Long-term in vitro cultivation of Echinococcus multilocularis metacestodes under axenic conditions
- First Online:
- Cite this article as:
- Spiliotis, M., Tappe, D., Sesterhenn, L. et al. Parasitol Res (2004) 92: 430. doi:10.1007/s00436-003-1046-8
We report here on the development of an in vitro system for the long-term cultivation of Echinococcus multilocularis larvae under axenic conditions. In the absence of feeder cells from the host, long-term survival of the parasite depended strictly on low oxygen conditions and the presence of reducing agents in the medium. Host serum supported survival of the parasite but the growth of metacestode vesicles and differentiation towards the protoscolex stage only occurred in the presence of culture medium that was preconditioned by hepatoma cells or several other immortal cell lines. On the basis of this in vitro system, future analyses on the identification of host-derived growth factors for E. multilocularis will be greatly facilitated.
Echinococcus multilocularis is an important human parasitic cestode for which the molecular mechanisms of development and the interaction with the host are only poorly understood. In order to alleviate these deficits, several in vitro cultivation systems for the parasite’s metacestode larval stage have been developed (reviewed by Hemphill et al. 2002). Basically, two different culture systems have been introduced. In the method described by Jura et al. (1996), metacestode vesicles are co-cultivated with host hepatocytes as feeder cells in the presence of a collagen matrix. The system introduced by Hemphill and Gottstein (1995) relies on the incubation of small tissue blocks or vesicle suspensions from experimentally infected mice in medium that contains calf serum. Although both methods are suitable for obtaining proliferating and differentiating metacestode vesicles, they suffer from the drawback that host cells are continuously present during culture. In the co-culture system, this is obvious. In the tissue block/vesicle suspension method, contaminating cells which derive from the laboratory host are not eliminated during culture (Hemphill and Gottstein 1995) and, at least in our hands, secondary growth of host material (mostly fibroblasts) in parasite cultures frequently occurs. In both studies, metacestode vesicles, once removed from the original culture, rapidly disintegrated and died. This could not be prevented by the addition of conditioned culture medium that had previously been incubated with feeder cells (Hemphill and Gottstein 1995; Jura et al. 1996). Although it has, on the basis of these results, been suggested that Caco2 and/or hepatocyte feeder cells produce growth factors for metacestode vesicles (Hemphill and Gottstein 1995; Jura et al. 1996), these data could also be interpreted in such a way that host cells simply remove compounds from the culture medium which are toxic for the parasite. Hence, whether host cells do, indeed, produce growth factors for the parasite has not been clearly demonstrated in these studies.
Several recent reports on in vitro cultivation systems for Schistosoma mansoni larvae showed that this parasite is highly susceptible to reactive oxygen intermediates which are formed during cultivation and which are obviously detoxified in the presence of Biomphalaria glabrata embryonic (Bge) feeder cells (Ivanchenko et al. 1999; Bixler et al. 2001; Bender et al. 2002). In these systems, axenic growth of schistosome larvae was only observed when the parasite was cultured under low oxygen conditions or in the presence of reducing agents. This prompted us to investigate similar mechanisms for E. multilocularis. For this, we designed different culture media as follows. Medium A (MedA) was Dulbecco’s minimal essential medium (DMEM; GIBCO BRL) with 100 U/ml penicillin G and 100 μg/ml streptomycin (Biochrom). Medium B (MedB) was MedA including 10% fetal calf serum (GIBCO BRL), and medium C (MedC) was MedB that had been conditioned in the presence of different feeder cell lines for 7 days. Furthermore, for each culture medium, we designed a formula for reduced conditions (MedA*, MedB*, MedC*) by including 0.01% 2-mercaptoethanol, 100 μM l-cysteine, and 10 μM bathocuproine disulfonic acid (all from Sigma). All media were sterilized by filtration before being added to parasite material. E. multilocularis metacestode vesicles were generated by in vitro co-culture in the presence of rat Reuber hepatoma cells [American Type Culture Collection (ATCC) no. CRL-1600], as described by Jura et al. (1996), for up to 3 weeks using MedB throughout the culture. Small vesicles of ~3 mm in diameter were then harvested by filtration through a 150 μM nylon mesh and were extensively washed using phosphate buffered saline (PBS) to remove contaminating host cells. The resulting vesicles were subsequently incubated for 2 weeks in a 25 ml culture flask at 37°C in MedC*. Medium was changed at intervals of 2 days and the oxygen gas phase of the flasks was replaced by nitrogen after each medium change. While RT-PCR analyses on feeder cell actin- and tubulin-encoding genes frequently detected residual host contamination at the beginning of this 2-week incubation period, no products were consistently obtained after this step (data not shown). After this incubation, parasite vesicles were again collected by filtration through a 150 μM nylon mesh and extensively washed with PBS. The metacestodes were then microscopically checked and only vesicles which were structurally intact and without inclusions or brood capsules were chosen for further analysis. Groups of five vesicles were then incubated at 37°C in 15 ml plastic tubes with 5 ml of the media as indicated above. The medium was changed every second or third day using, as the gas phase, nitrogen in the case of MedA*, MedB*, and MedC*, and 5% CO2/95% air in the case of MedA, MedB, and MedC. At intervals of 7 days, the total volume of the vesicles was determined and the metacestodes were microscopically analysed for protoscolex development.
In a further set of experiments, we were interested in the ability of different mammalian cell lines to support the growth and differentiation of E. multilocularis larvae. For this, cultures were established as indicated above using, as for MedC*, conditioned media from different sources: rat Reuber hepatoma cells (ATCC no. CRL-1600), HepG2 human hepatoma cells (HB-8065), human embryonic kidney cells HEK-293 (CRL-1573), SH-SY5Y human neuroblastoma cells (CRL-2266), CHO chinese hamster ovary cells (CCL-61), MMN02 human gastric epithelial adenocarinoma cells (Wagner et al. 1994), HeLa human epidermoid carcinoma cells (CCL-2), HEC1B human endometrial carcinoma cells (HTB-113), and the macrophage like cell line U937 (CRL-1593.2). As shown in Fig. 2, with the exception of U937 cells, all cell lines supported growth of E. multilocularis metacestodes to a certain extent. The best results were obtained for the rat Reuber hepatoma cell line, whereas all other sources resulted in intermediate growth. This indicates that growth factors for the parasite are not exclusively produced by cells of hepatocyte origin. Secretion of these factors could either be a widespread feature of mammalian cells or, alternatively, be a characteristic of a variety of immortal cell lines.
In conclusion, long-term axenic cultivation of Echinococcus metacestode vesicles is possible provided that reducing and hypoxic culture conditions are supplied. Similar to the situation in S. mansoni, the parasite is most probably highly susceptible to reactive oxygen intermediates which are formed during axenic cultivation, a mechanism which surely requires further investigation. Furthermore, we clearly show that host-derived feeder cells do secrete growth factors or molecules into the medium which are absolutely required for parasite development.
Based on the axenic culture system developed in this study, it will be possible in further investigations to directly study the influence of selected host factors on Echinococcus development for prolonged periods of time without the presence of interfering host cells. We recently demonstrated the presence of several signal transduction systems of the insulin- (Konrad et al. 2003), the epidermal growth factor- (Brehm et al. 2003; Spiliotis et al. 2003), and the transforming growth factor-β- (Zavala-Gongora et al. 2003) families in the parasite which show considerable structural homologies with corresponding systems of the host. The respective mammalian growth factors will be interesting candidates for studies using the axenic cultivation system. Furthermore, the system now also allows for the isolation of host-derived growth factors for Echinococcus using biochemical methods such as fractionation of conditioned media. Respective studies are currently under way and should lead to valuable insights into the host-parasite relationship during alveolar echinococcosis.
This work was supported by grant BR2045/1-1 (to K.B.) from the Deutsche Forschungsgemeinschaft. We are indebted to Matthias Frosch for continuous support and to Sebastian Suerbaum for providing us with the HMN02 cell line. We wish to thank Dominik Quest and Katja Klöpper for excellent technical assistance.