Description of Hymenolepis microstoma (Nottingham strain): a classical tapeworm model for research in the genomic era

Background Hymenolepis microstoma (Dujardin, 1845) Blanchard, 1891, the mouse bile duct tapeworm, is a rodent/beetle-hosted laboratory model that has been used in research and teaching since its domestication in the 1950s. Recent characterization of its genome has prompted us to describe the specific strain that underpins these data, anchoring its identity and bringing the 150+ year-old original description up-to-date. Results Morphometric and ultrastructural analyses were carried out on laboratory-reared specimens of the 'Nottingham' strain of Hymenolepis microstoma used for genome characterization. A contemporary description of the species is provided including detailed illustration of adult anatomy and elucidation of its taxonomy and the history of the specific laboratory isolate. Conclusions Our work acts to anchor the specific strain from which the H. microstoma genome has been characterized and provides an anatomical reference for researchers needing to employ a model tapeworm system that enables easy access to all stages of the life cycle. We review its classification, life history and development, and briefly discuss the genome and other model systems being employed at the beginning of a genomic era in cestodology.


Background
Species of Hymenolepis Weinland, 1858 (Platyhelminthes: Cestoda: Cyclophyllidea) have been used as tapeworm models in research and teaching since the 1950s when they were first domesticated in the laboratory of Clark P. Read [1]. Adult parasites of rodents with beetle intermediate hosts, they benefit from easy culture in vivo using natural hosts that are themselves model organisms (e.g. Mus musculus L., Tribolium confusum Jacquelin du Val). Research on Hymenolepis, and especially H. diminuta (Rudolphi, 1819), H. nana (von Siebold, 1852) and H. microstoma, is underpinned by an extensive literature that includes much of our classical knowledge of tapeworm biology [e.g. [2]]. A recently initiated effort sponsored by The Wellcome Trust Sanger Institute to characterize the genome and adult and larval transcriptomes of H. microstoma http://www. sanger.ac.uk/sequencing/Hymenolepis/microstoma/ has brought this classical model into the genomic era, greatly advancing its utility for researchers interested in employing a practical tapeworm system that allows access to all life cycle stages. In light of this development, and the fact that laboratory isolates can vary in features of their biology [3], it is desirable to have a description of the exact strain on which the genome is based, and to thus anchor the data to a well-defined entity.
Hymenolepis microstoma was first described from the bile ducts of mice in 1845 by Dujardin [4] who placed it in the genus Taenia L., 1758, which housed all tapeworms known at that time. In 1891, Blanchard [5] transferred the species to the genus Hymenolepis and provided an expanded description of the species. Although Bear and Tenora [6] suggested synonymy between H. microstoma and H. straminea (Goeze, 1782), species status of H. microstoma historically has been widely accepted, and molecular data have shown both species to represent independent, albeit closely related, lineages [7,8]. In contrast, the genus Hymenolepis has itself been overhauled on several occasions and its membership and internal structure remain controversial. For example, whereas Hughes [9,10] accepted the generic assignment H. microstoma by Blanchard, Spasskii [11] subdivided the genus and transferred H. microstoma to the genus Rodentolepis Spasskii, 1954, which he erected to house the rodent-hosted species of Hymenolepis with armed rostella. At the same time Spasskii erected the genus Vampirolepis Spasskii, 1954, which Schmidt subsequently considered a senior synonym of Rodentolepis, thus resulting in the new combination Vampirolepis microstoma (Dujardin, 1854) Schmidt, 1986 [12]. The genus Rodentolepis was retained by Czaplinski and Vaucher [13] in the most recent synoptic treatment of tapeworms [14], but this work did not consider species level taxa and therefore did not arbitrate on the generic assignment of H. microstoma. Thus although Vampirolepis microstoma [12] represents the most recent formal taxonomic assignment of the species, few investigators have adopted this name, and most reports refer to it as either a member of the genus Hymenolepis, or with less frequency, Rodentolepis. In our view, a natural circumscription of hymenolepid species will not be attained without the application of molecular data [15].
To this end, Haukisalmi et al. [8] recently used 28S rDNA to analyze phylogenetic relationships among 32 hymenolepidid species from rodents, shrews and bats, showing that both Hymenolepis and Rodentolepis represented paraphyletic assemblages. Although their work assigned H. microstoma to a 'Rodentolepis' clade, the lack of resolution and widespread paraphyly of the taxa in their analyses indicate that greater taxonomic representation and more robust data are needed before such nomenclatural circumscriptions can be made reliably. We therefore follow Blanchard [5] in recognizing the mouse bile duct tapeworm as a member of the genus Hymenolepis, employing the most common name in usage, whilst appreciating that a more comprehensive understanding of hymenlepidid interrelationships is likely to warrant generic reassignment.
Here we provide a description of a 'Nottingham' strain of H. microstoma based on light and scanning electron microscopy of laboratory-reared specimens from the same culture used to characterize the genome. History of the isolate, dating back to the laboratory of C. P. Read [1], suggests that it represents a model that has been widely employed and disseminated within the parasitological community for over 50 years, making the genome data directly relevant to a significant pre-existing literature on its biology.

Remarks
Hymenolepis species exhibit the well-documented 'crowding effect' in which overall size and egg production are inversely related to the intensity of infection [19]. Consequently, size is dependent not only on the age of the worms, but on the number of worms present in the host, and cannot be used diagnostically [20]. Crowding in H. microstoma has been shown to decrease linear growth, egg production and the rate of proglottide formation [21]. Moreover, we chose to document gravid adult specimens at an age and size most useful for laboratory manipulations in which larger worms pose unnecessary practical problems (e.g. assays involving whole mount in situ hybridization or in vitro culture). de Rycke [22] showed that H. microstoma is in rapid state of growth starting around 12-14 days post-infection in Mus musculus, and whereas our length measurements correspond to those reported by de Rycke for the relevant age class (see Table 1), they are obviously less comparable to reports based on older specimens, such as those stemming from natural infections.

Life history
Hymenolepis microstoma is most probably cosmopolitan in distribution [20] and is not known from human infections outside of a single report in which mixed infections of H. nana and H. microstoma were identified in four individuals from a remote region of Western Australia [23]. Reported natural definitive hosts include a large range of rodent genera that include mice (e.g. Apodemus Kaup, Dendromus Smith, Leggada Gray, Mastomys Thomas, Mus L.), gerbils (Meriones Illiger) and voles (Microtus Schrank) [9,12,24]. Infections in rats is controversial: whereas Joyeux and Kobozieff [25] reported successful infection of laboratory rats, Dvorak et al. [20] found rats to be refractory to H. microstoma, and Litchford [24] showed that rats became refractory with age. Similarly, although infections can be established in golden hamsters (Mesocricetus Nehring), they result in underdeveloped worms and cause severe pathology to the host [20,24]. Dvorak et al. [20] demonstrated that mice could not be infected via eggs, as is the case with H. nana (ie. auto-infection) [26]. However, in congenitally athymic mice, Andreassen et al. [27] found that autoinfection was possible, showing that oncospheres penetrated the intestinal tissues and developed into cysticeroids that subsequently excysted and developed normally in the bile duct and duodenum, in a manner similar to the direct cycle of H. nana. Autoinfection of BALB/c mice was also implied by the detection of stage-specific antigens [28].
The life history of H. microstoma ( Figure 3) has been described in detail previously [20,25,29] and is typical of other hymenolepid species, save its unusual location in the bile duct of the mammalian host. In brief, eggs containing patent oncospheres are expelled with faeces into the environment and may be ingested by either the adult or larval stage of an appropriate beetle host (e.g. Tribolium confusum, T. castaneum, Tenebrio molitor, and Oryzaephilus surinamensis). Oncospheral larvae (~20 μm; Figure 1D; Figure 2F) are released from their thin shells ( Figure 2E; n.b. appearing as a 'hymen' via light microscopy and the eponym of the genus) through the action of the host mouthparts, and after ingestion use their three pairs of hooks and proteolytic secretions [30] to enter the haemocoel. There they undergo a complete metamorphosis, reconstituting their bodies into cycsticeroid larvae [31] in approximately seven days, the phases of which have been documented by both Voge [32] and Goodchild and Stullken [33]. Upon infection of the definitive host, the combination of pepsin and HCl in the stomach act to dissolve the larval membranes, and juvenile worms are then activated in the duodenum in response to trypsin and bile salts. de Rycke [22] described adult growth and organogenesis in Mus musculus (summarized in Table 1): in the first three days the juveniles move anteriorly in the upper 20% of the small intestine and duodenum before establishing permanently in the bile duct, where they commence strobilation. Within approximately 14 days terminal segments are gravid and most of their strobila extends outside of the bile duct and into the duodenum. Thus the entire life cycle, from egg to gravid adult, can be completed in the laboratory in only three weeks. Although the germinative ('neck') region of tapeworms has the potential for 'immortality' as demonstrated in H. diminuta by Read [34], infections of H. microstoma in mice persist for an average of six months, whereas those in the intermediate host can remain infective for the life of the beetle (> one year).

The Hymenolepis genome
Through collaboration with The Wellcome Trust Sanger Institute, a draft genome of H. microstoma derived from the cultures described herein is now publically available: http://www.sanger.ac.uk/resources/downloads/helminths /hymenolepis-microstoma.html. The latest assembly (October 2010) includes more than 40× coverage of the estimated 140 Mb haploid genome and is based on data produced by a combination of Roche 454 and Illumina Solexa next-generation sequencing technologies. Gene annotation is presently being conducted using a combination of RNA-Seq [35] and automated gene prediction tools, revealing intron-exon structures and other aspects of their genomic organization, and additional tools are being used to characterize non-coding regions (M. Zarowiecki and M. Berriman, pers. comm.).
Hymenolepis microstoma is one of four tapeworm species to have complete genomes characterized: a reference genome of Echinococcus multilocularis Leukart, 1863 and draft genome of E. granulosus (Batsch, 1786) have been produced by the Sanger Institute (available from http:// www.sanger.ac.uk/resources/downloads/helminths/) in collaboration with Profs. Klaus Brehm and Cecelia Fernandez, respectively, and a consortium in Mexico are currently working to characterize the genome of Taenia solium L., 1758 [36]. These data herald the beginning of the genomic era in cestodology and are already accelerating advances in our understanding of tapeworm biology and infection. At present the only published platyhelminth genome is that of the human blood fluke, Schistosoma mansoni Sambon, 1907 [37]. However, genome data for Schistosoma Weinland, 1858 and Echinococcus Rudolphi, 1801, as well as the free-living flatworm    [40], Evans [41,42] and Seidel [43,44], but to our knowledge no report of research employing these techniques has been published subsequently. Our initial attempts to follow these protocols for the cultivation of adult worms resulted in only limited growth (3× increase in length) without the onset segmentation (unpub. data). However, as many of the reported media used by previous authors are no longer commercially available, more work is needed to develop contemporary protocols for in vitro culture. Among the most advanced in vitro systems available for tapeworm research today has been developed by Brehm and colleagues for Echinococcus [45][46][47][48], the genus on which most of our understanding of tapeworm molecular biology is based [49]. Development of an axenic culture system of the hydatid stage of E. multilocularis has allowed them to introduce transgenic and functional genomic techniques (e.g. RNAi) to cestodology, and their system is currently being used to pioneer research on stem-cells and developmental biology in parasitic flatworms [45,50].
Although not yet supported by genome characterization, another currently employed in vitro system is that of Mesocestoides Vaillant, 1863 [e.g. [51]] which are readily maintained in the larval tetrathyridial stage [31] and can increase their numbers in culture via asexual fission [52]. Adult worms have also been grown in vitro and induced to strobilate through the addition of bile salts [53]. However, as with species of Echinoccocus and Taenia, in vivo development of strobilar stages of Mesocestoides is prohibited by the legalities and expense of maintaining large vertebrate hosts in the laboratory. Rodent hosted Hymenolepis species therefore remain the most convenient systems for research on the biology of adult tapeworms, and for this reason we have been developing H. microstoma as a model to study the development and evolution of tapeworm segmentation [54].
Although the basic framework of cestode evolution has been revealed by previous molecular studies [55][56][57][58] and the interrelationships of select groups are now well resolved [59][60][61], there has yet to be a comprehensive molecular phylogenetic study of the largest and most important group of tapeworms with regard to human and animal health, the Cyclophyllidea. All of the tapeworm species for which genomes have been characterized thus far belong to this order and thus it is especially important that we elucidate the relative phylogenetic positions of the 350+ described genera [14]. Such knowledge will provide an evolutionary underpinning for comparative genomic studies within the group and allow us to identify the sister lineages whose genomes share the closest evolutionary histories to the species for which full genome data are now available.

Methods
A seed culture of Hymenolepis microstoma infected beetles was obtained from Nottingham University in 2005 courtesy of Prof. Jerzy Behnke and subsequently maintained in vivo at the Natural History Museum (London) using flour beetles (Tribolium confusum) and BKW outbred conventional mice (full protocols can found at http://www.olsonlab.com; please contact the corresponding author to enquire about seed cultures). Gravid, 14-16 day old specimens were removed from the bile ducts and duodenum of mice and quickly swirled in near-boiling 0.85% saline for~4 secs to fully extend the worms prior to fixation in cold 4% paraformaldehyde overnight at -4 C. Whole-mounted specimens were dehydrated in a graded ethanol series, stained using Gill's haematoxylon or left unstained, cleared in beachwood creosote and mounted in Canada balsam. Sections were prepared by paraffin embedding using standard histological techniques and stained with Mayer's Haemalum [62]. Measurements and illustrations were made under differential interference contrast on a Leica DM5000B compound microscope equipped with a camera lucida and digital documentation system. Specimens used for SEM were dehydrated as above, critically-point dried, sputter-coated with gold/palladium and viewed on a JEOL XL30 scanning electron microscope. Internal structures were imaged by SEM by cutting worms crudely using a razor blade.