ECHINOCOCCUS: BIOLOGY AND DISEASE

E. multilocularis and E. granulosus share some distinct features in their life cycle. The adult worms live in the intestine of their respective final host (dogs for E. granulosus, foxes, dogs and cats for E. multilocularis) where sexual reproduction and subsequent egg production takes place. Faecal shedding spreads the eggs into the environment, where they are accidentally taken up by suitable intermediate hosts, such as small rodents for E. multilocularis, and cattle and sheep for E. granulosus. These eggs contain the first larval stage, the oncosphere, and during stomach passage, the oncosphere is activated and leaves the protective egg. It actively penetrates the intestinal lining, and migrates via blood and lymphatic vessels to the visceral organs. These are primarily the liver for E. multilocularis, and the liver, lung and other target organs in the case of E. granulosus. There, these oncospheres develop into metacestodes, which represent the second larval stage. Within these metacestodes, protoscolex development takes place, and upon oral uptake by the respective final host, protoscoleces attach to the intestinal epithelium and develop into adult worms, thus concluding the life cycle (Rausch, Reference Rausch, Thompson and Lymbery1995).

Metacestodes of both species are fluid-filled vesicles, and represent the disease-causing stage. They are structured into an inner cellular and an outer acellular compartment (Fig. 1). The outer, acellular surface of the metacestode is formed by the laminated layer, a carbohydrate-rich structure synthesized by the parasite. The laminated layer covers the entire metacestode surface, and is much more prominent in E. granulosus metacestodes (Morseth, Reference Morseth1967; Gottstein and Hemphill, Reference Gottstein and Hemphill1997). The actual larval tissue is formed by the germinal layer, which itself is composed of a population of different cell types (Figs. 1 and 2). One part consists of the tegument, which is directly associated with the inner surface of the laminated layer. It is characterized by microvilli-like extensions, microtriches, which protrude into the matrix of the laminated layer and thereby increase the resorbing surface of the parasite (Fig. 2). In addition, the germinal layer contains highly differentiated cell types including connective tissue, muscle cells, and glycogen storage cells, as well as many undifferentiated cells (Fig. 2).

Fig. 1. SEM of E. granulosus (A, B) and E. multilocularis (C, D) metacestodes. A and C represent lower magnification views of broken vesicles, exoposing both the acellular laminated layer (LL) and the germinal layer (GL); bars=1·5 mm. B and D represent higher magnification views of the germinal layer of E. granulosus and E. multilocularis, respectively; bars=400 μm.

Fig. 2. TEM of E. multilocularis metacestodes cultured in vitro. A shows a low magnification view of a section through the metacestode wall, with laminated (LL) and germinal layer (GL); bar=7 μm. B shows a higher magnification view, demonstrating LL, GL-tissue with different cell types including undifferentiated cells with a large nucleus and nucleolus (uc), and the tegument with microtriches (arrows); bar=3·2 μm. C and D show higher magnification views of longitudinally sectioned (C; bar=0·56 μm) and cross-sectioned (D; bar=0·43 μm) microtriches revealing actin microfilaments as main structural components. Also note the filamentous meshwork of electron-dense thin filaments that is embedded into the matrix of the LL.

The fully mature E. granulosus metacestode (i.e. hydatid cyst) is a single-chambered or septate unilocular cyst that shows expansive growth and thereby causes compression of neighbouring tissue, tissue damage and organ dysfunction (Kern, Reference Kern2006). Accidental or traumatic cyst rupture can result in release of cyst fluid and dissemination of protoscoleces, leading to anaphylactic reactions and metastases (secondary echinococcosis) (Stey and Jost, Reference Stey and Jost1993). E. granulosus metacestodes are surrounded by a host fibrous capsule, the adventitial layer, which is composed of host connective tissue.

In E. multilocularis infection, metacestode development exhibits different characteristics. There is no limiting host-tissue barrier comparable to the adventitial layer. The metacestode tissue represents a multivesicular structure that is reproducing asexually, by formation and budding of daughter vesicles, with progressive tumour-like growth (Ali-Khan et al. Reference Ali-Khan, Siboo, Gomersall and Faucher1983). This leads to the formation of a large and heterogenous parasitic mass that is intermingled with host connective tissue, proliferative in the periphery, and often necrotic in the centre (Gottstein and Hemphill, Reference Gottstein and Hemphill1997). Metastases formation may occur in other organs due to the release of germinal layer cells into the blood or lymph system (Ali-Khan et al. Reference Ali-Khan, Siboo, Gomersall and Faucher1983; Eckert et al. Reference Eckert, Thompson and Mehlhorn1983; Mehlhorn et al. Reference Mehlhorn, Eckert and Thompson1983). Thus, AE can resemble a benign malignant tumour (Vuitton, Reference Vuitton2009). Protoscolex development in humans has been rarely observed.

BENZIMIDAZOLES FOR THE CHEMOTHERAPEUTIC TREATMENT OF CYSTIC ECHINOCOCCOSIS (CE) AND ALVEOLAR ECHINOCOCCOSIS (AE)

For many years, the preferred treatment strategy for CE has been radical resection of the parasite mass (Kern, Reference Kern2003, Reference Kern2006). Other options include image-guide percutaneous treatment (PAIR=puncture, aspiration, injection, reaspiration) (Brunetti et al. Reference Brunetti, Troia, Garlaschelli, Gulizia and Filice2004) and chemotherapy (El-On, Reference El-On2002). Both surgery and PAIR are always accompanied by chemotherapy and, for inoperable cases, chemotherapy remains the only option. These cases include patients exhibiting cysts in crucial organs such as spine, pelvis and other sites that are not easily accessible, or patients with multiple cysts in several organs. The currently used drugs are mebendazole and albendazole. These drugs clearly have the potential to kill metacestodes, and to cure the patients. Benzimidazoles exert their action by binding to tubulin, inducing microtubule depolymerization and inhibiting polymerization of this essential cytoskeletal element (reviewed by Hemphill and Müller, Reference Hemphill and Mueller2009). However, the efficacy of these benzimidazoles depends on size and type of cyst (benzimidazoles are more effective against smaller cysts), on the age of the patient (these drugs work better in younger than in older patients), and on which organ is affected (e.g. cysts localized in bones are less susceptible to chemotherapy than those in liver and lungs). The duration of treatment plays a crucial role, with prolonged drug administration producing the most favourable results for the patients (Horton, Reference Horton1997; Franchi et al. Reference Franchi, DiVico and Teggi1999; Vutova et al. Reference Vutova, Mechkov, Vachkov, Petkov, Georgiev, Handjiev, Ivanov and Todorov1999). Praziquantel, a heterocyclic pyrazinoisoquinoline derivative, has been proposed to be used alongside benzimidazoles in CE-patients. Praziquantel is well tolerated, less toxic and better absorbed than albendazole. Praziquantel is used against the adult stages of Echinococcus and many other cestodes, and was shown to exhibit a high efficacy against protoscoleces (Morris et al. Reference Morris, Richards and Chinnery1986) and metacestodes in animal experiments (Urrea-Paris et al. Reference Urrea-Paris, Moreno, Casado and Rodriguez-Caabeiro1999, Reference Urrea-Paris, Casado, Moreno and Rodriguez-Caabeiro2001). The mode of action of praziquantel is a matter of debate. Respective studies were mostly carried out with schistosomes, where praziquantel represents the only drug that is currently marketed. Possibilities of praziquantel toxicity include its actions on nucleoside uptake (Angelucci et al. Reference Angelucci, Basso, Bellelli, Brunori, Pica Maattoccia and Valle2007), inhibition of phospho-inositide turnover (Wiest et al. Reference Wiest, Li, Olds and Bowen1992), binding to parasite actin (Tallima and El Ridi, Reference Tallima and El Ridi2007) and the parasite myosin light chain, and possibly inhibiting its functional activity (Gnanasekar et al. Reference Gnanasekar, Salunkhe, Mallia, He and Kalyanasundaram2009), interference in glutathione-S-transferase activity (McTigue et al. Reference McTigue, Williams and Tainer1995), and stimulation of Ca²⁺ entry through voltage-operated Ca²⁺ channels (VOCCs) (Kohn et al. Reference Kohn, Roberts-Misterly, Anderson, Khan and Greenberg2003; Jeziorski and Greenberg, Reference Jeziorski and Greenberg2006). The most compelling evidence for the involvement of VOCCs was recently provided by Nogi et al. (Reference Nogi, Zhang, Chan and Marchant2009) : they used the free-living flatworm Dugesia japonicum and investigated regeneration of fragments excised from this planarian that have the ability to reform a complete body plan, an ability that is driven by a totipotent population of stem cells called neoblasts. In the presence of praziquantel, however, this regeneration process yielded complete duplication of the entire antierior-posterior axis, resulting in two-headed organisms with duplicated nervous and organ systems. This effect of praziquantel was selectively ablated by in vivo RNAi of VOCC beta-subunits, but not by knock-down of alpha-subunits. At higher doses of praziquantel, knock-down of VOCC beta-subunit also conferred resistance to praziquantel, confirming the critical involvement of this beta subunit in the action of the drug.

A combined treatment regimen with albendazole and praziquantel given during the month prior to surgery of E. granulosus-infected patients increased the number of human patients with non-viable protoscoleces as compared to therapy with albendazole alone (Cobo et al. Reference Cobo, Yarnoz, Sesma, Fraile, Aizcorbe, Trujillo, Diaz-de-Liano and Ciga1998). Thus, praziquantel is regarded as useful in cases where cyst content is spilled during surgery.

For the treatment of AE, surgery and chemotherapy are the two only treatment options (Ammann and Eckert, Reference Ammann, Eckert, Thompson and Lymbery1995). Spontaneous cure of AE, leading to calcified lesions, is possible, but it is not known how commonly this occurs (Gottstein and Hemphill, Reference Gottstein and Hemphill1997; Vuitton, Reference Vuitton2009). Radical surgery of viable lesions is carried out if possible, but can be difficult to achieve, and only about one third of all cases are actually operated on. In contrast to E. granulosus metacestodes, E. multilocularis metacestodes are almost exclusively located in the liver. However, metastases can occur, also involving distant sites including lungs, spleen and brain. Chemotherapy should last for at least 2 years post-surgery, and monitoring of patients should be continued for 10 years. Inoperable cases must undergo long-term chemotherapy, often life-long (Reuter et al. Reference Reuter, Jensen, Buttenschoen, Kratzer and Kern2000; Reference Reuter, Buck, Manfras, Kratzer, Seitz, Darge, Reske and Kern2004; Vuitton, Reference Vuitton2009). The experiences with long-term use of benzimidazoles has had an enormous impact on the use of surgery. Generally, palliative surgery should be avoided and replaced by benzimidazole treatment alone or combined with percutaneous or perendoscopic interventions to treat biliary or vascular complications (Bresson-Hadni et al. Reference Bresson-Hadni, Vuitton, Bartholomot, Heyd, Godart, Meyer, Hrusovsky, Becker, Mantion, Lenys and Miguet2000, Reference Bresson-Hadni, Delabrousse, Blagosklonov, Bartholomot, Koch, Miguet, André Mantion and Angèle Vuitton2006; Vuitton, Reference Vuitton2009). Extensive animal experimentation and clinical experience in human patients have both demonstrated that albendazole and mebendazole exhibit a parasitostatic rather than parasiticidal effect against E. multilocularis metacestodes (Reuter et al. Reference Reuter, Buck, Manfras, Kratzer, Seitz, Darge, Reske and Kern2004; Vuitton, Reference Vuitton2009). Thus benzimidazoles only prevent parasite growth, and the recurrence rates after interruption of therapy are high. Nevertheless, clinical studies have shown that chemotherapy has significantly increased the 10-year survival rate of inoperable or non-radically operated AE patients from 6–25% to 80–83% (Hemphill et al. Reference Hemphill, Spicher, Stadelmann, Mueller, Naguleswaran, Gottstein and Walker2007).

Benzimidazoles are generally well tolerated, but problems can occur. Adverse reactions include hepatotoxicity, alopoecia, gastrointestinal disturbances and leukopaenia, which is sometimes severe and irreversible (Horton, Reference Horton1997). Risks also include embryotoxicity and teratogenicity. In order to improve bioavailability, liposome-entrapped formulations and emulsions of albendazole have been tested (Wen et al. Reference Wen, New, Muhmut, Wang, Wang, Zhang, Shao and Craig1996; Chai et al. Reference Chai, Menghebat Wie, Deyu, Bin, Jincao, Chen, Xiong, Yiding, Xiuling, Dolikun, Guliber, Yanchun, Fanghua and Shuhua2004). The use of cimetidine is discouraged, since it increases intestinal absorption (Wen et al. Reference Wen, New, Muhmut, Wang, Wang, Zhang, Shao and Craig1996; Schipper et al. Reference Schipper, Koopmans, Nagy, Butter, Kaeger and Van Boxtel2000), which in turn could result in toxic effects.

Major problems associated with benzimdazoles are the intra- and inter-individual variations of the pharmacokinetics, which make it necessary to measure albendazole sulphoxide and mebendazole plasma levels on a regular basis in order to adjust drug dosage and to avoid toxicity (Vuitton, Reference Vuitton2009). This is only possible in specialized facilities, and thus precludes the use of benzimidazoles in those endemic areas where efficient drugs would be mostly needed. In addition, it is unknown whether there is a stage-specific response to exposure to benzimidazoles, and the actual concentration at the site where the drug exerts its action is not known. These pharmacological uncertainties, together with the difficulties in assessing metacestode viability in humans, show that novel and improved chemotherapeutical tools are needed in order to optimize the treatment of CE and AE.

THE ROLE OF IN VITRO CULTURE IN DRUG DISCOVERY

In order to identify novel potential alternatives for chemotherapy against echinococcosis, the strategy used most commonly has been whole organism screening of E. multilocularis and E. granulosus metacestodes and/or protoscoleces. In most cases, researchers have focused on broad-spectrum anti-infective agents and anti-cancer compounds, many of which also exhibited reasonable efficacy against Echinococcus (Hemphill and Müller, Reference Hemphill and Mueller2009).

In vitro chemotherapy studies on CE have mostly, but not exclusively, been carried out on protoscoleces, since these are easily obtained from cysts of an infected animal. The metabolism, and thus drug-susceptibility, of E. granulosus protoscoleces might however be different from the metacestode stage tissue. As an example, praziquantel is highly active against adult tapeworms and against protoscoleces, but only of limited efficacy against metacestodes. Conversely, albendazole is less active against protoscoleces, but of significantly higher efficacy against metacestodes (Taylor et al. Reference Taylor, Richards and Morris1989).

Infection of laboratory animals with E. granulosus protoscoleces results in the development of metacestodes, which mimics the process of secondary hydatid disease in humans, and this differentiation process can also be achieved in vitro (Walker et al. Reference Walker, Rossignol, Torgerson and Hemphill2004). E. granulosus metacestodes do not proliferate in vitro but, as in vivo, show a marked increase in size. Thus, extensive animal experimentation is necessary for constant supply of metacestode material. In contrast, E. multilocularis metacestodes, besides increasing in size, also proliferate asexually and, provided with the corresponding nutrients and growth factors, will form new vesicles either endogenously (from within the germinal layer) or exogenously (by budding of daughter metacestodes from older vesicles; Fig. 3). Thus, the fact that E. multilocularis metacestodes proliferate in vitro has rendered this model suitable for the experimental assessment of the effects of chemotherapeutically interesting compounds.

Fig. 3. Exogenous and endogenous proliferation of E. multilocularis metacestodes in vitro. Exogenous outgrowth of a newly formed vesicle (arrows in A–B) from a pre-exisiting parent vesicle appears to be followed by budding off of the daughter vesicle (C). Vesicles are also formed within older metacestodes, most likely by emerging out of the germinal layer, and these smaller metacestodes are released upon desintegration of the older metacestode (D, E).

Historically, the in vitro culture of E. multilocularis metacestodes was achieved by dissection of infected rodents and placing small pieces of infected tissue or vesicle suspension into a suitable culture medium at 37°C. After a few weeks, newly formed metacestodes emerged, which were infective when re-introduced back into rodents (Hemphill and Gottstein, Reference Hemphill and Gottstein1995). These were co-culture systems, as parasites could be maintained for several months in the presence of feeder cells, and in the absence of feeder cells, metacestodes degenerated within a few weeks. Jura et al. (Reference Jura, Bader, Hartmann, Maschek and Frosch1996) introduced a second in vitro co-culture system, which was based on co-incubation of homogenized parasite tissue obtained from infected rats with primary rat hepatocytes embedded in a collagen layer. Although useful in many instances (reviewed in Hemphill et al. Reference Hemphill, Stettler, Walker, Siles-Lucas, Fink and Gottstein2002, Reference Hemphill, Spicher, Stadelmann, Mueller, Naguleswaran, Gottstein and Walker2007; Brehm and Spilitois, Reference Spiliotis, Brehm, Rupp and Sohn2008a), these co-culture systems provided only moderate yields and only a very limited number of drugs could be studied. Thus, the amount of metacestodes that could be generated was not sufficient for larger scale investigations.

Recently, Spiliotis and Brehm (Reference Spiliotis, Brehm, Rupp and Sohn2008) have overcome this problem and established a large-scale cultivation system which utilizes homogenized metacestode tissue that is incubated in liquid culture together with rat hepatoma feeder cells. Using this system, high numbers of metacestodes with comparable sizes can be generated within a few weeks of incubation. Thus, a major task of feeder cells appears to be the provision and secretion of growth factors that are needed by the parasites to undergo proliferation, growth and differentiation. E. multilocularis metacestodes possess several receptors that interact with human epidermal growth factor (EGF), insulin, transforming growth factor-beta and steroid hormones (Brehm et al. Reference Brehm, Spiliotis, Zavala-Gongora, Konrad and Frosch2006; Brehm and Spiliotis, Reference Brehm and Spiliotis2008b), and these most likely allow the parasite to communicate with its host and adapt to alterations in living conditions, regulating gene expression to its favour. Of course, these receptors, and associated signalling pathways within the parasite, represent interesting targets for potential chemotherapeutical intervention (Brehm et al. Reference Brehm, Spiliotis, Zavala-Gongora, Konrad and Frosch2006; Gelmedin et al. Reference Gelmedin, Caballero-Gamiz and Brehm2008). In addition, Spiliotis et al. (Reference Spiliotis, Tappe, Sesterhenn and Brehm2004) developed the methodology to culture E. multilocularis metacestodes axenically, in the absence of feeder cells, under reducing and anaerobic conditions. In the presence of oxygen, these metacestodes degenerate within few weeks (Spilitois et al. Reference Spiliotis, Tappe, Sesterhenn and Brehm2004), illustrating the high sensitivity of E. multilocularis metacestodes to oxygen intermediates. The specific adaptation of the mitochondrial respiratory system to anaerobic environments has been shown also for E. multilocularis protoscoleces (Matsumoto et al. Reference Matsumoto, Sakamoto, Shiniyo, Kido, Yamamoto, Yagi, Miyoshi, Nonaka, Katakura, Kita and Oku2008). Thus, another important task of feeder cells is to remove toxic oxygen intermediates (Brehm and Spiliotis, Reference Brehm and Spiliotis2008a).

DRUG CANDIDATES FOR THE TREATMENT OF ECHINOCOCCOSIS

Cystic echinococcosis (CE)

A selection of drugs and potential drug candidates against CE is provided in Table 1. In many instances, in vitro studies on compounds active against E. granulosus have employed protoscoleces. Experimental prophylactic therapy of E. granulosus protoscoleces was carried out as a model that would mimic spillage during surgery, by treating protoscoleces with praziquantel (Urrea-Paris et al. Reference Urrea-Paris, Casado, Moreno and Rodriguez-Caabeiro2001) or a combination of praziquantel and albendazole (Casado et al. Reference Casado, Urrea-Paris, Moreno and Rodriguez-Caabeiro2001) prior to injection into mice. The combination of albendazole and praziquantel has been used successfully in the treatment of human CE (Cobo et al. Reference Cobo, Yarnoz, Sesma, Fraile, Aizcorbe, Trujillo, Diaz-de-Liano and Ciga1998). Other promising compounds with in vitro protoscolicidal actions against E. granulosus were cetrimide (Frayha et al. Reference Frayha, Bikhazi and Kachachi1981) and the ionophore monensin (Rogan and Richards, Reference Rogan and Richards1986), but these drugs were rather ineffective against metacestodes. Levamisole and ivermectin, which are classically used against nematode infections, exhibited in vitro activities similar to benzimidazoles (Martinez et al. Reference Martinez, Perez-Serrano, Bernadina and Rodriguez-Caabeiro1999; Elissondo et al. Reference Elissondo, Ceballos, Alvarez, Sánchez Bruni, Lanusse and Denegri2009). The direct injection of ivermectin into E. granulosus cysts in laparotomised patients has also shown parasiticidal effects (Ochieng'-Mitula and Burt, Reference Ochieng'-Mitula and Burt1996). The combination of ivermectin plus albendazole has shown synergistic effects in infected mice (Moreno et al. Reference Moreno, Casado, Urrea-París and Rodríguez-Caabeiro2002). Benzimidazoles vary considerably with regard to their protoscolicidal action. Albendazole and albendazole sulphoxide have been shown to be active against E. granulosus protoscoleces in vitro (Chinnery and Morris, Reference Chinnery and Morris1986; Morris et al. Reference Morris, Chinnery and Ubhi1987; Perez-Serrano et al. Reference Perez Serrano, Casado, Guillermo, Denegri and Rodriguez-Caabeiro1994), but the in vitro protoscolicidal action of these drugs is rather slow and requires a longer incubation period compared to mebendazole (Morris et al. Reference Morris, Chinnery and Ubhi1987; Rodriguez-Caabeiro et al. Reference Rodriguez-Caabeiro, Casado and Juarez-Pelaez1989) or oxfendazole, the major fenbendazole sulphoxide metabolite. Against E. granulosus infection in rodents, a combination of fenbendazole and netobimin (Garcia-Llamazarez et al. Reference Garcia-Llamazares, Alvarez-de-Felipe, Redondo-Cardena, Voces-Alonso and Prieto-Fernandez1997) showed synergistic effects, allowing the administration of lower drug dosages. Oxfendazole, like albendazole, is a benzimidazole, used in veterinary medicine for the treatment of nematode infections, and has a similar antimicrobial spectrum but a longer half-life. Experimental treatments of naturally E. granulosus-infected sheep and goats suggested that oxfendazole may be as efficacious as albendazole, but does not require daily uptake of the drug because of its prolonged bioavailability (Blanton et al. Reference Blanton, Wachira, Zeyhle, Njoroge, Magambo and Schantz1988; Dueger et al. Reference Dueger, Moro and Gilman1999). Nitazoxanide, a nitro-thiazole-analogue (Hemphill et al. Reference Hemphill, Mueller and Esposito2006) induced severe damage to E. granulosus protoscoleces and the germinal layer of E. granulosus metacestodes within few days of in vitro culture (Walker et al. Reference Walker, Rossignol, Torgerson and Hemphill2004). Nitazoxanide treatment was not effective against experimental E. granulosus infection in sheep, but oxfendazole treatment, and a combination of oxfendazole and nitazoxanide, significantly decreased the number of fertile cysts and increased the number of degenerated cysts (Gavidia et al. Reference Gavidia, Gonzalez, Lopera, Jayashi, Angelats, Barron, Ninaquispe, Villareal, Garcia, Verastegui and Gilman2009). On the other hand, a recent case report has suggested beneficial effects of nitazoxanide in the treatment of refractory bony hydatid disease in a human patient (Winning et al. Reference Winning, Braslins and McCarthy2009). This patient had been suffering from progressive disease despite treatment with albendazole and praziquantel, and the clinical response on nitazoxanide treatment showed marked improvement in the soft tissue cysts, with stable disease in the bony pelvis. Although further studies are required, this report suggests that nitazoxanide may be an effective treatment option in CE, particularly in patients with progressive disease who are receiving conventional treatment (Winning et al. Reference Winning, Braslins and McCarthy2009).

Table 1. A selection of drug candidates against cystic hydatid disease caused by Echinococcus granulosus.

Profound protoscolicidal activity was also reported for another benzimidazole-derivative, flubendazole (Ellissondo et al. Reference Elissondo, Dopchiz, Ceballos, Alvarez, Sanchez Bruni, Lanusse and Denegri2006, Reference Elissondo, Ceballos, Alvarez, Sánchez Bruni, Lanusse and Denegri2009). Further studies subsequently showed that flubendazole also exhibited anti-E. granulosus metacestocidal activities in vitro and in vivo in experimentally infected mice (Elissondo et al. Reference Elissondo, Ceballos, Dopchiz, Andresiuk, Alvarez, Sanchez Bruni, Lanusse and Denegri2007). The same authors, together with Kammerer and Perez-Esandi (Reference Kammerer and Perez-Esandi1975), showed that thymol, one of the major components of the essential oils of Thymus, was found to exhibit substantial protoscolicidal activity (Elissondo et al. Reference Elissondo, Albani, Gende, Eguaras and Denegri2008).

Cyclosporin A, employed mainly as an immunosuppressant during the management of organ transplants, also exhibits anti-echinococcal activity. The administration of cyclosporin A in five consecutive daily doses, beginning 2 days prior to the infection of mice with E. granulosus protoscoleces, resulted in a significant reduction in cyst numbers and cyst masses measured at 20 weeks after infection. However, no changes in cyst mass and numbers were recorded when the drug was administered 18 weeks after infection, but the wet weight was decreased by 42% compared with untreated controls. Ultrastructural examination of the germinal membrane and laminated layer of late-treated E. granulosus revealed abnormalities in all cysts studied whereas control and early-treated hydatids were normal (Hurd et al. Reference Hurd, Mackenzie and Chappell1993). Cyclosporin A has also been shown to affect E. granulosus protoscoleces in vitro (Colebrook et al. Reference Colebrook, Jenkins, Jones, Tatarczuch and Lightowlers2004).

Alveolar echinococcosis (AE)

A selection of drugs and potential drug candidates for the treatment of AE is provided in Table 2. Cyclosporin A, contrary to what was found in CE, lacked anti-parasitic activity against E. multilocularis infection in experimentally infected mice (Liance et al. Reference Liance, Bresson-Hadni, Vuitton, Lenys, Carbillet and Houin1992). Doxorubicin, or hydroxyldaunorubicin, a DNA-interacting drug used widely in the treatment of a wide range of cancers (Launchbury and Habboubi, Reference Launchbury and Habboubi1993), was bound to poly-isohexylcyanoacrylate nanoparticles (a colloidal biodegradable drug carrier) and applied in E. mutlilocularis-infected mice, resulting in the reduction of the development of the parasite in the liver and a reduced viability of the metacestode. Free doxorubicin or unbound nanoparticles had no antiparasitic activity (Liance et al. Reference Liance, Nemati, Bories and Couvreur1993). Animal experimentation in rodents demonstrated parasitostatic effects of mytomicin C, piperazine and quinolone derivates, alkylaminoethers and propargylic alcohols, either at a lower level or comparable to benzimidazoles (reviewed in Siles-Lucas and Hemphill, Reference Siles-Lucas and Hemphill2002). The efficacy of praziquantel was inadequate (Marchiondo et al. Reference Marchiondo, Ming, Andersen, Slusser and Conder1994), although showing some effects on protoscoleces in vitro (Taylor and Morris, Reference Taylor and Morris1988). Also, the treatment of E. multilocularis-infected mice with alpha-difluoromethylornithine was not successful (Miyaji et al. Reference Miyaji, Katakura, Matsufuji, Murakami, Hayashi, Oku, Okamoto and Kamiya1993).

Table 2. A selection of drug candidates against alveolar echinococcosis caused by Echinococcus multilocularis

* Parasitostatic

More recently, studies on chemotherapeutically interesting compounds have employed in vitro cultured parasites. Nitazoxanide, a broad-spectrum anthelminthic also used for treatment against enteric bacteria, Giardia and Cryptosporidium (Hemphill et al. Reference Hemphill, Mueller and Esposito2006), was identified as a compound inducing significant distortion of the germinal layer in vitro, and nitazoxanide-treated E. multilocularis metacestodes were non-viable when introduced into susceptible mice (Stettler et al. Reference Stettler, Fink, Walker, Gottstein, Geary, Rossignol and Hemphill2003). Reuter et al. (Reference Reuter, Manfras, Merkle, Harter and Kern2006) investigated the in vitro efficacy of a series of compounds against E. multilocularis metacestodes, including albendazole, artemether, caspofungin, itraconazole, ivermectin, methiazole, miltefosine, nitazoxanide, rifampicin and trimethoprim/sulfamethoxazole. They found that albendazole, itraconazole, methiazole and nitazoxanide effectively destroyed parasite vesicles in vitro. However, after drug discontinuation, re-growth of vesicles occurred, indicating a parasitostatic effect only. Combination treatment with albendazole/nitazoxanide at concentrations between 1 and 10 μg/ml for 3 weeks yielded no re-growth of parasites during 8 months of drug discontinuation, and the subsequent evaluation in a bioassay in gerbils did also not result in viable parasite infections. In this respect, Stettler et al. (Reference Stettler, Rossignol, Fink, Walker, Gottstein, Merli, Theurillat, Thormann, Dricot, Segers and Hemphill2004) showed that nitazoxanide, applied orally to E. multilocularis infected mice, either alone or in combination with ABZ, exhibited a profound anti-parasitic efficacy, with the albendazole/nitazoxanide combination yielding the most promising outcome in terms of reducing parasite weight. The pharmacokinetic analysis of corresponding serum levels in mice showed that the application of albendazole in combination with nitazoxanide increased considerably the levels and the half-life of albendazole sulfoxide (Stettler et al. Reference Stettler, Rossignol, Fink, Walker, Gottstein, Merli, Theurillat, Thormann, Dricot, Segers and Hemphill2004). Therefore, the increased efficacy observed in mice could be the result of an increased availability of albendazole sulfoxide in mice receiving the combination treatment. Despite these promising results, neither nitazoxanide monotherapy nor nitazoxanide-albendazole combination therapies were highly effective in human patients sufffering from AE (Kern et al. Reference Kern, Abboud, Kern, Stich, Bresson-Hadni, Guerin, Buttenschoen, Gruener, Reuter and Hemphill2008).

Amphotericin B desoxycholate (cAMB), an antifungal compound, effectively inhibited the growth of E. multilocularis metacestodes, first in vitro, and subsequently in human patients in vivo (Reuter et al. Reference Reuter, Merkle, Brehm, Kern and Manfras2003a, Reference Reuter, Buck, Grebe, Nüssle-Kügele, Kern and Manfrasb). A major limitation of cAMB is its mode of administration (intra-venous), which makes it unsuitable for prolonged use, except for salvage treatment (Reuter et al. Reference Reuter, Buck, Grebe, Nüssle-Kügele, Kern and Manfras2003b). Also, the action of cAMB is only parasitostatic and, since the drug is nephrotoxic, its widespread use is limited. Nevertheless, prolonged application of cAMB for months to years may be feasible in some cases, as side effects are mild and serious organ damage does not appear to occur (Reuter et al. Reference Reuter, Buck, Grebe, Nüssle-Kügele, Kern and Manfras2003b).

In vitro studies on E. multilocularis metacestodes and E. granulosus protoscoleces have shown that genistein, representing a major component of soya and the most prominent isoflavonoid, as well as a number of genistein derivatives, are also highly effective against these parasites. The molecular basis of the efficacy of genistein and its derivative Rm6423 have not yet been elucidated, but these compounds could interfere in signalling, for instance, through an inhibition of the tyrosine kinase activity associated with the epidermal growth factor receptor identified in E. multilocularis (see Brehm et al. Reference Brehm, Spiliotis, Zavala-Gongora, Konrad and Frosch2006). Recently, Gelmedin et al. (Reference Gelmedin, Caballero-Gamiz and Brehm2008) identified pyridinyl imidazoles as ATP-competitive inhibitors of a p38-like mitogen-activated protein kinase (MAPK) of E. multilocularis by adding them to in vitro cultures, demonstrating death of parasite vesicles at concentrations that did not affect cultured mammalian cells.

An endogenous metabolite of oestrogen with both anti-angiogenic and anti-tumour effects, 2-methoxyestradiol (2-ME) (reviewed by Schumacher and Neuhaus, Reference Schumacher and Neuhaus2001) was shown to induce severe damage to E. multilocularis metacestodes in vitro in a dose-dependent manner (Spicher et al. Reference Spicher, Naguleswaran, Ortega-Mora, Mueller, Gottstein and Hemphill2008a). However, 2-ME-treatment of experimentally infected mice did not result in a reduction in parasite weight compared to the control, demonstrating that in vitro and in vivo situations are not always comparable. Best results were achieved with a treatment using a combination of 2-ME and albendazole, which lead to a reduction in parasite weight compared to albendazole treatment alone, but results did not show statistical significance (Spicher et al. Reference Spicher, Naguleswaran, Ortega-Mora, Mueller, Gottstein and Hemphill2008a). In vitro treatment of E. multilocularis and E. granulosus larval stages with the antimalarials dihydroartemisinin and artesunate exhibited promising results, while 6 weeks of in vivo treatment of mice with infected E. multilocularis metacestodes had no effect. Again, combination treatments of both drugs with albendazole led to a substantial but statistically not significant reduction in parasite weight compared to results with albendazole alone (Spicher et al. Reference Spicher, Roethlisberger, Lany, Stadelmann, Keiser, Ortega-Mora, Gottstein and Hemphill2008b).

The in vitro effect of isoprinosine and its derivates has also been demonstrated against protoscoleces of E. multilocularis (Lawton et al. Reference Lawton, Walchshofer and Sarciron2001).

RELEVANCE OF ECHINOCOCCUS METACESTODE SCREENING MODELS FOR OTHER CESTODES AND TREMATODES

Numerous studies have demonstrated that many of those drugs that were active in E. granulosus drug screening models were also of significant relevance for E. multilocularis and vice versa (see above). In addition, there is ample evidence that most compounds with good in vitro and in vivo efficacy against Echinococcus metacestodes are also of relevance for combatting infections by other cestode larval stages. For instance, cysticerci of T. taeniaeformis were highly sensitive to praziquantel (Becker et al. Reference Becker, Mehlhorn, Andrews and Thomas1981), and the same applied to T. solium and T. pisiformis (Garcia-Dominguez et al. Reference Garcia-Dominguez, Correa, Rabiela and Flisser1991; Martinez Zedillo et al. Reference Martinez Zedillo, Hoyo Badillo, Amezcua and Gonzalez Barranco1992). Albendazole and mebendazole (or modified derivatives) were also active against T. taeniaeformis cysticerci in experimentally infected mice (Verheyen et al. Reference Verheyen, Vanparijs, Borgers and Thienpont1978; Jain et al. Reference Jain, Gupta, Katiyar, Maitra, Singh and Bhakuni1989). To date, albendazole and praziquantel, taken as short-term treatments (8–15 days) are the main drugs of choice for chemotherapeutic treatment of neurocysticercosis in humans (Garcia et al. Reference Garcia, Gonzales, Evans and Gilman2003; Shandera and Kass, Reference Shandera and Kass2006). A recent in vitro study on T. crassiceps indicates that a combination of nitazoxanide and albendazole could be used for the treatment of cysticercosis infections (Palomares-Alonso et al. Reference Palomares-Alonso, Piliado, Palencia, Ortiz-Plata and Jung-Cook2007), but in vivo evidence is still lacking. The effects of other benzimidazoles, such as flubendazole, were evaluated in T. solium-infected swine, with promising results (Tellez-Giron et al. Reference Tellez-Giron, Ramos and Montante1981) that were subsequently confirmed in human patients (Tellez-Giron, Reference Tellez-Giron, Ramos, Dufour, Montante, Tellez, Rodriguez, Gomez-Mendez and Mireles1984).

The activity of benzimidazole derivatives against Hymenolepis larvae was evaluated in the intermediate host Tribolium confusum, demonstrating clear effects with regard to larval development (Novak and Blackburn, Reference Novak and Blackburn1985). However, in the rodent model it was shown that benzimidazole drugs were only effective against H. nana oncosphere infection, but already developed cysticercoids were difficult to cure (Gupta et al. Reference Gupta, Katiyar and Sen1981; Maki and Yanagisawa, Reference Maki and Yanagisawa1985), and neither flubendazole nor thiabendazole could clear H. nana cysticercoid infections in mice. Mebendazole and a series of modified benzimidazole derivatives were effective against H. nana and H. diminuta cysticercoids (Dubey et al. Reference Dubey, Abuzar, Sharma, Chatterjee and Katiyar1985).

In vitro studies on the effects of exposure of Mesocestoides corti tetrathyridia to anti-parasitic drugs revealed that liposomized praziquantel and albendazole had a deleterious effect on the parasite morphology and development (Hrckova et al. Reference Hrckova, Velebny and Corba1998; Britos et al. Reference Britos, Dominguez, Ehrlich and Marin2000; Saldana et al. Reference Saldana, Marin, Fernandez and Dominguez2001). The efficacy and mode of action of praziquantel were studied in the mouse model, showing that application of praziquantel had an adverse effect on the tetrathyridia burden in the liver and peritoneum (Hrckova and Velebny, Reference Hrckova and Velebny1995, Reference Hrckova and Velebny1997). Following the oral administration of mebendazole to M. corti-infected mice, a parasiticidal effect was observed (Heath et al. Reference Heath, Christie and Chevis1975; Eckert and Pohlenz, Reference Eckert and Pohlenz1976). This is in contrast to other cestode larvae, against which mebendazole exerts a parasitostatic effect only. Deleterious actions of other drugs on Mesocestoides tetrathyridia, including cyclosporin A (Chappell et al. Reference Chappell, Wastling and Hurd1989) and albendazole (Terenina et al. Reference Terenina, Kosovatova, Gerasimova and Shalaeva1998), have also been reported.

Potentially, the Echinococcus screening models could also have some, albeit limited, relevance for a number of trematode species, including Schistosoma japonicum, S. manoni, Chlonorchis sinensis, Fasciola hepatica and Opistorchis viverrini. For instance, praziquantel is the drug of choice for the treatment of schistosomiasis (King, Reference King2007). The drug is highly effective against the adult worm (as for cestodes), but has only a minor activity against the larval schistosomula. Another class of drugs that show similarities between Echinococcus and trematodes are the artemisinins and synthetic trioxolanes. These antimalarial drugs possess a broad spectrum of activity against trematodes, causing profound damage in vitro and substantially reducing the worm burden in experimentally infected mice (reviewed by Keiser and Utzinger, Reference Keiser and Utzinger2007a, Reference Keiser and Utzingerb). Echinoccoccus metacestodes were also susceptible to artesunate and dihydroartemisinin in vitro, while the schistosomula were found to be particularly susceptible to artemether and artesunate (Utzinger et al. Reference Utzinger, Chollet, Tu, Xiao and Tanner2002; Spicher et al. Reference Spicher, Roethlisberger, Lany, Stadelmann, Keiser, Ortega-Mora, Gottstein and Hemphill2008b). The promising activity of mefloquine, another anti-malarial drug, in mice experimentally infected with S. japonicum and S. mansoni, has been reported (Keiser et al. Reference Keiser, Chollet, Xiao, Mei, Jiao, Utzinger and Tanner2009), and recent studies in our laboratory also revealed that mefloquine has a profound impact on in vitro cultured E. multilocularis metacestodes (A. Hemphill et al. unpublished observations).

On the other hand, albendazole and mebendazole, the main anti-echinococcal drugs, have no impact on S. mansoni (Schmidt, Reference Schmidt1998). In contrast, flubendazole, a mebendazole-derivative, was active against S. mansoni in experimentally infected mice (Nessim et al. Reference Nessim, Hassan, William and El-Baz2000; Williams et al. Reference Williams, Guirgus and Nessim2003), and another benzimdazole, triclabendazole, was active against Fasciola hepatica (Robinson et al. Reference Robinson, Hoev, Fairweather, Dalton, McGonigle and Trudgett2001). As for Echinococcus, F. hepatica was susceptible to nitazoxanide treatment in vitro, and clinical studies showed that nitazoxanide could be used for the treatment of fasciolasis in children and adults (reviewed in Hemphill et al. Reference Hemphill, Mueller and Esposito2006).

WHERE TO GO FROM HERE?

As outlined in this review, considerable efforts have been undertaken to improve the therapeutic options for the treatment of CE and AE (reviewed in Vuitton, Reference Vuitton2009). These efforts have largely concentrated on the establishment of procedures on how to manage the diseases and by setting up guidelines for treatment and classifications of disease status. Although most successful to a large extent, the current benzimidazole-based chemotherapy is far from optimal and, owing to the limited efficacy of this class of compounds, their side-effects and their costs, alternative drugs or drugs that could be integrated into a combination treatment are clearly needed. Thus, considerably more input and support is needed from academic institutions as well as pharmaceutical and biotechnological industries and governmental agencies to provide solutions for these neglected diseases. A recent survey on the financial resources going into research and development funding in 2007 has shown, that HIV/AIDS received 1·1 billion US dollars, malaria and tuberculosis obtained over 400 million US dollars each, and kinetoplastid diseases and diarrhoeal diseases were granted over 125 and 114 million dollars each. On the other hand, helminth infections as a whole, encompassing nematodes, cestodes and trematodes, received only 51·6 million dollars (Voelker, Reference Voelker2009).

But finances alone will not provide novel possibilities. Until recently, Echinococcus drug discovery has been based on rather anecdotal reports, where a limited number of drugs belonging to a certain compound class have been investigated. However, the Echinococcus drug discovery process has been lacking several important aspects that are compulsary for successfully identifying the best and most interesting compounds. First, we need to gain access to comprehensive compound libraries, and we need to be able to screen these libraries. This can only be done through the implementation of easy-to-handle and reliable medium-to-high-throughput in vitro assays that allow screening of larger numbers of anti-parasitic drugs in an efficient manner. Preferentially, these assays do not rely on subjective microscopic evaluation, but on objective criteria, such as specific markers (e.g. enzyme activities) that indicate parasite viability/intactness or non-viability/damage, and which could be detected preferentially in an (at least) semi-automated system. The development of such assays is ongoing, but should be intensified.

Secondly, large-scale drug screening activities are only possible if sufficient numbers of parasite organisms can be generated in vitro. This prerequisite has been fullfilled due to the pioneering work of Brehm and co-workers (see Spiliotis and Brehm, Reference Spiliotis, Brehm, Rupp and Sohn2008), who developed an E. multilocularis metacestode in vitro culture system that allows the generation of massive numbers of metacestodes out of a relatively small quantity of parasite tissue. This does not only enable researchers to carry out numerous drug-screening assays, but also provides the basis for biochemical studies, including the identification of drug targets for specific compounds by affinity chromatography and mass spectrometry-based sequencing.

Thirdly, it is important to have access to genomic and EST databases. In 2008, shotgun sequencing of the E. multilocularis genome was completed (http://www.sanger.ac.uk/Projects/Echinococcus/). This opens the door for increased use of in silico approaches for drug target identification, similar to the discovery of the anti-metacestodicidal activity of clarithromycin reported by Mathis et al. already in 2005. In addition, the availiability of a genome database permits drug target identification by affinity chromatography of parasite extracts on drug-coupled matrices and subsequent protein identification by mass spectroscopy (Müller et al. Reference Müller, Naguleswaran, Müller and Hemphill2008a, Reference Müller, Sidler, Nachbur, Wastling, Brunner and Hemphillb).

Furthermore, a system is required to verify putative drug targets and investigate their functional role, preferentially by genetic means, such as over-expression or silencing of genes of interest. The basis for this was set in 2008, when a method for the long-term in vitro cultivation and proliferation of primary cells isolated from axenically grown E. multilocularis metacestodes was established by Spiliotis et al. (Reference Spiliotis, Lechner, Tappe, Scheller, Krohne and Brehm2008). Isolated E. multilocularis cells were transiently transfected with a plasmid carrying the gene coding for the cyano-fluorescent-protein (CFP), and the corresponding gene product was expressed and detected by Western blot analysis. When co-cultured with hepatocytes, cultured E. multilocularis cells form aggegates, and eventually undergo complete in vitro regeneration of metacestode vesicles. Prospectively, this could well lead to the development of transgenic larval stages, and even adult E. multilocularis worms, and genetic tools can be exploited to elucidate the exact functional role of putative drug targets in different stages of development (reviewed in Brehm and Spiliotis, Reference Brehm and Spiliotis2008a).

Finally, the in vitro activity of a compound that exhibits outstanding performance can be verified in vivo, initially preferentially in a relevant small laboratory animal model. The murine or gerbil models for primary and secondary AE (Stettler et al. Reference Stettler, Rossignol, Fink, Walker, Gottstein, Merli, Theurillat, Thormann, Dricot, Segers and Hemphill2004; Reuter et al. Reference Reuter, Manfras, Merkle, Harter and Kern2006) represent reliable tools for such in vivo studies.