Introduction

Toxoplasma gondii tachyzoites replicate through a specialized form of internal budding known as endodyogeny, wherein two daughter cells (DCs) are assembled within the cytoplasm of the mother cell (Francia and Striepen, 2014; Hu et al., 2002). This unique replication mechanism demands precise orchestration of organelle inheritance and spatial organization within the parasitophorous vacuole (PV). Apicomplexan parasites, including T. gondii, exhibit a highly polarized cellular architecture defined by specialized secretory organelles—micronemes, rhoptries, and dense granules—essential for host cell invasion and intracellular survival. These are accompanied by canonical organelles such as the Golgi apparatus, endoplasmic reticulum (ER), a single mitochondrion, and the apicoplast, a non-photosynthetic plastid. Notably, unlike most eukaryotes, T. gondii is enveloped by a pellicle consisting of the plasma membrane and an underlying inner membrane complex (IMC) (Ouologuem and Roos, 2014), a structure composed of flattened vesicles tightly associated with the subpellicular cytoskeleton (Harding and Meissner, 2014).

As DCs emerge during endodyogeny, they progressively encapsulate nearly all maternal cytoplasmic contents. While the sequence of organelle acquisition has been extensively characterized (Nishi et al., 2008), the extent to which maternal organelles and proteins are recycled remains unclear. We previously showed that maternal micronemal proteins, such as MIC2, are almost entirely recycled into daughter cells via an F-actin–dependent mechanism and are trafficked through the intravacuolar network (IVN), which structurally organizes the residual body (RB) (Periz et al., 2019). Historically viewed as a passive remnant of maternal cytoplasm, the RB has emerged as a key structure in coordinating replication, facilitating organelle recycling, and maintaining cytoplasmic continuity across individual parasites (Periz et al., 2017b). The RB forms at the posterior end of dividing parasites, physically linking them through the actin-rich IVN and enabling cytoplasmic exchange and synchronized development within the PV (Periz et al., 2017b; Tosetti et al., 2019). The RB originates from the collapse of the maternal parasite during daughter cell budding and occupies the space previously occupied by the mother cell. As a highly dynamic structure, its morphology and organization change throughout parasite replication. At the end of the first replication cycle, the newly formed RB is initially visible as a thin F-actin-rich connection between newly formed daughter parasites and subsequently becomes more prominent as recycled organelles and other cellular material traffic through this compartment. Although no RB-specific molecular marker has yet been identified, the RB is consistently characterized by a dense F-actin network connecting parasites within the parasitophorous vacuole (Periz et al., 2017b). Recent studies have shed some light on the molecular architecture and regulation of the IVN. We demonstrated that F-actin filaments organize the IVN, with disruption of actin (ACT1) resulting in parasite disorganization and impaired microneme recycling (Periz et al., 2019; Periz et al., 2017b). The RB and IVN also require the function of unconventional myosins MyoI and MyoJ, which are implicated in cell-cell communication (Frenal et al., 2017). Additionally, the actin-nucleating factors Formin-2 and Formin-3 are required for IVN formation (Stortz et al., 2019; Tosetti et al., 2019). However, the mechanisms driving the physical transport of maternal organelles remain poorly defined. F-actin could mediate this process either by serving as a track for myosin-dependent vesicular transport, as described for myosin VI in other eukaryotes (Frank et al., 2004), or through the intrinsic mobility and dynamic association of actin bundles themselves, which could transiently interact to promote vesicle exchange (Das et al., 2021; Khaitlina, 2014; Moore et al., 2021).

Although the biogenesis of daughter cell structures has been well described, the fate of maternal organelles during replication remains incompletely understood. Evidence suggests that maternal components such as MIC2 and the IMC protein GAP40 can be recycled via actin-dependent mechanisms (Ouologuem and Roos, 2014; Periz et al., 2019). However, a systematic investigation of organelle recycling during endodyogeny has yet to be performed.

In this study, we employ a dual-labeling strategy using the HaloTag system (Urh and Rosenberg, 2012) to investigate the fate of maternal organelles during T. gondii replication. Through fluorescence intensity-based analysis, we identify three distinct fates for maternal proteins: full inheritance of intact organelles, incorporation of newly synthesized components into expanded maternal structures, and degradation without recycling. Furthermore, we demonstrate that the residual body functions as a key recycling hub, with this process dependent on the activity of the class XXII myosin F (MyoF). These findings provide new mechanistic insights into organelle inheritance and reveal how intracellular organization is maintained during parasite replication.

Results

HaloTag-Based pulse chase labelling allows discrimination of recycled and de novo material. To evaluate HaloTag as a tool for distinguishing maternal from newly synthesized proteins during endodyogeny, two marker proteins with known fates were used: MIC2, which undergoes extensive recycling (Periz et al., 2019), and IMC1, primarily synthesized de novo (Ouologuem and Roos, 2014). Using CRISPR/Cas9, endogenous loci were tagged to generate MIC2-Halo and IMC1-Halo strains (Singer et al., 2023).

Both strains were sequentially labeled by incubation with membrane permeable dyes for 1h: maternal proteins were marked with Halo JF646 before invasion (M-MIC2, M-IMC1), and 24 hours later after replication, by Halo JF549 for 1h to label newly synthesized proteins (n-MIC2, n-IMC1) (Figure 1A). As Toxoplasma gondii replicates in an asynchronous manner, after 24 hours of replication, vacuoles containing 1, 2, 4, and 8 parasites can be observed in a single dish.

Figure 1:

Fluorescence intensity (FI) analysis across those replication stages (Figure 1B–D, S1) revealed a progressive decline of M-MIC2 and an increase in n-MIC2, confirming protein recycling. In contrast, M-IMC1 FI dropped from 100% to 14% between stages 1 and 2, with a corresponding rise in n-IMC1 to ∼80%, indicating predominant de novo synthesis. These distinct FI dynamics demonstrate HaloTag’s utility for resolving protein origin and support the de novo assembly of the IMC.

HaloTag profiling reveals three distinct patterns of protein inheritance

To further investigate protein inheritance during endodyogeny, endogenous HaloTag fusions were generated for a set of organelle-associated proteins: RON2 and ROP1 (rhoptries) (Besteiro et al., 2009; Striepen et al., 2001), GRA1 (dense granules) (Carruthers and Sibley, 1997), TIC20 (apicoplast) (Carruthers and Sibley, 1997), SortLR (Golgi) (Sloves et al., 2012), ANKER1 (Barylyuk et al., 2020), GAPM1a (IMC) (Harding et al., 2019), and MyoA (glideosome) (Meissner et al., 2002) (Figure 2A, Figure S2).

Figure 2:

Following the same sequential labeling strategy used for MIC2 and IMC1, fluorescence intensity (FI) of maternal proteins was quantified across the different replication stages (Figure 2B, C, S2C-G). Analysis revealed three distinct inheritance profiles:

• Group 1: RON2 and ROP1 exhibited stable maternal FI, with minimal decline of FI within a single rhoptry, indicating recycling of whole organelles. In good agreement with this hypothesis, detailed analysis of M-RON2 inheritance demonstrated that the maternal rhoptries are separated during the distribution and that some daughter cells did not obtain maternal organelles after successive rounds of replication (Figure 3D).

• Group 2: Proteins from diverse organelles (SortLR, ANKER1, TIC20, MyoA) showed a stepwise ∼50% reduction in maternal FI per division, suggesting that recycled, maternal material and de novo material end up in the same organelle

• Group 3: IMC proteins (IMC1 and GAPM1a) displayed a sharp FI loss between the first and second replication cycles, consistent with predominant de novo synthesis of the IMC and degradation of maternal material.

Figure 3:

Interestingly, MIC2 exhibited intermediate behavior, prompting further analysis to refine its classification. These patterns highlight distinct organelle-specific protein inheritance mechanisms in T. gondii.

Micronemes and rhoptries are recycled as intact organelles during Toxoplasma endodyogeny

Micronemes are known to comprise distinct subpopulations (Kremer et al., 2013), and prior work has shown that recycled and de novo MIC2 localize to separate microneme subsets (Periz et al., 2019). Based on these observations, we hypothesized that both micronemes and rhoptries are recycled as intact organelles, with new organelles formed independently via de novo synthesis.

To test this, we conducted detailed analyses of MIC2 and rhoptry inheritance. As expected, recycled (M-MIC2) and newly synthesized MIC2 (n-MIC2) were largely segregated into distinct micronemes with minimal colocalization (Figure 3A, S3)(Periz et al., 2019). Focusing on sparsely distributed micronemes outside the apical region (Figure 3A, boxed area), we measured fluorescence intensity (FI) of individual M-MIC2-positive micronemes at replication stages 2, 4, and 8 (Figure 3A, zoomed region; 3C). The relatively stable FI of these isolated structures supports whole organelle recycling, consistent with the FI profiles of rhoptries (Figure 2B, C).

Rhoptry analysis further corroborated this model. As with MIC2, M-RON2 and n-RON2 localized to distinct rhoptry populations (Figure S3), with divergence becoming more apparent at stage 8 (Figure 3D, Figure S4A). M-RON2 FI remained relatively constant through replication (Figure 3 F), while n-RON2 gradually increased from stage 2 onward (Figure S4B). Due to their low copy number (8–12 per tachyzoite), maternal rhoptries were separated and often absent in some daughter cells after successive divisions, consistent with stochastic whole-organelle inheritance illustrated by decrease in the average area of the M-RON2 fluorescence signal and the constancy of the total surface area (Figure S4C, D). As a consequence, a progressive decline in M-RON2-positive parasites per vacuole can be observed (Figure 3E, D; Figure S4A).

For the micronemes, even in larger parasitophorous vacuoles, all daughter parasites retained a mix of recycled and newly synthesized micronemes (Figure S5). Interestingly, the maternal micronemes were observed to be relatively well distributed among the daughter cells (Figure S5A, B). Considering that a single mother parasite contains approximately 64 micronemes and that successive rounds of endodyogeny generate up to 32 daughter cells, a purely random segregation of existing micronemes would be unlikely to produce the relatively uniform distribution observed (∼0–3 maternal micronemes per tachyzoite). Simple probabilistic modelling indicates that the observed distribution would occur in less than 1% of vacuoles under a random inheritance scenario. Yet similar distributions were consistently observed across multiple vacuoles and independent biological replicates (Figure S5C, D). These observations strongly argue against stochastic segregation and instead support the existence of a regulated mechanism that promotes balanced microneme inheritance between daughter parasites. Together, these results demonstrate that both micronemes and rhoptries are recycled as intact organelles. De novo organellogenesis acts in parallel to maintain a full complement in each daughter cell, ensuring faithful inheritance during endodyogeny.

Golgi inheritance involves coordinated expansion and partitioning of maternal and de novo proteins

Next, we analyzed SortLR, a Golgi marker, using dual HaloTag labeling. Maternal SortLR (M-SortLR) fluorescence declined by ∼50% per replication round, while newly synthesized SortLR (n-SortLR) increased proportionally, without abrupt loss (Figure 4A–B). This gradual dilution contrasts sharply with the rapid decay seen in Group 3 proteins.

Figure 4:

M-SortLR and n-SortLR exhibited strong colocalization across replication stages, and all daughter cells retained M-SortLR (Figure 4C). The total Golgi area expanded from ∼2 µm² to ∼10 µm² by stage 8 (Figure 4D), consistent with coordinated Golgi elongation and medial fission, as described previously (Pelletier et al., 2002). These results demonstrate that the Golgi undergoes duplication, incorporating both maternal and de novo proteins during its growth and partitioning. Similar signal evolution was observed for the ER, MyoA and apicoplast (Figure S2, S6). Thus, Group 2 organelles follow a distinct inheritance mode, involving regulated expansion, integration of new material, and equal distribution to progeny—distinct from the whole-organelle recycling of rhoptries/micronemes and the complete turnover seen in IMC proteins.

Maternal IMC proteins are degraded in the RB during replication

Group 3 proteins, typified by IMC components like GAPM1a and IMC1, exhibit a sharply distinct inheritance profile, characterized by rapid loss of maternal fluorescence during replication. Using HaloTag-based dual labeling, we observed a ∼90% reduction in maternal GAPM1a signal by stage 2, with minimal retention through later stages (Figure 5A–C), indicative of active degradation rather than recycling. M-GAPM1a and n-GAPM1a did not colocalize, and residual maternal signal localized transiently at the posterior pole before disappearing entirely, suggesting disposal via the residual body (RB) (Figure 5D).

Figure 5:

Live-cell imaging of IMC1 further supported this model: maternal IMC1 (M-IMC1) condensed at the posterior pole during daughter budding, followed by a marked fluorescence drop, without redistribution to other regions (Figure 5E, F and Figure S7, Video S1). This signal loss was replication-specific, as non-dividing parasites retained stable M-IMC1 levels (Figure S7D). Quantification across regions revealed a transient increase at the posterior, followed by complete disappearance, consistent with degradation (Figure 5F, Figure S7C).

Interestingly, DCs IMCs exhibited brighter fluorescence than the maternal IMC (Figure S7E, F), especially during early stages of elongation (Figure S7G, H). FI peaked when DCs reached ∼3 μm in length, suggesting the IMC is fully assembled before emergence and subsequently unfolds without requiring additional material. This preformation explains the absence of maternal IMC recycling. These findings align with earlier observations that IMC components are synthesized de novo during elongation (Ouologuem and Roos, 2014).

Whole-organelle inheritance of micronemes and rhoptries occurs via similar recycling pathways

Our data suggest that the organelles trafficking through the residual body (RB) have different fate: while microneme and rhoptry proteins are routed through the RB, IMC proteins are primarily degraded within this compartment. In contrast, recycling of maternal Golgi, ER, and apicoplast proteins appears to occur directly, bypassing the RB.

To investigate this further, we performed live-cell imaging of rhoptry and microneme inheritance (Figure 6A, B, Video S2, S3). MIC2-Halo parasites co-expressing IMC1-YFP, showed that maternal MIC2 (M-MIC2) transits through the RB (Figure 6A, white arrows). Notably, after endodyogeny was completed, M-MIC2 was redistributed from the RB to the apical tip of the daughter cells (Figure 6A, 11:30, 16:00 h), confirming that the RB serves as a temporary reservoir during microneme recycling (Periz et al., 2019).

Figure 6:

For rhoptries, we examined the inheritance of maternal RON2 (M-RON2) in RON2-Halo parasites co-expressing IMC1-YFP. In approximately 90% of parasites undergoing replication, M-RON2 was integrated into daughter rhoptries prior to mother cell collapse and formation of the RB (Figure 6B, 4:15–4:30 h and 10:30–10:45 h). Like M-MIC2, M-RON2 was occasionally detected in the RB, though less prominently, suggesting more rapid, tightly regulated or easier recycling due to their lower number.

To test whether recycled rhoptries associate with F-actin, as previously reported for micronemes (Periz et al., 2019), we stably integrated the F-actin marker Cb-Emerald (Periz et al., 2017b) in the RON2-Halo line (Figure 6C, D). While m-MIC2 displayed clear F-actin–associated movement along filaments in the residual body (Figure 6C), colocalization of M-RON2 with Cb-Emerald in the residual body was infrequent. On the other hand, both secretory organelles were found to be associated with cytoplasmic F-actin (Figure 6D).

In summary, both rhoptries and micronemes are inherited as intact organelles during T. gondii replication. These findings highlight the organelle-specific complexity of recycling pathways and support whole-organelle segregation of micronemes and rhoptries into the daughter cells. Given the F-actin dependency of this recycling pathway demonstrated previously for the micronemes (Periz et al., 2019), we hypothesized that a conserved apicomplexan myosin mediates this process. MyoF, a class XXII myosin is conserved across Apicomplexa (Jacot et al., 2013) and has been implicated in diverse functions—from apicoplast inheritance to F-actin–dependent vesicular transport (Carmeille et al., 2021; Heaslip et al., 2016; Kellermeier and Heaslip, 2024).

We endogenously Halo-tagged MIC2, RON2, and SortLR in the auxin-inducible degron line MyoF-mAID (Carmeille et al., 2021) and performed dual HaloTag labelling to differentiate maternal (M) from de novo (n) protein pools. Live imaging revealed that MyoF depletion resulted in the accumulation of maternal but not de novo MIC2 and RON2 in the RB during early replication stages (Figure 7A, B). As Toxoplasma replicates asynchronously, different replication stages coexist within the same dish after 24 h. The second labeling step marks all proteins synthesized since the beginning of the experiment, allowing discrimination between proteins present in the initial mother parasite and those synthesized during subsequent replication cycles. Yet, during each round of replication, daughter cells form within and inherit material from their mother cell. As replication proceeds, proteins synthesized during earlier cycles are progressively inherited and recycled in subsequent stages. Under MyoF depletion, this results in the accumulation of green-labeled protein pools, initially synthesized de novo during the previous cycle, within the residual body, rather than their redistribution to daughter cells starting from the second round of replication (Figure 7 Stage 4).Immunofluorescence with α-AMA1, α-MIC4, and α-MIC8 confirmed that this phenotype broadly affects the entire repertoire of micronemal proteins (Figure S8, S9). Quantification showed that approximately 90% of vacuoles exhibited RB accumulation of maternal material upon auxin treatment (Figure 7D), indicating that MyoF specifically mediates the recycling of maternal micronemes and rhoptries. Interestingly, although our time lapse analysis indicated that rhoptries only rarely traffic via the RB, upon depletion of MyoF, they accumulate in the RB, supporting the hypothesis that micronemes and rhoptries are transported in the same manner during recycling. In contrast, Golgi inheritance (marked by SortLR) remained unaffected by MyoF depletion (Figure 7C). Neither accumulation in the RB nor separation of maternal and de novo SortLR signals was observed, consistent with previous reports that Golgi duplicates and partitions independently (Pelletier et al., 2002). Yet, similar fragmentation of the Golgi as previously reported was observed.

Figure 7:

To confirm that this phenotype results directly from MyoF depletion and not broader disruption of endomembrane architecture, we performed a rescue experiment. MIC2-Halo MyoF-mAID parasites were cultured in auxin for 24 hours to induce M-MIC2 retention in the RB, followed by auxin chase and continued replication (Figure 8A). Following auxin washout, M-MIC2 was redistributedto daughter parasites during subsequent replication cycles, although occasionally in an uneven manner (Figures 8A, 8C, and S10A). To determine whether this redistribution depended on restoration of MyoF expression, we monitored MyoF recovery by IFA using the HA epitope fused to the MyoF-mAID protein (Figure 8B). Following auxin washout, MyoF expression was restored. Quantification revealed a strong association between the presence of MyoF and the redistribution of maternal micronemes, supporting the conclusion that microneme recycling from the RB is dependent on MyoF function (Figure S10B).

Figure 8:

Live-cell imaging confirmed these dynamics (Video S4, S5). In untreated parasites, M-MIC2 entered the RB ∼7 h post-replication and was later redistributed to daughter cells (Figure 8D, Video S4). Under auxin treatment, M-MIC2 accumulated post-budding and failed to redistribute (Figure 8E; Video S5). Washout of auxin rescued microneme redistribution during the subsequent replication cycle, although live visualization required intensity adjustments due to differences in signal intensity between the accumulated and redistributed microneme pools (Figure 8F; Video S6). Redistribution was not observed immediately upon MyoF re-expression, but rather during the following replication cycle. This delay suggests that restoration of MyoF-dependent trafficking may require sufficient recovery of MyoF function before efficient redistribution of maternal micronemes can occur.

To assess whether microneme degradation occurred during retention within the residual body, the fluorescence intensity of individual maternal micronemes was compared between control vacuoles and vacuoles subjected to MyoF depletion followed by auxin washout (Figure S10C, D). No significant difference in microneme fluorescence intensity was detected between control micronemes and micronemes that had remained within the RB for 24 h prior to redistribution. These results indicate that retention within the RB does not result in detectable loss of maternal microneme signal. Together, these findings establish that MyoF is essential for recycling maternal micronemes and rhoptries via the RB, redefining the RB as a dynamic trafficking hub, not merely a degradative compartment. MyoF acts as a mechanistic bridge between F-actin dynamics and organelle inheritance, orchestrating precise redistribution of maternal cargo.

Discussion

Use of dual labelling to differentiate maternal and de novo material

Understanding the complexity of organelle inheritance in Toxoplasma gondii requires tools capable of resolving both the spatial and temporal dynamics of protein trafficking. Conventional single-labelling approaches have been limited to static snapshots, obscuring key processes such as organelle biogenesis, recycling, and selective degradation. To overcome these challenges, we employed a dual HaloTag pulse-chase labelling strategy to distinguish maternally inherited from de novo–synthesized proteins. This approach, previously underutilized in apicomplexan parasites (Koreny et al., 2023; Periz et al., 2019), allowed us to resolve protein fates across replication cycles and map distinct inheritance routes for multiple organelles (Figure 9A).

Figure 9:

The residual body as a regulatory hub

Our findings reaffirm the emerging view of the RB as an active, multifunctional structure rather than a passive cytoplasmic remnant (Frenal et al., 2017; Periz et al., 2017a). First described over 50 years ago (Sheffield and Melton, 1968), the RB was long believed to serve as a repository for discarded material, often called a “waste bin” (Nishi et al., 2008). However, we show that the RB temporarily stores maternal secretory organelles, such as micronemes and rhoptries, which are later redistributed to daughter cells in a MyoF-dependent manner (Figure 9B). Live-cell imaging revealed that organelles in the RB are not static but can re-enter functional pathways (Figures 8), establishing the RB as a trafficking hub (Periz et al., 2019). This active role is further supported by the dependence of RB-mediated recycling on F-actin and the class XXII myosin MyoF, which we show to be essential for retrieval of maternal MIC2 and RON2 but dispensable for Golgi inheritance although we noticed a fragmentation of the Golgi, which has been described to depend on MyoF (Carmeille et al., 2021). The RB is also critical for cytoplasmic continuity across parasites, synchronizing replication and facilitating inter-parasite protein exchange (Frenal et al., 2017; Muniz-Hernandez et al., 2011; Periz et al., 2017a). Its formation is dependent on actin-nucleating factors and myosins, indicating a complex and regulated origin rather than a passive leftover of cytokinesis (Stortz et al., 2019; Tosetti et al., 2019). Given these parallels, the RB may be functionally analogous to the mammalian midbody remnant, which regulates post-mitotic signaling and intracellular trafficking (Kuriyama et al., 2025; Peterman and Prekeris, 2019).

Fate determination: inheritance or degradation?

An important question raised by our study is how the parasite determines whether maternal proteins are recycled or degraded. Our results indicate that this decision is selective rather than stochastic (Figure 9). For instance, maternal micronemes persist for at least six replication cycles in the absence of MyoF, without signs of degradation (Figure S10C, D). In contrast, IMC proteins like GAPM1a and IMC1 are lost within a single cycle and show no evidence of recycling. Our data are consistent with degradation occurring within the RB., raising questions about the underlying mechanism. Although T. gondii encodes homologs of lysosomal and autophagic degradation machinery (Besteiro et al., 2011; Smith et al., 2021; Thaprawat et al., 2025), none have been localized to the RB or directly linked to IMC degradation. However, ubiquitin ligases and proteasomal components have been identified in the RB proteome (O’Shaughnessy et al., 2023; Que et al., 2002). IMC proteins are also part of the ubiquitinated proteome (Silmon de Monerri et al., 2015), and recent work showed that deletion of the kinase ERK7 prevents sequestration of the E3 ligase CSAR1, resulting in aberrant degradation of conoids (O’Shaughnessy et al., 2023). These findings support a model in which protein fate within the RB is determined through selective ubiquitination and regulated proteolysis.

A call to reassess recycling factors in Toxoplasma gondii

The recognition of the residual body (RB) as a key site for organelle recycling has fundamentally revised our understanding of intracellular trafficking in T. gondii (Periz et al., 2019; Tosetti et al., 2019). Earlier models emphasized de novo synthesis of organelles and proposed a repurposing of classical endocytic machinery for secretory functions (Tomavo, 2014). However, our HaloTag-based pulse-chase studies show that secretory organelles such as micronemes and rhoptries are extensively recycled via the RB, in a process dependent on F-actin and the unconventional myosin MyoF.

This shift in our understanding calls for a systematic re-evaluation of trafficking factors traditionally implicated in protein targeting, vesicle transport, and organelle biogenesis. Many of these, particularly Rab-GTPases, SNAREs, and dynamins were previously studied without consideration of recycling pathways. Notably, accumulation of microneme material in the RB has been observed upon perturbation of several trafficking regulators, including Rab5A (Kremer et al., 2013), ArlX3 (Klinger et al., 2024), SORTLR (Sloves et al., 2012), VPS8 (Morlon-Guyot et al., 2018), and VPS11 (Morlon-Guyot et al., 2015). Yet, whether these phenotypes reflect a block in secretion, synthesis, or recycling remains unclear. To resolve this, a targeted reanalysis of trafficking factors using our dual-labelling HaloTag assay combined with the splitCas9 system (Li et al., 2022) would allow high-throughput functional screening to specifically distinguish defects in recycling from those in de novo biogenesis, offering a path to redefine the roles of classical and lineage-specific trafficking regulators in T. gondii.

Material and methods

Parasite culture and genetic manipulation

Growth and generation of transgenic T. gondii

T. gondii tachyzoites from the RH strain and derived lines, including RH Δku80/TATi and RH Δku80/Tir1, were maintained at 37 °C with 5% CO₂ in human foreskin fibroblasts (HFFs; ATCC, SCRC 1041) cultured in Dulbecco’s Modified Eagle Medium (DMEM; Sigma, D6546) supplemented with 10% fetal bovine serum (FBS; BioSell, FBS.US.0500), 4 mM L-glutamate (Sigma, G7513), and 20 μg/mL gentamycin (Sigma, G1397) as previously described (Gras et al., 2019).

Generation of transgenic parasites

New strains were generated using CRISPR/Cas9 as previously described (Li et al., 2022). Guide RNAs (gRNAs) targeting the regions of interest were designed using EuPaGDT (Alvarez-Jarreta et al., 2024). All gRNA and primer sequences are listed in Supplementary Table 1. Briefly, gRNA oligos were annealed, ligated into the Cas9-YFP vector, and verified by sequencing (Eurofins Genomics). Repair templates were generated by PCR amplification of the Halo or YFP tag flanked by 50 bp homology arms to the target gene, using Q5 High-Fidelity DNA Polymerase (New England BioLabs). PCR products were purified with a PCR purification kit (Blirt, EM26.1). Tachyzoites were mixed with the repair template and 10 µg of the corresponding Cas9 vector, and transfected using the Amaxa 4D-Nucleofector system (Lonza, AAF-1003X). Transfected parasites were allowed to invade fresh HFFs and replicate for 48 h. Following manual egress and filtration through a 3 μm filter, Cas9-YFP-expressing parasites were enriched by FACS (FACSAria III, BD Biosciences) and sorted into 96-well plates. Correct integration of the repair template was confirmed by PCR.

Labelling and characterization

Maternal and de-novo protein discrimination

Fresh tachyzoites expressing Halo-tagged reporters were mechanically egressed, filtered through a 3 μm filter, and resuspended in cold DMEM containing a membrane-permeable Halo dye, Janelia Fluor® 646 (1:1000, Promega), for 1 h. Parasites were centrifuged at 2,500 rpm for 5 min and washed three times with fresh medium to remove unbound dye. Parasites were then seeded onto HFF-covered Ibidi live-cell dishes for overnight replication. After replication, a second labeling was performed with a new membrane-permeable Halo dye, Janelia Fluor® 549 (1:1000, Promega) for 1 h at 37 °C, followed by three washes before imaging.

Fluorescence intensity across the replication

Parasites were labeled as described above and allowed to replicate for 24 h on HFF-coated Ibidi live-cell dishes. Approximately 15 fields of view were imaged using Z-stacks spanning 3 μm centered on the vacuoles (Figure S11A). For each replication stage (1, 2, 4, and 8 parasites per vacuole), individual tachyzoites were sampled across multiple vacuoles.

Maximum-intensity projections were generated from non-deconvolved images. Fluorescence intensity (FI) was quantified as the maximum gray value measured within regions of interest (ROIs) drawn on individual tachyzoites from each vacuole stages (Figure S11B). ROIs excluded overlapping parasites, neighboring vacuoles, and regions with atypical signal intensity (Figure S11C). For each biological replicate, up to 25 tachyzoites per replication stage were analyzed, and the mean FI value was calculated for each stage. The highest mean FI observed among the stages within a replicate was defined as 100%, and FI values for the other stages were expressed relative to this maximum. Relative FI values were then averaged across three independent biological replicates. Data are presented as mean ± SD. A total of 272, 286, 251, 276, 223, 284, 265, 300 and 260 vacuoles were analysed for RON1, ROP1, SortLR, MyoA, TIC20, ANKER1, GAPM1a, IMC1 and MIC2 respectively.

Fluorescence intensity of the daughter cells vs new mother cells

Images were processed as described above. Fluorescence intensity was quantified as the maximum gray value measured within regions of interest (ROIs) drawn on individual daughter cells and on newly formed mother cells. To reliably discriminate newly formed mother IMC from daughter IMC (both in the same channel colour, Jan. 549), the channel corresponding to the initial maternal IMC (Jan. 646) was intentionally overexposed, allowing unambiguous identification of the new mother cell IMC signal. ROIs were drawn on non-overlapping regions and excluded areas containing overlapping parasites, neighboring vacuoles, or atypical signal intensity, following the same exclusion criteria used for replication-stage measurements (Figure S11D). Quantification was performed on maximum-intensity projections generated from non-deconvolved Z-stacks. Data are presented as mean ± SD from vacuoles pooled across three independent biological replicates.

Measurement of the signal area of the fluorescence

To quantify the average and total fluorescent area, images were analyzed in Fiji. Up to 25 isolated vacuoles per stage (1 to 8) were cropped and analyzed individually. A threshold was applied to the maternal signal, and both average and total signal area were measured. Three independent biological replicates were performed, and values are reported as mean ± SD.

Quantification of percentage of parasite per vacuole with maternal protein

Images were analyzed in Fiji to determine the proportion of parasites per vacuole with maternal protein signal. Up to 25 vacuoles per stage (1 to 8) were cropped and analyzed individually. Vacuoles where all parasites retained maternal signal were scored as 100%. If some parasites lacked signal, the percentage of maternal-positive parasites was calculated. The experiment was performed in three independent biological replicates. Values are reported as mean ± SD.

Live replication assay

Parasites were labeled with Janelia Fluor® 646 (1:1000, Promega) for 1 h and washed three times before transfer to Ibidi live-cell dishes covered with HFFs. Parasites were allowed to invade for 1–2 h to obtain ∼10 parasites per field of view. Excess parasites were removed by washing three times. Imaging was performed on a Leica DMI8 microscope at 37 °C in 5% CO₂. Images were acquired every 15–30 min for 12 h using minimal laser power and exposure. Experiments were performed in three biological replicates and analyzed in Fiji.

Colocalization assays: ANKER1-Halo with HDEL-GFP

Parasites expressing ANKER1-Halo were transiently transfected with an HDEL-GFP plasmid to label the ER. After overnight replication in HFF-coated Ibidi dishes, parasites were labeled with Janelia Fluor® 646 (1:1000, Promega) for 1 h and washed three times prior to imaging. Z-stacks of HDEL-GFP-positive parasites were acquired. The proportion of vacuoles displaying ER localization of ANKER1-Halo was assessed manually in Fiji/ImageJ by inspection of full Z-stacks and maximum-intensity projections. In addition, the degree of colocalization between ANKER1-Halo and HDEL-GFP was quantified using Pearson’s correlation coefficient calculated with the Coloc2 plugin in Fiji/ImageJ. Pearson analysis was performed on 30 individual parasites obtained from three independent biological replicates (10 parasites per replicate). The mean Pearson’s correlation coefficient was 0.92 ± 0.04 (mean ± SD).

Colocalization assays: MIC2/RON2 with Cb-Emerald

For analysis of association with the F-actin network, MIC2 or RON2 was endogenously tagged and labeled in a strain stably expressing Cb-Emerald. Freshly egressed parasites were labeled with Janelia Fluor® 646 (1:1000, Promega) for 1 h, washed, and seeded onto HFF-coated live-cell dishes. After 24 h of replication, parasites were fixed in 4% paraformaldehyde and imaged. Z-stacks of vacuoles displaying a clearly defined F-actin network were acquired, and the association of MIC2- or RON2-positive structures with F-actin was assessed in both the parasite cytoplasm and the RB. Analysis was performed manually in Fiji/ImageJ by inspection of full Z-stacks and maximum-intensity projections. The RB was defined by the presence of thick bundles of F-actin located at the basal pole of the parasites and forming a continuous structure connecting parasites within the vacuole, consistent with the characteristic morphology of the RB. Cytoplasmic F-actin was defined as all actin signal located within the parasite body excluding this RB-associated region. Three independent biological replicates were analyzed, and data are reported as mean ± SD.

Quantification of the maternal microneme distribution

To assess the distribution of maternal micronemes at stage 8, parasites were labeled as described in “Maternal and de novo protein discrimination.” Twenty-five stage 8 vacuoles were selected. Using the de novo MIC2 signal, the outline of each tachyzoite was traced, and the number of maternal MIC2-positive parasites was counted. Tachyzoites were numbered 1 to 8 from left to right. Three biological replicates were performed. Data are reported as mean ± SD. Images were analyzed in Fiji.

Probability of maternal microneme inheritance

To evaluate whether maternal micronemes could be distributed among daughter parasites by stochastic segregation alone, we modeled the allocation of 64 micronemes from a mother parasite to 32 daughter tachyzoites. We specifically assessed the probability that each daughter would receive between 0 and 3 micronemes, corresponding to the range observed experimentally. As an approximation, microneme allocation was modeled as an independent random process following a Poisson distribution:

where X is the number of micronemes inherited by a daughter and .

This gives:

Assuming independent inheritance events, the probability that all 32 daughters within a vacuole receive between 0 and 3 micronemes is:

Finally, assuming independence between vacuoles, the probability of observing this outcome in 20 consecutive vacuoles is:

These calculations indicate that the experimentally observed microneme distribution would be highly unlikely under a simple stochastic segregation model and are therefore consistent with the existence of a regulated mechanism promoting balanced microneme inheritance.

Phenotypic assays

Induction of MyoF KD

MyoF-mAID parasites were genetically modified to endogenously tag MIC2, RON2 and SortLR. As previously described for MyoF mAID parasites (Carmeille et al., 2021), MyoF-mAID MIC2/RON2 or SortLR-Halo parasites were induced +/- auxin for 4h prior to labelling and experiments.

Quantification of the accumulation of protein in the residual body

Parasites were treated ± auxin for 4 h and labeled as described above. After 24 h replication on live-cell dishes, ∼100 vacuoles were imaged across ∼15 fields of view. The percentage of vacuoles showing accumulation of material in the residual body was calculated. The experiment was performed with three independent biological triplicates. The values were then expressed as the mean values of the three independent experiments ± SD. Images were analysed via Fiji.

Auxin chased experiment

Parasites were induced ± auxin for 4 h prior to labeling as described above in “Maternal and de novo protein discrimination”. Following transfer to live-cell dishes, parasites were allowed to replicate for 48 h in the continued presence or absence of auxin before imaging. For the auxin chase condition, parasites were first maintained in auxin-containing medium for 24 h. The medium was then removed, and dishes were washed three times with fresh medium to ensure complete auxin removal before addition of fresh medium lacking auxin. Parasites were subsequently allowed to replicate for an additional 24 h prior to imaging.

To assess recovery of MyoF expression following auxin washout, parallel samples were fixed in 4% paraformaldehyde and processed for immunofluorescence using an anti-HA antibody to detect the MyoF-mAID-HA fusion protein. Vacuoles were categorized according to the presence or absence of detectable MyoF signal and correlated with the corresponding microneme distribution phenotype.

Approximately 15 fields of view were imaged per condition, corresponding to ∼100 vacuoles. The percentage of vacuoles displaying microneme accumulation within the residual body was determined. In addition to the “accumulated” and “normal” phenotypes, a third category termed “uneven distribution” was defined for vacuoles in which micronemes were no longer retained within the residual body but had not yet reached the homogeneous distribution observed in control parasites. The frequency of each phenotype was quantified. Experiments were performed in three independent biological replicates, and values are presented as mean ± SD. Image analysis was performed using Fiji/ImageJ.

MyoF-mAID Live replication assay

Parasites were induced +/- auxin for 4h prior labelling. Parasites were labelled with the Halo membrane permeable dye Halo 646 (Janelia Fluor®646 1:1000 Promega) for 1 h and washed three-time prior transfer onto a live cell dish covered with HFF. Parasites were allowed to invade for 1h-2h to obtain enough invasion events to reach an average of 10 parasites per FOV. After the invasion, the excess of parasites was removed by 3 washes. A new membrane permeable dye Halo 549 (Janelia Fluor®549 1:10000 Promega) was added to the media to record the de-novo generation of the POI live. Live cell dishes were then transferred to the Leica-DMI8, heated at 37°C under a chamber containing 5% CO2. Laser power and exposure were adjusted to the lowest values allowing reliable imaging. Imaged were taken every 15-30 minutes to follow the tagged protein behaviour during replication for 12h. For the Chased assay, the imaging started after the removal of the auxin after 24h of replication as described above in “Auxin chased experiment.” _The experiments were performed with three independent biological triplicates. Images were analysed via Fiji.

Imaging

Widefield microscopy

Unless stated otherwise, all images were acquired on a Leica-DMI8, objective 100x with the LasX software (v3.7.4). Fiji (v1.53c) was used to analyse the picture and all counts were made manually. LasX software (v.) from Leica was used to obtain parasite imaging data and All images and movies were processed using Fiji (ImageJ) software v Image Processing Software (Schindelin et al., 2012).

Software

Fiji ( FIJI ImageJ v1.54f) was used to analyze the picture and all counts were made manually.

Data analysis

All data were plotted using Microsoft Excel

Statistical analysis

Two-tailed unpaired Student’s t-tests were performed to evaluate statistical significance.

-For Figures 2C, 2E, 2F, 3B, 3D, 4B, 4C, 5C (stage 1–2), and Supplementary Figures S1A, S1B, S2C–S2G, S4B–S4D, differences between stages 1–2, 2–4, and 4–8 were assessed.

-For Figures 2B, 4C, and 5C (stage 2–8), variance could not be calculated because values were constant (0 or 100%), and these were therefore labeled as non-significant.

-For Figures 7D, S8D, and S9B, differences between uninduced controls and MyoF knockdown (KD) induced samples were analyzed.

-For Figure 8C, the distribution of phenotypes (normal, accumulated, and redistributed micronemes) was compared between control, auxin-induced (48 h), and auxin-washed (chased) conditions.

-For Figure S5B, microneme distribution across positions 1–2, 2–3, 3–4, 4–5, 5–6, 6–7, and 7–8 was analyzed; no statistically significant differences were observed, and all were labeled as non-significant.

-Finally, for Figure S7F, the difference in fluorescence intensity between daughter and mother IMC signals was assessed.

Data are presented as mean ± SD. Statistical significance ns: not significant (p ≥ 0.05); * : p < 0.05; ** : p < 0.01; *** : p < 0.001.

Supplementary figures

Figure S1:

Figure S2:

Figure S3:

Figure S4:

Figure S5:

Figure S6:

Figure S7:

Figure S8:

Figure S9:

Figure S10:

Figure S11:

Acknowledgements

This work was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) grant GR 5696/2-2 to Simon Gras, grant ME 2675/6-2 and DFG Equipment grant INST 86/1831-1 to Markus Meissner. We thank VEuPathDB for their invaluable Informatics Resources. The MyoF-mAID strain was kindly provided by Prof. Aoife Heaslip (University of Conneticut), MIC4, MIC8 and AMA1 antibodies were kindly provided by Prof Domique Soldati-Favre (University of Geneva) and Prof. Gary Ward (University of Vermont) respectively. We thank Dr. Elena Jimenez-Ruiz for the useful discussions.

Additional files

Image quantifications for indicated figures.

Video S1. Live degradation of the maternal IMC1. Magenta: M-IMC1 (Jan646), Green: Cb-Emerald.

Video S2. Live inheritance of the maternal RON2. Magenta: M-RON2 (Jan646), Green: IMC1-YFP.

Video S3. Live inheritance of the maternal MIC2. Magenta: M-MIC2 (Jan646), Green: IMC1-YFP.

Video S4. Live replication of MyoF mAID MIC2-Halo without auxin. Magenta: M-MIC2 (Jan646), Green: n-MIC2 (Jan549).

Video S5. Live replication of MyoF mAID MIC2-Halo with auxin. Magenta: M-MIC2 (Jan646), Green: n-MIC2 (Jan549).

Video S6. Live replication of MyoF mAID MIC2-Halo after auxin chase. Magenta: M-MIC2 (Jan646), Green: n-MIC2 (Jan549).

Additional information

Funding

Deutsche Forschungsgemeinschaft (DFG) (426682790)

• Markus Meissner

Deutsche Forschungsgemeinschaft (DFG) (404794945)

• Markus Meissner