Developmental Sensitivity in Schistosoma mansoni to Puromycin To Establish Drug Selection of Transgenic Schistosomes

ABSTRACT

Schistosomiasis is considered the most important disease caused by helminth parasites, in terms of morbidity and mortality. Tools to facilitate gain- and loss-of-function approaches can be expected to precipitate the discovery of novel interventions, and drug selection of transgenic schistosomes would facilitate the establishment of stable lines of engineered parasites. Sensitivity of developmental stages of schistosomes to the aminonucleoside antibiotic puromycin was investigated. For the schistosomulum and sporocyst stages, viability was quantified by fluorescence microscopy following dual staining with fluorescein diacetate and propidium iodine. By 6 days in culture, the 50% lethal concentration (LC₅₀) for schistosomula was 19 μg/ml whereas the sporocysts were 45-fold more resilient. Puromycin potently inhibited the development of in vitro-laid eggs (LC₅₀, 68 ng/ml) but was less effective against liver eggs (LC₅₀, 387 μg/ml). Toxicity for adult stages was evaluated using the xCELLigence-based, real-time motility assay (xWORM), which revealed LC₅₀s after 48 h of 4.9 and 17.3 μg/ml for male and female schistosomes, respectively. Also, schistosomula transduced with pseudotyped retrovirus encoding the puromycin resistance marker were partially rescued when cultured in the presence of the antibiotic. Together, these findings will facilitate selection on puromycin of transgenic schistosomes and the enrichment of cultures of transgenic eggs and sporocysts to facilitate the establishment of schistosome transgenic lines. Streamlining schistosome transgenesis with drug selection will open new avenues to understand parasite biology and hopefully lead to new interventions for this neglected tropical disease.

INTRODUCTION

Schistosomiasis is considered the most significant of neglected tropical diseases (NTDs) caused by helminth parasites, in terms of morbidity and mortality (1). The genome sequences of the schistosome species parasitizing humans, Schistosoma mansoni, S. japonicum, and S. haematobium, have been reported (3–5), and therefore, functional genomics tools that allow both gain- and loss-of-function approaches are needed to interrogate these sequences to discover new intervention targets (6–11). Transgenesis is a powerful functional genomics tool for both forward genetics, such as insertional mutagenesis and promoter trapping (7, 9), and reverse genetics, such as gene silencing mediated by vector-based RNA interference (RNAi) (2, 12). More recently, clustered regularly interspaced short palindromic repeats (CRISPR)–CRISPR-associated protein 9 (CRISPR-Cas9) has shifted the landscape for manipulating the genome by introducing specific mutations in the DNA (13). Transgenesis coupled with CRISPR-Cas9-driven genome editing likely would advance functional genomics for this NTD pathogen (14).

In like fashion to more-tractable systems, including Caenorhabditis elegans and Plasmodium falciparum (15, 16), transgenesis for schistosomes would represent a powerful tool to investigate new interventions (7, 17). We developed a platform for transgenesis in schistosomes using mammalian retroviruses, pseudotyped murine leukemia virus (MLV), and lentivirus to insert transgenes into schistosomes, including reporter genes and short hairpin RNA (shRNA)-expressing cassettes (7, 18–27). The pseudotyped virions transduce developmental stages of S. mansoni, including schistosomula and eggs cultured in vitro, leading to random integration through all eight chromosomes of the flatworm. By transfecting the egg stage, retroviral transgenes reach the germ line and are transmitted to filial generations, through both the asexual and sexual developmental cycles of the schistosome (22, 26). This protocol has successfully led to the establishment of stable lines of transgenic schistosomes (26). This route inserts the transgene into schistosome chromosomes, through the activity of the retroviral integrase enzyme and, by focusing on the egg stage, delivers transgenes to the germ line, thereby avoiding mosaicism after establishing the transgenic lines. Moreover, it has been shown that lentivirus expressing microRNA-adapted short hairpin RNAs targeting genes expressed in schistosome eggs could be employed to transfect eggs and study the inflammatory response of the transduced eggs in mouse lung tissue (28). Recently, we demonstrated that transduction of schistosomula with pseudotyped HIV-1 led to integration of viral cDNA into schistosome chromosomes (27).

Genetic selection systems that combine antibiotics and antibiotic resistance markers allow conditional enrichment of manipulated cells (29). Drug selection finds broad application in tissue culture of human cells, microbial pathogens, and other systems. In like fashion, antibiotic selection for schistosome transgenesis is desired because it would rapidly hasten progress for functional genomics by stringent enrichment of transgenic worms. Although transgenesis has long been employed with molecular genetics of C. elegans, surprisingly, antibiotic selection was deployed only in the last decade (15, 29–32). An absence of information on sensitivity to commonly used laboratory antibiotics was likely responsible for the delay. A key review in 2010 (29) highlighted this development with C. elegans and summarized the three essential elements needed to be considered to develop a transgenesis system for a new target species: (i) genomic DNA transformation strategies, (ii) stable expression of transgenes, (iii) specific selection conditions to enrich the population of the transgenic organisms. Here we focused on the last element by investigating the toxicity of the aminonucleoside antibiotic puromycin, widely utilized in selection of mammalian cells (33), against developmental stages of S. mansoni.

RESULTS

Mammalian and molluskan schistosome larvae exhibit differential sensitivity to puromycin.

Schistosomula were cultured in the presence of indicated concentrations of puromycin for 10 days. The culture media and antibiotic were replaced every second day (Fig. 1A). Sensitivity to puromycin was evaluated by both microscopic appearance and dual-fluorescence bioassay (Fig. 1B; see also Fig. S1 in the supplemental material) (34). Differences between parasites cultured in complete Basch's medium and parasites cultured with 0.05 mM HEPES vehicle control were not apparent in any of the experiments (Fig. 1C). Schistosomula cultured in puromycin at 0.25 μg/ml showed minimal toxicity, whereas those in concentrations of ≥25 μg/ml exhibited mortality rates of approximately ≥20% from day 2, which increased as time progressed (Fig. 1C to E and S1). However, from day 6 onwards, differences became more evident with toxicity dramatically increasing as the 50% lethal concentration (LC₅₀) plunged to 19.32 μg/ml (Fig. 1C to E and S1), and by days 8 and 10, mortality had plateaued with LC₅₀s of 1.01 and 0.85 μg/ml, respectively. These assays defined a time window and puromycin concentration range in which dose-dependent responses were elicited for schistosomula. This information facilitated downstream rescue assays involving murine leukemia virus (MLV)-transduced schistosomules (described in “PuroR-expressing schistosomula rescued when cultured in puromycin” below).

FIG 1.

Schistosomula of Schistosoma mansoni cultured in the presence of puromycin. (A) Schistosomula were cultured for 10 days in the presence of increasing concentrations of antibiotic. Culture media and puromycin were replaced when indicated (P). CT, cercarial transformation; P, puromycin. (B) Representative micrographs of schistosomula cultured for 8 days in 0 μg/ml (top), 2.5 μg/ml (center), or 12.5 μg/ml (bottom) puromycin and costained with propidium iodide (PI) and fluorescein diacetate (FDA) to label dead (red) and live (green) parasites, respectively. Selected concentrations induced minimal harm from no treatment, substantial harm from 2.5 μg/ml, and complete killing from 12.5 μg/ml. Bar, 200 μm. (C) Survival of schistosomula over time displayed as means ± standard errors of the means (SEM). For clarity, selected curves only are presented. (D) Dose-response curves (DRC) of schistosomula exposed to puromycin (means ± SEM). (E) Fifty percent lethal concentration (LC₅₀) over time indicating 95% confidence intervals.

The intramolluskan developmental stage primary sporocysts, obtained by collecting miracidia from liver eggs and further transformation into sporocysts in vitro, were cultured in appropriate culture media under hypoxic conditions at 25°C. The sensitivity of this less well studied developmental stage to puromycin was assessed by culturing the parasite in the presence of increasing concentrations of the antibiotic and replacing the medium only once a week to ensure that hypoxic conditions were maintained (Fig. 2A). The sensitivity to puromycin was evaluated by both microscopic appearance and dual-fluorescence bioassay (Fig. 2B). Whereas less than 40% of the larvae cultured with 1,000 μg/ml puromycin remained viable at day 6, concentrations of ≤500 μg/ml of the antibiotic did not induce significant mortality in sporocysts compared to controls until day 21 (Fig. 2B to D). In contrast to the schistosomula with plummeting LC₅₀s, the LC₅₀s only slowly dropped from day 6 at 872 μg/ml to day 21 at 408 μg/ml (Fig. 2E).

FIG 2.

Schistosoma mansoni sporocysts cultured in the presence of puromycin. (A) Primary sporocysts were cultured for 21 days in increasing concentrations of the antibiotic. Culture media and puromycin were replaced when indicated (P). ST, sporocyst transformation; P, puromycin. (B) Representative micrographs of primary sporocysts cultured in 0 μg/ml (top), 250 μg/ml (center), or 1,000 μg/ml (bottom) puromycin and costained with propidium iodide (PI) and fluorescein diacetate (FDA) to stain dead (red) and live (green) parasites, respectively. Bar, 200 μm. (C) Sporocyst survival over time displayed as means ± SEM. (D) Dose-response curves (DRC) of sporocysts exposed to puromycin (means ± SEM). (E) Fifty percent lethal concentration (LC₅₀) over time indicating 95% confidence interval error bars.

To summarize, both the intramammalian schistosomulum and the intramolluskan primary sporocyst developmental stages of S. mansoni were sensitive to increasing concentrations of puromycin. However, these developmental stages displayed a marked disparity in tolerance to puromycin, with the schistosomulum stage displaying far greater sensitivity, with a difference in sensitivity of one order of magnitude between these larval stages. By 6 days in culture, the 50% lethal concentration (LC₅₀) for schistosomula was 19 μg/ml whereas the sporocysts were 45-fold more resilient with an LC₅₀ of 872 μg/ml (Fig. 1E and 2E).

Puromycin impairs the development and hatching of schistosome eggs.

Eggs laid in vitro by female adult worms kept in culture (in vitro-laid eggs [IVLEs]) are of particular interest, given that they represent a targetable developmental stage to derive germ line-stable transgenic lines of schistosomes (22, 26). It is of interest to apply antibiotic selection to transduced eggs in order to enrich the transgenic population of parasites (22, 27). Accordingly, egg sensitivity to puromycin was analyzed by incubating IVLEs collected within 72 h after the recovery of adult worms from experimentally infected mice in the presence of increasing concentrations of the antibiotic (Fig. 3A). The culture medium and antibiotic were replaced by day 5 after perfusion, and the eggs were washed and incubated in water to induce hatching 7 days after perfusion. From day 4 after perfusion, pictures of nonoverlapping fields were recorded each day for egg development analysis (Fig. 3B and C). Almost 100% of the eggs cultured with puromycin concentrations of ≥200 μg/ml displayed impaired development and halted normal maturation within 24 h (not shown). However, eggs cultured in lower concentrations of puromycin exhibited clear, concentration-dependent developmental impair (Fig. 3B to D). Eggs incubated with no antibiotic develop within 7 days according to the Vogel and Prata's staging system of egg maturation (35, 36), and 24.5% of the eggs reached stage V of development and/or hatched when exposed to water. Eggs cultured in the presence of 1 μg/ml puromycin or greater failed to mature beyond stages III and IV, and more than 80% of the eggs remained in stages I and II by day 7 (Fig. 3B and C). Dose-response curves based on stage V and hatched eggs (stage V + hatched) show that the LC₅₀ remained moderately stable (67 to 134 ng/ml) over the 4 days following puromycin addition (days 4 to 7 postperfusion) as shown in Fig. 3D and E. These findings revealed a concentration range of antibiotic, i.e., from 0.1 to 1 μg/ml, for future rescue experiments with MLV-transduced eggs expressing the puromycin-resistant marker.

FIG 3.

Puromycin impairs the egg maturation. (A) Schematic showing the experimental design. Eggs laid in vitro (in vitro laid eggs [IVLEs]) by female adult schistosomes cultured for 3 days after perfusion were collected and exposed to the indicated concentrations of puromycin or vehicle control for 4 days. The culture media and puromycin were replaced on day 5 after perfusion, and 2 days later eggs were induced to hatch by transfer into water and exposure to bright light. (B) Percentages of eggs at different maturation stages as categorized using the staging system of Vogel and Prata (35) for the indicated days after perfusion and puromycin concentrations. (C) Representative pictures of IVLEs cultured in the presence of 0, 1, and 100 μg/ml of puromycin at 7 days after perfusion. Bar, 200 μm. (D) Dose-response curves (DRC) of IVLEs exposed to puromycin (means ± SEM). (E) Fifty percent lethal concentration (LC₅₀) over time; 95% confidence intervals are indicated.

Eggs isolated from infected livers have also been previously employed to derive transgenic lines, although less efficiently than when IVLEs were employed (22). On the other hand, ∼70% of the eggs from liver are already mature, i.e., at stage V of Vogel and Prata's staging system of egg maturation, and hatch as soon as they are transferred to water and incubated under light (17). Therefore, we reasoned that these eggs could also be employed for antibiotic selection and decided to test the effect of puromycin on egg hatching. Eggs incubated for 5 days in the presence of the indicated concentrations of the antibiotic were washed and transferred to water under light for 1 h to induce hatching. A reduction in the hatching rate of eggs cultured with 250 μg/ml of antibiotic was evident compared to control eggs cultured with vehicle control, i.e., 2 mM HEPES (Fig. 4A and B). Higher concentrations, i.e., 500 and 1,000 μg/ml, did not further reduce the hatching rate, and the dose-response curve yielded an LC₅₀ of 387 μg/ml (Fig. 4C).

FIG 4.

Toxicity of puromycin for schistosome eggs isolated from mouse livers. (A) Representative micrographs of eggs cultured in the presence of 0, 125, and 1,000 μg/ml puromycin for 5 days, hatched under bright light for 2 h, and fixed. Bar, 200 μm. (B) Eggs isolated from liver were cultured for 5 days in increasing concentrations of puromycin or controls (untreated and vehicle control, i.e., 1 mM HEPES). Thereafter, the eggs were washed, transferred to water, induced to hatch under bright light for 2 h, fixed, and counted. Eight to 10 nonoverlapping field pictures showing an average of 12 eggs per field were inspected in each group, i.e., 96 to 120 eggs per treatment group. Each dot within the groups indicates the percentage of hatched eggs in each nonoverlapping field, and the means of hatched eggs for treatments group are presented. One-way ANOVA among the groups: P ≤ 0.05; t test between each group and HEPES control: *, P ≤ 0.05; **, P ≤ 0.01. (C) Dose-response curves (DRC) of relative hatched eggs exposed to increasing concentrations of puromycin (means ± SE). Fifty percent lethal concentration (LC₅₀) is shown.

These analyses revealed that both eggs laid in vitro and eggs isolated from liver were sensitive to puromycin, impairing the maturation and reducing the hatching rate. Consequently, we predict that antibiotic selection could be employed to enrich the population of transduced eggs expressing the puromycin-resistant marker (5, 32). Interestingly, IVLEs were dramatically more susceptible (2,888- to 5,671-fold) to the effect of the drug (Fig. 3E and 4C).

Puromycin kills adult schistosomes in vitro.

The xCELLigence worm real-time motility (xWORM) assay was employed to evaluate the sensitivity of adult worms to three applications of puromycin over time (Fig. 5A) (37–40). Ten male or female adult worms were individually incubated in increasing concentrations of the antibiotic, and the motility/vitality of the parasites was quantified. Figure S2A in the supplemental material shows the cell index output of the xWORM assay revealing the constant large amplitude of the untreated control and the amplitude at 1,000 μg/ml puromycin that falls over time as the drug reduces the worm motility. During xWORM analysis, the standard deviation (SD) of the cell index output, 120 min (Fig. S2A), was employed to establish the motility index and the adult motility percentage compared to those of control schistosomes cultured without the antibiotic. Adult schistosomes were sensitive to puromycin at concentrations of ≥10 μg/ml (Fig. 5B and C). These findings were corroborated by direct microscopic inspection, whereby the motility/vitality of the schistosomes dropped in a concentration-dependent manner (Fig. 5B; see also the videos in the supplemental material). Reduced motility/viability of adult worms was evident during culture in puromycin compared to drug-free controls, with both female (Fig. 5C and D) and male (see Fig. S2B and C in the supplemental material) worms. Some differences were observed between males and females, but overall, schistosomes of both sexes succumbed to puromycin, with the LC₅₀ dropping below 20 μg/ml after only 48 h and one puromycin treatment (Fig. 5E). These findings suggested that selection of puromycin resistance-expressing worms might be feasible, both in vitro and in vivo.

FIG 5.

Puromycin kills adult schistosomes in vitro. (A) Experimental design: 1 day after collection (C) the adult worms were separated into males and females and exposed to increasing concentrations of puromycin (P). The medium and antibiotic were changed every other day. (B) Representative micrographs of adult female worms cultured in the presence of 1,000 μg/ml of antibiotic (left) or HEPES-treated control group (right). Bar, 250 μm. (C) Adult motility measured by xWORM assay of female worms exposed to the indicated concentrations of puromycin over time. (D) Dose-response curves (DRC) of survival relative to control cultured in the absence of antibiotic at three different time points as indicated (means ± SE). (E) Fifty percent lethal concentration (LC₅₀) over time indicating 95% confidence intervals for female and male adult schistosomes 48 h after exposure to the antibiotic.

PuroR-expressing schistosomula were rescued when cultured in puromycin.

We have previously reported selection of murine leukemia virus-transduced schistosomula expressing NeoR when cultured in the presence of neomycin, i.e., G418 (5, 22), an aminoglycoside antibiotic. Here, we investigated drug selection with the aminonucleoside antibiotic puromycin and PuroR-expressing schistosomula that were transduced by MLV virions. MLV-transduced schistosomula expressing PuroR were cultured with 2.5 μg/ml of puromycin for 6 days. Significantly more MLV-transduced schistosomula survived than did controls (Fig. 6A and B) when exposed to puromycin for 6 days (51.7% versus 36.4%, transduced parasites versus controls, respectively). In addition, PuroR transgene expression was detected by reverse transcription-quantitative PCR (RT-qPCR) in MLV-transduced schistosomula (Fig. 6C). When higher concentrations of puromycin, ≥25 μg/ml, or longer incubation times were tested for these experiments, rescue of PuroR-expressing schistosomula was not evident (not shown).

FIG 6.

Rescue from puromycin toxicity of PuroR-expressing schistosomula. (A) Micrographs of control (top) or MLV-transduced parasites expressing puroR (bottom) cultured in the presence of puromycin for 6 days. Bar, 100 μm. (B) Percentage of live control or MLV-transduced schistosomula expressing PuroR cultured in the presence of 2.5 μg/ml of puromycin for 6 days (means ± SD; SD_control = 0.09, SD_{MLV_puroR} = 0.14; n = 5 micrographs showing nonoverlapping fields containing ≥50 parasites). Student's t test between groups: *, P ≤ 0.05. (C) Relative PuroR transgene expression in control or MLV-transduced schistosomula cultured in 2.5 μg/ml of puromycin for 6 days. The experiment was repeated at least twice with different batches of virions.

DISCUSSION

The genome sequences of the major schistosome species, Schistosoma mansoni, S. japonicum, and S. haematobium, have been reported (3–5), and functional genomics tools, including gain-of-function/loss-of-function approaches, are now needed to interrogate these genomes in the hopes that they will lead to new interventions against these neglected tropical disease pathogens (6–11, 14). Genetic selection systems that combine antibiotics and antibiotic resistance markers allow conditional enrichment of manipulated cells (29). Drug selection finds wide application in tissue culture of human cells, microbial pathogens, and other systems. Antibiotic selection for schistosome transgenesis is desired because it would rapidly hasten progress by stringent enrichment of transgenic worms. Although transgenesis has long been employed with molecular genetics of C. elegans, antibiotic selection was, surprisingly, deployed only in the last decade (15, 29–32). We have developed platforms using a mammalian retrovirus, vesicular stomatitis virus glycoprotein (VSVG)-pseudotyped MLV and VSVG-pseudotyped HIV-1 to introduce transgenes into schistosomes, including reporter genes and shRNA-expressing cassettes (5, 18–21, 23–25, 27, 41). The tractability of the approach might be markedly improved by inclusion of selectable resistance markers. Drug selection of genetically modified schistosomes can be expected to improve the efficiency of functional genomic tools applied to this neglected tropical disease pathogen.

The aminonucleoside antibiotic puromycin is widely used as a selective agent in cell culture, is active against both prokaryotic and eukaryotic cells, and is also an antineoplastic agent. Puromycin inhibits protein synthesis in vivo as well as in cell-free systems. Whereas its mode of action has not been definitively established, in structure puromycin resembles aminoacyl-tRNA and would release uncompleted polypeptide chains prematurely from the ribosome. Puromycin also inhibits the metalloprotease cytosol alanyl aminopeptidase and the serine protease dipeptidyl peptidase (42–46). Resistance to puromycin is conferred by the pac gene, which encodes a puromycin N-acetyl-transferase (PAC) isolated from Streptomyces alboniger. The pac (PuroR) gene is widely used as a selectable marker for enrichment of transformed cells in culture in the presence of puromycin (47). Here, we investigated the sensitivity of discrete developmental stages of the schistosome to puromycin. Schistosomula, eggs isolated from liver or laid in vitro by adults, sporocysts, and adult worms of S. mansoni were cultured in the presence of increasing concentrations of puromycin. The intramammalian larvae schistosomula and intramolluskan larvae sporocysts were scored as live or dead by dual-fluorescence bioassay (34, 48). The viability of schistosome eggs isolated from liver was evaluated by the hatching assay on days 5 and 10, whereas the development of the eggs laid in vitro was monitored microscopically every day and by the hatching assay on day 7 after collection. Although we had observed that eggs were insusceptible to the aminoglycoside antibiotic neomycin at doses of up to 10 mg/ml (G. Rinaldi, A. E. Folley, V. H. Mann, and P. J. Brindley, unpublished data), all the developmental stages examined in the current study were sensitive to puromycin. However, the sensitivity to puromycin varied dramatically among the developmental stages investigated here. Notable among the differences, eggs laid in vitro showed impaired development with concentrations of puromycin at 1/100 the concentrations of puromycin that induced a hatching rate reduction of eggs isolated from liver of mice and very different (>5,600-fold) LC₅₀s.

To quantify motility and viability of parasitic helminths in vitro, schistosomes can be investigated using the xCELLigence worm real-time motility (xWORM) assay (39, 40, 48), among other objective assays (34). The xWORM assay was employed to quantify the motility/viability of adult schistosomes in escalating concentrations of puromycin (39). Both male and female worms were negatively affected by puromycin; however, males were more sensitive than females. This sensitivity profile is similar for praziquantel (49). Schistosomula and IVLEs were markedly more sensitive to puromycin than the liver eggs and sporocysts. Regarding the disparate sensitivities to puromycin between the intramammalian and the intramolluskan larvae, we speculate that they relate to (i) the divergent culture conditions, e.g., different culture media and oxygen levels, since the sporocysts were cultured under hypoxic conditions, and/or (ii) intrinsic differences between developmental stages, e.g., tegument permeability that may affect drug accessibility to the parasite or divergent metabolic and protein synthesis, among others. It has been described that certain antibiotics, such as tunicamycin, may display a reduced activity during hypoxia (50). Both the high concentrations of puromycin (average LC₅₀, ∼620 μg/ml) and the long interval (≥6 days in the presence of 1 mg/ml of antibiotic) required to kill sporocysts in vitro, at least under the conditions tested here, might be disadvantageous for the application of drug selection in this developmental stage. Moreover, high concentrations of antibiotic may induce nonspecific toxicity in the cell, which would impede efficiency of drug selection (51). In the future, we aim to test other culture conditions, including more-frequent replacement of medium and antibiotic and the addition of compounds to increase the permeability of the tegument, e.g., mild detergents (32). Puromycin effectively halts the development (and indeed kills) IVLEs. The IVLEs mature in 1 week, and under in vitro conditions about 15 to 20% fully develop into the miracidia that hatch upon contact with water and can infect the intermediate host snail by the natural route of direct penetration of the head-foot tissues (22, 41). High metabolic and protein synthesis activities are required for embryogenesis of the miracidium (35), and hence puromycin, which blocks protein synthesis, is expected to induce damage even at low concentrations. On the other hand, more than 70% of the eggs isolated from liver of experimentally infected mice are already mature and ready to hatch. Eggs recovered from mouse liver may display lower metabolic and protein synthesis activity than the IVLEs, being more resistant to the antibiotic, i.e., much higher LC₅₀s were observed for liver eggs than for IVLEs. Antibiotic selection applied to IVLEs transduced with retrovirus during development in vitro can be expected to facilitate enrichment of transgenic lines, providing a key additional selection step in the protocol to generate transgenic lines of schistosomes reported previously (22). We consider that this will enhance the efficiency of the approach through enrichment of transgenic parasites and maintenance of stable transgenic lines, as shown with C. elegans (15). To our knowledge, eggs either isolated from liver of mice or laid in vitro by female worms and mother sporocysts obtained by in vitro transformation of miracidia have been the only two developmental stages successfully employed to target and transform the germ line of schistosomes (22, 41, 52). Therefore, we hypothesize that the eggs and sporocysts are key developmental stages to establish germ line transgenesis for schistosomes, presumably, given the high germ line-to-soma cell ratio in these developmental stages (35). Surprisingly, the eggs were exceedingly resistant to G418 (see above), and hence, this antibiotic could not be employed to apply selection in such a relevant developmental stage. In contrast to their reaction to G418, eggs were indeed sensitive to the aminonucleoside antibiotic puromycin, which additionally displays other advantages, i.e., puromycin has been shown to be faster and more efficient than G418 in mammalian cells (53) and has been already validated in invertebrate models such as C. elegans (32) and Drosophila sp. strain Schneider S2 cells (54).

Previously, we reported that schistosomula expressing NeoR could be partially rescued and survived longer than control wild-type schistosomula when cultured in the presence of the aminoglycoside neomycin (55). Moreover, a germ line transgenic progeny (schistosomula), i.e., transgenic schistosomula obtained from snails infected with miracidia that had been hatched from MLV-transduced eggs, also exhibited resistance to G418 (22). Promisingly, here we showed that schistosomula transduced with MLV virions expressing the PuroR gene were partially rescued when cultured in the presence of puromycin. However, it appears that at least in MLV-transduced schistosomula, where somatic transgenesis was obtained, neomycin-based selection (55) was more efficient to enrich for transduced parasites than the present puromycin/PuroR system, through which puromycin killed ∼65% of nontransformed parasites and less than 50% of transformed parasites. We have reported that 1,000 μg/ml of G418 killed ∼40% of nontransformed parasites and ∼20% of NeoR-expressing parasites at day 4 and that 500 μg/ml of G418 killed ∼40% of nontransformed parasites and only ∼10% of NeoR-expressing parasites at day 8 (55). On the other hand, although they are naturally resistant to G418, we now report that eggs were sensitive to puromycin, allowing to speculate that puromycin selection might be a better choice to be applied in eggs, enriching for transformed miracidia before infecting snails. Consecutively, schistosomula derived from these snails might be subjected to neomycin selection. Even though puromycin is indeed effective to kill the diverse developmental stages examined here, neomycin may be better to drive the selection of specific developmental stages such as schistosomula (22, 55). Further experiments that compare the neomycin/NeoR selection system with the puromycin/PuroR selection system not only in MLV-transduced schistosomula but also in schistosomula derived from MLV-transduced eggs may refine the most effective selection strategy, at least for schistosomula in culture.

These findings can be expected to facilitate “dual selection” of schistosomes with neomycin and puromycin. The neomycin/NeoR system is more efficient in schistosomula, but the puromycin/PuroR selection system might be appropriate in eggs. Accordingly, a “consecutive dual selection” might be applied to enrich transgenic schistosomes, applying puromycin selection in eggs and subsequent neomycin selection in schistosomula from the same transgenic line. Given that both intramolluskan and intramammalian developmental stages are sensitive to the antibiotic puromycin, we can speculate that in addition to in vitro selection of transgenic parasites, in vivo approaches whereby schistosome-infected snails and/or mice are treated with antibiotic to enrich for PuroR-expressing parasites might be applied in the future. Challenges such as reaching effective concentrations of the antibiotic in serum to select transgenic schistosomes in vivo might be encountered. However, at the present time, similar approaches are routinely applied to select transgenic protozoan parasites (56–58). Moreover, schistosomicidal drugs, including oxamniquine and praziquantel, have been employed to treat infected snails (59), paving the way for similar approaches with antibiotics, e.g., puromycin and neomycin. An essential condition to apply selection is to achieve a constant, strong, and reliable expression of transgenes. Insulator sequences protect the expression of transgenes against chromatin-silencing effects (23), and we are currently optimizing transgene expression by identifying and cloning schistosome regulatory elements. Finally, manipulation by transgenesis, genome editing, knockout, and/or perturbation by CRISPR-Cas9-related approaches (60–62) can be predicted to enhance understanding of these pathogens, their somatic stem cells (63), reproduction, longevity in infected hosts (64, 65), and novel intervention targets. These approaches, including transgenes encoding drug resistance markers, may also facilitate establishment of sex-biasing gene drives to block the spread of schistosomiasis (66).

MATERIALS AND METHODS

Ethics statement.

Mice infected with Schistosoma mansoni were obtained from the Biomedical Research Institute (BRI), Rockville, MD, and housed at the Animal Research Facility of the George Washington University Medical School, which is accredited by the American Association for Accreditation of Laboratory Animal Care (AAALAC no. 000347) and has an Animal Welfare Assurance on file with the National Institutes of Health, Office of Laboratory Animal Welfare, OLAW assurance number A3205-01. All procedures employed were consistent with the Guide for the Care and Use of Laboratory Animals. Maintenance of the mice and recovery of schistosomes were approved by the Institutional Animal Care and Use Committee of the George Washington University. For the experiments performed in the Wellcome Sanger Institute (WSI), all animal experiments were conducted under Home Office Project Licenses no. 80/2596 and P77E8A062. All protocols were presented and approved by the Animal Welfare and Ethical Review Body (AWERB) of the WSI. The AWERB is constituted as required by the UK Animals (Scientific Procedures) Act 1986 Amendment Regulations 2012.

Schistosome developmental stages.

Biomphalaria glabrata snails and mice infected with the NMRI (Puerto Rican) strain of S. mansoni were supplied by the Biomedical Research Institute, Rockville, MD, USA, or maintained at the WSI, Hinxton, Cambridgeshire, UK, where the complete life cycle of S. mansoni (NMRI strain) has been established. Developmental stages of Schistosoma mansoni were obtained and cultured as described previously (17). Schistosomula were obtained by mechanical transformation of cercariae. In brief, snails were washed, transferred to water (∼100 ml), and exposed to light to induce cercarial shedding for 2 h, replacing the water and collecting cercariae after 30 min. Cercariae were concentrated by centrifugation (800 × g for 15 min) and resuspended in Dulbecco's modified Eagle medium (DMEM) supplemented with 200 U/ml penicillin, 200 μg/ml streptomycin, 500 ng/ml amphotericin B, and 10 mM HEPES (schistosomula wash medium). Tails of cercariae were removed by ∼20 passes back and forth through a 22-gauge (22-G) emulsifying needle, bodies of schistosomula were separated from the tails by Percoll gradient centrifugation, and the parasites were cultured at 37°C in modified Basch's medium under 5% CO₂ in air (17, 67). Mixed-sex adults of S. mansoni were perfused from mice 6 to 9 weeks after infection and maintained as described previously (17).

Eggs were recovered from two sources (22, 41): (i) eggs isolated from experimentally infected mouse livers digested with collagenase (68) and cultured at 37°C under 5% CO₂ in air in DMEM supplemented with 10% fetal bovine serum and 100 U of penicillin and streptomycin (69); (ii) eggs laid in vitro by female worms, i.e., in vitro-laid eggs (IVLEs) (22, 41). In brief, after recovery from mice, adult worms were washed and transferred into a 74-μm-diameter mesh Netwell, in 6-well tissue culture plates, where they were maintained in culture at 37°C under 5% CO₂ in air in modified Basch's medium (17, 70) for 48 h. (Eggs laid after the females have been in culture for >48 h progressively develop less well [41, 71].) The IVLEs from these worms fell through the mesh and were collected on the surface of the culture plate and concentrated by filtering through an 8-μm-mesh Transwell (41). Thereafter, IVLEs were maintained in modified Basch's medium, where they developed and matured within a week. At that point, they were transferred to sterile water and exposed to bright light to induce hatching. Approximately 20% of the IVLEs hatched within 120 min, releasing motile miracidia (22, 41).

Primary sporocysts were obtained by transferring miracidia hatched from eggs isolated from livers into sporocyst medium, a chemically defined medium termed MEMSE-J (17, 72, 73), in 6-well tissue culture plates maintained in a hypoxia chamber (catalogue no. MIC-101; Billups-Rothenberg, Del Mar, CA) under 5% O₂, 5% CO₂, and 90% N₂ at 25°C for ≥14 days (74).

Dose-response curves to puromycin by schistosomula, sporocysts, and eggs.

Schistosomula were obtained as described above and transferred 2 days later into 24-well tissue culture plates (∼1,000 parasites per well) and cultured in the presence of puromycin at concentrations ranging from 0.25 μg/ml (0.46 mM) to 500 μg/ml (920 mM) as indicated (see below). Puromycin hydrochloride was supplied at 10 mg/ml in 20 mM HEPES buffer (Invitrogen, USA). Wells containing parasites in medium only or medium with matched concentrations of HEPES were included as controls. We arbitrarily divided the dose-response curve assays into the following two groups: low-concentration curves, i.e., ranging from 0.25 μg/ml (0.46 mM) to 25 μg/ml (46 mM), and high-concentration curves, i.e., ranging from 25 μg/ml (46 mM) to 500 μg/ml (920 mM). Culture medium and puromycin were replaced every second day from days 0 to 10 (Fig. 1A). Primary sporocysts obtained and maintained for 3 days, as described above, were transferred to 24- or 6-well plates (∼1,000 to 3,000 parasites per well) and cultured in the presence of increasing concentrations of puromycin ranging from 100 μg/ml (184 mM) to 1,000 μg/ml (1,840 mM) in sporocyst medium. Medium and puromycin were replaced weekly, as indicated (Fig. 2A).

After the schistosomula or sporocysts were exposed to increasing concentrations of puromycin, at the indicated time points, viability of the parasites was ascertained and scored. About ∼200 schistosomula or sporocysts for each treatment and each time point were transferred from culture to a 24-well plate well containing 500 μl of 1× phosphate-buffered saline (PBS) and costained with propidium iodide (PI; 2 μg/ml; Sigma-Aldrich, St. Louis, MO) and fluorescein diacetate (FDA; 0.5 μg/ml; Sigma-Aldrich) as described previously (34, 48). Four or five micrographs of the costained parasites were taken using an epifluorescence Zeiss Axio Observer A.1 inverted microscope fitted with a digital camera (AxioCam ICc3; Zeiss, Germany). Manipulation of digital images, limited to insertion of scale bars, adjustments of brightness and contrast, cropping, and the like, was undertaken with the AxioVision release 4.6.3 software (Zeiss). Micrographs of each puromycin concentration recorded nonoverlapping regions of the tissue culture well.

In vitro-laid eggs (IVLEs) were collected, transferred to an 8-μm-diameter mesh Transwell in 24 well-tissue culture plates, and cultured in modified Basch's medium in increasing concentrations of puromycin ranging from 0.1 μg/ml (0.184 mM) to 1,000 μg/ml (1,840 mM) as indicated. The culture medium and puromycin were replaced every second day, and the eggs were induced to hatch 7 days after perfusion (Fig. 3A). The development of IVLEs in culture was assessed each day by microscopic observation and documented pictorially in micrographs of nonoverlapping regions of the culture. The IVLEs were counted, classified according to the Vogel and Prata's staging system of egg maturation (35, 36), and clustered in three main groups from the most immature to the most developed eggs, i.e., immature (stages I and II), intermediate (stages III and IV), and developed eggs (stage V and hatched eggs stage). Approximately 100 eggs isolated from livers of experimentally infected mice, i.e., liver eggs (LEs), were transferred into 96-well tissue culture plates and cultured in puromycin at concentrations that ranged from 125 μg/ml (230 mM) to 1,000 μg/ml (1,840 mM) in fresh DMEM supplemented with 10% fetal bovine serum and 100 U of penicillin and streptomycin as described above. Wells containing LEs in medium alone or medium containing HEPES were included as controls. The culture media and puromycin were replaced every second day. After treatment with puromycin for 5 or 10 days, the LEs were washed three times in 1× PBS, transferred to 500 μl distilled water, and exposed to bright light to induce hatching. Two hours later, 10% formaldehyde in 1× PBS was added to fix the eggs. The ratio of nonhatched eggs to total eggs in each group was determined from visual inspection of the micrographs; 8 to 10 nonoverlapping field images displaying on average 12 eggs per field were inspected in each treatment group in each experiment, i.e., 96 to 120 eggs per treatment group were examined. At least three biological replicates of each puromycin dose-response curve experiment were performed for each developmental stage.

Real-time motility assay of S. mansoni adults cultured in the presence of puromycin.

The effects of increasing concentrations of puromycin on the motility of adult stage worms were quantified and evaluated employing the xCELLigence system (ACEA Biosciences, San Diego, CA), designed to monitor cellular events in real time by measuring electrical impedance across interdigitated microelectrodes integrated on the bottom of tissue culture E-plates (https://aceabio.com/) (75). The motility assay was performed as described previously (37–39); male or female worms, as indicated, were placed individually into each well of 8-well E-plates. In brief, the motility of flukes was registered for ∼6 h before adding puromycin at increasing concentrations from 10 μg/ml (18.3 mM) to 1,000 μg/ml (1,840 mM) or HEPES as vehicle control, in 200 μl medium per well and including ≥3 replicates per condition. The culture medium and antibiotic were replaced every 48 h, and determination of 50% lethal concentration (LC₅₀) was undertaken with log (concentration) versus normalized response with variable slope using Graphpad prism 5.0 software (75). In parallel, adult worms were incubated with matched concentrations of puromycin or HEPES in 24-well culture plates for microscopic observation. In some experiments, adult male or female worms were cultured in the presence of HEPES control, 100, 500, or 1,000 μg/ml in triplicate for 4 days, with medium replaced each 48 h, and at day 4 micrographs were taken and videos recorded, with ≥3 replicates per condition.

Rescue of schistosomes expressing puromycin resistance.

Rescue experiments of vesicular stomatitis virus glycoprotein (VSVG)-pseudotyped murine leukemia virus (MLV)-transduced schistosomula were performed according to protocols previously described (22). Briefly, VSVG-MLV particles were produced in GP2-293 cells cotransfected with both VSVG-plasmid and the retroviral construct pLNHXcHS4_puroR, where the original neomycin resistance marker (NeoR) was replaced with the puromycin antibiotic resistance gene, i.e., PuroR (42). Thereafter, the virion particles were concentrated and viral titers determined (26). Two days after cercarial transformation, ∼ 5,000 to 10,000 schistosomula were exposed to virions in 1 ml of medium and 8 μg/ml of Polybrene; a viral titer of 2.4 × 10⁶ CFU (∼10⁹ virions based on quantitative PCR [qPCR] estimation) was employed. Schistosomula exposed to Polybrene alone served as controls. The parasites were incubated in the presence of the virions for 18 h at 37°C, 5% CO₂, washed, and transferred into medium containing 2.5 μg/ml of puromycin. Micrographs of the schistosomes costained with PI and FDA (see above) were collected on days 2, 4, and 6. The parasites were harvested by day 6, washed, and processed for RNA isolation and PuroR expression analysis by qRT-PCR. In brief, cDNAs were synthesized from ∼50 ng of total RNA isolated from MLV-transduced or control schistosomules, using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). The relative expression of PuroR was investigated by qRT-PCR using primers specific for PuroR (forward primer, 5′-TGCAAGAACTCTTCCTCACG-3′, and reverse primer, 5′-CGATCTCGGCGAACACC-3′) and SmGAPDH (GenBank accession M92359) as a reference gene (23). The PCR efficiencies for primer pairs were estimated by titration analysis to be 100% ± 5% (76), and the reactions were performed using the iQ SYBR green Supermix (Bio-Rad), in triplicate in 96-well plates, and a thermal cycler (iCycler; Bio-Rad) fitted with a Bio-Rad iQ5 detector to scan the plates in real time, with denaturation at 95°C for 3 min followed by 40 cycles of 30 s at 95°C and 30 s at 55°C, followed by a melting curve. Relative quantification was assessed by the 2^−ΔΔCT method (77), with the outcome plotted as the normalized fold change of expression of the PuroR gene relative to SmGAPDH (where GAPDH is glyceraldehyde-3-phosphate dehydrogenase), considering the PuroR relative expression level measured in MLV-transduced schistosomules (calibrator sample) as 1.

Statistical analysis.

Levels of statistical significance among treatments were determined using analysis of variance (ANOVA) and Student's t test; P values of ≤0.05 were considered to be significant between two indicated experimental groups. The survival rate of schistosomula or sporocysts in each tissue culture well was calculated as follows:

$$\%\hspace{0.17em}\text{live parasites}=\frac{\text{live parasites}\hspace{0.17em}(\text{FDA fluorescence})}{\text{live parasites}\hspace{0.17em}(\text{FDA fluorescence})+\hspace{0.17em}\text{dead parasites}\hspace{0.17em}(\text{PI fluorescence})}\times 100\%$$

The hatching rate of LEs in each tissue culture well was calculated as follows:

$$\%\hspace{0.17em}\text{egg hatching rate}=\frac{\text{number of hatched eggs}}{\text{total number of eggs}}\times 100\%$$

$$\%\hspace{0.17em}\text{relative hatched eggs}=\frac{\text{average}\hspace{0.17em}\%\hspace{0.17em}\text{egg hatching rate}}{\text{average HEPES negative control}\hspace{0.17em}\%\hspace{0.17em}\text{egg hatching rate}}\times 100\%$$