Echinococcus

Desmond Smyth pioneered in vitro cultivation techniques to support the development of the taeniid cestode Echinococcus in the laboratory (reviewed by Smyth, Reference Smyth1967; Smyth and Davies, Reference Smyth and Davies1974a; Howell and Smyth, Reference Howell, Smyth, Thompson and Lymbery1995). His detailed observations of the stages of development of Echinococcus induced in vitro and their inherent plasticity generated a wealth of information about developmental and physiological processes in cestodes. Most importantly, they provided the basis for generating a series of thought-provoking, seminal publications that raised numerous questions and hypotheses (Smyth, Reference Smyth1968, Reference Smyth1969, Reference Smyth1971, Reference Smyth, Taylor and Muller1972; Smyth and Smyth, Reference Smyth and Smyth1964, Reference Smyth and Smyth1968; Smyth and Davies, Reference Smyth and Davies1975; Smyth and Barrett, Reference Smyth and Barrett1979; Smyth et al. Reference Smyth, Howkins and Barton1966, Reference Smyth, Miller and Howkins1967). At the time, Smyth's work challenged views on the simplicity of cytodifferentaition in platyhelminths. He also proposed that the Echinococcus in vitro system could be a model for both invertebrate and vertebrate studies, which is only now really being appreciated and exploited. For example, the multipotential, stem cell-like nature of the postulated germinal cells of Echinococcus, which form part of the parasite's ill defined, syncytial ‘germinal layer’, have only recently attracted the attention they deserve (Spiliotis et al. Reference Spiliotis, Lechner, Tappe, Scheller, Krohne and Brehm2008).

However, perhaps the most important observation made by Desmond Smyth concerned differences in the development of adult Echinococcus from the larval protoscolex. In contrast to protoscoleces of E. granulosus collected from hydatid cysts in sheep, those collected from hydatid cysts in horses failed to develop in in vitro culture, even though both sheep and horse protoscoleces were grown in exactly the same medium (Smyth and Davies, Reference Smyth and Davies1974b). This fairly simple observation resulted in radical shifts in our appreciation of host specificity in Echinococcus, our understanding of the epidemiology of hydatid disease, and the taxonomy of the aetiological agents.

The significant differences in the ability of E. granulosus of sheep and horse origin to develop in identical culture media led Smyth to refer to the phenomenon as physiological strain differences (Smyth and Davies, Reference Smyth and Davies1974b; Smyth, Reference Smyth1982). This revolutionized the concept of a strain in parasitology, and demonstrated the importance of combining phenotypic and genetic differences in the characterization and description of parasites at the intraspecific level (Thompson and Lymbery, Reference Thompson and Lymbery1990, Reference Thompson and Lymbery1996).

The extrapolation that such strain differences in development in vitro might reflect differences in host specificity, with the sheep strain of E. granulosus adapted to sheep and the horse strain to horses, suggested that the epidemiology of infections caused by these 2 strains could be different. This was very important at the time in Great Britain where there had been a dramatic increase in the incidence of hydatid disease in horses (Thompson and Smyth, Reference Thompson and Smyth1974). However, knowledge that we could be dealing with 2 different organisms with different intermediate host preference, meant that epidemiological studies should consider the operation of 2 distinct maintenance cycles, one involving sheep and the other horses. This was shown to be the case, with quite different husbandry practices supporting the two cycles (Thompson and Smyth, Reference Thompson and Smyth1975). The existence of a distinct cycle of transmission involving horses and dogs was subsequently demonstrated in other countries (Kumaratilake et al. Reference Kumaratilake, Thompson and Eckert1986). Importantly, epidemiological evidence has not only demonstrated distinct differences in intermediate host specificity but also that, unlike the sheep strain, the horse strain of E. granulosus does not appear to be infective to humans (Thompson and Lymbery, Reference Thompson and Lymbery1988, Reference Thompson and Lymbery1991).

The concept of host-adapted strains of E. granulosus led to studies on other forms of the parasite in other species of intermediate hosts, such as cattle, pigs, camels and cervids. These studies not only confirmed the existence of a number of host-adapted life cycles in different parts of the world but also provided additional data on developmental differences between strains (reviewed by Thompson and Lymbery, Reference Thompson and Lymbery1988; Thompson et al. Reference Thompson, Lymbery and Constantine1995). These ‘strain differences’ were recognized as having a direct impact on the life cycle in nature, which could affect control strategies. For example, the cattle strain of E. granulosus was shown to develop more quickly in dogs than the sheep strain (Thompson et al. Reference Thompson, Kumaratilake and Eckert1984) resulting in an earlier release of infective eggs into the environment. Thus, if anti-cestodal drugs were used to ‘break the cycle’ in dogs by being given at regular intervals to prevent adult worms reaching patency, they have to be given at least a week earlier in areas where the cattle strain is endemic compared to the sheep strain (Thompson, Reference Thompson, Thompson and Lymbery1995).

The development and application of a range of molecular tools to these different host-adapted strains of E. granulosus added another element to studies on variation in Echinococcus. The application of molecular tools has convincingly demonstrated that E. granulosus comprises a series of genetically distinct host-adapted strains or genotypes which phylogenetic analysis indicates warrant taxonomic status (Thompson et al. Reference Thompson, Lymbery and Constantine1995; Lymbery and Thompson Reference Lymbery and Thompson1996; Thompson, Reference Thompson, Gillespie and Pearson2001, Reference Thompson2008a; Thompson and McManus, Reference Thompson, McManus, Eckert, Gemmell, Meslin and Pawlowski2001, Reference Thompson and McManus2002; McManus and Thompson, Reference McManus and Thompson2003; Lavikainen et al. Reference Lavikainen, Lehtinen, Meri, Hirvelä-Koski and Meri2003; Jenkins et al. Reference Jenkins, Romig and Thompson2005; Romig et al. Reference Romig, Dinkel and Mackenstedt2006; Moks et al. Reference Moks, Jogisalu, Valdmann and Saarma2008). Consequently, the taxonomy of Echinococcus has been revised and what were referred to as strains or genotypes are now recognized as separate species with well-defined morphological characteristics and host ranges. Interestingly, the nomenclature used for these ‘species’ conforms to that proposed by observational parasitologists in the 1920s–1960s, before molecular tools were available to confirm their morphological descriptions and epidemiological observations (Thompson et al. Reference Thompson, Lymbery and Constantine1995; Thompson and McManus, Reference Thompson and McManus2002). Furthermore, molecular characterization has confirmed that certain morphological features are distinct between species and strains, and thus microscopy can be used as a convenient, low cost and ‘low tech’ method in field studies on the epidemiology of Echinococcus infections (Hobbs et al. Reference Hobbs, Lymbery and Thompson1990; Harandi et al. Reference Harandi, Hobbs, Adams, Mobedi, Morgan-Ryan and Thompson2002).

I am convinced that had Smyth not undertaken his seminal observations by comparing the growth of E. granulosus of sheep and horse origin in vitro, subsequent comparative studies on the parasite of different intermediate host origin would not have been undertaken in such a timely and systematic fashion. Smyth undoubtedly provided the stimulus for such studies.

Giardia

Giardia is a ubiquitous enteric protozoan pathogen of vertebrates, frequently parasitizing mammals, that is characterized by the unusual presence of 2 morphologically similar, transcriptionally active diploid nuclei, no mitochondria or peroxisomes, and a unique attachment organelle called the ‘ventral sucking disc’ (Thompson and Monis, Reference Thompson and Monis2004; Morrison et al. Reference Morrison, McArthur, Gillin, Aley, Adam, Olsen, Best, Cande, Chen, Cipriano, Davids, Dawson, Elmendorf, Hehl, Holder, Huse, Kim, Lasek-Nesselquist, Manning, Nigam, Nixon, Palm, Passamaneck, Prabhu, Reich, Reiner, Samuelson, Svard and Sogin2007). Phylogenetic relationships remain controversial. One view suggests that the genus comprises one of many divergent eukaryotic lineages that adapted to a microaerophilic life-style rather than diverging before the endosymbiosis of the mitochondrial ancestor and the other that Giardia is a basal eukaryote (Thompson and Monis, Reference Thompson and Monis2004; Morrison et al. Reference Morrison, McArthur, Gillin, Aley, Adam, Olsen, Best, Cande, Chen, Cipriano, Davids, Dawson, Elmendorf, Hehl, Holder, Huse, Kim, Lasek-Nesselquist, Manning, Nigam, Nixon, Palm, Passamaneck, Prabhu, Reich, Reiner, Samuelson, Svard and Sogin2007). As with Echinococcus, the question of host specificity and taxonomy has proved to be controversial with Giardia that occur in mammals, and particularly with respect to the question of zoonotic potential (Monis and Thompson, Reference Monis and Thompson2003; Traub et al. Reference Traub, Monis, Robertson, Irwin, Mencke and Thompson2004; Caccio et al. Reference Caccio, Thompson, McLauchlin and Smith2005; Leonhard et al. Reference Leonhard, Pfister, Beelitz, Wielinga and Thompson2007).

Early workers suspected that there were a number of species of Giardia restricted to certain species of mammalian host. However, the lack of morphological characters to reliably distinguish between these species led Filice to rationalize the number of species putting those with no evidence of morphological distinctness into the 1 species, G. duodenalis. Some 20 years later, Meyer (Reference Meyer1970) reported the first axenic cultivation of Giardia from mammals, and in 1976 from a human (Meyer, Reference Meyer1976). Another 20 years after this, Binz et al. (Reference Binz, Thompson, Meloni and Lymbery1991) succeeded in establishing cloned lines of different isolates of Giardia in in vitro culture. These developments were critical in enabling the laboratory amplification of defined isolates of Giardia from various sources – both human and non-human – so that sufficient quantities of the different isolates would be available for comparative studies in the laboratory, initially for those using allozyme electrophoresis (Mayrhofer et al. Reference Mayrhofer, Andrews, Ey and Chilton1995; Meloni et al. Reference Meloni, Lymbery and Thompson1995; Monis et al. Reference Monis, Andrews, Mayrhofer, Mackrill, Kulda, Isaac-Renton and Ey1998, Reference Monis, Andrews, Mayrhofer and Ey1999).

In retrospect, observations on the in vitro cultivation of Giardia isolates can now be seen to be very important in the interpretation of the genetic data that have been generated over the last decade. However, initially it quickly became clear that some isolates would grow in in vitro culture whereas others would not. Non-human isolates, for example from dogs, cats and livestock, would not grow in culture, but not all human isolates would grow in culture either and it was thought this may be due to the age of the isolate, condition of the cysts etc. (reviewed by Thompson and Monis, Reference Thompson and Monis2004). We now know these observations largely reflect genetic data indicating that certain genetic groups (assemblages) only appear to occur in certain species of host, and more correctly should be recognized as distinct host adapted species thus validating the observations of taxonomists in the early part of the 1900s (Thompson and Monis, Reference Thompson and Monis2004; Caccio et al. Reference Caccio, Thompson, McLauchlin and Smith2005; Wielinga and Thompson, Reference Wielinga and Thompson2007; Monis et al. Reference Monis, Caccio and Thompson2009). Thus, the failure of isolates from dogs and livestock to grow in culture is now considered to reflect differences in host specificity, since the media used for growing Giardia in in vitro culture was principally developed to support growth of Giardia from humans. Future research is needed to develop culture media that support the growth of Giardia from carnivores and herbivores, and in this respect comparative proteomics may give clues to important biochemical differences between the different species of Giardia (Steuart et al. Reference Steuart, O'Handley, Lipscombe and Thompson2008).

The fact that some isolates from humans fail to establish in culture is likely to reflect the extensive genetic variability that we now know exists within the 2 main genetic groupings of G. duodenalis, assemblages A and B, that affect humans, as well as other mammals (Thompson and Monis, Reference Thompson and Monis2004; Caccio et al. Reference Caccio, Thompson, McLauchlin and Smith2005; Monis et al. Reference Monis, Caccio and Thompson2009). Assemblage A isolates appear to have a selective advantage under axenic in vitro culture conditions compared with assemblage B isolates (Andrews et al. Reference Andrews, Mayrhofer, Chilton, Boreham and Grimmond1992; Binz et al. Reference Binz, Thompson, Lymbery and Hobbs1992; Thompson and Lymbery, Reference Thompson and Lymbery1996). In addition, differences have been reported in metabolism and biochemistry, DNA content, in vitro and in vivo growth rates, drug sensitivity, predilection site in vivo and duration of infection, pH preference, infectivity, susceptibility to infection with a dsRNA virus, and clinical features (reviewed by Thompson and Monis, Reference Thompson and Monis2004; Caccio et al. Reference Caccio, Thompson, McLauchlin and Smith2005; Wielinga and Thompson, Reference Wielinga and Thompson2007; Monis et al. Reference Monis, Caccio and Thompson2009). It is particularly interesting that, generally, isolates from assemblage B grow more slowly than those from assemblage A, and are more difficult to establish in in vitro culture. This may correlate with differences in clinical features, with assemblage B isolates more commonly associated with chronic infections, and a postulated ability to establish more persistent infections that may be refractory to routine chemotherapy than infections with assemblage A isolates (Hopkins et al. Reference Hopkins, Constantine, Groth, Wetherall, Reynoldson and Thompson1999; Thompson, Reference Thompson, Olson, Olson and Wallis2002, Reference Thompson2008b; Thompson and Monis, Reference Thompson and Monis2004).

In relation to this, a recent study using comparative proteomics has found distinct differences in several proteins between Giardia isolates from assemblages A and B (Steuart et al. Reference Steuart, O'Handley, Lipscombe and Thompson2008). One of these, alpha 2 giardin, appears to be an assemblage A-specific protein of human infective G. duodenalis. Alpha 2 giardin is a structural protein and associates with the caudal flagella and the plasma membrane, and is thought to have a role in adhesion and motility (Weiland et al. Reference Weiland, McArthur, Morrison, Sogin and Svard2005). Its absence in assemblage B isolates indicates that different processes and proteins may be involved in these key functional activities of the infection process – adhesion and motility (Steuart et al. Reference Steuart, O'Handley, Lipscombe and Thompson2008). As such, this finding may provide a basis for better understanding the differences in pathogenesis associated with infections caused by assemblage A and assemblage B isolates of Giardia.

As with Echinococcus, the ability to culture and observe Giardia isolates in the laboratory has been pivotal in our understanding of the variability that is inherent in this parasite, and has provided phenotypic data crucial in our interpretation of genetic data.

Cryptosporidium

Although described from mice in the early 1900s by Tyzzer (Reference Tyzzer1912), Cryptosporidium species did not become a major focus of research until the mid-1970s when the first human cases of cryptosporidiosis were reported. This was closely followed by the emergence of Cryptosporidium as a life-threatening opportunistic infectious agent in AIDS patients (Hunter and Thompson, Reference Hunter and Thompson2005) for which the lack of any curative treatment remains an obstacle in limiting the clinical impact of Cryptosporidium in endemic areas such as Africa. A major impediment to research on Cryptosporidium for many years was the inability to maintain the life cycle in in vitro culture. This prevented the ability to directly observe, and confirm the sequential development of life-cycle stages that had previously been described principally from in situ observations on tissues from animals. It was not until 2001 that Hijjawi succeeded in establishing culture conditions that would support the complete development of C. parvum, C. hominis and C. andersoni and allow their long-term maintenance in vitro, in both cell-associated and cell-free culture (Hijjawi et al. Reference Hijjawi, Meloni, Morgan and Thompson2001, Reference Hijjawi, Meloni, Morgan, Olson and Thompson2002, Reference Hijjawi, Meloni, Ng'anzo, Ryan, Olson, Cox, Monis and Thompson2004). As a consequence of these advances, it was possible to study the sequential development of Cryptosporidium over extended periods in culture. These studies not only confirmed the existence of previously described developmental stages, but also demonstrated novel stages not previously reported to occur in the life cycle (Hijjawi et al. Reference Hijjawi, Meloni, Morgan and Thompson2001, Reference Hijjawi, Meloni, Morgan, Olson and Thompson2002, Reference Hijjawi, Meloni, Ng'anzo, Ryan, Olson, Cox, Monis and Thompson2004; Thompson et al. Reference Thompson, Olson, Zhu, Enomoto, Abrahamsen and Hijjawi2005). Furthermore, studies in vitro combined with observations in mice (Hijjawi et al. Reference Hijjawi, Meloni, Morgan and Thompson2001, Reference Hijjawi, Meloni, Morgan, Olson and Thompson2002, Reference Hijjawi, Meloni, Ng'anzo, Ryan, Olson, Cox, Monis and Thompson2004) revealed that the development of Cryptosporidium comprises an extracellular phase, thus showing that Cryptosporidium is not an obligatory intracellular pathogen as previously thought. These studies also emphasized the superficial, epicellular nature of the relationship the parasite has with its host cell (Barta and Thompson, Reference Barta and Thompson2006; Butaeva et al. Reference Butaeva, Paskerova and Entzeroth2006; Borowski et al. Reference Borowski, Clode and Thompson2008; Valigurova et al. Reference Valigurova, Jirku, Koudela, Gelnar, Modry and Slapeta2008).

In addition to challenging accepted views on the life cycle of Cryptosporidium, the discovery of novel developmental stages in vitro (Hijjawi et al. Reference Hijjawi, Meloni, Morgan and Thompson2001, Reference Hijjawi, Meloni, Morgan, Olson and Thompson2002, Reference Hijjawi, Meloni, Ng'anzo, Ryan, Olson, Cox, Monis and Thompson2004), and more recently in vivo in mice and amphibia (Valigurova et al. Reference Valigurova, Jirku, Koudela, Gelnar, Modry and Slapeta2008), and observations on their behaviour, including a process similar to syzygy (association of gamonts (pre-gametes) end to end or in lateral pairing prior to the formation of gametes, found in most gregarine protists and perhaps, piroplasms (Barta and Thompson, Reference Barta and Thompson2006)), further challenged the parasite's coccidian affinities (Thompson et al. Reference Thompson, Olson, Zhu, Enomoto, Abrahamsen and Hijjawi2005; Barta and Thompson, Reference Barta and Thompson2006; Valigurova et al. Reference Valigurova, Hofmannova, Koudela and Vavra2007). Although believed to be coccidia for many years (Levine, Reference Levine1988), Cryptosporidium species were always viewed as atypical in light of their unusual autoinfective oocyst, lack of sporocysts, strange extracytoplasmic association with their host cell, unique (for coccidia) feeding organelle and insensitivity to anti-coccidial drugs (reviewed in Thompson et al. Reference Thompson, Olson, Zhu, Enomoto, Abrahamsen and Hijjawi2005; Barta and Thompson, Reference Barta and Thompson2006). An unexpected report of serological cross-reactivity with Monocystis sp., a gregarine (Bull et al. Reference Bull, Chalmers, Sturdee, Curray and Kennaugh1998) served to further question this relationship with coccidia. The question of this relationship has now been resolved, with molecular tools providing convincing evidence that Cryptosporidium species share a common ancestor with what were thought to be distantly related apicomplexan protists, the gregarines, rather than with the coccidia (Carreno et al. Reference Carreno, Martin and Barta1999). Thus, the coccidia may be considered as being more closely related to malaria and haemogregarine blood parasites than they are to Cryptosporidium species, despite the latter being traditionally considered ‘coccidia’ (Barta and Thompson, Reference Barta and Thompson2006).

The similarities among cryptosporidian and gregarine parasites highlight some fascinating aspects that they have in common. Consideration of these may reveal insights into the interfacial relationship between Cryptosporidium and its host, particularly the function of its feeding organelle, a structure which is not seen in other coccidia. It has been proposed that the feeding mechanism employed by Cryptosporidium is a form of myzocytosis (predatory mode of feeding in which the parasite cell pierces the cell wall and/or membrane of the prey (host) cell with a feeding tube, and sucks out the cellular contents and digests it (Leander and Keeling, Reference Leander and Keeling2004; Barta and Thompson, Reference Barta and Thompson2006)). However, Cryptosporidium species' ancestral myzocytotic feeding has evolved into an epicellular association, with the vertebrate epithelial cell characterized by an elaborate membranous feeding organelle that develops from the apical region of the zoite after internalization of the parasite within the host cell (Barta and Thompson, Reference Barta and Thompson2006; Butaeva et al. Reference Butaeva, Paskerova and Entzeroth2006; Valigurova et al. Reference Valigurova, Hofmannova, Koudela and Vavra2007, Reference Valigurova, Jirku, Koudela, Gelnar, Modry and Slapeta2008). Evolutionary studies suggest that primitive apicomplexan parasites along with dinoflagellates use myzocytosis in their parasitic and micropredatory roles respectively (Leander and Keeling, Reference Leander and Keeling2004). This suggests that the feeding organelle that is seen in Cryptosporidium species may represent an evolutionary modification to the ancestral myzocytotic morphological adaptations (Barta and Thompson, Reference Barta and Thompson2006). Apart from size, the only difference between the two modes of feeding is that Cryptosporidium has evolved a way to induce the host cell to overlay it with an extension of the host cell membrane (Barta and Thompson, Reference Barta and Thompson2006; Borowski et al. Reference Borowski, Clode and Thompson2008). Physical ingestion of host cell cytoplasm has not been observed in Cryptosporidium as it has for dinoflagellates, perkinsids, colpodellids and some gregarines. Barta and Thompson (Reference Barta and Thompson2006) have proposed that the intimate host association of the epimerite of gregarine trophozoites and the development of the feeding organelle by trophozoites of Cryptosporidium species, is derived evolutionarily from the micropredatory feeding methods of their shared common myzozoic ancestor, with nutrients apparently taken up via transmembrane transport in the apical region.

Future studies on the evolution of feeding mechanisms employed by the dinoflagellates and apicomplexan parasites will not only provide clues about the evolution of intracellular parasitism but, in the case of Cryptosporidium, will provide a better understanding of the host parasite relationship. We also know very little about the pathogenesis of cryptosporidial infections and the relative contribution of ‘intracellular’ and extracellular phases of development to disease processes and epidemiology (Borowski et al. Reference Borowski, Clode and Thompson2008). In this respect, and as pointed out by Dionisio (Reference Dionisio2002), it is interesting that no correlation has been found between histological intensity of Cryptosporidium infections and clinical severity (Manabe et al. Reference Manabe, Clark, Moore, Lumadue, Dahlman, Belitsos, Chaisson and Sears1998). The close relationship between Cryptosporidium and gregarines opens up a fertile area of research, not only with respect to better understanding host parasite relationships, but also parasite diversity and environmental ecology. Cryptosporidium has a broad host range which, given its affinities with gregarines, may extend further than lower vertebrates.

As with Echinococcus and Giardia, it has been direct observational studies on the parasite in vitro that has not only confirmed Cryptosporidium's affinities with gregarine protozoa but has provided the basis for a re-assessment of phylogenetic relationships and the nature of the host-parasite relationship.