The moving junction protein RON8 facilitates firm attachment and host cell invasion in Toxoplasma gondii

Abstract

The apicomplexan moving junction (MJ) is a highly conserved structure formed during host cell entry that anchors the invading parasite to the host cell and serves as a molecular sieve of host membrane proteins that protects the parasitophorous vacuole from host lysosomal destruction. While recent work in Toxoplasma and Plasmodium has reinforced the composition of the MJ as an important association of rhoptry neck proteins (RONs) with micronemal AMA1, little is known of the precise role of RONs in the junction or how they are targeted to the neck subcompartment. We report the first functional analysis of a MJ/RON protein by disrupting RON8 in T. gondii. Parasites lacking RON8 are severely impaired in both attachment and invasion, indicating that RON8 enables the parasite to establish a firm clasp on the host cell and commit to invasion. The remaining junction components frequently drag in trails behind invading knockout parasites and illustrate a malformed complex without RON8. Complementation of Δron8 parasites restores invasion and reveals a processing event at the RON8 C-terminus. Replacement of an N-terminal region of RON8 with a mCherry reporter separates regions within RON8 that are necessary for rhoptry targeting and complex formation from those required for function during invasion. Finally, the invasion defects in Δron8 parasites seen in vitro translate to radically impaired virulence in infected mice, promoting a model in which RON8 has a crucial and unprecedented task in committing Toxoplasma to host cell entry.

Figure 1. RON8 is the only RON/MJ protein to be disrupted in the Apicomplexa.

A) Schematic of the two-step targeting strategy used to ablate RON8 from Δku80Δhpt parasites. Homologous recombination at the RON8 locus replaced the portion of the gene encoding residues 1-1716 with selectable marker HPT and concurrently removed the negative screening marker GFP. This generated the intermediate strain Δron8 (1-1716, +HPT). To remove the remaining RON8 genomic sequence and the HPT cassette, a 2^nd targeting construct without HPT and a 3′ flank downstream of the RON8 stop codon was homologously inserted into the Δron8 (1-1716, +HPT) genome, producing Δron8 parasites. PCR1 amplifies sequence inclusive of C-terminal exons, while PCR2 amplifies sequence between the regions upstream and downstream of the RON8 locus. (B) PCR1 confirms the loss of RON8 coding sequence from the Δron8 strain, and PCR2 establishes the bridging of RON8 flanking regions through removal of RON8. C) IFAs of wildtype (Δku80Δhpt parental) and Δron8 intracellular parasites show the absence of RON8 staining in the rhoptry necks of knockout parasites. RON4 is used for colocalization to the rhoptry necks for each strain. D) Western analysis of lysates made from wildtype and Δron8 parasites demonstrates the loss of RON8 expression in the knockout strain, where RON2 serves as a loading control. An ∼230 kDa major RON8 breakdown product previously seen is also visible. E) The moving junction can still be detected in invading Δron8 parasites, as seen by RON4 staining (white arrow). The cartoon illustrates the direction of invasion for this Δron8 parasite (black arrow).

Figure 2. Complementation of RON8 by targeting a RON8 expression cassette to the KU80 locus.

A) Diagram of the complementation construct used to generate R8c parasites. The HA-epitope-tagged RON8 coding sequence is driven by ∼1.8 kb of RON8 promoter sequence (dark green) and GRA2 3'UTR (not shown) with the downstream selectable marker HPT. The construct was targeted to the ablated KU80 locus by homologous recombination using KU80 flanking regions (blue). B) R8c parasites demonstrate restored RON8 expression in the rhoptry necks of intracellular parasites (as shown by colocalization with RON4 by IFA, top panel) and restored RON8 traffic to the moving junction during invasion (arrowhead, bottom panel). C) Western analysis of wildtype, Δron8, and R8c parasite lysates indicates slightly greater levels of RON8 expression in complemented parasites than the wildtype strain; equal lysate loads are shown by the ROP13 control (bottom panel). D) Immunofluorescence showing that the C-terminal HA tag of R8c parasites is often not detected in intracellular parasites (top panel). In parasites where HA is detected, RON8 colocalization shows two more posterior spots that are suggestive of pro-rhoptries (arrows, bottom panel). E) Detection of the C-terminal HA tag in the pro-rhoptries of R8c parasites is confirmed by IFA colocalization (arrows) by colocalization with anti pro-ROP4 antibodies. F) Comparison of the size of RON8 (detecting primarily the mature form) and HA (detecting the unprocessed form in the pro-rhoptries) via Western blot reveals no detectable shift in size between the unprocessed form and the mature form of the protein. This indicates that the C-terminal cleavage event does not remove a large portion of this region of the protein.

Figure 3. Parasites lacking RON8 are deficient in invasion likely through increased detachment from host cells.

A) Quantification of invasion using red/green assays demonstrates a substantial invasion defect for Δron8 parasites that is rescued upon complementation. Green bars represent internal/penetrated parasites, while red bars depict attached/extracellular parasites for wildtype, Δron8, or R8c strains allowed to invade fibroblast monolayers for 1 hour. For each strain, at least 250 total parasites were counted from nine random fields per sample, and values are presented as internal (Int) or external (Ext) parasites per field. Data are mean values +/− SEM (error bars) for two independent experiments performed in triplicate. The asterisk indicates that parasite penetration is significantly lower (p value = 0.0245 using a Student's two-tailed t-test) in Δron8 parasites compared to wildtype. B) Initial stages of attachment are unaffected in Δron8 parasites. Equal numbers of wildtype, Δron8, or R8c parasites were preincubated with 1 µM cytochalasin D for 15 min prior to incubation with host fibroblasts in the presence of cytochalasin D, then fixed and stained in detergent-free conditions with rabbit antisera against SAG1. For each strain, values are displayed as total numbers of parasites counted divided over nine random fields. The data is expressed as mean values +/− SEM for two independent experiments performed in triplicate.

Figure 4. Disorganized secretion of junction components in Δron8 parasites.

A) Pulse invasion assays using wildtype parasites were carried out for 5 min on host fibroblasts prior to IFA with antibodies against RON4 and RON2, showing the typical punctate spot of junction proteins (arrow) after the parasite has fully invaded the host cell but the vacuole has likely not yet detached from the host membrane (see cartoon). B) Similar assays conducted with Δron8 parasites show a trail of RON4 secreted from the posterior end of the newly invaded parasite (arrow). The orientation of the parasite is shown by costaining with anti-ISP1 which detects the apical cap of the IMC. C) Other MJ/RONs detected by anti-RON2 (top panel) or anti-RON5C antibodies which (bottom panel) colocalize with RON4 in secreted trails deposited by Δron8 parasites (arrows).

Figure 5. Selective complementation of Δron8 parasites reveals regions of RON8 necessary for RON8 targeting, complex formation, and function.

A) Schematic and IFA of complementation with full-length 1–2980 HA-tagged RON8 targeted to the RON8 locus. Shown are the RON8 flanks used for targeting by homologous recombination, and the coding sequence, signal peptide (SP), N-terminal prodomain (pro) and C-terminal HA tag (HA). The selectable marker HPT is also shown. IFA of stable parasite clones shows correct targeting to the rhoptry necks as shown by colocalization with RON5N. B) The N-terminal 262 amino acids of RON8 are only partially sufficient for targeting RON8 to the rhoptry necks. The construct contains the first 262 amino acids fused to the mCherry reporter and is targeted to the RON8 locus using the same method as the full length control in “A”. IFA shows that some of the fusion is trafficked properly to the rhoptry necks (arrows) as assessed by colocalization with RON1. However, a significant amount of mistargeted material is also seen in punctate spots in the apical end of the parasite and also in more diffuse patches in the posterior portion of the parasite. C) Addition of the C-terminal portion of RON8 (residues 1318–2980) restores rhoptry neck targeting. Diagram of the construct and IFA showing restoration of rhoptry neck targeting upon inclusion of the C-terminal region of the protein. Colocalization is shown using antisera against RON1. D) Immunoprecipitation of the MJ complex from parasites expressing the full length RON8 in “A” shows efficient purification of all members of the MJ complex. Western blots show an enrichment of all members compared to whole cell lysates (equivalent amounts of lysate and elution are loaded for each). Anti-RON2 antibodies are used for the immunoprecipitation. E) The R8_promCherryR8C fusion is incorporated into the MJ complex. Immunoprecipitation of the MJ complex with anti-RON2 precipitates the R8_promCherryR8C fusion as well as other members of the MJ complex. Thus, the RON8 prodomain plus C-terminal region are sufficient for rhoptry neck targeting and complex association.

Figure 6. RON8-deficient parasites are severely compromised in establishing disease in vivo.

A–C) Groups of 4 CD1 mice were infected with (A) 50 wildtype, (B) 50, 500, 5×10³, 5×10⁴, or 5×10⁵ Δron8, or (C) 50, 500, or 5×10³ R8c parasites and monitored for 28 days. All surviving mice from Δron8 parasites were protected against a lethal challenge of 10⁴ wildtype parasites (not shown).

Figure 7. Model of the Toxoplasma moving junction in wildtype and Δron8 parasites.

The model shows a diagram of a partially invaded parasite with moving junction proteins at the interface of the host and invading parasite. The transmembrane proteins RON2 and AMA1 form the bridge between the host cell and the invading parasite with the soluble proteins RON4, RON5 (processed into N and C fragments) and RON8 exposed to the host cell cytoplasm. RON8 forms a stable intracellular clamp that commits the parasite to invasion, potentially by binding host elements in the cortical cytoskeleton. In the absence of RON8, the MJ is frequently unstable, leading to frequent abortive attachment and invasion.