Diphyllobothrium sprakeri n. sp. (Cestoda: Diphyllobothriidae): a hidden broad tapeworm from sea lions off North and South America

Background The systematic of several marine diphyllobothriid tapeworms of pinnipeds has been revised in recent years. However, 20 species of Diphyllobothrium from phocids and otariids are still recognized as incertae sedis. We describe a new species of Diphyllobothrium from the intestine of California sea lions Zalophus californianus (Lesson) (type-host) and South American sea lions Otaria flavescens (Shaw). Methods Zalophus californianus from the Pacific coast of the USA and O. flavescens from Peru and Argentina were screened for parasites. Partial fragments of the large ribosomal subunit gene (lsrDNA) and the cytochrome c oxidase subunit 1 (cox1) mitochondrial gene were amplified for 22 isolates. Properly fixed material from California sea lions was examined using light and scanning electron microscopy. Results A total of four lsrDNA and 21 cox1 sequences were generated and aligned with published sequences of other diphyllobothriid taxa. Based on cox1 sequences, four diphyllobothriid tapeworms from O. flavescens in Peru were found to be conspecific with Adenocephalus pacificus Nybelin, 1931. The other newly generated sequences fall into a well-supported clade with sequences of a putative new species previously identified as Diphyllobothrium sp. 1. from Z. californianus and O. flavescens. A new species, Diphyllobothrium sprakeri n. sp., is proposed for tapeworms of this clade. Conclusions Diphyllobothrium sprakeri n. sp. is the first diphyllobothriid species described from Z. californianus from the Pacific coast of North America, but O. flavescens from Argentina, Chile and Peru was confirmed as an additional host. The present study molecularly confirmed the first coinfection of two diphyllobothriid species in sea lions from the Southern Hemisphere. Supplementary Information The online version contains supplementary material available at 10.1186/s13071-021-04661-1.

composed of two orders, Tetrabothriidea with only seven members of the genus Anophryocephalus Baylis, 1922, and Diphyllobothriidea with around 30 species of several genera [7,8]. Waeschenbach et al. [4] revised the systematic of diphyllobothriids based on morphological as well as molecular data and proposed systematic changes mainly in the polyphyletic genus Diphyllobothrium Cobbold, 1858, which now includes only seven species parasitizing cetaceans. Non-monophyletic Diphyllobothrium (designated as 'Diphyllobothrium' by Waeschenbach et al. [4]) provisionally comprises 20 species mainly from pinnipeds (eight of which have been characterized

Background
The systematic of the tapeworms (Cestoda) has been under detailed revision since the last decade including adding new molecular data [1]. However, current knowledge and systematics of the tapeworms of marine mammals are still superficial [2][3][4][5][6][7]. The pinnipeds are well-known hosts of several groups of tapeworms

Sample collection
A total of 39 CSLs, 9-10 months to 16 years old, were collected stranded on the Pacific coast of central California (36°57′-38°32′N, 121°95-123°00′W), USA, between 2012 and 2018 (see [18] for details). CSLs died in the Marine Mammal Center (Sausalito, California, USA) from different causes and were necropsied using a standard procedure [26]. Two recently dead SASLs were collected from two localities in South America: (i) a subadult male from Bellavista beach (12°04′S, 77°07′W), Callao City, Callao, Peru, in October 2017; (ii) a subadult female from Playa Unión (43°19′S, 65°03′W), Chubut, Argentina, in October 2013. The intestines of fresh SASLs were excised from the carcasses, opened and washed with tap water through a series of sieves. Intestinal contents were placed in Petri dishes with saline and examined under a dissecting microscope. Tapeworms were washed in saline and killed with hot (90 °C) tap water and fixed in 70% ethanol. A few posterior proglottids of selected specimens were cut off and fixed in molecular-grade ethanol (99%) for DNA sequencing before killing the worm.

Molecular data and phylogenetic analyses
Three diphyllobothriid specimens from CSLs and 18 from SASLs (17 from Peru and one from Argentina) were selected for molecular studies. Pieces of strobila were used for DNA isolation and sequencing. The remaining parts of the worms were stained and mounted in Canada balsam and were kept as molecular vouchers (i.e. hologenophores sensu Pleijel et al. [27]). Total genomic DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. Phylogenetic relationships of the studied diphyllobothriids were evaluated based on two molecular markers: the large subunit nuclear ribosomal RNA gene (lsrDNA) and the cytochrome c oxidase subunit 1 (cox1) mitochondrial gene.
Sequences were assembled and inspected for errors using Geneious v.11 and deposited in the GenBank database (accession numbers MW600336-MW600339 for the lsrDNA sequences and MW596661-MW596682 for the cox1 sequences). Newly generated sequences were aligned in two independent datasets following the alignments from Hernández-Orts et al. [2] and Waeschenbach et al. [4]. Sequences from other diphyllobothriids were retrieved from GenBank and aligned with our novel sequences using default settings of MUSCLE [34] implemented in Geneious (Additional file 1: Table S1). The extremes were trimmed resulting in an alignment with 1574 bp for the lsrDNA (Additional file 2: lsrDNA_alignment) and 1571 bp for the cox1 (Additional file 3: cox1_alignment). A combined lsrDNA + cox1 alignment (3145 bp; Additional file 4: lsrDNA_cox1_alignment) was also constructed using only taxa with sequences for both markers available from GenBank (Additional file 1: Table S1).
Bayesian inference (BI) and maximum likelihood (ML) analyses were performed for each dataset. jMod-elTest 2.1.10 software [35] was used to select the best nucleotide substitution model under the Akaike information criterion. The TIM2 + I + G model was chosen for the lsrDNA and cox1 datasets and the TIM2 + I + G & TIM1 + I + G for the combined dataset. Bayesian inference analyses were constructed using MrBayes 3.2.6 [36]. The BI analyses were estimated via two independent Markov Chain Monte Carlo runs of four chains with standard settings for 10,000,000 generations with a sampling frequency of 1000th generations. Burn-in periods were set to 25% of generations. The ML analyses were run with raxmlGUI v.2.0 [37]. Bootstrap nodal support values were computed by running 1000 bootstrap resamples. The resulting trees for BL and ML were visualized in FigTree 1.4.4 [38]. Genetic distances (uncorrected p-distance) were calculated with MEGA 10.1.8 [39] from the total number of nucleotide differences from the lsrDNA alignment and from the full cox1 alignment excluding partially characterized sequences (i.e. < 1400 bp).

Morphological examination
For morphological evaluation, selected tapeworms were stained with Mayer's hydrochloric carmine, dehydrated through an ethanol series, cleared with eugenol and mounted in permanent slides in Canada balsam. Selected pieces of the strobila were embedded in paraffin wax, cross-sectioned (thickness 15 µm), stained with hematoxylin-eosin, and mounted in Canada balsam. Mounted specimens were examined with an Olympus BX51 microscope (Olympus Corp., Tokyo, Japan). Measurements were taken from digital images with the QuickPHOTO CAMERA 3.2 image analysis software (Promicra Ltd., Prague, Czech Republic). Measurements are expressed in micrometers unless otherwise stated and presented as the range (minimum-maximum), with the mean followed by the standard deviation (SD) and the number of measured specimens or structures in parentheses. Detailed line drawings were made using a drawing tube attached to an Olympus BX51 microscope.
Selected scoleces and proglottids were prepared for scanning electron microscopy (SEM). Specimens were dehydrated through an ethanol series, transferred to hexamethyldisilazane (Ted Pella, Inc., Redding, CA, USA) and allowed to air dry. Samples were mounted on aluminium stubs on double-sided adhesive carbon tape, gold sputter-coated and examined with a JEOL JSM 7401-F scanning electron microscope (JEOL Ltd., Tokyo, Japan) at 4 kV at the Laboratory of Electron Microscopy, Institute of Parasitology, Biology Centre, Czech Academy of Sciences. Terminology of microtriches follows Chervy [40].
Specimens of the type series and voucher specimens from the present study are deposited in the Hel-

Results
More than 150 tapeworm specimens (including immature and gravid specimens) were collected in the intestine of CSLs. In the SASLs from Argentina and Peru, 13 and 18 immature tapeworms were collected, respectively.

Phylogenetic relationships and genetic divergence
A total of four partial lsrDNA sequences (1450-1481 bp long; three isolates from CSLs, USA, and an isolate from SASL, Argentina) were generated. Additionally, one almost complete (1565 bp long; isolate from SASL, Argentina) and 21 partial cox1 sequences (415-420 bp long; four isolates from CSLs, USA, and 17 isolates from SASL, Peru) were generated (see Additional file 1: Table S1).
Both ML and BI analyses of the cox1 dataset (59 taxa) resulted in generally similar topologies (Fig. 2), but slightly differed from the recent phylogenetic study of Waeschenbach et al. [4]. Four newly generated sequences from isolates from CSLs and 14 novel sequences from SASLs (13 from Peru and 1 from Argentina) formed a strongly supported clade with three sequences identified as 'Diphyllobothrium' sp. 1., including one sequence from an isolate from CSL (KY552890) and two from SASLs (MF893274 and KY945922) (Fig. 2). The genetic divergence between isolates from CSLs and SASLs in the cox1 gene was 2.87-3.07%. In the phylogenetic tree, the clade formed by our novel sequences + 'Diphyllobothrium' sp. 1. appeared also as sister to the sequence of Diphyllobothriidae gen. sp. from T. bernacchii (KY552888). Genetic divergence between our newly generated sequences + 'Diphyllobothrium' sp. 1 and the isolate reported as Diphyllobothriidae gen. sp. ranged from 14.56-14.94%.
The phylogenetic trees inferred with the combined (lsrDNA + cox1) dataset (33 taxa) were similar to the topologies inferred for the lsrDNA and cox1 datasets (Additional file 5: Fig. S1). Our newly generated sequences formed a well-supported clade with sequences of 'Diphyllobothrium' sp. 1 from CSL of California [4] and 'Diphyllobothrium' sp. 1 from SASL of Chile [19].
In summary, both ML and BI analyses for the lsrDNA, cox1 and the combined datasets revealed that most of our newly generated sequences from isolates from CSLs and SASLs belong to an as yet undescribed species reported as 'Diphyllobothrium' sp. 1 [4]. Genetic variation was detected between sequences from isolates of the putative new species from CSLs and SASLs. However, these values were somewhat lower, especially for cox1, than the interspecific variation between other 'Diphyllobothrium' species of pinnipeds included in our phylogenetic analyses. Therefore, we consider isolates of the putative new species of 'Diphyllobothrium' from CSLs and SASLs from MW600336 'Diphyllobothrium' sprakeri n. sp. ex Otaria flavescens (Argentina) 1

Description
[Based on 20 specimens from CSLs. All tapeworms collected from SASL were immature specimens and were not included in the description]. Diphyllobothriidea, Diphyllobothriidae. Specimens anapolytic, long worms up to 2 m (holotype 56 cm; Fig. 3a) in length; maximum width ca. 8 mm. Immature worms may reach 1 m in length. Scolex surface covered with capilliform filitriches and coniform spinitriches (Fig. 4a, b). Microtriches on strobila surface not observed.

Remarks
The new species is placed in the genus Diphyllobothrium because of its typical shape of the scolex, composition of genital organs, absence of transverse papilla-like tegumental protuberances on the ventral surface of the proglottids and using pinnipeds as its definitive host (see [4]).
Kuchta and Scholz [3] reported 20 species of 'Diphyllobothrium' (incertae sedis) from the intestine of pinnipeds.  (Shipley, 1907) by having an enlarged thick-walled sac-like structure in the terminal part of the uterus in mature and gravid proglottids [41].
Our molecular analyses reported immature specimens of A. pacificus and D. sprakeri n. sp. in the intestine of a SASL from Peru. Immature specimens of both species are morphologically indistinguishable. Adult specimens of A. pacificus can be distinguished from the new species by the presence of the papilla-like protuberances anterior to the male gonopore [2] and the absence of a thick-walled sac-like structure in the terminal part of The voucher material identified as 'Diphyllobothrium latum' from CSLs off Mexico and the USA deposited in London (NHML) was substantially decomposed. These specimens were immature without developed genital organs crucial for identification. However, D. sprakeri n. sp. is similar to these specimens in the shape of the scolex and the absence of a neck. 'Diphyllobothrium' sprakeri n. sp. has been the only species of 'Diphyllobothrium' reported from the intestine of CSLs ( [18], present study). Based on this evidence, the voucher material deposited in London is tentatively conspecific with the new species. One voucher specimen deposited in London (NHML 1980.6.3.189) was collected from a CSL stranded on the Mexican coast. In Mexico, CSLs are distributed along the east and west coasts of the Baja California Peninsula [47]. Therefore, this peninsula may represent an additional locality for D. sprakeri n. sp.

Discussion
The taxonomy of diphyllobothriids is insufficiently resolved. Identification of individual species is complicated because of their uniform strobilar morphology, the high amount of intraspecific and intraindividual variation for most morphological characters and incomplete original descriptions [3,8].
Diphyllobothriids of pinnipeds have been revised by the present authors for more than 10 years based on detailed morphological examination of well-fixed material combined with molecular data (see [2, 4-6, 8, 40]). Our previous studies suggested that otariids are only definitive hosts of three diphyllobothriid species: A. pacificus, widely distributed in both hemispheres, and D. tetrapterum and Pyramicocephalus phocarum (Fabricius, 1780), limited to the Northern Hemisphere [2,5,8]. However, a revision of the metazoan parasites of CSLs recognized a new undescribed diphyllobothriid tapeworm which is different from these three species [4,5,18]. This species showed a uterine sac-like structure in mature and gravid proglottids, which is an uncommon character in diphyllobothriids. 'Diphyllobothrium' sprakeri n. sp. is the first diphyllobothriid species described from CSLs and, with A. pacificus, the second valid species of otariids from the Southern Hemisphere.
Yurahkno and Maltsev [45] described D. lobodoni from the intestine of crabeater seals Lobodon carcinophagus (Hombron & Jacquinot) from Antarctica. This diphyllobothriid species differs from D. scoticum by the size of the strobila, scolex, neck and uterine sac-like structure, the shape of proglottids, thickness of the tegument and muscle layer, number of testes and the position of the cirrus sac and external seminal vesicle. However, the size of the strobila and scolex and number of testes and other structures may depend on the fixation methods, host species, its size, physiological state or intensity of infection (see [5], and references therein) and are not suitable characters in species delimitation [2,3]. Moreover, most of the used discriminant characteristics of D. lobodoni overlap with those of D. scoticum reported from leopard seals (type-host) by other authors (Table 1). Further studies of the type material and molecular data from the typehost of D. lobodoni are necessary to confirm the validity and systematic position of this species.
Our study suggests that D. sprakeri n. sp. has a wide geographical distribution in both hemispheres, including the Pacific and Atlantic Oceans, and infects at least two otariid species. The distribution of D. sprakeri n. sp. in the Northern Hemisphere is limited to the Pacific coast of California, USA, and Baja California, Mexico. Interestingly, our new species has not been recorded in the North Pacific coast, where the diversity of diphyllobothriid tapeworms from otariids has been comprehensively evaluated in recent years [2,5,49,50]. In the Southern Hemisphere, D. sprakeri n. sp. is more widely distributed, occurring in temperate waters of the Pacific coast of South America (Peru and Chile) and the Southwest Atlantic along the Patagonian coast of Argentina.
The life cycle of D. sprakeri n. sp. probably includes marine fishes as the second intermediate hosts. Recently, Mondragón Martínez [51] reported plerocercoids of A. pacificus and an unidentified species of Diphyllobothrium in marine fishes from Peru based on partial cox1 sequences. According to the phylogenetic analysis of Mondragón Martínez [51], unidentified diphyllobothriid plerocercoids, collected from anchoveta Engraulis ringens Jenyns and Pacific jack mackerel Trachurus symmetricus (Ayres), formed a clade sister to A. pacificus and Diphyllobothrium spp. Unfortunately, partial cox1 sequences of these diphyllobothriid plerocercoids are not available in the GenBank dataset. These plerocercoids probably belong to D. sprakeri n. sp.; however, sequences generated from these plerocercoids need to be analyzed in a more robust phylogenetic context for reliable species identification.
The Pacific broad tapeworm Adenocephalus pacificus is considered the most important causative agent of diphyllobothriosis among humans in South America [52]. Diphyllobothriosis caused by this species has been reported predominantly in Peru, where human infections are associated with the habits of consuming raw or undercooked marine fishes [23,52]. Our new species Table 1 Comparison of selected biometrical data among 'Diphyllobothrium' sprakeri n. sp., 'Diphyllobothrium' scoticum and 'Diphyllobothrium' lobodoni. The incomplete description of D. scoticum from leopard seals from Macquarie Island, Antarctica, by Johnston [44] is not included. Measurements in micrometers, unless otherwise stated a Specimens without scoleces referred to as 'Diphyllobothrium scoticum-like cestode' b Testes measured in transverse section c Metrical data on Dibothriocephalus pygoscelis Rennie & Reid, 1912, which is considered a junior synonym of D. scoticum (see Johnston [44], Markowski [22]), is included d Fuhrmann [43]