Since the initial characterization of a member of the insulin receptor family in E. multilocularis, EmIR1 , few studies have been conducted to investigate the effects of mammalian insulin on flatworm parasite insulin signalling pathways and development. In each of the related parasites Schistosoma mansoni and S. japonicum, two EmIR1-like tyrosine kinases of the insulin receptor family were identified and, as originally shown for EmIR1, the possibility of an interaction of these receptors with host insulin was verified using the yeast two-hybrid system [11, 13]. These studies did, however, not address whether host-derived insulin would stimulate (or generally influence) parasite development and/or establishment within the host. Although Ahier et al.  later investigated effects of host insulin on glucose uptake of S. mansoni in vitro, significant stimulation was only achieved using hormone concentrations of 1 μM, which can be considered non-physiologically high since plasma levels of insulin in humans and animals usually range between 1 to 2 nM [34, 35]. Likewise, in studies on cestode systems conducted by Canclini and Esteves  (Mesocestoides corti) and Escobedo et al.  (Taenia crassiceps), effects on glucose metabolism or parasite development (T. crassiceps budding) were only observed at insulin concentrations several magnitudes higher than physiological concentrations. Hence, although several investigations had already addressed the possibility of insulin-based hormonal cross communication between flatworm parasites and mammalian hosts, it is still unclear to date whether host insulin at physiological concentrations indeed influences parasite development and metabolism or whether such effects are mediated by evolutionarily conserved insulin signalling systems of these parasites.
In the present study, we concentrated on a cestode, E. multilocularis, the larval stage of which displays a strong organ-tropism towards the liver where the highest insulin concentrations (up to 15 nM) within mammals can be measured [4–6]. Several independent lines of evidence clearly indicate that E. multilocularis larvae are responsive to exogenously added host-insulin at physiological concentrations. First, 10 nM insulin significantly increased the production of metacestode vesicles from parasite stem cells as well as the re-differentiation of protoscoleces towards metacestode vesicles, and also significantly stimulated parasite stem cell proliferation in primary cell cultures and metacestode vesicles, as measured by the incorporation of BrdU. Second, the uptake of radioactively labelled glucose by metacestode vesicles was significantly stimulated in the presence of 10 nM host insulin. Third, exogenously added host insulin clearly affected the phosphorylation profiles of components of the PI3K/Akt signalling pathway in the metacestode. On the basis of these data, we propose that insulin constitutes an important host factor that influences the development and physiology of E. multilocularis larvae within the liver. The observed effects were most striking for initial metacestode development from stem cells, which could aid the parasite in establishing itself early during an infection, when it is most vulnerable to attacks by the host immune system . Compared to primary cells, somewhat lower effects were observed on the proliferation of mature metacestode vesicles, which could be due to the fact that this stage contains significantly lower proportions of stem cells that are capable of proliferation than the primary culture system (Koziol et al., submitted for publication). On the other hand, given the important role of glycogen as the main energy source for larval cestode metabolism, the observed effects of host insulin on glucose uptake by E. multilocularis could be important for long-term persistence of the parasite within the host. Whether the insulin-stimulated re-differentiation of protoscoleces towards the metacestode is important in vivo still remains to be determined. Protoscolex re-differentiation in experimental secondary echinococcosis or following accidental or surgery-induced rupture of parasite cysts is a well described phenomenon [2, 3] and at least for E. granulosus it is thought that parasite persistence within the host is aided by re-differentiation of existing protoscoleces once the mother hydatid cyst experienced physical damage . In this regard, the influx of elevated concentrations of host insulin into ruptured parasite cysts, followed by increased re-differentiation of protoscoleces, may well contribute to prolonged parasite survival. However, whether these mechanisms are also relevant to E. multiocularis infections is still not clear. In any case, the observed effects of 1 nM and 10 nM insulin on protoscolex re-differentiation again demonstrate that E. multilocularis larvae are well responsive to physiological concentrations of insulin.
Since our data revealed that insulin significantly stimulates metacestode vesicle formation from primary cell cultures in a system that mimics the natural oncosphere-metacestode-transition, it is, of course, tempting to speculate that the relatively strict organ-tropism of E. multilocularis towards the host liver [1, 2] may, at least in part, depend on the high insulin concentrations usually present in this organ. Although this is supported by our data showing that host insulin stimulates proliferation of E. multilocularis stem cells, which is in line with the role of insulin signalling in proliferation control of neoblasts in free-living flatworms , further experiments addressing insulin effects on naturally isolated oncospheres are necessary to obtain a conclusive picture. This would also require comparative analyses on oncospheres from E. granulosus, which display a relaxed liver organ tropism, and those of Taenia solium (or Taenia saginata), which usually don’t develop in the host liver, despite an entry route into the host comparable to that of E. multilocularis. It is interesting to note in this context that Escobedo et al.  did not observe effects on T. solium cysticerci under high insulin treatment conditions that stimulated larval budding in T. crassiceps. However, care has to be taken in the interpretation of their results, since for T. solium the authors measured scolex evagination which is not, per se, a developmental process.
According to the theory of hormonal host-helminth cross-communication, endo- and paracrine hormonal systems of mammals (or even invertebrates) could influence the physiology and development of metazoan parasites through stimulation of evolutionarily conserved signalling systems [2, 37–39]. This theory has thus far been supported by several in vitro studies showing that parasite surface receptor kinases of the insulin-, the EGF- and the TGF-β-families can principally bind respective host-derived hormones [10, 11, 13, 37–41]. One of the most convincing examples supporting this theory has been brought up by Vicogne et al.  who demonstrated that human EGF can activate an EGF-receptor, such as tyrosine kinase of S. mansoni in vitro and at the surface of schistosomes, and that exogenously added EGF also influences protein and DNA synthesis in the parasite. We now propose the host-insulin-E. multilocularis-EmIR1 system as another example that supports this theory. Again, several lines of evidence clearly indicate that at least some of the effects of host insulin on E. multilocularis development and physiology involve binding of the host hormone to the insulin-receptor-like tyrosine kinase EmIR1. First, exogenously added host insulin influences EmIR1 phosphorylation patterns in the metacestode which is prevented in the presence of an anti-insulin-receptor inhibitor. Second, host insulin particularly influenced the phosphorylation of components of the PI3K/Akt pathway, which is known to act downstream of insulin-receptor tyrosine kinases in many organisms [7–9], and this was prevented in the presence of an insulin receptor inhibitor. Since the stimulation of the PI3K/Akt pathway through insulin receptors requires IRSs as intermediate signalling molecules , a binding site for which is present in EmIR1 (but not in EmIR2), the activation of this pathway in E. multilocularis most likely involves EmIR1. Third, although E. multilocularis encodes ILPs, the expression levels of the respective genes in the metacestode are very low and none of the parasite ILPs interacted with EmIR1 in yeast two-hybrid assays, indicating that host insulin is the only EmIR1 activating hormone present in significant concentrations around the growing metacestode. In this respect, it is even tempting to speculate that EmIR1 entirely lost the capacity to be stimulated by parasite-encoded ILPs since it is most active in parasite stages that have contact with elevated concentrations of host insulin. We, thus, propose that several of the actions of insulin on the E. multilocularis metacestode, particularly the stimulation of glucose uptake and the stimulation of metacestode proliferation, are mediated by direct binding of the host hormone to EmIR1, followed by subsequent activation of insulin-dependent parasite signalling pathways. This should be particularly relevant in the Echinococcus GSCs, which display the highest expression levels of EmIR1 and are the cell type responsible for carbohydrate storage.
Although EmIR1 at the protein level was not detected in the E. multilocularis primary cell cultivation system, we could observe clear effects of host insulin on the formation of metacestode vesicles from parasite stem cells. These effects are, thus, most probably mediated independently of EmIR1 and in the present study we identified a second E. multilocularis insulin receptor molecule, EmIR2, which could be involved in the effects on parasite stem cells. On the one hand, our histochemical analyses showed that EmIR2 expression is dispersed through primary cell aggregates, which contain a large number of parasite stem cells . Furthermore, the in situ hybridization experiments presented in this work clearly indicate that at least in developing protoscoleces, emir2 transcripts are closely associated with the proliferation zone where parasite stem cells are most active (Koziol et al., submitted for publication), indicating a link between EmIR2 and stem cell proliferation or differentiation. The presence of two insulin receptor encoding genes in E. multilocularis closely resembles the situation in the related schistosomes, which also express two molecules of this class [11–13]. As with the schistosome receptor LBDs, which interacted with human insulin in the yeast two-hybrid system [11, 13], we herein demonstrated that in addition to EmIR1, EmIR2 can also interact with the host hormone. Since the Echinococcus emilp2 gene was expressed at low, but detectable, levels in primary cells and since the encoded peptide, EmILP2, interacted with EmIR2 in the yeast two-hybrid system, we cannot exclude that a certain level of stimulation of EmIR2 by EmILP2 in primary cells could contribute to initial parasite development within the liver. However, our experiments clearly indicate that physiological levels of human insulin, that should be present at the site of initial parasite development from the oncosphere, can significantly add to these effects. Hence, it is conceivable that during the oncosphere-metacestode transition both EmILP2 and human insulin bind to EmIR2, which could lead to higher activation of the parasite receptor than through EmILP2 alone, and which could thus promote rapid parasite establishment. Whether this indeed occurs in vivo and which parasite signalling pathways act downstream of EmIR2, given that it lacks the conserved NPXY motif, still remains to be established. Unfortunately, and in contrast to the metacestode vesicle culture system, membrane fractionation and insulin stimulation studies are very difficult to carry out on the stem cell cultivation system due to the fragility of stem cell aggregates and their high sensitivity to serum-free cultivation conditions. Nevertheless, given that EmIR2 is capable of interacting with human insulin in the yeast two hybrid system and that it is expressed as the only parasite insulin receptor in the primary cell system, hormonal host-parasite cross-communication through insulin-binding to EmIR2 could indeed play a significant role in parasite establishment within the liver.
Ahier et al.  and You et al.  previously used inhibitors specifically designed to bind to insulin receptor-like kinases and observed deleterious effects on the uptake and consumption of glucose by schistosomes, indicating that at least the mechanisms of glucose uptake, similar to Echinococcus as shown in this study, are under the control of insulin signalling in these parasites. In the present study, we employed HNMPA(AM)3, the same inhibitor used by You et al. , and observed various effects on the development of metacestode vesicles from primary cells, on the survival of mature metacestode vesicles and on the re-differentiation process from protoscoleces towards the metacestode. In mature metacestode vesicles, only relatively high concentrations (100 μM) of HNMPA(AM)3 led to killing and we suggest that this mostly involved binding of the drug to EmIR1, accompanied by defects in glucose uptake and consumption. That the drug can principally bind to EmIR1 is supported by our in silico analyses showing that the parasite receptor’s ATP-binding pocket is capable of harbouring HNMPA(AM)3 with considerable affinity. Compared to mature metacestode vesicles, the effects of HNMPA(AM)3 on primary cells were much more dramatic. Already at a concentration of 25 μM, the insulin receptor inhibitor completely prevented the formation of metacestode vesicles from parasite stem cells. Since EmIR1 is not expressed in this parasite stage, we suggest that EmIR2 is also capable of binding HNMPA(AM)3, maybe even with higher affinity than EmIR1. Indeed, in a recent report Vanderstraete et al.  demonstrated that HNMPA(AM)3 inhibits the schistosome receptor SmIR1 (which is the ortholog to EmIR2) with much higher efficacy (10 to 100 fold more) than SmIR2 (ortholog to EmIR1). When applied to the Echinococcus system, this could explain the relative resistance of the (EmIR1 expressing) metacestode to the drug when compared to the (EmIR2 expressing) primary cell system. However, care has to be taken in the interpretation of data on insulin inhibitor effects on flatworms since Vanderstraete et al.  also showed that these can affect a structurally diverse family of receptor kinases that are composed of an extracellular Venus FlyTrap (VFT) motif and an intracellular, insulin receptor-like TKD. In S. mansoni, two of these kinases, named SmVKR1 and SmVKR2, are expressed and a panel of available insulin receptor inhibitors that showed effects on SmIR1 and SmIR2 also affected SmVKR1 and SmVKR2 in a similar manner . In the E. multilocularis genome, only one gene encoding such a tyrosine kinase, EmVKR, is present and transcriptome data indicate that it is expressed in a similar manner as emir2 (data not shown). For the inhibitor data concerning EmIR1 phosphorylation upon addition of insulin, we do not see interpretation problems since this was carried out specifically for EmIR1, immunoprecipitated from membrane fractions. However, at least some of the effects we observed on entire Echinococcus larvae after application of HNMPA(AM)3 could indeed be due to inhibition of EmVKR rather than EmIR1 and EmIR2. Unfortunately, it is presently not possible to clearly distinguish between these possibilities since highly selective inhibitors for the parasite insulin receptors versus the VKR receptors are not available  and since RNAi methodology for E. multilocularis is still in its infancy. Nevertheless, our data indicate that the insulin signalling system of E. multilocularis, including insulin receptor kinases, EmVKR, and downstream signalling components, might be a fruitful target for the development of novel chemotherapeutics, as has previously been argued in the case of schistosomes [11–13, 31].
In summary, our data indicate an important role of host insulin on the development of E. multilocularis larvae within the host’s liver. We also showed that this involves hormonal host-parasite cross-communication via evolutionarily conserved signalling systems, which is particularly striking for EmIR1 concerning glucose uptake in GSCs of the metacestode and, most likely, also applies to EmIR2 in the primary cell system. Using a well-known inhibitor of insulin receptor signalling, we also demonstrated clear effects on parasite survival and, particularly, development. Although HNMPA(AM)3 might not be as efficient as other kinase inhibitors, such as pyridinyl imidazoles  or imatinib , in inducing killing of the metacestode, which is the main target of chemotherapy against alveolar echinococcosis, our study now opens the way for the development of more specific inhibitors that could be used to affect glucose uptake by the parasite during development. Furthermore, due to their obvious effects on parasite stem cell proliferation, insulin receptor inhibitors might be used to inhibit asexual multiplication of an already established parasite mass or to prevent metastasis formation from stem cells in advanced cases of the disease . Since the somewhat lower efficacy of HNMPA(AM)3 to inactivate metacestode vesicles (when compared to primary cells) could at least in part be due to problems in penetrating the laminated layer which surrounds the parasite cells, issues of improved tissue penetration should also be considered in studies on the development of anti-insulin signalling drugs against AE.