Caira et al., 2014). A total of 87 species of proteocephalids in 36 genera have been reported from South American catfishes (Alves et al., 2017a; de Chambrier et al., 2017), but new taxa are still being described and new genera erected (e.g. Alves et al., 2015, 2017b; Arredondo et al., 2017). Likewise, evaluation of type-specimens and newly collected material made it possible to elucidate the taxonomic status of some previously insufficiently known genera, such as Brayela Rego, 1984, Megathylacus Woodland, 1934 and Chambriella Rego, Chubb & Pavanelli, 1999 (de Chambrier et al., 2014; Alves et al., 2017b).
Goezeella Fuhrmann, 1916 was erected to accommodate Goezeella siluri Fuhrmann, 1916 from the blue whale catfish Cetopsis coecutiens (Lichtenstein) (Siluriformes: Cetopsidae) in the Amazon River basin (Fuhrmann, 1916). This author distinguished Goezeella from Monticellia La Rue, 1911, which has also the testes, ovary, vitelline follicles and uterus in the cortex, by its possession of a well-developed metascolex, formed by an enlargement of the neck (proliferation zone) and the posterior part of the scolex (Fuhrmann, 1916).
Woodland (1933a) confused tapeworms found in the pimelodid catfish Brachyplatystoma vaillantii (Valenciennes) (Pimelodidae) in the Brazilian Amazonia with G. siluri, even though they had eggs with polar projections, whereas G. siluri described by Fuhrmann (1916) possessed ‘typical’ eggs of proteocephalids, i.e. without polar projections. Woodland (1933a) apparently studied a mixture of G. siluri from C. coecutiens and the species later described as Amphoteromorphus praeputialis Rego, Santos & Silva, 1974 [now Brooksiella praeputialis (Rego, Santos & Silva, 1974) Rego, Chubb & Pavanelli, 1999]. For cestodes from B. vaillantii, which are devoid of polar projections on their eggs, Woodland (1933a) proposed a new species, Goezeella piramutab Woodland, 1933; this species was later synonymized with G. siluri by de Chambrier et al. (2004a).
Brooks & Deardorff (1980) reported G. siluri from the auchenipterid catfish Ageneiosus caucanus Steindachner (= syn. of A. pardalis Lütken) in Colombia, but these cestodes differed from G. siluri in several morphological traits, such as the arrangement of the vitelline follicles, number of the testes and relative position of the vaginal sphincter (de Chambrier et al., 2004a). Therefore, the latter authors proposed a new name, G. danbrooksi de Chambrier, Rego & Mariaux, 2004, to accommodate the specimens from A. pardalis, thus adding the second species to the genus.
As part of long-term studies on the diversity, host associations and interrelations of proteocephalid tapeworms in the Neotropical Region (e.g. de Chambrier & Vaucher, 1999; Alves et al., 2015; de Chambrier et al., 2015a), new material of Goezeella spp. from Pinirampus pirinampu (Spix & Agassiz) (Pimelodidae) and Pimelodella cristata (Müller & Troschel) (Heptapteridae) was collected. Evaluation of this new material, supplemented by examination of type-specimens of G. siluri and G. danbrooksi, made it possible to revise the genus including emendation of its diagnosis, to redescribe its type-species and to describe a new species from P. cristata. In addition, an identification key to the three species of the genus is provided.
MATERIAL AND METHODS
The present study is based on the evaluation of the syntypes of Goezeella siluri and the holotype of G. danbrooksi deposited at the Natural History Museum, Geneva and the National Museum of Natural History, Washington, D.C., respectively, and newly collected specimens from Pimelodella cristata and Pinirampus pirinampu.
The newly collected tapeworms were removed from the host's intestine and placed in 0.9% NaCl solution. Several posteriormost proglottids were excised and fixed in 96% molecular-grade ethanol for sequencing of the partial large subunit nuclear ribosomal RNA gene (lsrDNA gene; D1–D3 domains) (see Brabec et al., 2012 for methodology). The anterior parts (hologenophores sensu Pleijel et al., 2008) were placed in a small amount of saline and hot (almost boiling) 4% formaldehyde solution was immediately added to keep the worms straight; after two weeks they were transferred to 70% ethanol before further processing.
Specimens for morphological studies were stained with Mayer's hydrochloric carmine solution, dehydrated in an ethanol series, cleared with eugenol (clove oil) and mounted in Canada balsam. Eggs were drawn in distilled water. For histological sections, pieces of the strobila were embedded in paraffin wax, cross sectioned at 12–15 μm, stained with Weigert's haematoxylin and counterstained with 1% eosin B (acidified with five drops of pure acetic acid for 100 ml solution) following protocol outlined by de Chambrier (2001). For scanning electron microscopy (SEM) observations, scoleces of each species were dehydrated through a graded ethanol series, dried in hexamethyldisilazane, coated with gold and examined in a JEOL JSM-740 1F scanning electron microscope at the Institute of Parasitology, Biology Centre of the Czech Academy of Sciences. Microthrix terminology follows Chervy (2009). All measurements are given in micrometres unless otherwise indicated; abbreviations used in descriptions are: x = mean; n = number of measurements. The fish names follow Froese & Pauly (2017).
Museum acronym abbreviations:
CHIOC
Helminthological Collection of the Instituto Oswaldo Cruz in Rio de Janeiro, Brazil
IPCAS
Institute of Parasitology, Biology Centre of the Czech Academy of Sciences, České Budějovice, Czech Republic
MHNG-PLAT
Platyhelminthes collection of the Natural History Museum of Geneva, Switzerland
USNM
National Museum of Natural History, Washington, D.C., USA
TAXONOMIC ACCOUNT
Goezeella Fuhrmann, 1916
Amended diagnosis: Onchoproteocephalidea, Proteocephalidae. Testes, ovary, vitelline follicles and uterus cortical. Small to medium-sized, robust worms, strobila acraspedote, with longitudinal and transverse grooves; proglottids variable in shape and size. Scolex conical, with rounded to quadrangular apical part, without apical organ, wider than proliferative zone (neck); metascolex well-developed, wrinkled, formed by enlargement of neck. Suckers robust, biloculate; loculi variable in shape and size, interlocular septum conspicuous or not. Internal longitudinal musculature well-developed, with numerous individual fibres not forming bundles; fibres more abundant on lateral sides of proglottids. Testes in one irregular field, in one or two layers, dorsally overlapping cirrus-sac and vitelline follicles. Cirrus-sac elongated to pear-shaped, internal sperm duct thick-walled, strongly coiled in proximal half of cirrus-sac. Genital pores markedly pre-equatorial, irregularly alternating. Genital atrium present, deep. Ovary with medullary isthmus and 2 follicular (grape-like) lobes penetrating inner longitudinal musculature to dorsal cortex; lobes with numerous dorsal outgrowths. Vagina anterior to cirrus-sac, surrounded by numerous chromophilic cells in distal (terminal) part (pars copulatrix vaginae); vaginal sphincter present. Vitelline follicles only on ventral or ventrolateral side of cortex, arranged in two uninterrupted lateral bands, widened towards ovarian level. Uterine development type 2 according to de Chambrier et al. (2004b). Parasites of siluriform catfishes in the Neotropical Region.
Type species: Goezeella siluri Fuhrmann, 1916.
Additional species: Goezeella danbrooksi de Chambrier, Rego & Mariaux, 2004; Goezeella mariae sp. nov.
Remarks: The validity of Goezeella has been questioned since its establishment, mainly due to different assumptions on the taxonomic usefulness of two morphological traits, i.e. the presence or absence of a metascolex and the distribution of the internal organs in relation to the internal longitudinal musculature. Woodland (1925) synonymized Goezeella with Monticellia because they share the cortical position of all internal organs, whereas Harwood (1933) synonymized it with Corallobothrium Fritsch, 1886 based on the presence of a metascolex in species of both genera (see Scholz et al., 2011 for data on Corallobothrium). Based on a cladistic analysis inferred from morphological characters, Brooks (1995) synonymized Choanoscolex La Rue, 1911, Goezeella, Jauella Rego & Pavanelli, 1985 and Peltidocotyle Diesing, 1850 with Spatulifer Woodland, 1934, questioning validity of Peltidocotyle, which should have priority, because of the assumed similarity of the metascolex development shared by these genera (see Brooks & Rasmussen, 1984). However, this synonymy has never been accepted by most of the authors, including Rego et al. (1999), de Chambrier & Vaucher (1999) and de Chambrier et al. (2004a), who recognized Goezeella as a valid genus. Recent molecular data suggest that the above-mentioned features may not be useful to assess the interrelations among proteocephalids and a more natural classification, reflecting the evolutionary history of the group, should be proposed (de Chambrier et al., 2004b, 2015b; Scholz et al., 2013).
As a result, the subfamilial classification proposed by W. N. F. Woodland and widely used for proteocephalids (Rego, 1994; de Chambrier & Vaucher, 1999; Rego et al., 1999), based on the position of the testes, uterus and vitelline follicles in relation to the internal longitudinal musculature (see de Chambrier et al., 2009 and references therein) is not considered in the present account. Nevertheless, for practical reasons, we compared Goezeella with the 11 genera previously placed in the Monticelliinae (Table 1), which are typified by the cortical position of the testes, ovary, vitelline follicles and uterus (Rego, 1994).
Goezeella can be readily differentiated from all proteocephalid genera by the vitelline follicles present only in the ventral cortex, i.e. completely absent dorsally, with lateral bands widened at the ovarian level, and by the possession of the inner longitudinal musculature formed by individual muscle fibres rather than forming compact bundles.
Goezeella siluri
Fuhrmann, 1916
Figs 3, 4, 10-19
Goezeella siluri Fuhrmann, 1916: 385.
Monticellia siluri. – Woodland, 1925: 714.
Corallobothrium siluri. – Harwood, 1933: 140.
Goezeella piramutab Woodland, 1933a: 488.
Monticellia piramutab. – Woodland, 1935: 222.
Spatulifer piramutab. – Brooks & Deardorff, 1980: 17.
Spatulifer siluri. – Brooks, 1995: 365.
Material examined
Syntype: MHNG-PLAT-36375, a whole-mounted specimen (2 slides) and 12 slides of serial cross-sections. Additional specimens: MHNG-PLAT-19858, a whole-mounted specimen (4 slides) and 6 slides of serial cross-sections, collected on 13.04.1992, host field no. Br 237a. – CHIOC 38858, a whole-mounted specimen (one slide) collected on 30.09.1995, host field no. Br 436. – MHNG-PLAT-21845, 21877, 2 uncomplete whole-mounted specimens (3 slides) and 3 slides of serial cross-sections collected on 02.09.1995, host field no. Br 472. – MHNG-PLAT-21879, CHIOC 38859, a whole-mounted specimen (4 slides) and 6 slides of serial cross-sections collected on 19.10.1995, host field No. Br 811; all specimens from P. pirinampu from Itacoatiara, State of Amazonas, Brazil (3°09′S, 58°26′W), collected by A. A. Rego and A. de Chambrier. – MHNG-PLAT-85161, a whole-mounted specimen (3 slides), collected on 07.10.2011, host field no. PI 819a; hologenophore; specimen from P. pirinampu from Iquitos, Region of Loreto, Peru (3°47′S, 73°20′W).
Type locality: Amazon River basin (specific locality unknown).
Other localities: Delta of the Orinoco River, Venezuela; Amazon River, Itacoatiara, State of Amazonas, Brazil; Amazon River, Iquitos, Region of Loreto, Peru.
Type host: Blue whale catfish Cetopsis coecutiens (Lichtenstein, 1819) (Siluriformes: Cetopsidae).
Other definitive hosts: Cetopsis othonops (Eigenmann) (Siluriformes: Cetopsidae) and Brachyplatystoma vaillantii (Valenciennes) (Siluriformes: Pimelodidae) (doubtful host – see remarks below); Pinirampus pirinampu (Spix & Agassiz) (Siluriformes: Pimelodidae).
Site of infection: Anterior intestine.
Prevalence of infection: Fuhrmann (1916) found 4 tapeworms in one C. coecutiens, but did not provide the total number of fish examined; 43 P. pirinampu examined/8 fish infected (19%) from Itacoatiara, Brazil.
Morphological description: Fuhrmann (1916), de Chambrier et al. (2004a), present study.
Representative DNA sequences: A fragment of 1491 bp long of the lsrDNA (D1–D3 domains) (GenBank MF370207). The newly generated sequence of an isolate of G. siluri from P. pirinampu is 483 bp longer than the sequence from the same host and river basin available in the GenBank (AJ388612).
Redescription (based on 6 whole worms, serial cross-sections of mature proglottids and 1 scolex studied using SEM from P. pirinampu; measurements taken from the syntype in brackets): Proteocephalidae. Testes, ovary, vitelline follicles and uterus cortical; medium-sized worms. Total body length 90–230 mm (n = 3), maximum width up to 3 mm (n = 3). Strobila acraspedote, anapolytic, with longitudinal and transverse grooves, consisting of about 335–360 proglottids: 195–200 immature (up to appearance of spermatozoa in vas deferens), 45–55 mature (up to appearance of eggs in uterus), 50–55 pregravid (up to appearance of hooks in oncospheres) and 45–50 gravid. Immature and mature proglottids much wider than long (length: width ratio 0.10–0.20), pregravid proglottids wider than long (length: width ratio 0.45–0.55), gravid proglottids wider than long to longer than wide (length: width ratio 0.80–1.93).
Scolex 1.18-1.60 × 1.45-1.94 mm (n = 5), much wider than neck (proliferation zone), 1.77-1.80 × 1.40-1.45 mm, bearing 4 robust suckers, biloculate, with loculi variable in size; anterior loculus 210-295 (x = 255; n = 5) in diameter, posterior loculus 205-255 (x = 217; n = 5) in diameter; conspicuous septum separating each loculus (Figs 11, 15). Metascolex present, uniformly wrinkled. Apex conical, lacking apical organ, with numerous gland-cells (Figs 3, 4, 11, 15). Apex of scolex and lumen of suckers covered with acicular filitriches, similar in density (not shown); surface between suckers, base of metascolex and neck covered with capilliform filitriches, similar in density (not shown); pregravid proglottids covered with acicular filitriches (Fig. 10).
Inner longitudinal musculature well-developed, composed by numerous, individual muscle fibres, concentrated on lateral sides of proglottids (Figs 13, 17). Osmoregulatory canals situated at same level, median to lateral-most testes and vitelline follicles, almost straight (Figs 16, 17); ventral osmoregulatory canal wider than dorsal one (Figs 16, 17).
Testes numerous, spherical, small, 55–80 in diameter, in 1 or 2 irregular layers, 282–366 (x = 314; n = 13) [380–430] per mature proglottids (Figs 12, 16, 18). Testes form 1 irregular field on dorsal side, less numerous in median line of proglottids (uterine stem), usually surpassing osmoregulatory canals, dorsally overlapping cirrus-sac, vitelline follicles and sometimes ovary (Figs 12, 13, 16, 18). Testes present also in gravid proglottids.
Vas deferens coiled, with loops forming elongated field reaching to, but not crossing, median line of proglottid (Figs 12, 16, 18). Cirrus-sac pear-shaped, thin-walled, slightly widened towards distal (terminal) part (Figs 12, 13, 16, 18), 220-340 × 95-145 (n = 13) [250-300 × 80-110], its length representing 10–21% (x = 15; n = 13) [14–25%] of proglottid width. Sperm duct (internal vas deferens) sinuous. Cirrus muscular, reaching up to 82% (n = 13) [50%] of cirrus-sac length. Common genital atrium narrow, deep (Figs 12, 13, 16, 18). Genital pores alternating irregularly, markedly pre-equatorial, situated at 3–10% (x = 6; n = 13) [5–14%] of proglottid length from anterior margin (Figs 12, 16, 18).
Ovary with wide isthmus in medulla and 2 follicular, grape-like lobes penetrating inner longitudinal musculature to dorsal cortex; numerous dorsal outgrowths present (Figs 13, 17). Length of ovary represents 27–37% (x = 30%; n = 13) [33–45%] of proglottid length, its width representing 57–76% (x = 68%; n = 13) [74–86%] of proglottid width (Figs 16, 17). Mehlis' gland about 110–165 in diameter, representing 7–20% of proglottid width (n = 13). Relative ovarian size, i.e., percentage of ovary surface to total surface of mature or pregravid proglottids (see de Chambrier et al., 2012), 11–16% (x = 13%; n = 10).
Vaginal canal slightly sinuous, surrounded by chromophilic cells, wider in terminal part (pars copulatrix vaginae), possessing a terminal vaginal sphincter (Figs 12, 16, 17). Vagina anterior to cirrus-sac (n = 35). Vitelline follicles cortical, ventral, forming 2 long uninterrupted bands, occupying large triangular field, widened and confluent posteriorly at ovary level (Figs 13, 16, 17). Length of bands represents 77–94% (x = 86%) [94–98%] and 84–98% (x = 90%; n = 13) [91–98%] of length of proglottid on poral and aporal side, respectively (Figs 12, 16).
Uterus cortical, with development of type 2 (see de Chambrier et al., 2004b, 2015b); uterine stem and diverticula (lateral uterine branches) in mature and pregravid proglottids lined with numerous chromophilic cells, extended much beyond branches (Fig. 16). Uterus with 20–26 [17–22] lateral diverticula on each side (Fig. 16). Eggs oval, outer envelope 22-23 × 19-20, bilayered embryophore 17-18 × 12-14 [21-22 ×14-22], oncosphere 11-12 × 9-10 [10-11 × 7-8], embryonic hooks 6-7 long (Fig. 19).
Remarks: The original description of Goezeella siluri was detailed and contained basic measurements and illustrations (Fuhrmann, 1916). However, the only complete specimen preserved is slightly decomposed, contracted and twisted on the slide. Moreover, Fuhrmann (1916) overlooked an important feature of the scolex morphology, i.e. the presence of biloculate, rather than uniloculate, suckers (compare the same scolex drawn in his Fig. 2 and Fig. 11 of the present study). This characteristic was first reported by Rego (1975; see his fig. 8), but not by Woodland (1933a) as reported by Brooks & Deardorff (1980). In fact, morphology of the suckers may be difficult to observe, especially if the suckers are hidden within wrinkles of a contracted metascolex (see Fig. 3).
The specimens from P. pirinampu are considered conspecific with G. siluri even though they differ from those found in C. coecutiens described by Fuhrmann (1916) in the lower number of the testes (282–366 vs. 380–430), a longer cirrus (reaching up to 82% of the cirrus-sac length vs. reaching only up to 50%), and in the anterior extent of the bands of vitelline follicles on the poral side, which represents 77–94% of proglottids length vs. 94–98% in the syntype. These differences are considered to be accounted for by intraspecific variability, but poor quality of the type material of G. siluri should also be considered. Therefore, new material of G. siluri from its type host, C. coecutiens, is needed for confirmation of measurements taken from the type specimen.
Specimens from Pseudocetopsis othonops (syn. of Cetopsis othonops) in the Orinoco River, Venezuela, reported by Brooks & Rasmussen (1984) as G. siluri, possess the vagina anterior or posterior to the cirrus-sac, unlike the exclusively anterior position in all species of Goezeella including G. siluri and the newly described species (see below). This feature has been broadly used to differentiate species of proteocephalid cestodes (Arredondo & Gil de Pertierra, 2010; Gil de Pertierra & de Chambrier, 2013), but Alves et al. (2017b) described conspecific cestodes (genetically identical individuals based on the lsrDNA) with both the anterior and posterior vagina in relation to the cirrus-sac. Therefore, this character should be considered with caution. Since all specimens from P. othonops were fixed in AFA (alcohol-formalin-acetic acid), they could not be used for DNA sequencing.
Brooks & Rasmussen (1984) also reported immature cestodes identified as G. siluri from B. vaillantii, but reliable identification of juvenile specimens is almost never possible because they lack key morphological traits that are present only in their mature forms. Two of the present authors (TS and AdC) have not found G. siluri in any of the 25 B. vaillantii examined (de Chambrier et al., 2015a); therefore, this record of G. siluri in a species of Brachyplatystoma Bleeker requires verification.
Table 1.
Selected morphological traits of the genera previously placed in the Monticelliinae (Cestoda: Proteocephalidae), parasites of freshwater fishes in the Neotropical region.
Figs 1–10.
Scanning electron micrographs of Goezeella spp. (1, 2, 5–9) Goezeella mariae n. sp. ex Pimelodella cristata (IPCAS C-759). (1, 2) Scoleces, dorsoventral and apical views, respectively. (5–9) Microtriches on the apex of scolex, lumen of suckers, surface between suckers and base of metascolex and neck, respectively. (3, 4, 10) Goezeella siluri Fuhrmann, 1916 ex Pinirampus pirinampu (MHNG-PLAT-21908). (3, 4) Scoleces, dorsoventral and apical views, respectively. (10) Microtriches on the surface of pregravid proglottids. Note: small black numbers correspond to the figures showing higher magnification images of these surfaces.
Figs 11–14.
Goezeella siluri Fuhrmann, 1916 ex Cetopsis coecutiens (syntype MHNG-PLAT-36375). (11) Scolex, apical view. (12) Terminal genitalia, dorsal view. (13) Cross-section at ovary level. (14) Eggs drawn in distilled water. Abbreviations: al – anterior sucker loculus; cc – chromophilic cells; ci – cirrus; cs – cirrus-sac; em – bilayered embryophore; ga – genital atrium; lm – internal longitudinal musculature; oe – outer envelope; on – oncosphere; oo – ovary outgrowths; ov – ovary; pl – posterior sucker loculus; te – testes; vi – vitelline follicles; vs – vaginal sphincter.
Fig. 15.
Goezeella siluri Fuhrmann, 1916 ex Pinirampus pirinampu (MHNG-PLAT-21845). Scolex, dorsoventral view. Abbreviations: al – anterior sucker loculus; gc – glandular cells; pl – posterior sucker loculus.
Goezeella danbrooksi
de Chambrier, Rego & Mariaux, 2004
Figs 20, 21
Goezeella siluri. – Brooks & Deardorff, 1980: 15.
Holotype: USNM 1370061 (USNPC 74498), a whole-mounted specimen (2 slides) – USNM 1370107 (USNPC 74544), fragments on 19 slides of serial cross- and frontal sections.
Type and only known locality: Magdalena River near San Cristóbal, Province of Bolívar, Colombia.
Type and only known host: Ageneiosus pardalis Lütken (= A. caucanus Steindachner) (Siluriformes: Auchenipteridae).
Site of infection: Anterior intestine.
Prevalence of infection: Unknown.
Morphological description: Brooks & Deardorff (1980).
Remarks: The description of this species, which was originally identified as G. siluri by Brooks & Deardorff (1980), was based on a single specimen, which is partially decomposed. However, de Chambrier et al. (2004a) observed several morphological differences between this specimen and G. siluri, such as the position of the vitelline follicles, which are ventrolateral in G. danbrooksi (vs. only ventral in G. siluri), fewer testes (183–310 vs. 380–430), and the position of the vaginal sphincter (at a distance from the genital atrium in G. danbrooksi vs. terminal, i.e. close to the genital atrium in G. siluri). Based on these differences, de Chambrier et al. (2004a) proposed the new name, G. danbrooksi, for the tapeworm from A. pardalis.
Brooks & Deardorff (1980) reported the vitelline follicles to be present both in the ventral and dorsal cortex. However, the study of the holotype including its histological sections revealed that the follicles are only on the ventral side of the cortex, reaching only to its lateral margin (Figs 20, 21).
The study of the type material of G. danbrooksi also showed that some fibres of the inner longitudinal musculature are close together, thus somewhat resembling muscle bundles (Figs 20, 21). However, the tissue of histological sections is partially decomposed and thus a new, well fixed material of the species has to be examined to reveal the actual structure of the inner longitudinal muscles of G. danbrooksi.
Goezeella mariae
sp. nov.
Figs 1, 2, 5–9, 22-30
Holotype: CHIOC 38860a-f, a whole-mounted specimen (1 slide) and 5 slides of serial cross-sections, collected on 25.05.2013, host field no. BR AMP 46a. – MHNG-PLAT-97017 (2 slides of cross-sections).
Paratypes: IPCAS C-759, a whole-mounted specimen (1 slide; hologenophore), host field no. BR AMP 106b. – CHIOC 38861, IPCAS C-759, MHNG-PLAT-86883, a whole-mounted specimen (one slide), 6 slides of serial cross-sections and 2 slides of sagittal sections of scolex, host field No. BR AMP 109a. – MHNG-PLAT-97016, a whole-mounted specimen (one slide; SEM voucher), host field no. BR AMP 111b; all specimens collected on 25.05.2013.
Type and only known locality: Lowermost Amazon River near Macapá, State of Amapá, Brazil (00°01′N, 50°59′W).
Type and only known host: Pimelodella cristata (Müller & Troschel) (Siluriformes: Heptapteridae).
Site of infection: Anterior intestine.
Prevalence: 7 fish examined/4 fish infected (57%).
Representative DNA sequences: A fragment 1491bp long of the lsrDNA (D1–D3 domains) (GenBank MF370208).
Etymology: The species is dedicated to the first author's mother, Maria Thereza Vieira Pinto Alves, for providing continuous support for his studies.
Description: (based on 4 whole-mounted worms; 13 slides with serial cross-sections of mature proglottids and 2 slides with sagittal sections of 1 scolex; 1 scolex studied using SEM). Proteocephalidae. Testes, ovary, vitelline follicles and uterus cortical; small-sized worm. Total body length 14–38 mm (n = 3), maximum width up to 1.3 mm (n = 3). Strobila acraspedote, anapolytic, with longitudinal and transverse grooves, consisting of about 40–90 proglottids: 27–32 immature, 4–8 mature, 10–15 pregravid and 23–39 gravid. Immature and mature proglottids much wider than long (length: width ratio 0.17–0.35), pregravid proglottids markedly wider than long (length: width ratio 0.40–0.76) and gravid proglottids slightly wider than long to much longer than wide (length: width ratio 0.80–2.60).
Scolex 0.68-0.83 × 0.91-1.15 mm (n = 3), much wider than neck (proliferation zone), 0.83-1.20 × 0.75-0.81 mm, bearing 4 biloculate suckers, with loculi unequal in size; anterior loculus 158–161 (x = 160; n = 3) in diameter, posterior loculus 123–126 (x = 125; n = 3) in diameter; loculi separated by inconspicuous interlocular septum (Figs 1, 2, 22, 23). Metascolex present, more wrinkled than neck (Figs 1, 2, 22). Apex rounded, lacking apical organ, with few gland cells (Figs 1, 2, 22, 23). Apex of scolex, lumen of suckers, surface between suckers and base of metascolex covered with acicular filitriches, less dense on lumen of suckers (Figs 5–8); neck covered with capilliform filitriches (Fig. 9).
Inner longitudinal musculature well-developed, formed by numerous, individual muscle fibres not forming compact bundles, more concentrated laterally (Figs 28-30). Osmoregulatory canals situated at same level of lateral-most testes, median to vitelline follicles, markedly sinuous (Figs 24–26, 28-30); ventral osmoregulatory canal wider than dorsal one (Fig. 30).
Testes numerous, spherical to oval, small, 34–46 in diameter, in 1 irregular layer, 103–167 (x = 134; n = 13) per mature proglottid (Figs 24, 25). Testes form 1 irregular field on dorsal side, less numerous alongside median line of proglottids (uterine stem), usually reaching laterally to osmoregulatory canals, dorsally overlapping cirrus-sac, vitelline follicles and sometimes ovary (Figs 24–26, 28-30). Testes present also in gravid proglottids.
Vas deferens coiled, with loops forming elongate field reaching to, but not crossing, median line of proglottid (Figs 24–26). Cirrus-sac elongated to pear-shaped, thin-walled (Figs 24–26, 28), 130–213 × 59–85 (n = 13), its length representing 11–24% (x = 17; n = 13) of proglottid width. Sperm duct (internal vas deferens) sinuous (Figs 24–26, 28). Cirrus muscular, reaching up to 64% (n = 13) of cirrus-sac length. Common genital atrium narrow, deep (Figs 24–26, 28). Genital pores alternating irregularly, markedly pre-equatorial, situated at 7–17% (x = 11; n = 13) of proglottid length from anterior margin (Figs 24–26).
Ovary with wide isthmus in medulla and two follicular, grape-like lobes penetrating to dorsal cortex; numerous dorsal outgrowths present (Figs 24, 25, 30). Length of ovary represents 21–31% (x = 26%; n = 13) of proglottid length, its width representing 57–77% (x = 66%; n = 13) of proglottid width (Figs 24, 25). Mehlis' gland about 60–138 in diameter, representing 8–11% of proglottid width (n = 13). Relative ovarian size, i.e., percentage of ovary surface to total surface of mature or pregravid proglottids (see de Chambrier et al., 2012), 10–15% (x = 12%; n = 13).
Vaginal canal slightly sinuous, surrounded by chromophilic cells, wider in terminal part (pars copulatrix vaginae); terminal vaginal sphincter present (Figs 24–26). Vagina anterior to cirrus-sac (n = 32). Vitelline follicles cortical, ventral, forming 2 long uninterrupted bands, occupying large triangular field, widened and confluent posteriorly at ovary level; lateral to lateral-most testes (Figs 24–26, 29). Length of bands represents 73–91% (x = 83%) and 72–91% (x = 83%) of length of proglottid on poral and aporal side, respectively (n = 13) (Figs 24, 25). Uterus cortical, with development of type 2 (see de Chambrier et al., 2004b, 2015b); uterine stem and diverticula (lateral uterine branches) in mature and pregravid proglottids lined with numerous chromophilic cells, extended much beyond branches (Fig. 24). Uterus with 14–25 lateral diverticula on each side (Figs 24, 25). Eggs oval, outer envelope 21-25 × 18-19, bilayered embryophore 19-20 × 13-15, oncosphere 9-10 × 11-12, embryonic hooks 5-6 long (Fig. 27).
Remarks: Goezeella mariae sp. nov. differs from G. siluri and G. danbrooksi in having fewer testes (103–167 vs. 183–310 and 282–366 in G. danbrooksi and G. siluri, respectively) and inconspicuous interlocular septum (not obvious in SEM images; see Figs 1, 2), rather than the septum conspicuous as in the two other species. The new taxon can be further distinguished from G. siluri by its smaller dimensions, such as the total body length (14–38 mm vs. 90–230 mm), scolex width (0.91–1.15 mm vs. 1.45–1.94 mm) and the length of the cirrus-sac (130–213 μm vs. 220–340 μm) as well as the appearance of the metascolex, which is more wrinkled in G. mariae sp. nov. compared to that of G. siluri; compare Figs 1, 2 with Figs 3, 4. Moreover, G. mariae sp. nov. possesses a terminal, rather than markedly subterminal, vaginal sphincter as it is in G. danbrooksi.
The new species differs in its sequence of the partial lsrDNA gene (D1–D3 domains) from that of G. siluri from P. pirinampu in 14 nucleotides, i.e. genetic divergence 0.9%. A phylogenetic analysis (data not shown) revealed both taxa clustered in a clade comprising also both known species of Gibsoniela Rego, 1984, i.e. G. mandube (Woodland, 1935) and G. meursaulti de Chambrier & Vaucher, 1999, parasites of the auchenipterid catfish Ageneiosus inermis (Linnaeus, 1766) in the Neotropical Region, but interrelations within this lineage remain unresolved. Close relationship of species of Goezeella with those of the genus Gibsoniela is not evident based on their morphology, because they differ in the position of the internal organs in relation to the inner longitudinal musculature (previously used to distinguish individual subfamilies – see Rego, 1994), but also by the morphology of the scolex (no metascolex in the latter genus) and their suckers (biloculate in Goezeella vs. triloculate in Gibsoniela) (Rego, 1984; de Chambrier & Vaucher, 1999).
To the best of our knowledge, this is the first parasite found in Pimelodella cristata. This heptapterid catfish was described from a tributary of the Branco River, Guyana (Bockmann & Guazzelli, 2003) and is distributed throughout the Amazon River basin, inhabiting the sand bottom of creeks and rivers (Reis & Lima, 2009). Proteocephalus bagri Holcman-Spector & Mañé-Garzón, 1988 and P. rhamdiae Holcman-Spector & Mañé-Garzón, 1988, both from Rhamdia sapo (Valenciennes) [syn. of Rhamdia quelen (Quoy & Gaimard)] in Uruguay, are the only other proteocephalids known from heptapterids in South America (Holcman-Spector & Mañé-Garzón, 1988). In addition, Proteocephalus brooksi García-Prieto, Rodríguez & Pérez-Ponce de León, 1996 was described from Rhamdia guatemalensis (Günther) in Mexico by García-Prieto et al. (1996).
Figs 16–19.
Goezeella siluri Fuhrmann, 1916 ex Pinirampus pirinampu. (16) Pregravid proglottid, ventral view (MHNG-PLAT-21877). (17) Cross-section at ovary level (MHNG-PLAT-21879). (18) Terminal genitalia, dorsal view (MHNG-PLAT-21879). (19) Eggs drawn in distilled water (MHNG-PLAT-19858). Abbreviations: cc – chromophilic cells; ci – cirrus; cs – cirrus-sac; do – dorsal osmoregulatory canal; em – bilayered embryophore; ga – genital atrium; ln – longitudinal nerve cord; oe – outer envelope; on – oncosphere; oo – ovary outgrowths; te – testes; vi – vitelline follicle; vo – ventral osmoregulatory canal; vs – vaginal sphincter.
Figs 20–21.
Goezeella danbrooksi de Chambrier, Rego & Mariaux, 2004 ex Ageneiosus pardalis (holotype, USNM 1370061). (20) Cross-section at middle part of proglottid. (21) ovary level. Abbreviations: do – dorsal osmoregulatory canal; lm – internal longitudinal musculature; ov – ovary; st – subtegumental muscle fibres; su – subtegumental cells; te – testes; tg – tegument; ut – uterus; vi – vitelline follicles; vo – ventral osmoregulatory canal.
Figs 22–27.
Goezeella mariae sp. nov. ex Pimelodella cristata. (22) Scolex, dorsoventral view (holotype, CHIOC 38860a). (23) Scolex, sagittal section (paratype, CHIOC 38861). (24) Pregravid proglottid, ventral view (paratype MHNG-PLAT-97016). (25) Gravid proglottid, dorsal view (holotype, CHIOC 38860a). (26) Terminal genitalia, dorsal view (holotype, CHIOC 38860a). (27) Eggs drawn in distilled water (paratype IPCAS C-759). Abbreviations: al – anterior sucker loculus; cc – chromophilic cells; ci – cirrus; cs – cirrus-sac; do – dorsal osmoregulatory canal; em – bilayered embryophore; ga – genital atrium; oe – outer envelope; on – oncosphere; pl – posterior sucker loculus; te – testes; vi – vitelline follicles; vo – ventral osmoregulatory canal; vs – vaginal sphincter.
Fig. 28.
Goezeella mariae sp. nov. ex Pimelodella cristata. Cross-section at the cirrus-sac level (holotype, CHIOC 38860d). Abbreviations: cc – chromophilic cells; ci – cirrus; cs – cirrus-sac; ed – ejaculatory duct; va – vas deferens.
Figs 29–30.
Goezeella mariae sp. nov. ex Pimelodella cristata. (29) Cross-section at middle part of proglottid (holotype, CHIOC 38860d). (30) Cross-sections at ovary level (paratype MHNG-PLAT-86883). Abbreviations: do – dorsal osmoregulatory canal; lm – internal longitudinal musculature; ln – longitudinal nerve cord; ov – ovary; st – subtegumental muscle fibres; su – subtegumental cells; te – testes; tg – tegument; ut – uterus; vi – vitelline follicles; vo – ventral osmoregulatory canal.
DISCUSSION
The present study provides a new insight into the taxonomy, species composition and host associations of Goezeella, one of insufficiently known genera of Neotropical proteocephalid cestodes. An important feature newly added into an amended generic diagnosis is the unique arrangement of the inner longitudinal musculature, which does not form compact bundles as usually in proteocephalids (Rego, 1994). In contrast, the musculature in species of Goezeella is formed by numerous individual muscle fibres not forming such bundles. The exclusively ventral or ventrolateral position of the vitelline follicles and the possession of a metascolex with numerous, largely longitudinal wrinkles and biloculate suckers, represent other typical characteristics of this genus.
Species of Rudolphiella Fuhrmann, 1916 and Ephedrocephalus Diesing, 1850 also have the vitelline follicles restricted to the ventral cortex (not in the dorsal field as given in the key of de Chambrier et al., 2009; see Rego et al., 1999 and Gil de Pertierra & de Chambrier, 2000). However, they differ markedly from Goezeella by their topography of the genital organs in relation to the inner longitudinal musculature (see Rego, 1994) and possess the uniloculate, rather than biloculate, suckers (Mola, 1906; Gil de Pertierra & de Chambrier, 2000). Spatulifer maringaensis Pavanelli & Rego, 1989, which also shares the same distribution of the internal organs in relation to the inner musculature with Goezeella (both were previously placed in the Monticelliinae), and possesses a metascolex and the vitelline follicles limited to the ventral cortex (Arredondo & Gil de Pertierra, 2008), bears only uniloculate suckers, rather than biloculate as in species of Goezeella.
Identification key to the species of Goezeella Fuhrmann, 1916
1A Vaginal sphincter at a distance from genital atrium; vitelline follicles ventrolateral. In auchenipterids (Ageneiosus pardalis) G. danbrooksi
1B Vaginal sphincter near genital atrium (terminal); vitelline follicles only on ventral side of cortex, missing laterally 2
2A Much more than 200 testes per proglottid. In cetopsids (Cetopsis) and pimelodids (Pinirampus) G. siluri
2B Much fewer than 200 testes per proglottid. In heptapterids (Pimelodella cristata) G. mariae sp. nov.
Species of Goezeella including G. mariae sp. nov. have been found in the Amazon and Orinoco River basins, whereas no published record exists from the Paraná River basin. The spectrum of definitive hosts of this genus, i.e. fishes of four siluriform families (Auchenipteridae, Cetopsidae, Heptapteridae and Pimelodidae) is markedly wider than typical for comparably species-rich genera of Neotropical proteocephalids, which are usually specific to only a single fish family (Rego et al., 1999; de Chambrier et al., 2015a).
Several attempts have been made to assess the phylogenetic relationships of the species-rich fauna of Neotropical proteocephalids, but their actual interrelations remain unclear (Zehnder & Mariaux, 1999; de Chambrier et al., 2004b; Hypša et al., 2005; de Chambrier et al., 2015a). Neither the addition of terminal taxa, nor the use of new molecular markers seem to have improved the phylogenetic signal of the large polytomy of the ‘Neotropical superclade’ (de Chambrier et al., 2015a), where G. siluri ex P. pirinampu appears together with 37 taxa (including 29 taxa from Neotropical fishes). This lack of phylogenetic resolution was also observed in the recent published studies by Alves et al. (2017b) and Arredondo et al. (2017). Therefore, no tree is presented in this paper. Addition of a new sequence (a representative of G. mariae n. sp.) also did not increase resolution of this clade, but revealed this new taxon within a small clade comprising G. siluri and both species of Gibsoniela. In contrast, molecular data on G. danbrooksi are not available to assess its interrelations to the remaining congeners and thus new material should be collected, similarly as specimens of G. siluri from the type host. All attempts of the present authors, who examined seven C. coecutiens between 1992 and 1995, to find this cestode in this fish host have failed.
Caira et al. (2014) proposed a new order, Onchoproteocephalidea Caira, Jensen, Waeschenbach, Olson & Littlewood, 2014, to group together phylogenetically closely related, but morphologically and ecologically conspicuously dissimilar hook-bearing parasites of elasmobranchs, placed previously in the ‘tetraphyllidean’ family Onchobothriidae Braun, 1900, with parasites of freshwater bony fishes, amphibians and reptiles of the former order Proteocephalidea Mola, 1926. The authors speculated that the allegedly synapomorphy of both distinct groups could be the presence of gladiate spinitriches on the strobila. However, these spinitriches were observed neither in G. siluri nor G. mariae sp. nov., similarly as they are absent in other proteocephalids studied using SEM (e.g. Arredondo et al., 2017). Therefore, identification of morphological synapomorphies for members of this newly erected, conspicuously heterogeneous order is still pending (Arredondo et al., 2014).
To help identify the species of Goezeella, a simple morphology-based key is presented above.
ACKNOWLEDGEMENTS
We are grateful to Jean Mariaux (Subject Editor of RSZ) and two anonymous reviewers for valuable comments, and to Amilcar Arandas Rego (Instituto Oswaldo Cruz, Rio de Janeiro, Brazil), Vivian Suane (Universidade Federal Rural do Estado do Rio de Janeiro, Seropédica, Brazil), Aristides Ferreira (Universidade Federal do Amapá, Macapá, Brazil) and Marcos Brito (Universidade Federal do Oeste do Pará, Santarém, Brazil) for invaluable help with collection and dissection of fish. Thanks are also due to Roman Kuchta (Institute of Parasitology, České Budějovice, Czech Republic) for help with SEM micrographs, to Janik Pralong and Gilles Roth (Natural History Museum, Geneva, Switzerland) for technical help and to Marcos Tavares-Dias (Embrapa Amapá, Macapá, Brazil) for providing the facilities during the field trip to Macapá in 2013. The first author (PVA) thanks Jean Mariaux (Natural History Museum of Geneva, Switzerland) for enabling him to realise a research stay at the Museum. TS, whose stay in the USA in 2017 was supported by the Fulbright Commission, is much obliged to Anindo Choudhury (St. Norbert College, De Pere, Wisconsin) and Anna Phillips (Curator, National Museum of Natural History, Washington, D.C.), to make him possible to study the holotype of G. danbrooksi. This study was supported by the ‘Ciência sem fronteiras' Brazilian program – visitant researcher modality (No. 135/2012) (stays of TS in Brazil at the Universidade Federal Rural de Rio de Janeiro), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) grants to José Luis Luque (Nos. 474077/2011-0, 304254/2011-8, 402665/2012-0), National Science Foundation (PBI awards Nos. 0818696 and 0818823), Institute of Parasitology (institutional support RVO 60077344) and Czech Science Foundation (P505/12/G112). PVA was supported by postgraduate fellowship from CNPq. This study is part of the PhD thesis of the first author (PVA).
