Calmodulin-specific small interfering RNA induces consistent expression suppression and morphological changes in Echinococcus granulosus

Among parasitic helminths, biological features of Echinococcus granulosus have been a focus of particular interest in biology and medicine. The determinants and underlying molecular mechanisms of Echinococcus development in different host settings is largely unknown. The phenomenal bi-directional development of E. granulosus protoscoleces into multi-proglottid and/or microcysts, is a fascinating feature of the parasite cultivation. Calmodulin (CaM) is the major intracellular Ca2+ binding protein in plant and animal organisms. Many Ca2+-related processes in the physiology of eukaryotic organisms are CaM-dependent, however little is known on the role of CaM in platyhelminths growth and development. Small interfering (si) RNA-induced manipulations of the genes involving in the parasite development is an opportunity to explore novel approaches for cystic echinococcosis (CE) prevention and management. Regarding the fundamental role of CaM in cellular function of the parasites, in this study, we investigated the molecular and morphological changes induced by siRNA on CaM in different in vitro stages of E. granulosus. Three developmental stages of the tapeworm, protoscoleces, microcysts and strobilated worms, were cultivated in vitro in mono- and di-phasic media and three delivery methods, i.e. electroporation, soaking and electro-soaking, were used for RNA interference. The level of mRNA suppression as well as the phenotypic changes of the parasites were measured. Following RNA interference, EgCaM mRNA suppressions of 65–99% were recorded in different stages of the tapeworm as compared to untreated/unrelated siRNA controls. Lower viability, growth retardation, morphological abnormalities as well as EgCaM expression suppression were documented in the parasite implying potential of siRNA technology for the prevention and management of CE.

extracellular matrix in the middle called laminated layer that is unique to the genus Echinococcus, and an external host-derived adventitial layer formed in response to the parasite 2 . As a model organism Echinococcus granulosus has been demonstrated as a valuable biological entity influencing advancements in developmental biology of flatworms 3 .
Following Echinococcus cultivation in appropriate diphasic and/or monophasic media the phenomenon of bi-directional development of E. granulosus protoscoleces into multi-proglottid and/or microcysts, is a fascinating feature of Echinococcus biology. This provides us an opportunity to study genes involving in the parasite development and to manipulate the parasite for disease management and prevention [4][5][6] . Understanding the determinants and molecular mechanisms in the parasite development in different host settings offers suitable tools for prevention and control of CE 2 .
Calmodulin (CaM), is a multifunctional intermediate calcium sensor protein expressed in all eukaryotic organisms. The functions of CaM include Ca2+ binding and alteration of calcium signal transduction pathway to regulate a multitude of biological processes, such as cytoskeletal assembly/reorganization, activation of phosphorylase kinase, abiotic stress responses, neurotransmission, smooth muscle contraction, metabolism and cell motility 7,8 . Wang et al. have used Immunohistochemical localizations on different stages of the parasite, suggesting the expression of E. granulosus CaM (EgCaM) in the tegument tissues, protoscoleces parenchymal region, germinal layer as well as in the adult stage. The study showed physiologically active regions for EgCaM expression and suggested the crucial role of Ca2+ signaling pathway in the growth and development of the helminth and its function between E. granulosus and its host 9 .
In recent decade, there has been increasing interest in RNA interference (RNAi) technology in the field of biology and development. Our understanding of the genetic basis of development in invertebrates is limited. Small interfering RNAs (siRNAs) provides us a valuable tool to improve our knowledge on the developmental biology of Platyhelminthes 10 . Molecular methods based on RNAi or Post-Transcriptional Gene Silencing (PTGS) are the new millennium tools that provide information on gene function and characterization of the genes involving in regeneration and development. RNA interference mediated by siRNA with perfect homology to their target can cause silencing of specific genes for experimental and therapeutic purposes 11,12 . RNAi has been successfully applied to numerous helminth organisms to date. Silencing of the gene encoding Fasciola hepatica leucine aminopeptidase (LAP) with siRNAs as well as double-stranded RNAs (dsRNA), suppressed mRNA and protein expression for up to 3 weeks in the newly excysted juveniles 13 . RNAi has been used to suppress a number of E. multilocularis endogenous genes expression in protoscoleces 14 . Protoscoleces electroporation of genes encoding the 14-3-3 and elp with siRNA resulted in reduced expression of up to 21.8 and 35.5% of the genes respectively, compared to untreated control. In addition, the target proteins have been significantly reduced on day 15 14 .
CaM-specific RNAi technology has been utilized to determine the functional importance of calmodulin in several helminth parasites, however there are very few data on the role of CaM function in E. granulosus. There are reports for schistosomes 15,16 , the liver flukes Fasciola gigantica and F. hepatica 17,18 and Caenorhabditis elegans 19 . Suppression of calmodulin mRNA by feeding Schistosoma mansoni with dsRNA resulted in a phenotype characterized by waves of contraction in adult worms but not in schistosomula 16 . RNAi has also been used to determine the biological function of F. hepatica calmodulin in vitro. Stunted growth and lower viability were documented after treating the in vitro cultured flukes with dsRNAs and siRNA 20 . The purpose of the present study was to determine the effect of RNA interference on EgCaM expression in different developmental stages of E. granulosus in vitro.

Results
In vitro cultivation of protoscoleces. The parasite was characterized using PCR-sequencing and identified as the G1 genotype of E. granulosus (Accession Number MG832791). E. granulosus protoscoleces were successfully cultivated in both monophasic and diphasic media. The protoscoleces started to develop proglottization in diphasic culture and the first proglottids were observed on day 28. However, the full grown three-proglottid strobilated forms were developed after 55 days of cultivation. Egg production in the terminal proglottid was not observed under the culture conditions (Fig. 1). In monophasic condition the protoscoleces were developed into conspicuous microcysts after 45 days. The culture was characterized by wide proliferation of parasite vesicles with no development of new protoscoleces. www.nature.com/scientificreports www.nature.com/scientificreports/ The effects of suppressed calmodulin on the protoscoleces. Previous reports had demonstrated the activity of experimental RNAi in protoscoleces, either by electroporation or soaking 14,21 . Initially, we focused on uptake of fluorescently labeled siRNA in protoscoleces by fluorescent microscope. Fluorescently labeled siRNA was detected in treated parasites comparing to no or very low level of autofluorescence in untreated parasites (Fig. 2a).
Following RNA interference, 69 to 99% EgCaM mRNA suppression was recorded as compared to untreated controls (Fig. 3a). We found that electro-soaking method exerted the strongest transcript level suppression among three delivery methods (Table 1). Hence the gene expression data for negative control siRNA (siR-Ctrl) were only demonstrated for electro-soaking method. There was no significant difference in gene silencing between the negative control siRNA and no-treatment control groups.
The effect of EgCaM suppression on viability and size changes in protoscoleces was evaluated after siRNA treatment. Our results indicated the remarkable effect of EgCaM suppression on the viability of protoscoleces on day 8 post treatment. As shown in Fig. 3d, the viability of the parasites was decreased to 3-49% depending on different delivery methods. The protoscoleces were most affected by RNA interference using electro-soaking so that no protoscoleces were viable after three days post treatment.
Morphometric data are indicative of growth inhibition in siRNA-treated protoscoleces. A significant growth retardation was observed after day 8 in the protoscoleces exposed to siRNA compared to the control parasites (Fig. 3e). The protoscoleces size in the control groups were increased towards microcyst development (mean size of protoscoleces: control = 389 ± 14.5 μM; negative siRNA = 375 ± 17.4 μM) while the growth and development of siRNA-treated parasites were inhibited (mean size = 224 ± 10.07 μM).
The effects of suppressed calmodulin on the microcyst. Figure 2b shows the fluorescence produced by siRNA incorporation into the microcyst wall. Relative expression of EgCaM eight days after siRNA delivery was reduced by 99% (Fig. 3b, Table 1).
Some morphological changes were noticed in siRNA-treated microcysts compared to the control groups. As shown in Fig. 4a the treated microcysts were noticeably darker with irregular outer layer compared to the controls with transparent microcysts and smooth surface. The electro-soaked microcysts were the most affected group when compared with microcysts treated by soaking or electroporation.
The effects of suppressed calmodulin on the strobilated forms. There was successful uptake of fluorescently labeled target siRNA by strobilated forms cultured in vitro. This was visible throughout the helminth body. In comparison to the metacestode stages we observed high levels of autofluorescence in the strobilated www.nature.com/scientificreports www.nature.com/scientificreports/ control groups (no-treatment and unrelated negative siRNA controls), however the fluorescence pattern was distinct and readily distinguishable in the parasites tegument treated with siRNA. (Fig. 2c).
siRNA treatment of the strobilated forms induced a range of 65 to 89% suppression of mRNA for EgCaM as assessed after 8 days post treatment (Fig. 3c, Table 1). As shown in Fig. 4b EgCaM silencing led to severe morphological abnormalities and reduced activity compared to the both controls. Seven days after EgCaM silencing the helminths were severely damaged after electroporation and electro-soaking.
As shown in Fig. 3f significant decrease in body contractions in the strobilated worms was noted, indicating the contribution of EgCaM silencing by siRNA. In the electro-soaking group while on day 2, on average 20 Data was compared to the controls: control protoscoleces (Ctrl) and negative siRNA control (siR-Ctrl). The gene expression data for siR-Ctrl were only demonstrated for electro-soaking method. Bars show the mean ± standard deviation (SD) derived from duplicates experiments. (**P < 0.01, ***P < 0.001, ****P < 0.0001). www.nature.com/scientificreports www.nature.com/scientificreports/ contractions per minute were recorded, on day 5 no body contractions were found, compared to the untreated control and negative control siRNA with an average of 47 and 46 contractions per minute respectively.

Discussion
RNA-dependent gene silencing has been increasingly utilized as a reverse-genetics approach in a variety of helminth organisms to generate functional gene knockdowns 22 . RNAi provides us with a powerful tool for predicting potential roles of genes involved in host-parasite interactions and vital processes essential for survival in several helminths such as Brugia malayi [23][24][25] Ascaris suum 26,27 C. elegans 19,28 , Schistosoma species 29-31 and F. hepatica 13,32 . Although RNAi has been widely used for silencing genes in nematodes and trematodes, there are few studies applying RNAi on cestode parasites e.g. adult Moniezia expansa 33 , and protoscoleces of E. multilocularis 14 and E. granulosus 21 . In the present study RNA interference technology was applied to suppress CaM in E. granulosus. Although CaM has been extensively studied and well characterized in several parasitic and free-living helminths 7,16,19 , there are very few data on the function and biological properties of CaM in E. granulosus. In the current investigation, we used siRNA sequences for EgCaM suppression to observe effects on different developmental stages of E. granulosus in vitro. All developmental stages of E. granulosus delivered with EgCaM-specific siRNA showed a robust suppression in EgCaM gene expression leading to obvious phenotypic changes (Figs 3  and 4). These results suggested a probable role for EgCaM in the growth and development ofE. granulosus.
Suppression in CaM gene expression by siRNA and dsRNA has been investigated in a number of platyhelminthes including S. mansoni 16 , F. hepatica 7,20 as well as the free-living nematode, C. elegans 19 . The results indicate CaM, as a transporter of Ca 2+ , plays an important role in the helminths growth, motility and metabolism.
Using different siRNA delivery methods produced a consistent suppression at mRNA level. The strongest effect (at least 92% suppression) was observed following electro-soaking in all developmental stages of E. granulosus (Table 1). It has been shown that electroporation produces pores in some biological membranes, favoring entrance of siRNA in the transfection buffer 13 .
We observed a robust suppression in EgCaM gene expression, resulted in phenotypic changes and viability and growth inhibition in siRNA-treated protoscoleces (Fig. 3). Mizukami et al. successfully suppressed 14-3-3 and elp genes expression by 21.8% and 35.5% respectively. Also, in the analysis of viability, 14-3-3, and elp siRNA-treated samples displayed 58.0 ± 23.0, and 55.1 ± 14.6% viability on day 15, respectively, compared to the untreated control 14 . This is in line with our data on the viability of protoscoleces after siRNA-mediated CaM suppression (Fig. 3d).
Our findings indicate that EgCaM specific siRNA in the protoscoleces induces growth retardation and inhibits protoscoleces transformation to microcysts compared to the controls (Fig. 3e). This imply that the reduction in CaM expression may prevent growth and development processes as shown in previous studies that linked CaM with the growth and development of other helminth species including miracidial transformation to sporocysts in S. mansoni 15 and reduced growth in the free-living nematode C. elegans 19 and the liver fluke Fasciola hepatica 20 . The growth retardation induced by RNAi has been demonstrated also in several other genes in S. manson [34][35][36] . Tegument malformations have been documented after using Eg-TSP1-specific siRNA on the soaked protoscoleces of E. granulosus 21 . www.nature.com/scientificreports www.nature.com/scientificreports/ In the current study RNAi-based interventions were associated with strong suppression of EgCaM in siRNA-treated microcysts, causing shrinkage and darkness in the outer wall of the microcysts (Figs 3b and 4a). Wang et al. showed EgCaM expression in the germinal layer and speculated that an unknown calcium-dependent mechanism occurs in this layer 9 . More in-depth investigations are needed on the probable role of CaM in the stability of the microcyst wall and germinal layer.
After 8 days of siRNA treatment on the strobilated forms, 65-89% EgCaM expression suppression was observed (Table 1). In a recent study conducted on S. mansoni by Guidi et al., Sm-CaM suppressions ranging from 70 to 95% were documented using dsRNA. Waves of contraction/dilation in the adult worms were also reported in the treated worms 16 . Another study showed siRNA-mediated knockdown of a 24 kDa calcium-regulated heat-stable protein (CRHSP-24) in juvenile S. japonicum, inducing death or changes in the parasite morphology 37 .
In the strobilated worms we noted significant phenotype changes including severe morphological abnormalities and reduced body contractions from day 2 onwards (Fig. 4b). It seems the suppression of EgCaM in the strobilated forms of E. granulosus leads to decreased motility and strobilar contractions (Fig. 3f). While parasites in the untreated control and negative siRNA groups were quite active with normal contractions, no body contractions were found in their siRNA-treated counterparts. The association of motility and CaM expression is well-documented in the literature. In 1990s, Ashizawa et al. were first to demonstrate disturbed motility of spermatozoa after experimental disruption of CaM function 38 . In general, Ca 2+ -dependent protein kinases like Calmodulin, MAPK and Polo-Like Kinase (PLK) are proved to play a key role in the motility-related functions in invertebrates 16,39,40 . However, using both RNAi and the CaM inhibitor, Trifluoperazine, McCammick et al. showed a significant increase in motility, migration and movement of the juvenile F. hepatica 20 .
Generalizations of in vitro data to in vivo experiments is difficult. Therefore, cautions have to be made in the interpretation of culture data at the host level 41 . As in vitro-reared E. granulosus do not develop as much as they do in the hosts, we could not precisely predict the behavior of the helminth in its natural habitat in the host 42 . As the dsRNAs are generally induced more persistent effects on target genes, this study would be supplemented by using an EgCaM-specific dsRNA. However, using dsRNAs has its own limitations and challenges 43,44 . Another prospect for RNAi-induced gene silencing in E. granulosus is to investigate the interference impacts at protein level using western blotting and other protein-based tools as well as further in-depth in vivo studies followed by deep sequencing analyses.
The present study used siRNA for suppressing EgCaM in different developmental stages of E. granulosus. The results indicate the significant effect of siRNA treatments on the phenotype as well as the gene expressions in all development stages of E. granulosus cultured in mono-and di-phasic media using different delivery methods. The study demonstrated that EgCaM is essential for viability, growth and development of the protoscoleces. Different phenotypic changes and transcript knockdowns observed in the present study encourage further investigations towards the development of novel therapeutic agents against cystic echinococcosis.

Methods
In vitro cultivation of protoscoleces. Parasite specimens were obtained by dissection of livers from naturally infected sheep at the municipal abattoir of Kerman, southeastern Iran with consent from animal handlers and the abattoir veterinary officer. Animals were slaughtering as part of the normal daily practice in the abattoir. The study was approved by the University Ethics Review Committee (code 95000288). All procedures were performed according to the University guidelines and regulations. The infected organs were immediately transferred to the Helminthology Lab, Dept of Medical Parasitology, Kerman University of Medical Sciences. After careful examinations of fertility and viability, one single viable cyst was finally selected for further experiments. Genotyping has been carried out using mitochondrial cox1 PCR-sequencing and the sequence was submitted to GenBank. Protoscoleces used for in vitro cultivation, were obtained from a single sheep liver cyst. Therefore, the microcysts and strobilated worms were derived from the cultivation of protoscoleces obtained from the same cyst.
Under sterile conditions the hydatid fluid containing protoscoleces was aspirated with a 50 ml syringe and then the laminated/germinal layer were removed. The aspirate as well as the cyst layers were carefully washed five times with PBS containing 100 U/ml penicillin and 100 μg/ml streptomycin (PBS-PS). The number of protoscoleces per ml was adjusted to 1 × 10 4 protoscoleces in 0.9% NaCl solution with a viability rate of at least 95%. Before cultivation the viability of the protoscoleces was checked by 0.1% aqueous eosin under a light microscope 44,45 . Two layers of sterile gauze were used to release protoscoleces from the brood capsules.
The protoscoleces were used for in vitro culture in diphasic and monophasic media to reach the strobilated as well as the microcyst forms, respectively, according to the method described by Smyth et al. 5,45 . The diphasic medium was S.10E.H, consisted of two phases: (i) the liquid phase, containing 260 mL of CMRL 1066 medium (Gibco, Grand Island, NY,), 100 mL of heat-inactivated fetal calf serum (FCS, Gibco-BRL, Gaithersburg, MD), 36 mL of 5% yeast extract (Sigma-Aldrich, St. Louis, MO) in CMRL 1066, 5.6 mL of 30% glucose (Sigma-Aldrich) in distilled water, 1.4 mL of 5% dog bile in PBS, 20 mM HEPES (Sigma-Aldrich), 10 mM NaHCO3 supplemented with penicillin (100 IU/mL), streptomycin (100 mg/mL), and (ii) the solid phase that is bovine serum coagulated at 76 °C for 20-30 min 42,46 . The medium was changed every 7 days and viability and morphological development of the parasite was observed under an invert microscope (TCM 400, Labomed Inc., CA).
For monophasic culture the protoscoleces were cultivated in Dulbecco's minimal essential medium (DMEM) (Gibco, Grand Island, NY) containing 10% heat-inactivated FBS (Gibco), 2 mM glutamine (Sigma-Aldrich), penicillin (100 IU/ml), and streptomycin (100 mg/ml) at the bottom of 25 cm 2 flasks incubated at 37 °C with 5% CO2. The medium was changed every 8 days. The flasks containing the protoscoleces were monitored weekly under an optical microscope to check the growth status of the microcysts 5,42,47,48 . Designing and Synthesis of small interfering RNA. We used several online softwares for siRNA design i.e. siDirect (sidirect2.rnai.jp/), siRNA Design (Integrated DNA Technologies, IDT), BLOCK-iT RNAi Designer (www. invitrogen.com/rnai), and RNA wizard (https://www.invivogen.com/sirnawizard). After prediction, all www.nature.com/scientificreports www.nature.com/scientificreports/ siRNAs were manually re-checked relative to their position, target site, length of siRNA, nucleotide content and specificity of siRNA (off targets), and finally the best was chosen. The siRNA was synthesized commercially by TAG Copenhagen A/S (Copenhagen, Denmark). The sequences of the EgCaM siRNA were as follows: sense 5′UCGUUAAAGUCAAUAACACCC3′-fluorescein and antisense 5′GUGUUAUUGACUUUAACGAAU3′fluorescein. Another negative (non-silencing) control siRNA was purchased as an unrelated siRNA (Qiagen, Germany). siRNA delivery and phenotype studies. Five sets of experiments were designed for each of three developmental stages of E. granulosus: (1) no-treatment group without any RNAi intervention, (2) unrelated/irrelevant Negative Control siRNA (siR-Ctrl) group that was transfected into the parasites using three transfection techniques i.e. soaking, electroporation and electro-soaking, (3) EgCaM-specific siRNA group delivered by soaking, (4) EgCaM-specific siRNA group delivered by electroporation and (5) EgCaM-specific siRNA group delivered by electro-soaking 13,14,21,49 . EgCaM-specific and Negative Control siRNAs were delivered into the parasites of corresponding treatment groups using three transfection techniques. No interventions were made on the parasites in no-treatment groups. For siRNA delivery, in 6-well plates different stages of the parasite with three separate batches of 5000 protoscoleces, 15 microcysts and 15 strobilated worms were treated by soaking, electroporation and electro-soaking in the presence of 50 nM/ml of fluorescent siRNA 50 . For soaking, protoscoleces were cultured in 2.5 ml DMEM supplemented with siRNAs and a lab-made transfection reagent was then added to each well and maintained for 6 hours 51 . For electroporation, protoscoleces were removed from the culture, resuspended in 200 μl of electroporation buffer containing the fluorescently labeled siRNA to give a final concentration of 50 nM, transferred to a 2 mm gap cuvette, and were subjected to time constant protocol (125 V, 20 ms) using Gene Pulser II (Bio-Rad, CA) 13,16,31 . For electro-soaking following electroporation, the parasites were maintained in a culture media containing siRNAs for 6 hours. After treatment, the parasites were rinsed thoroughly in PBS to remove unincorporated siRNA molecules. A fluorescent microscope (HB-10101AF, Nikon, Japan) was used to examine siRNA delivery. The parasites viability, as revealed by flame cell activity, were recorded daily for eight days and any phenotypic changes e.g. motility, transparency and tegumental alterations were examined. In addition, the size of the protoscoleces was measured using a calibrated microscope. The motility of the strobilated worms from each group of target siRNA and controls were measured as the number of body contractions per minute (see the Supplementary Fig. S1).
Gene expression assay. Total RNA was extracted from different developmental stages of the parasite (protoscoleces, microcysts and three or more proglottids) using a commercial kit (RNeasy Mini Kit, Qiagen, Germany). Furthermore, RNA was quantified by measuring the absorbance ratios at 260 and 280 nm by spectrophotometry (Nano Drop ND-1000, Nano Drop Technologies, Wilmington, DE). cDNA was then synthesized from 100 ng RNA in a total volume of 20 μl using miscript ® II Reverse Transcriptase Kit (Qiagen, Germany) according to the manufacturer's instructions.
For relative quantification, 2 −ΔΔCT method was employed, using β-actin as the reference gene for each sample. Results obtained from the parasites treated with the unrelated Negative Control siRNA were used as calibrators.
Negative Control siRNA target gene was unknown, as it is the proprietary information of Qiagen ® . According to the manufacturer, the unrelated siRNA has been proven to have no significant effect on cell proliferation, viability, or morphology. Statistical analysis. Differences between and within groups were assessed for statistical significance using one-way or two-way ANOVA test using GraphPad Prism 7.0 Software (www.graphpad.com). P values of less than 0.05 were considered significant.