Gene silencing in Cryptosporidium: A rapid approach to identify novel targets for drug development

Background Cryptosporidiosis is a major cause of diarrheal disease. However, the only drug approved for cryptosporidiosis does not work well in high risk populations. Therefore, novel drugs are urgently needed. Then, the identification of novel is necessary to develop new therapies against this parasite. Recently, we have developed a rapid method to block gene expression in Cryptosporidium by using pre-assembled complexes of Cryptosporidium antisense RNA and human protein with slicer activity (Argonaute 2). We hypothesized that structural proteins, proteases, enzymes nucleotide synthesis and transcription factors are essential for parasite development, thus in this work we knock down expression of 4 selected genes: Actin, Apicomplexan DNA-binding protein (AP2), Rhomboid protein 1 (Rom 1) and nucleoside diphosphate kinase (NDK) and elucidated its role during invasion, proliferation and egress of Cryptosporidium. Methods We used protein transfection reagents (PTR) to introduce pre-assembled complexes of antisense RNA and human Argonaute 2 into Cryptosporidium parvum oocysts, the complexes blocked expression of Actin (Act), Transcription factor AP2 (AP2), nucleoside diphosphate kinase (DKN), and rhomboid protein 1 (Rom1). After gene silencing, we evaluated parasite reduction using In vitro models of excystation, invasion, proliferation and egress. We evaluated the potency of ellagic acid, a nucleoside diphosphate kinase inhibitor for anti-cryptosporidial activity using a model of in vitro infection with human HCT-8 cells. Results Silencing of Act, AP2, NDK and Rom1 reduce significantly invasion, proliferation and egress of Cryptosporidium. We showed that silencing of NDK markedly inhibited Cryptosporidium proliferation. This was confirmed by demonstration that ellagic acid reduced the number of parasites at micro molar concentrations (EC 50 =15-30 µM) without showing any toxic effect on human cells. Conclusions Overall the results confirmed the usefulness RNA silencing can be used to identify novel targets for drug development against Cryptosporidium. We identified ellagic acid (EA), a nucleoside diphosphate kinase inhibitor also blocks Cryptosporidium proliferation. Since EA is a dietary supplement approved for human use, then this compound should be studied as a potential treatment for cryptosporidiosis. Author summary The World Health Organization reports diarrhea kills around 760,000 children under five every year. Cryptosporidium infection is a leading cause of diarrhea morbidity and mortality. Current therapies to treat this infection are suboptimal, therefore novel treatments are urgently needed. We used genetic tools to identify novel targets for drug development, thus in this work we evaluated the role of 4 genes during Cryptosporidium infection. We demonstrated that silencing of nucleoside-diphosphate kinase (NDK) drastically reduced invasion, proliferation and egress of this parasite. To validate these finding we used the Ellagic acid (EA) an inhibitor of NDK to treat infected intestinal cells. Our results confirmed that the EA blocks parasite proliferation on infected cells. Interestingly we observed that the ellagic acid also has anti cryptosporidial activity by inducing apoptosis. Since EA is a dietary supplement already approved for human use, then this compound has potential to be used as a rapid alternative to treat Cryptosporidiosis.


109
For transfection experiments, we used oocysts (Iowa isolate) purchased from University of 110 Arizona (Sterling Laboratories, Tucson, AZ). First the oocyst were prepared for transfection: the 111 oocysts (1x10 6 for each target) were transferred to 1.5 ml tubes, the samples were diluted with  (Table S2).

235
Silenced targets are not involved on excystation of Cryptosporidium parasites. We 236 evaluated the role of silenced genes during excystation of Cryptosporidium sporozoites 237 measuring the excystation rate by fluorescence (Fig 1A). Silencing did not produce a significant 238 reduction on excystation rate for any of the targets (Fig. 1B).

240
Gene silencing of selected genes blocks parasite entry.

241
We evaluated gene silencing by flow cytometry using an invasion model (Fig 2A), for these 242 experiments we measured the proportion of CSFE-labeled parasites that failed to invade HCT-8 243 cells (Fig 2A). First, we defined sporozoites population by flow cytometry (Fig S1A). gated cells, meaning that sporozoites did not invaded the host cells ( Fig 2B). Silencing of NDK,

248
Ap2 and Actin also showed a partially effect on sporozoite invasion ( Fig 2B). We did not 249 observed differences between untreated parasites and transfected parasites with PTR (data not 250 shown).

NDK inhibition reduces parasite proliferation.
To evaluate the effects of gene silencing on 253 parasite division we used a proliferation model (Fig 3), parasites were labeled with CFSE and 254 collected at 16 hours post-infection prior to the time of egress (Fig 3). For controls, cell 255 proliferation led to decreased CFSE signal, such that 89% of cells had deceased signal by 16 256 hours ( Fig 3A). By contrast, after silencing of NKD, Ap2 or Actin, only 28-35% of cells had decreased CFSE signal ( Figure 3B). Silencing of CP23 and Rom1 had intermediate values,

258
suggesting partial inhibition of proliferation (54-55%). In additional studies we tested the effect of 259 EA acid on sporozoites viability, we did not observe killing effect of EA on sporozoites (Fig S1B).

261
Silencing of Rom1 and AP2 reduced parasite egress. To test the effects of silencing on 262 egress, we transfected intracellular parasites on infected HCT-8 cells. Transfected complexes 263 did not affect the viability of HCT-8 cells (Fig S2), however we observed a reduction on 264 expression in all tested targets (Fig S3). After confirmation of silencing, fresh media was added 265 to collect merozoites released between 16-19 hrs (Fig S4). To evaluate the egress, we 266 conducted qRT-PCR to quantify the relative number of merozoites in supernatants of treated 267 samples and untreated samples (Fig 4B). There was a significant reduction in the number of 268 merozoites observed in the supernatant of silenced samples (Fig 4B). Therefore, these results

269
indicated that silencing of DKN and Actin reduced both parasite proliferation and egress. By

295
After initial analysis, we identified 100 potential candidates (Fig S1). We developed antisense 296 ssRNA sequences to silence selected genes. In these experiments silencing rates were among 297 30-94% (Data not shown). Since partial silencing (30-75%) may not be optimal for phenotypic 298 studies, then in this work we only used antisense ssRNA that induces >75%. Selected genes for 299 silencing included: 1) Actin, an structural essential for Cryptosporidium motility 7 , 2) NDK an  was partially affected by parasite proliferation (Fig 4B). Silencing of NDK, AP2 and Actin 335 blocked proliferation leading to a reduction in the number of merozoites (Fig 3B). In contrast, 336 silencing with Rom 1 had an even great effect on egress. Since that protein only had moderate 337 effect in proliferation (Fig 3B), it is likely Rom1 is implicated in parasite egress through a 338 proteolytic mechanism. In Plasmodium, release of merozoites from schizonts resulted in the    Table S2. Cryptosporidium antisense ssRNA sequences used in this study. Table S3. Target genes and primer sequences used for RT-PCR analysis.