Host Proteins in E. multilocularis Metacestodes

1. Introduction

The closely related zoonotic cestodes Echinococcus multilocularis (fox tapeworm) and E. granulosus (dog tapeworm) cause the neglected diseases alveolar echinococcosis (AE) and cystic echinococcosis (CE), respectively, in both humans and animals. In Europe, AE is the most significant foodborne parasite in humans and ranks among the top three globally [1], with an annual burden exceeding 688,000 Disability Adjusted Life Years (DALYs) [2]. CE, while ranking fourth in Europe, is the second most impactful globally, with 184,000 DALYs. Though relatively rare (~18,250 new human cases annually [3]), AE incidence is rising [4,5], CE affects over 188,000 people each year [2], and poses a major economic burden on global livestock industries, causing losses exceeding US$ 2 billion annually [6]. E. multilocularis follows a predator-prey transmission cycle involving carnivores as definitive hosts and rodents as intermediate hosts, whereas the E. granulosus species complex primarily involves larger herbivorous mammals as intermediate host species such as bovines or sheep [7]. Accidental hosts of E. multilocularis, the focus of this study, include humans, monkeys, and dogs [8]. The parasite’s metacestode stage develops in the liver, causing cancer-like growth and metastasis [9]. Without treatment, AE is fatal in >90% of cases within 10–15 years [9]. Curative treatment requires radical metacestode removal and drug therapy [9], but surgery is only viable in 20–50% of cases [10,11]. If surgery is not an option, patients receive lifelong benzimidazoles [11,12], which halt parasite growth but are not parasiticidal, thus massive doses have to be administered throughout life [9].

The fluid-filled E. multilocularis metacestodes are vesicular structures comprised of an acellular laminated layer, the tegument, and the inner germinal layer, collectively called vesicle tissue (VT). The metacestodes of E. granulosus follow a similar structure, yet they present rather as unilocular cysts. The outer laminated layer of Echinococcus metacestodes is a unique carbohydrate-rich extracellular matrix covering the surface [13]. The syncytial tegument is anchored to the interior surface of the laminated layer, and the germinal layer is composed of connective tissue, muscle cells, nerve cells, glycogen/lipid storage cells, subtegumentary cytons and undifferentiated stem cells [14]. The vesicle fluid (VF) inside the metacestodes plays a role in nutrition and exchange of metabolites between the parasite and its surroundings. Stem cells are the only cells of the metacestode with an unlimited proliferative potential, rendering the parasite immortal, and drugs need to destroy the stem cells to achieve parasiticidal activity [15].

Larval stages of Echinococcus spp. and of other cestodes evolve in an environment characterized by the presence of host micro- and macrometabolites including proteins. To which extent host proteins are taken up into these stages and what may be their role? Early investigations have revealed that isolated E. granulosus metacestodes and other larvae take up I¹²⁵-labelled immunoglobulins [16]. Furthermore, native human serum albumin and IgG are found in E. granulosus hydatid cyst fluid from a patient [17]. Finally, proteomic analyses of cysts and adult stages from intermediate and definitive hosts have revealed numerous host proteins within cyst fluid and other compartments of the parasite [18,19,20]. The read-out of studies with E. granulosus is, however, hampered by the fact that they are based on ex vivo, thus patient-borne material, since standardized in vitro culture systems for this parasites have been only recently developed [21].

The situation with E. multilocularis metacestodes is different. Using a culture medium preconditioned with rat hepatoma cells, they can be cultivated in vitro during several months and therefore provide a suitable standardized system to study host-parasite interactions [22]. A proteomic study using this system has led to the identification of 54 culture medium thus host-based proteins in E. multilocularis VF [23]. Employing the same culture system, we have succeeded in characterizing the metabolomes of metacestode VF and of metacestode culture medium [24]. Moreover, we have characterized the E. multilocularis proteomes in metacestode VT, VF, and in culture medium with a special emphasis on secreted antigen B isoforms [25]. In the present study, we shed light on proteins of host origin within E. multilocularis VT and VF asking the question whether host proteins in VT and VF correspond qualitatively and quantitatively to the respective levels in the medium. Our null hypothesis is that there is no selectivity resulting in equal distribution of host proteins between medium, VT and VF. Moreover, we compare the results concerning the VF proteome obtained in vitro to the proteome of VF collected ex vivo.

2. Results

2.1. The Overall Protein Concentration in Metacestode Vesicles Remains Constant

When analyzing the protein concentration in VF of metacestode vesicles of different sizes and from different batches, it turned out that there was no decrease in protein concentration, i.e. a dilution of proteins, nor an increase in protein concentration, i.e. an accumulation of proteins, as a function of vesicle diameter the concentration fluctuating around 2 g/l in all samples. An exception is, however, observed when regarding vesicles obtained ex vivo, where up to 6 g/l were measured in VF. In these ex vivo vesicles, a tendency to an increased protein concentration in larger vesicles could even be observed (Figure 1).

To investigate whether the protein concentration of the VF is correlated with the protein concentration of the medium, metacestode vesicles of three different size classes were incubated for three days in culture medium (CM) containing various amounts of fetal bovine serum. The protein concentrations of the corresponding VFs ranged in all samples between 2.1 and 5.3 g/l, irrespectively of the metacestode vesicle size or the protein concentration of the culture medium (Table 1).

2.2. Uptake of Host Proteins by Metacestode Vesicles

When intact metacestode vesicles were incubated in vitro in the presence of various Cy3 or FITC labelled proteins including IgG, the following were identified in VF as well as in VT: labelled thyreoglobulin (665 kDa), catalase (226 kDa), bovine serum albumin (66 kDa), chymotrypsin (25 kDa), as well as free Cy3 (0.8 kDa) (Figure 2 A). Moreover, incubation of metacestode vesicles in the presence of FITC-labelled IgG of both mouse and goat origins resulted in detection of the label in VF in all cases and in VT in the case of a labelled goat-anti-rabbit IgG (Figure 2 B).

2.3. Spectrum of Host Proteins Differs Between Compartments

To understand the composition of host proteins in metacestode vesicles and medium, shotgun LC-MS/MS analyses of these three compartments were performed. The overall results are summarized in Table 2.

Since proteins of both bovine and rat origin are in the medium and since both organisms represent potential hosts of Echinococcus spp., these proteins were referred to as “host proteins” in combination in this study.

As expected, the most abundant protein in the conditioned medium was bovine serum albumin amounting to more than 20% of the total abundance of host proteins, followed by rat serotransferrin and hemiferrin at one magnitude lower abundance (Table 3). The full dataset of medium proteins is presented as supplemental Table S1. These proteins were still present in VT and VF, but at much lower levels. The most abundant proteins showed a different pattern in these compartments as compared to culture medium. The most abundant host protein in VT was histone H4 (ca. 34 % of total proteins) followed by actin and apolipoprotein A-1 (both 13-14%). In VF, the most abundant host protein was alpha2-HS-glycoprotein with ca. 40% of total host proteins followed by alpha-antiproteinase or serpin a1 and serum albumin (both around 17%), as shown in Table 3. The full dataset is presented as supplemental Table S2.

The comparison of the five most abundant proteins suggested an absence of direct correlation between host protein contents in the three compartments CM, VT and VF. This view was fostered by subjecting the rankings of protein abundances in the respective compartments to correlation analysis. According to our zero hypothesis, there should be a concentration balance between the compartments. Consequently, the relative abundances should be correlated. To address this, we compared the abundance rankings of proteins found both in CM and VF. As shown in Figure 3, there is only a weak correlation (r=0.33) between the rankings in both compartments.

When analyzing the abundance rankings of host proteins in VF and VT, even a complete absence of correlation (r = 0.03) was noted. However, two distinct subsets of host proteins could be identified in these compartments, one overrepresented in VF and at or below the detection limit in VT (Figure 4, group A), one overrepresented in VT as compared to VF (Figure 4, group B).

In VF, the three most abundant group A proteins are alpha-fetoprotein, vitamin D binding protein and transthyretin with mean IBaq values around 5x10¹¹ (Figure 5A). These proteins were below the detection limit in VT. Furthermore, amongst the seven most abundant group A proteins were two serin protease inhibitors, namely serpinA3-2 and inter-alpha-trypsin inhibitor heavy chain H4. In total, that six other serin protease inhibitors were in this group of proteins (Table S2).

Conversely, group B representing proteins overrepresented in VT and almost absent in VF was dominated by apolipoprotein A-II (IBaq 2x10¹¹) followed by four other proteins with nearly one quarter of this abundance. These four proteins were hemoglobin subunit alpha, a ribosomal protein and two histones (Figure 5B).

3. Discussion

In this study, we have shown that host proteins pass from the surrounding medium into E. multilocularis VF irrespectively of their size or charge, thereby confirming previous results. Alone or combined with endogenous proteins, these host proteins may have various functions. First, the proteins and other polymers within the vesicle adsorb water thereby increasing the colloid-osmotic or oncotic pressure, which – together with the osmotic pressure (proportional to the molarity of solutes only) – causes an influx of water maintaining the vesicle turgid and with an open lumen within the surrounding tissue (see e.g. [26] for general considerations concerning water potentials). For instance, serum albumin, the most abundant protein in host serum [27], as well as an abundant host protein in vesicles, has a water binding capacity of ca. 170 mg tightly sorbed water per gram of albumin [28]. This space is a prerequisite for protoscolex formation within the metacestode thus contributing to Darwinian fitness. The metacestode thus compares to a vertebrate amnion filled with amniotic fluid. The protein concentration of the metacestode fluid remains constant during growth even in the presence of high protein concentrations in the CM. This is different from the situation e.g. in human amniotic fluid, where the protein concentration first increases and then markedly decreases towards the term of gestation at the benefit of the embryo [29,30]. The increase of volume of amniotic fluid is thus rather due to an influx of electrolytes increasing the osmotic pressure [31], than to accumulation of proteins increasing the oncotic pressure.

A second function of host proteins within the metacestode may reside in providing nutrients thereby fueling growth and differentiation into protoscoleces both aspects contributing to Darwinian fitness, once again. The host proteins may serve as food by directly providing (essential) amino acids to the growing and differentiating metacestode after cleavage by endo- and exopeptidases. In fact, metacestodes contain cathepsin-like cysteine peptidases [32] acting on a variety of host proteins including serum albumin and immunoglobulins [33], and not only on preselected peptides of specific interest such as e.g. eotaxin [34]. Moreover, metalloproteinases have been identified in E. granulosus [35], and leucine aminopeptidases in E. granulosus [36] and E. multilocularis [37] metacestodes. Conversely, the presence of serpins (serine protease inhibitors) in metacestodes [38,39] suggests a preference for the above-mentioned peptidases over serine proteases for host protein degradation. This view is fostered by our own observation that serpins and other serin protease inhibitors of host origin are enriched in the VF over VT.

Furthermore, host proteins taken up into metacestodes may transport micromolecular nutrients. The most predominant host protein in VF, apha-2-HS-glycoprotein, interacting with fetuin, another abundant host protein, forms calciprotein particles preventing the unwanted precipitation of calcium phosphate. These particles may contribute to the import of calcium into and storage within the vesicle [40,41]. Besides its multiple other properties [42], serum albumin is a notorious fatty acid binding protein [43] and therefore, in all likelihood, contributes to fueling the metacestode with nutrients. Other, more specific nutrient binding proteins found in metacestodes are the apolipoproteins AI and AII, both enriched in VT over VF. Both proteins are components of high-density lipoproteins [44] and thus involved in the transport of cholesterol from the periphery to the liver [45,46]. The uptake of cholesterol-loaded lipoproteins into the metacestode could contribute to fueling proliferating cells in the germinal layer. Serotransferrin, one of the more abundant proteins found in VF as well as in CM, is the predominant iron transport protein in human plasma [47]. Since a constant supply with iron is essential for proliferating germinal layer cells as well as for host cells, the serotransferrin within the VF can be considered as a reservoir maintaining this essential iron homeostasis. Moreover, host derived hormone and vitamine transporters are accumulated in VF, in particular transthyretin, the major thyroid hormone binding (and therefore iodine carrier) protein [48], and a vitamin D binding protein, the primary transporter of the steroid vitamin D [49].

A third major host protein group of interest identified in metacestode vesicles – although not abundant in the CM – are histones. Principal constituents of chromatin under physiological conditions, histones may be released to the medium after cellular damage and act as mediators of innate immunity by stimulating cytotoxicity, cells of the immune system and coagulation [50]. Consequently, interference with extracellular histones contributes to dimming inflammatory reactions [51]. Another inflammation-related protein scavenged by metacestodes is the protein S100-A9. This protein is considered as an alarmin [52] promoting of inflammation via nuclear factor-kappaB activation [53,54].

Therefore, it is not surprising that E. multilocularis, a master in avoidance of innate and acquired immune responses (see e.g. [55,56] for review), avidly scavenges inflammation stimulating proteins from the medium. Scavenging histones means hitting two birds with one stone, namely dimming host inflammatory reactions and fueling cell proliferation and ultimately differentiation into protoscoleces with the essential amino acid lysine, a major constituent of histones and by stimulating the uptake of serum albumin [57].

This leads to our initial question whether the uptake of host proteins into the metacestode occurs unspecifically. Labelled proteins with diverse physico-chemical properties are taken up into metacestode vesicles under artificial conditions, and in vitro and in vivo, the most abundant proteins in the medium, serum albumin in particular, are also amongst the most abundant proteins in metacestode VF. Nevertheless, comparing the respective amounts of these proteins in CM, VT and VF, we do not see, a strict correlation thereby falsifying our zero hypothesis that this uptake occurs unspecifically. How this selectivity may be explained? As shown by a biotin labelling study, uptake of supercharged proteins from the extracellular medium into endosomes is favored over e. g. cell-penetrating peptides [58]. The fact that the endophilin homolog protein p29 is amongst the most abundant E. multilocularis proteins in the VT [25] suggests a major role of endocytosis in protein uptake from the medium into germinal layer cells. Consequently, the host proteins have to cross the laminated layer, where they may be passively retained (and stored) as a function of their charges (e.g. positively charged histones by carboxyl groups of the extracellular matrix). Unfortunately, besides its role as a potential vaccine documented in numerous articles (see e.g. [59,60,61]), p29 has not been characterized so far with respect to its biological function.

Moreover, host proteins may cross the laminated and germinal layers by alpha-helical channels favoring the passage of proteins over small molecules [62]. At specific sites within the metacestode tissue, such hypothetical channels may translocate proteins directly from the medium into the metacestode lumen. Moreover, E. multilocularis and other helminths may possess genuine protein uptake mechanisms unknown, so far, in other organisms. Figure 6 summarizes the roles and potential uptake mechanisms of host proteins in Echinococcus metacestodes.

So far, the uptake mechanisms mentioned above are hypothetical since appropriate investigation tools are lacking in Echinococcus, in the opposite to the situation in model mammalian cell lines or yeast. A future strategy could be to express candidate Echinococcus genes in these model systems and to look for gain-of-function phenotypes (e.g. increased uptake of one or several host proteins investigated in this study). One should, however, be aware that the transformation procedure by itself may cause artifacts, and include appropriate controls in this type of experiment (see e.g. [63] for review).

Finally, we raise the question whether our results are of importance for the improvement of AE and CE therapies. Targeted drug delivery, i.e using selected host proteins loaded with a suitable drug as a kind of guided missile, may sound – prima vista – seducing from an academic point of view, but is closer to Science Fiction than to science as a part of a realistic treatment. Overall, the current treatment of echinococcosis comprising surgery and/or chemotherapy based on benzimidazoles may be improved by an appropriate diet, as e.g. in the case of anti-cancer therapy [64]. Our results presented here as well as previously published findings [24,65] suggest that proteins together with free amino acids and lipids play an important role in fueling proliferation and differentiation of metacestodes. If this is the case, a diet with reduced input of these nutrients in combination with current treatments might enhance the therapeutic success.

4. Materials and Methods

4.1. Chemicals

If not stated otherwise, all chemical used were purchased form Sigma (St. Louis, MO, USA). Cell culture media and fetal bovine serum (FBS) were from Bioswisstec (Schaffhausen, Switzerland).

4.2. In Vitro Culture of E. multilocularis Metacestodes

Metacestode in vitro growth medium was prepared by preconditioning Dulbecco’s Modified Essential Medium (DMEM) with Reuber rat hepatoma cells (RH) as described [24]. The studies presented here were performed using E. multilocularis isolate H95 metacestode vesicles grown in vitro for 12 weeks. Labelling and proteomic analyses were performed using vesicles with a diameter between 2 and 4 mm. Protein concentrations of VF harvested by syringe were determined using metacestode vesicles with diameters between 0.5 and 10 mm by bichinolinic acid (Thermo Fisher Scientific, Waltham, MA, USA). Metacestode vesicle integrity was visually confirmed during the experiment and before collection of samples.

4.3. Parasite Maintenance and Ex Vivo Sample Preparation

Animals for parasite maintenance were purchased from Charles River Laboratories (Sulzheim, Germany) and used for parasite maintenance after 2 weeks of acclimatization. BALB/c mice were maintained in a 12 h light/dark cycle, controlled temperature of 21 °C – 23 °C, and a relative humidity of 45 % – 55 %. Food and water were provided ad libitum. All animals were treated in compliance with the Swiss Federal Protection of Animals Act (TSchV, SR455), and were approved by the Animal Welfare Committee of the canton of Bern under the license numbers BE2/22 and BE30/19. E. multilocularis metacestodes (isolate H95) were grown in intraperitoneally infected mice for approximately 4 months. The mice were euthanized with CO₂ after 3-4 months of parasite growth. VF of five independently infected animals was aseptically removed by syringe and stored at – 80°C until further analysis.

4.4. Fluorescence Labelling of Marker Protein

To perform uptake studies with defined, labelled xenoproteins, purified chymotrypsinogen A (25 kDa), bovine serum albumin (67 kDa), catalase (226 kDa), and thyreoglobulin (665 kDa) from a marker kit for size exclusion chromatography (Thermo Fisher Scientific, Waltham, MA, USA) were labelled with Cy3 (0.8 kDa) using a commercial kit and the amount of incorporated label was quantified according to the manufacturer’s instructions (Promega, Madison, WI, USA). Moreover, FITC labelled mouse monoclonal antibodies g11 and IFA, as well as FITC labelled goat-anti-mouse and goat-anti-rabbit IgG ((Thermo Fisher Scientific, Waltham, MA, USA) were used for uptake studies.

4.5. Proteomics

For subsequent analysis by LC-MS/MS, the metacestode CM of the above-described setup was collected, centrifuged for 10 min at 500 x g and 4 °C, and immediately stored at -80 °C. The metacestode vesicles were washed three times in 4 °C cold NaCl (0.9 %), broken with a 1 mL pipette tip, and centrifuged at 9’000 × g for 20 min at 4 °C. The supernatant was removed and was centrifuged again at 12’000 × g for 20 min, 4 °C yielding VF. The remaining metacestode tissue was washed with NaCl (0.9 %, 4 °C) and centrifuged as described above yielding VT. Both preparations were stored at -80 °C for subsequent protein analysis by LC-MS/MS performed as described [25]. The original MS/MS data are available on Pride ProteomeXchange with access number PXD040274

(https://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD040274). The data were analyzed using the uniport database (www.uniprot.ch; last access January 2025).