|Home | About | Journals | Submit | Contact Us | Français|
The intravascular trematode Schistosoma mansoni is a causative agent of schistosomiasis, a disease that constitutes a major health problem globally. In this study we cloned and characterized the schistosome tegumental phosphodiesterase SmNPP-5 and evaluated its role in parasite virulence. SmNPP-5 is a 52.5-kDa protein whose gene is rapidly turned on in the intravascular parasitic life stages, following invasion of the definitive host. Highest expression is found in mated adult males. As revealed by immunofluorescence analysis, SmNPP-5 protein is found prominently in the dorsal surface of the tegument of males. Localization by immuno-electron microscopy illustrates a unique pattern of immunogold-labeled SmNPP-5 within the tegument; some immunogold particles are scattered throughout the tissue, but many are clustered in tight arrays. To determine the importance of the protein for the parasites, RNA interference (RNAi) was employed to knock down expression of the SmNPP-5-encoding gene in schistosomula and adult worms. Both quantitative real-time PCR (qRT-PCR) and Western blotting confirmed successful and robust gene suppression. In addition, the suppression and the ectolocalization of this enzyme in live parasites were evident because of a significantly impaired ability of the suppressed parasites to hydrolyze exogenously added phosphodiesterase substrate p-nitrophenyl 5′-dTMP (p-Nph-5′-TMP). The effects of suppressing expression of the SmNPP-5 gene in vivo were tested by injecting parasites into mice. It was found that, unlike controls, parasites whose SmNPP-5 gene was demonstrably suppressed at the time of host infection were greatly impaired in their ability to establish infection. These results demonstrate that SmNPP-5 is a virulence factor for schistosomes.
Blood-dwelling worms of the genus Schistosoma are the causative agents of schistosomiasis, a tropical disease that is prevalent in Africa, South America, the Arabian Peninsula, China, the Philippines, and Indonesia (14). The disease infects over 200 million people globally, and over 600 million live at risk of infection (34). Schistosomiasis mansoni is characterized by the presence of adult worms within the mesenteric venous plexus of an infected host. The disease is characterized clinically by chronic hepatic and intestinal fibrosis, portal hypertension, and anemia (14).
Adult schistosomes, living as male-female pairs, can survive for many years in the vasculature and appear recalcitrant to elimination by the immune system. The worm surface likely plays an important role in immune evasion, as it is a site of intimate host-parasite interaction. This surface is structurally unique and is multilaminate in appearance (21). In one model of fine tegument morphology, the surface is interpreted as an apical plasma membrane that is covered by a laminate secretion called the membranocalyx (35). Specialized membranous bodies called multilamellar vesicles (MLVs) are synthesized in cell bodies that lie beneath the peripheral muscle but connect to the tegument through thin cytoplasmic connections. These MLVs migrate from the cell bodies to the tegument proper and have been reported to fuse with the tegumental apical plasma membrane to contribute material to the overlying membranocalyx (35).
Using proteomics, the major protein components of the tegumental membranes of Schistosoma mansoni have been identified (6–9). These proteins belong to a variety of classes, including nutrient transporters, receptors, and enzymes and several proteins of unknown function. In these proteomic studies, a putative phosphodiesterase was identified. The protein was not only greatly enriched in the tegument but was available for surface biotinylation of adult worms, highlighting its surface exposure (8). Its presence had been earlier implied by the detection of phosphodiesterase activity in schistosome tegumental extracts (11). Recently, the cDNA encoding this enzyme was cloned and characterized and its sequence was shown to exhibit greatest homology with sequences of members of the nucleotide pyrophosphatase-phosphosdiesterase 5 (NPP-5) family (26).
In this work we show that production of this protein (designated SmNPP-5) is rapidly upregulated at the time of definitive host invasion. We confirm the surface localization of SmNPP-5 and find that it displays a unique localization pattern in the tegument as revealed by immuno-electron microscopy (immuno-EM). We show that suppressing SmNPP-5 gene expression by the use of RNA interference (RNAi) impairs the ability of larval schistosomes to establish infection in vivo, revealing this molecule to be important for parasite virulence.
The Puerto Rican strain of Schistosoma mansoni was used. Schistosomula were prepared from cercariae that were released from infected snails, and those were cultured in Basch medium (lacking red blood cells [rbcs]) at 37°C, in an atmosphere of 5% CO2, as described previously (2). These schistosomula were used in the RNAi work described below. Adult male and female parasites were recovered by perfusion from Swiss Webster mice that had been infected with 125 cercariae 7 weeks previously. Adult worms were maintained in Basch medium (lacking rbcs). Parasite eggs were isolated from infected mouse liver tissue (15). Miracidia were recovered from the eggs and transformed to sporocysts that were cultured for 24 h, as described previously (15). Snails infected with a single miracidium each were obtained from Fred Lewis, Biomedical Research Institute, Rockville, MD. Cercariae emerging from these snails are all of one sex, either male or female. Cercariae from individual snails were used to infect mice; 7 weeks later, worms of a single sex were recovered from the mice by perfusion. These single-sex adult worms were used exclusively in developmental expression analysis and not for any RNAi work.
Proteomic analysis of the S. mansoni tegument revealed a phosphodiesterase homolog (S. mansoni ORESTES database accession number C606073.1) (7). BLAST interrogation of the S. mansoni genome (version 3) with this sequence led to contig 0018606. Using this sequence as a guide, the predicted 5′-most exon was extended in silico to a potential initiator methionine. Next, using oligonucleotides designed from the predicted 5′ untranscribed region (5′UTR) just upstream of this methionine (PDE1f [5′-GTTATCGAAAAGCCAGTCGTAG-3′]) and the 3′UTR downstream of the last predicted exon (PDE3r [5′-TGGCAACAACAATTCATTCATTAG-3′]) together with adult parasite cDNA in a PCR, we amplified and then sequenced the complete SmNPP-5 coding DNA at the Tufts University Core Facility.
NH2-TLKNKGAHGYDPDYK-COOH, a peptide comprising SmNPP-5 amino acid residues 354 to 369, was synthesized by Genemed Synthesis, Inc., San Antonio, TX. A cysteine residue was added at the amino terminus to facilitate conjugation of the peptide to bovine serum albumin (BSA). Approximately 500 μg of the peptide-BSA conjugate mixed with Freund's complete adjuvant was used to immunize two New Zealand White rabbits subcutaneously. The rabbits were given booster treatments of 100 μg of peptide alone in incomplete Freund's adjuvant 20, 40, and 60 days later. Ten days after the last treatment, serum was recovered from both rabbits and pooled, and anti-SmNPP-5 antibodies were subjected to affinity purification using a peptide-ovalbumin conjugate and dialyzed against phosphate-buffered saline (PBS), as previously described (28).
The levels of expression of the SmNPP-5 gene in different life stages of the parasite and in parasites treated with gene-specific small interfering RNAs (siRNAs) were measured by quantitative real-time PCR (qRT-PCR), using a custom TaqMan gene expression system (Applied Biosystems, Carlsbad, CA). First, parasite samples were lysed by the addition of 50 μl of cell disruption buffer (PARIS kit; Ambion, Austin, TX). Samples were homogenized on ice using an RNase-free pestle for ~1 min, and the parasite homogenate was split into two halves. One half was used for isolating RNA and the other for protein analysis. RNA was isolated from the parasite homogenate by the use of the PARIS kit per the manufacturer's guidelines. Residual DNA was removed by DNase digestion using a TurboDNA-free kit (Applied Biosystems). cDNA was synthesized using 1 μg of RNA, an oligo(dT)20 primer, and SuperScript III reverse transcriptase (Invitrogen, CA). The levels of expression of the SmNPP-5 gene at different life stages were measured by qRT-PCR using the housekeeping gene encoding triose phosphate isomerase as the endogenous control (17).
Primer sets and reporter probes labeled with FAM (6-carboxyfluorescein; Applied Biosystems, Carlsbad, CA) were used for qRT-PCR. To detect SmNPP-5 expression, the following primers and probe were used: primers SmPhosphod-F (5′-GGACGATTATTGCTGACAGAACGT-3′) and SmPhosphod-R (5′-TGGAGACATCTCTTTGTAATCTGGATCA-3′) and probe SmPhosphod-M2 (5′-FAM-TTTATTTTTCAGGGTTATCCC-3′). Each qRT-PCR was performed using cDNA equivalent to 10 ng of total parasite RNA (in a final volume of 25 μl) according to the manufacturer's PCR protocol for universal conditions. All samples were processed in triplicate and underwent 40 amplification cycles on a StepOnePlus real-time PCR system instrument. For quantifications, the ΔΔCt method was employed (20) and the gene encoding schistosome tubulin was used as the within-stage, endogenous control (13).
To monitor expression of the SmNPP-5 protein, parasite samples were first homogenized on ice in ice-cold cell disruption buffer and protein content was measured using a BCA protein assay kit (Pierce Biotechnology, Rockford, IL) according to the manufacturer's instructions. Five micrograms of protein from each sample was resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions and blotted using a polyvinylidene difluoride (PVDF) membrane. The membrane was then incubated in a 5% skim milk–PBS solution containing 0.1% Tween 20 (PBST) for 1 h at room temperature. Next, the membrane was probed overnight at 4°C with affinity-purified rabbit anti-SmNPP-5 antibody at 1:200. After the membrane was washed twice in PBST, bound primary antibody was detected using horseradish peroxidase-labeled anti-rabbit IgG (GE Healthcare, NJ) (1:5,000) and a TMB membrane peroxidase system (from Kirkegaard and Perry Laboratories Inc., Gaithersburg, MD), following the instructions of the manufacturers. The membrane was exposed to X-ray film and images were captured using a Kodak Image Station 2000RT system. To monitor protein loading per lane, a duplicate gel was stained with Coomassie brilliant blue.
To measure phosphodiesterase activity in live parasites, ~500 SmNPP-5-suppressed and control schistosomula (processed in triplicate) were incubated in assay buffer (50 mM Tris-HCl buffer [pH 8.9], 120 mM NaCl, 5.0 mM KCl, 60 mM glucose) containing 0.5 mM phosphodiesterase chromogenic substrate p-nitrophenyl 5′-dTMP (p-Nph-5′-TMP) (Sigma-Aldrich, MO), as described previously (26). Changes in optical density at 405 nm were monitored continuously for ~3 h.
To prepare schistosomulum extracts, parasites were harvested at 0, 2, and 4 days after siRNA treatment, washed three times with PBS, and homogenized on ice in ice-cold cell disruption buffer (Paris kit; Ambion) (30 μl). Protein content was measured using the BCA protein assay kit as described above. Schistosomulum extract (3 μg in 100 μl of phosphodiesterase assay buffer) was used in the enzyme assay, following the protocol described above.
Adult worm sections 7 μm thick were obtained using a cryostat and fixed in cold acetone. Immunofluorescent detection of SmNPP-5 was carried out using affinity-purified rabbit anti-SmNPP-5 antibody diluted 1:25 and Alexa Fluor 488-conjugated goat anti-rabbit IgG (Invitrogen), essentially as described earlier (29). Control parasite sections were treated with secondary antibody alone.
Freshly perfused adult parasites were fixed overnight with 2% glutaraldehyde–0.1 M cacodylate buffer at 4°C. The samples were then dehydrated in a graded ethanol series and then infiltrated and embedded in LR White acrylic resin (London Resin Company). Ultramicrotomy was performed using a Leica Ultracut R ultramicrotome, and the sections were collected on gold grids. Grids were immunolabeled in a two-step process according to the following procedure. The grids were conditioned in PBS three times for 5 min each time at room temperature, followed by blocking of nonspecific labeling for 30 min at room temperature using 5% nonfat dry milk–PBS. After rinsing, the grids were exposed to primary antibody diluted 1:30 for 1 h at room temperature followed by washing in PBS, then incubated with secondary antibody diluted 1:30 (10-nm-diameter-gold-particle-labeled goat anti-rabbit IgG [H&L; GE Healthcare]) for 1 h at room temperature, and finally rinsed thoroughly in water. Control parasite preparations were treated with secondary antibody alone. Grids were exposed to osmium vapor and/or lightly stained with lead citrate to improve contrast and were examined and photographed using a Philips CM 10 electron microscope at 80 kV.
Adult worms were treated either with a synthetic siRNA targeting SmNPP-5 (SmNPP-5 siRNA1 [5′-TTGATGGATTTCGTTATGATTACTTTG-3′]) or with control siRNAs. Control siRNAs were used in two forms. The first siRNA control type targets unrelated schistosome genes by the use of schistosome siRNA control 1 (5′-AAACAGCATACACCAATTTATTTGGCT-3′), which targets the SmATPdase 1 gene (GenBank accession number AY323529), and schistosome siRNA control 2 (5′-AAGAAATCAGCAGATGAGAGATTTAAT-3′), which targets the SmAP gene (GenBank accession number EU040139). The second siRNA control type targets no sequence in the schistosome genome (5′-CTTCCTCTCTTTCTCTCCCTTGTGA-3′). Delivery of siRNAs to the parasites was performed by electroporation as described previously, using 10 μg of each siRNA (18, 22). Gene suppression was assessed posttreatment as described above by comparing mRNA levels (using qRT-PCR) and protein levels (by Western blotting and enzyme activity measurements) in target versus control groups.
One-day-old cultured schistosomula were electroporated with either SmNPP-5, or control or no siRNA. Some parasites were then used to infect female Swiss Webster mice by adding ~1,000 schistosomula to 100 μl of RPMI medium without phenol red and injecting the inoculum into the thigh muscle of the animals using a 1-ml tuberculin syringe and a 25-gauge, 1-in needle (19). The remaining parasites were cultured in vitro to allow an assessment of gene suppression levels at different time points. Three different infection protocols were followed. In protocol 1, parasites were used to infect mice immediately after RNAi treatment (day 0) and worms were recovered by vascular perfusion 28 days later. In protocol 2, parasites were maintained in culture for 2 days prior to being washed and counted and used to infect mice. Those worms were recovered 14 days after infection. In protocol 3, parasites were maintained in culture for 4 days before being washed and counted and used to infect mice. These worms were recovered 28 days after infection. All recovered worms were counted and examined under a light microscope; subsequently, their SmNPP-5 gene expression levels were determined by qRT-PCR, as described earlier. To compare parasite sizes, images of the worms were captured using an inverted microscope (TH4-100; Olympus, Tokyo, Japan) equipped with a Retiga 1300 camera (Q Imaging, Surrey, British Columbia, Canada), and the area occupied by each individual parasite was measured using ImageJ 1.41 software (U.S. National Institutes of Health, Bethesda, MD).
For data generated by qRT-PCR and for the determinations of enzyme activity in worm extracts, one-way analysis of variance (ANOVA) and Tukey's honestly significant difference test were used for post hoc analysis. To analyze the live worm enzyme activity data, two-way repeated-measurement ANOVA was used. To assess worm recovery and worm size data, one-way ANOVA, Student's t test, or the Mann-Whitney test was used, as appropriate. In all cases, P values < 0.05 were considered significant.
The GenBank accession number for SmNPP-5 is EU769293.
The cloned SmNPP-5 cDNA potentially encodes a 458-amino-acid protein with a molecular weight (MW) of 52,563 and a pI of 6.28. This sequence is similar to that of a cDNA independently cloned by Rofatto et al. (26) but extends the 5′ end by 9 amino acids. This encompasses the 27-amino-acid, N-terminal potential signal peptide sequence 1MYCIETMQKMIILLLICFFPYIERIYA27 (per SignalP 3.0 software [http://www.cbs.dtu.dk/services/SignalP-3.0/]). The sequence identity of SmNPP-5 to animal homologs (e.g., humans [ENPP5, GenBank accession no. AAQ88878] and in the sea anemone Nematostella vectenis [GenBank accession no. XP001641023]) is 29 to 35%; the sequence identity is somewhat lower (24%) to a Saccharomyces cerevisiae enzyme (GenBank accession no. NP009955). In SmNPP-5, the proposed catalytic site 84TLTFPSH90 is conserved, as are amino acids reported to be important in metal binding (D45, D207, H211, D255, H256, and H362). The protein has a single predicted transmembrane domain at the carboxyl terminus (436LSIIFIIKFIILSIFMV452). The 355TLKNKGAHGYDPDYK369 peptide was used to generate anti-SmNPP-5 antibodies.
The developmental expression of SmNPP-5 was examined in several schistosome life stages by qRT-PCR, and the results are shown in Fig. 1A. Of the various life stages tested, the relative gene expression of SmNPP-5 was negligible in eggs, cercariae, and sporocysts. In essence, the gene appears to be turned on following invasion of the definitive host, with relatively high expression in the intramammalian life stages (schistosomula and adults) and highest expression in the adult male parasites. Females from single-sex infections (Fig. 1A, lane F′) expressed larger amounts of SmNPP-5 than females from a mixed infection (Fig. 1A, lane F).
Concurrent findings were observed at the protein level, as determined by Western blotting and shown in Fig. 1B. SmNPP-5 was detected at about its predicted size (~55 kDa) (arrow, Fig. 1B) in extracts of 14-day cultures of schistosomula (Som14) and all adult male and female worms (from both mixed and single-sex infections). SmNPP-5 protein was not detected in egg and cercarial (Cer) extracts, and levels were barely detectable in freshly transformed schistosomula (Som0). The strip of a Coomassie brilliant blue-stained duplicate gel (Fig. 1B, bottom panel) demonstrates that protein was present in all lanes.
Figure 2A shows the immunolocalization pattern of SmNPP-5 in a section of an adult schistosome pair. It is clear that the protein is prominently expressed in the tegument of males. The most intense staining is seen in the dorsal surface (arrow, Fig. 2A). Localization of SmNPP-5 by immunogold electron microscopy (Fig. 2B and C) confirmed that the protein is distributed on the host interactive tegumental membrane. Two distinctive patterns of immunogold particle localization within the tegument—scattered and clustered—are discernible. Scattered individual immunogold particles are seen distributed widely throughout the section. In addition, clusters of from 5 to 20 or more immunogold particles are also widely apparent (arrowheads, Fig. 2B and C). The inset shows a cluster at higher magnification (white arrow, Fig. 2C). The top left inset (Fig. 2C, black arrow) is interpreted to represent a cluster dispersed at the parasite surface. Parasites treated with secondary antibody alone demonstrated no tissue staining (data not shown).
SmNPP-5 gene expression was targeted for suppression in adult parasites in vitro by introduction of a specific siRNA via electroporation. Figure 3A shows the robust (>95%) suppression of SmNPP-5, measured by qRT-PCR, 8 days after treatment. Western blotting demonstrated that the siRNA treatment also resulted in a substantial diminution in SmNPP-5 protein production (Fig. 3B). Note that, in this instance, SmNPP-5 resolves as two bands: a dominant lower band of about the expected size of SmNPP-5 (~55 kDa) (arrow, Fig. 3B) and an upper, fainter band of lower electrophoretic mobility (arrowhead, Fig. 3B). That fainter band likely represents a posttranslationally modified (perhaps gylcosylated) SmNPP-5 variant. Notably, lower levels of both forms of the SmNPP-5 protein were detected in extracts from SmNPP-5 siRNA-treated parasites (left lane, Fig. 3B) versus controls (middle and right lanes, Fig. 3B) at 8 days posttreatment. The lower panel in Fig. 3B shows a fragment of a Coomassie brilliant blue-stained polyacrylamide gel, distant from the location of SmNPP-5, to illustrate that all lanes contained roughly equivalent amounts of parasite protein. The robust suppression of SmNPP-5 did not result in any detectable change in schistosomulum or adult parasite morphology or behavior. However, live schistosomula whose SmNPP-5 expression was suppressed by RNAi treatment, unlike controls (treated with the irrelevant control siRNA or left untreated), had a significantly diminished ability to cleave the exogenously added, synthetic phosphodiesterase substrate p-Nph-5′-TMP (P < 0.05) (Fig. 3C).
To investigate whether RNAi-mediated gene silencing of SmNPP-5 had any impact on the parasites in vivo, we infected mice with 1-day-old SmNPP-5-suppressed or control schistosomula. Using protocol 1, mice were infected at ~1 h after RNAi treatment. At 28 days postinfection, worm burdens were compared across the groups; the data are shown in Fig. 4A. In this experiment, no significant differences in worm burden were found in the SmNPP-5-suppressed group compared to either control group (Fig. 4A). Furthermore, no phenotypic differences between the worms from each group were observed. SmNPP-5 gene expression analysis was undertaken using the worms recovered from the infected mice and the parasites that had been maintained in vitro for 7 or 28 days. SmNPP-5-suppressed parasites cultured for 7 days exhibited close to 100% suppression of SmNPP-5 (Fig. 4B, left panel). Even after 28 days in culture, the mRNA levels in the SmNPP-5 siRNA-treated worms were still markedly (~90%) suppressed (Fig. 4B, middle panel). In contrast, parasites recovered from the vertebrate host were no longer suppressed; SmNPP-5 transcript levels had returned to control levels (Fig. 4B, right panel).
Using an alternative RNAi treatment and infection protocol (protocol 2), schistosomula were maintained in culture for 2 days after RNAi treatment before we introduced them into mice. Those mice were perfused 14 days later to recover any worms present. A total of 125 parasites were recovered from the control group versus 72 parasites from the SmNPP-5 knockdown group. Since the numbers of individual worms per mouse were not counted, no statistical analysis can be performed on those data. However, the sizes of the parasites from both groups of mice were compared (Fig. 5A). The mean size of the parasites from the SmNPP-5-suppressed group was significantly less than that of parasites from the control group (P < 0.05). SmNPP-5 gene expression was compared in the worms recovered from the infected mice versus parasites that had been maintained in vitro for the same time period (14 days). As Fig. 5B shows, parasites maintained for 14 days in vitro exhibited almost 100% target gene suppression (Fig. 5B, left pair of bars) whereas SmNPP-5-suppressed parasites recovered from the infected mice now exhibited ~75% suppression (Fig. 5B, right pair of bars). This amount is still substantial compared to controls but is significantly less than that seen in the parasites maintained in culture for the entire experiment (P < 0.05).
Using a final RNAi treatment and infection method (protocol 3), schistosomula were cultured for longer (4 days) after RNAi treatment before introducing them into mice, which were perfused 28 days later. Data from two experiments performed using this protocol yielded essentially the same results. As shown in Fig. 6A and B, using protocol 3, there was a significant reduction in worm burden in the SmNPP-5-suppressed group compared to the burden seen with a control group treated with an irrelevant siRNA (Fig. 6A; 1 ± 0.8 worms versus 20 ± 6 worms [P < 0.05]) or to that seen with a control group that had received no siRNA treatment (Fig. 6B; 0.4 ± 0.2 worms versus 6 ± 3 worms [P < 0.05]). We detected no significant difference in the mean size of the small number of SmNPP-5 siRNA-treated parasites recovered from mice versus the mean size of control parasites. Next, the expression levels of SmNPP-5 in the recovered worms were compared. As noted above, the SmNPP-5-suppressed parasites that had been cultured for 28 days still exhibited profound (>95%) gene suppression (Fig. 6C, left pair of bars). In contrast, the small number of parasites recovered from infected mice after 28 days were no longer suppressed; SmNPP-5 transcript levels had effectively returned to control levels (Fig. 6C, right pair of bars).
In final experiments, we sought to determine whether the 3 different RNAi protocols used in this work had any bearing on the level of target enzyme activity at the time of infection. Therefore, phosphodiesterase activity assays were carried out using parasite extracts generated at 0, 2, and 4 days post-RNAi treatment; these data are shown in Fig. 7. It is clear that activity increased in the control parasite group (gray bars) after day 0, with day 2 and 4 parasites exhibiting approximately twice the activity of their day 0 counterparts. This finding corroborates the Western blot studies described above (Fig. 1B), which demonstrated SmNPP-5 expression increasing as cercariae transformed and schistosomula matured. Figure 7 (left panel) demonstrates that enzyme activity in the day 0 suppressed-parasite extract did not differ significantly from the corresponding control value. In contrast, at day 2 (Fig. 7, middle panel), the controls exhibited about twice the activity of their suppressed counterparts (P < 0.05). It is clear that SmNPP-5 gene suppression prevented the surge in activity seen in the control day 2 schistosomula. In a similar vein, extracts of day 4 SmNPP-5-suppressed schistosomula exhibited low enzyme activity compared to the day 4 controls (P < 0.05) (Fig. 7, right panel). The suppressed day 4 worms exhibit only ~30% of the enzyme activity of controls.
Schistosomes are globally successful intravascular parasites. Our laboratory seeks to understand how the molecules expressed at the host-interactive surface contribute to their success (27). Among the enzyme activities detected at the schistosome surface is that of phosphodiesterase, which was first described in a study of the schistosome tegument published over 30 years ago (11). A cDNA encoding such an enzyme was recently cloned and designated SmNPP-5 (26). We have independently cloned this cDNA, and here we confirm and extend previous observations. Sequence analysis catalogs SmNPP-5 as a type 1 transmembrane protein; it is predicted to be a single-pass transmembrane protein, with its N terminus, and the majority of the protein, being external to the cell.
Expression of the SmNPP-5 gene accompanies cercarial transformation, which coincides with vertebrate host invasion. This suggests that the protein performs a function for the intravascular worms. Maximal SmNPP-5 expression is seen in the mature adults—particularly males from a mixed-sex infection. Upon comparing SmNPP-5 gene expression results determined for adult parasites from single-sex infections, we found that females from a single-sex infection expressed ~3 times more SmNPP-5 than females from a mixed infection. Mated females reduced their expression of SmNPP-5, whereas males increased theirs. This suggests that the male “takes over” this function upon mating, perhaps permitting the mated female to divert the resources necessary to express SmNPP-5 to other tasks, notably expression of egg-laying genes. The sexual dimorphism of schistosome adults suggests that adult males and females have separate and distinctive functions in vivo, a major role for males being, e.g., to transport the female, and for females (lying in the male's gynecophoric canal) being, e.g., to produce and release eggs (3). However, the precise function of SmNPP-5 is not known and the advantage to the schistosome couple for the male to begin to monopolize this activity is unclear.
In adult schistosome sections, the tegument stains strongly with anti-SmNPP-5 antibodies, demonstrating that the protein localizes there, including at the host-parasite interface. This tegumental localization of SmNPP-5 has been reported by other research groups using proteomic approaches (7, 33) and immunofluorescence localization (26); also, as noted earlier, phosphodiesterese activity in tegument-enriched fractions of adult worms was previously reported (10, 11, 23, 25). The localization of SmNPP-5 as determined using immunogold electron microscopy and reported here confirms and extends those data. We noted two patterns of immunogold particle localization—clustered and scattered—in the tegument. While scattered immunogold particles have been seen in other immunolocalization experiments involving tegumental molecules (4, 16, 17), the report of the clustered pattern presented here is, to our knowledge, unique. We speculate that the clustered particles overlay membranous bodies within the tegument called multilaminate vesicles (MLV). These structures are difficult to visualize in sections chemically prepared for immuno-EM analysis. MLVs have been proposed to play an important role in surface membrane formation and turnover (27). In one model of tegument morphology, the surface (multilaminate in appearance) is interpreted as an apical plasma membrane that is covered by a laminate secretion called the membranocalyx (35). MLVs have been reported to fuse with the apical plasma membrane to contribute material to the overlying membranocalyx (35). If, as we propose, SmNPP-5 enzymes are clustered within MLVs, much of the enzyme would be delivered into the membranocalyx to participate intimately in host-parasite interactions. Immunogold labeling is certainly apparent at the host-parasite interface, and SmNPP-5 can clearly access exogenous substrate. Suppressing expression of SmNPP-5 by the use of RNAi greatly impedes the ability of live worms to cleave a synthetic phosphodiesterase substrate added to the medium. However, the idea of placing SmNPP-5 in the membranocalyx is at odds with recent work involving the identification of schistosome tegumental proteins that can be removed from live parasites by external application of the protease trypsin (9). In that work, a number of tegumental proteins, perhaps located in the membranocalyx, are accessed by trypsin, but not SmNPP-5 (9). Thus, the definitive localization of SmNPP-5 within the tegumental membranes requires further experimental analysis.
To investigate the importance of SmNPP-5 for schistosomes, gene expression was suppressed using RNAi. Introducing siRNAs targeting SmNPP-5 by electroporation resulted in very robust (>95%, at the RNA level) gene suppression in adult parasites and schistosomula. Robust suppression at the protein level, as determined by Western blotting and by comparative enzyme activity assays, was also observed. Despite this finding, suppressing SmNPP-5 gene expression did not cause any visible morphological or behavioral change in the SmNPP-5-suppressed versus control parasites even after prolonged culture for up to 28 days. This was not because the RNAi effect wore off in the cultured parasites; even after 28 days in vitro, SmNPP-5 gene expression levels were >95% lower than the control value. This suggests that SmNPP-5 does not play an important role for schistosomes in vitro.
In order to test the effects of knocking down expression of the SmNPP-5 gene in vivo, we first followed a protocol in which schistosomula were used to infect mice immediately after siRNA treatment (12, 19, 24, 32). Using this method (which we designate “protocol 1”), no differences in worm burden in the test versus control groups were found after 28 days and no obvious morphological differences between the recovered worms from the different groups were seen. Furthermore, the levels of expression of the SmNPP-5 gene in the worms recovered from infected mice were all similarly high. In other words, the SmNPP-5-suppressed parasites overcame the suppression in vivo, in contrast to their counterparts that were maintained in culture. We speculate that the rapid growth and metabolic vigor of schistosomes in an infected animal, compared with the generally observed stunting of the parasites maintained in culture, may contribute to the ability of the parasites within mice to overcome gene suppression. In support of this notion, it was previously reported that rapid cell multiplication can quickly dilute an RNAi effect (1). Since suppressing SmNPP-5 had no impact on schistosomes introduced into mice, this may mean that (as is the case in vitro) the gene plays no essential role for the worms in vivo. However, we reasoned that, as a result of the use of protocol 1, the parasites may have been introduced into the mice before gene suppression was well established and that this allowed the worms that were growing in vivo to overcome any incipient RNAi effect. We inferred, therefore, that maintaining the suppressed parasites in vitro for several days would allow mRNA and protein levels to decline and permit us to better gauge the real importance of SmNPP-5.
Using a new procedure (protocol 2), RNAi-treated parasites were first cultured for 2 days after treatment before they were introduced into mice. The mice were perfused 14 days later. In contrast to protocol 1, this protocol led to a clear and demonstrable gene suppression effect on the recovered parasites. The test group showed a reduced total number of parasites compared to the control group, and worms that were recovered from the test group were generally smaller in size than those from the control group. Thus, impeding SmNPP-5 function in this case impaired parasite development. In addition, expression of the SmNPP-5 gene in worms recovered from the test group remained low (~75% lower than the expression level seen with control parasites). Unlike the SmNPP-5-suppressed worms recovered following implementation of protocol 1 as described above, here the SmNPP-5-suppressed worms had not effectively overcome the RNAi effect at the 14-day time point.
Our final gene suppression and infection method (protocol 3) differed from the earlier version in that parasites were cultured for longer (4 days) prior to infection and were recovered from mice 28 days later. Using this protocol, we saw the most drastic differences between the different groups in worm recovery results. Very few parasites were recovered from the SmNPP-5-suppressed group (a mean of ~10-fold fewer than were recovered from the control groups). In other words, when protocol 3 was implemented, SmNPP-5-suppressed schistosomula were significantly impaired in their ability to establish infection. This outcome demonstrates that SmNPP-5 is an important virulence determinant for schistosomes.
To test the hypothesis underpinning these infection experiments, namely, that retaining parasites in culture after RNAi treatment significantly diminishes SmNPP-5 enzyme activity compared to that seen with controls, we set out to establish the time dependency of suppression in schistosomula. Studies were undertaken to determine the levels of phosphodiesterase substrate cleavage activity in extracts of control and test parasites at day 0, day 2, and day 4 after siRNA exposure. The results validate the hypothesis. As expected, there were no significant differences in enzyme activity in extracts of control or RNAi-treated parasites that were tested immediately after siRNA exposure. In contrast, at day 2 (and more so at day 4), the differences in enzyme activity between suppressed and control parasites were stark. Therefore, when suppressed parasites that have been maintained in culture are later injected into mice, they display substantially less enzyme activity than controls. We argue that this enzymatic impairment is responsible for the decreased ability of the suppressed worms to establish a vigorous infection.
As noted earlier, following implementation of protocol 3, the level of SmNPP-5 suppression remained substantial in cultured parasites even after a month in vitro. In contrast, in equivalent parasites recovered from infected mice 4 weeks after RNAi treatment, the SmNPP-5 gene was no longer suppressed. The few parasites that survived in vivo had SmNPP-5 mRNA levels that had returned to control levels. A possible explanation for this observation is that RNAi is variably effective in different parasites and/or that different individuals in the treated parasite population received different amounts of siRNA. Those in which SmNPP-5 knockdown was least effective, or that received less double-stranded RNA (dsRNA), would have survived because expression of their SmNPP-5 gene was minimally impaired. Another possibility is that some worms are more metabolically robust in vivo and that this leads to a shorter half-life of the dsRNA and/or its downstream effectors.
What is the function of SmNPP-5, an essential gene product for infecting schistosomes? We proposed earlier that tegumental phosphodiesterase may participate with other schistosome tegumental phosphatase homologs in the catabolism of extracellular nucleotides (5). However, sequence analysis of this tegumental enzyme shows that it clearly belongs to the pyrophosphatase/phosphosdiesterase-5 (NPP-5) family (26), and members of this family have not been reported to metabolize nucleotides (30, 31). In agreement with this sequence analysis, in our unpublished work, we found that SmNPP-5-suppressed parasites, though impaired in their ability to cleave the synthetic substrate p-nitrophenyl 5′-dTMP, are not impaired in their ability to degrade exogenous ATP, ADP, or AMP (data not shown). No natural substrate of any member of the NPP-5 family has been identified to date (30, 31); thus, the function of SmNPP-5 for schistosomes has not yet been determined.
Only by maintaining the suppressed parasites in vitro prior to using the parasites to infect mice was the importance of SmNPP-5 made clear. However, we and other groups have shown an impact of gene suppression in schistosomes by introducing parasites immediately after RNAi treatment (i.e., using protocol 1) (12, 19, 24). This apparent discrepancy may be due to differences in the ability of parasites to recover from suppression of different gene targets and/or due to the higher susceptibility of schistosomula to a diminution of a particular target gene product during the initial phase of infection.
In summary, we report here the characterization of the S. mansoni tegumental enzyme SmNPP-5. The SmNPP-5 gene is rapidly upregulated following host invasion. The protein exhibits an intriguing clustered distribution in the tegument as revealed by immuno-EM localization. Suppressing SmNPP-5 gene expression impairs the ability of living schistosomes to cleave exogenous phosphodiesterase substrate but exerts no overt morphological effect on the worms. This illustrates that SmNPP-5 is not essential for schistosomes in culture. In contrast, parasites whose SmNPP-5 gene is demonstrably suppressed at the time of host infection are greatly impaired in their ability to establish infection. This demonstrates that SmNPP-5 is a virulence factor for schistosomes. While almost certainly involved in some intimate aspect of host parasite relations, the natural substrate of this surface enzyme and its precise molecular function await discovery.
This work was supported by grant AI-056273 from the National Institutes of Health—National Institute of Allergy and Infectious Diseases (NIH-NIAID).
Schistosome-infected snails were provided by the Biomedical Research Institute through NIH-NIAID contract HHSN272201000009I. We thank Chuck Shoemaker for helpful discussions, Phyllis Mann for assistance with statistical analysis, and John Nunneri for help with electron microscopy.
Published ahead of print on 8 August 2011.