Molecular and functional characterisation of an extracellular Argonaute protein secreted by a gastrointestinal nematode

Parasites manipulate their hosts to promote infection by secreting bioactive molecules including proteins, lipids and RNAs. Helminths (parasitic worms) secrete a plethora of such molecules possessing immunogenic and immunomodulatory properties and some of these are co-packaged within extracellular vesicles (EVs) and can be directly transferred to host cells. We have found that the mouse-infective gastrointestinal nematode Heligmosomoides bakeri, a close relative of the human hookworm Necator americanus and the animal-infective nematodes Teladorsagia circumcincta and Haemonchus contortus, secretes an RNA-binding protein which belongs to the family of Argonaute (AGO) proteins. AGO proteins are at the heart of RNA interference, a mechanism involved in gene regulation. By associating with small RNA guides, AGO proteins are directed to messenger RNA targets. This interaction usually leads to gene silencing. The AGO protein secreted by H. bakeri, termed exWAGO, is found in two forms: a vesicular form (detected inside EVs) and a non-vesicular form that does not co-purify with EVs. One goal in this thesis is to understand whether and how these two forms of exWAGO are different in terms of their sRNA guides and potential targets. It is known that H. bakeri EVs are internalised by mouse host cells, such that the parasite-derived cargo could directly interact with and interfere with host gene expression. The EVs predominantly contain 5'PPP secondary short interfering RNAs (siRNAs) and EV uptake by mouse cells was shown to result in suppression of host genes involved in immunity and inflammation. As AGO proteins coordinate gene silencing mechanisms, we hypothesise that exWAGO is directly involved in mediating changes in host gene targets. The goal in this thesis is to build understanding on the putative role of exWAGO in mediating cross-species gene silencing by identifying the RNA and protein interaction partners of exWAGO. A further goal is to determine the importance of exWAGO in parasite survival by blocking it through vaccination and testing the consequence to subsequent infections. To identify the small RNAs (sRNAs) that associate with exWAGO we immunopurified exWAGO and generated sRNA libraries from three different sample types: adult worms (to detect the intra-parasite exWAGO), exWAGO from the excretory-secretory products that is found in EVs, and exWAGO from the excretory-secretory products that does not co-purify with EVs. sRNA sequencing analyses show that the intra-parasite exWAGO and the two extracellular forms of exWAGO bind 22-23G 5'PPP secondary siRNAs originating from transposable elements and novel repeats, but are depleted from microRNAs (miRNAs), Y-RNAs, transfer RNAs and ribosomal RNAs. This suggests that exWAGO is bound specifically to secondary siRNAs, and we hypothesise that exWAGO is required for the export of these sequences from the parasite. To study the binding selectivity of exWAGO, we used gel shift assays and found that exWAGO binds 5'PPP guide RNAs with high affinity in contrast to the mammalian mouse AGO2 (mAGO2). Direct comparison of the secondary siRNA sequences bound by the vesicular and non-vesicular exWAGO indicates that the two exWAGO forms have some overlap in which secondary siRNAs they bind, however there are differences in the relative abundance of different siRNAs and there are some siRNAs only found in one exWAGO form. This suggests that the vesicular and non-vesicular forms of exWAGO might target different genes and have different functional properties. Identification of the host transcripts targeted by exWAGO is crucial in understanding the role that exWAGO might play in cross-species gene silencing. To examine this, we developed a method to immunopurify the exWAGO protein in vivo from the gut of H. bakeri-infected mice and sequenced the RNAs with which it associates. Our initial dataset using the mAGO2 as a positive control suggests that we can successfully detect gene targets with this method. The preliminary results indicate that RNAs associated with exWAGO map to intronic regions in the mouse transcriptome, in contrast to mAGO2 where reads map to 3' untranslated regions and coding regions (consistent with location of mAGO2 target sites). Further experiments are required to test if the putative targets of exWAGO identified are true targets. Development of a bioinformatics pipeline to examine this complex dataset would also allow us to test if the method enabled formation of guide-target chimeric reads, hence permitting direct identification of guide sRNA-host target interactions. To understand the mechanism by which exWAGO might mediate gene silencing in host cells, we tested whether exWAGO possesses the ability to cleave "slice" targets in vitro. Our results show that exWAGO does not possess slicer activity, consistent with its lack of a catalytic motif required by other AGOs for slicer activity. To explore what other gene regulation mechanism(s) might be employed by exWAGO inside host cells we set out to identify the protein interactors of exWAGO in vitro and in vivo using liquid chromatography-tandem mass spectrometry. The proteomic analysis identified some putative interactors that have been reported to interact with other known mammalian AGO proteins as well as some proteins with no previous literature linking these to gene silencing. From these data we can now formulate hypotheses that exWAGO might be localised to specific compartments in host cells to perform its function. We also discovered a putative candidate protein that might be involved in the internalisation of exWAGO inside host cells and in host AGO trafficking. Further experiments are required to validate the protein interactions identified. Finally, to understand the importance of exWAGO in infection and test if exWAGO can serve as a vaccine candidate, we immunised mice with recombinant exWAGO protein. Vaccination with recombinant exWAGO protein led to a strong induction of IgG1 antibodies against the protein and resulted in decreased egg and worm burdens (59.0% and 66.7% respectively, calculated as average across three experiments). These data suggest that blocking exWAGO in vivo reduces the ability of the parasite to survive. As exWAGO is highly conserved amongst Clade V gastrointestinal worms, including the human hookworm and sheep parasites, we propose that exWAGO could be a vaccine candidate for these nematodes. In summary, this thesis presents a first insight into the mechanistic basis of exWAGO in terms of what sRNA guides it binds, what its putative target host genes are, and what proteins it interacts with. We also explore the properties and characteristics of the two extracellular forms of exWAGO for the first time. By understanding if and how exWAGO may facilitate communication between parasite sRNAs and host targets, we expand our understanding of how helminths manipulate their hosts. This thesis offers a putative intervention for gastrointestinal worm infections via exWAGO vaccination, but further studies could also lead to development of other therapeutic strategies or molecular tools based on exWAGO.