Cryptosporidium aspartyl protease 2 is required for host cell egress of merozoites and male gametes

Abstract

The parasite Cryptosporidium causes severe diarrheal disease that can be life-threatening, and effective treatments are sorely lacking. Recently, aspartyl proteases (ASP) have emerged as targets with significant therapeutic potential in several related parasites, resulting in the development of multiple potent leads. ASPs are critical to the proteolytic activation and maturation of secretory proteins that parasites rely on to invade, manipulate, and upon completion of their replication cycle, exit the host cells in which they reside. The Cryptosporidium genome encodes five ASPs, which have not been previously studied. Here, we explore two of these enzymes and in genetic experiments find one, CpASP2, to be essential to parasite growth. Conditional deletion of the gene encoding this protease leads to arrest at two distinct points in the lifecycle. Cell biological studies of the mutant phenotype demonstrate that CpASP2 is required for egress of both asexual merozoites and male gametes. Mutant parasites appear to complete intracellular development yet are paralyzed and incapable of responding to stimuli that trigger motility and egress in wild-type. Ablation of CpASP2 in infected mice leads to rapid parasite clearance, highlighting the promise of CpASP2 and likely additional related enzymes as multistage targets of therapy.

• Aspartyl proteases are prominent targets for antiparasitic drug development; however, little is known about these enzymes and their cellular functions in Cryptosporidium.

• Using genetic experimentation, we find Cryptosporidium parvum ASP2 to be required for parasite host cell egress and essential at multiple points in the lifecycle, and ablation of CpASP2 leads to rapid cure of infected mice.

• Cryptosporidium is a leading cause of diarrheal disease in young children, and effective drug treatment is lacking. This work identifies a promising target for the development of urgently needed therapeutics and provides fundamental insight into the cell biology of this important pathogen.

INTRODUCTION

The intestinal parasite Cryptosporidium is a leading cause of diarrheal disease prevalent around the globe. Severe disease in young children is strongly associated with malnutrition as a risk factor (Kotloff et al., 2013; Gilbert et al., 2023), while infection with the parasite is also a cause of malnutrition, leading to developmental delay and stunting (Khalil et al., 2018). Cryptosporidium is a significant veterinary pathogen (Thomson et al., 2017), and spillover from multiple domestic animal species to humans results in frequent zoonotic transmission and numerous outbreaks (Lange et al., 2014; Guo et al., 2021).

Efforts to curtail the impact of the disease on children are hindered by the lack of efficacious drug treatments or preventative vaccines. Currently, only a single drug, nitazoxanide, is approved by the US Food and Drug Administration for treatment of cryptosporidiosis. However, nitazoxanide was found to be of no benefit in immunocompromised individuals and showed only limited efficacy in malnourished young children (Amadi et al., 2002). There is thus a pressing need for the development of novel treatments, particularly for those most at-risk. The disproportionate burden of cryptosporidiosis in lower-income countries challenges the traditional for-profit model of drug development (Gilbert et al., 2023). Repurposing therapeutics that have already benefited from significant investment for a new use in the treatment of neglected diseases is a way to reduce cost and circumvent this serious roadblock (Janes et al., 2018). In the related parasite Plasmodium, the etiological agent of malaria, protease inhibitors used to treat HIV infection were found to be effective in vitro, in mouse models, and in human clinical trials (Andrews et al., 2006; Achan et al., 2012). These drugs target aspartyl proteases (ASP), which in Plasmodium are also called plasmepsins (PM (Mishra et al., 2024)). Many of the PMs studied initially act in the hemoglobin digestion pathway of the malaria parasite (Kolakovich et al., 1997), and targeting these enzymes had little effect on parasite survival due to their overlapping and redundant function (Omara-Opyene et al., 2004). However, additional ASPs identified more recently appear to play nonredundant roles. They function as protein maturases in the secretory pathway at different stages in the lifecycle (PMVI-VIII; Ecker et al., 2008; Li et al., 2016; Mastan et al., 2017), enable effector transport to the host cell (PMV, TgASP5; Boddey et al., 2013; Hammoudi et al., 2015; Coffey et al., 2018; Coffey et al., 2015), or are critical to host cell invasion and egress (PMIX, PMX, TgASP3; Dogga et al., 2017; Nasamu et al., 2017). Several of these ASPs were found not only to be essential but also druggable (Dogga et al., 2017; Nasamu et al., 2020). The unique structural properties of these enzymes enhance the selectivity of ASP inhibitors, and multiple promising compound classes have been identified. For PMX, these include the peptidomimetic 49c (Pino et al., 2017), the aminohydantoin CWHM-117 (Nasamu et al., 2017), and a series of cyclic guanidines, with one, WM382, inhibiting both PMIX and PMX (Favuzza et al., 2020; Hodder et al., 2022; Lowe et al., 2022). The recently identified peptidomimetic 7k displayed subnanomolar potency along with high bioavailability and significant protection in mouse models (Kovada et al., 2023). The multistage, multitarget properties of several ASP inhibitors add to their overall potential therapeutic value.

Several Cryptosporidium proteins have been recently found to be secreted during or following invasion (Dumaine et al., 2021; Guérin et al., 2023; 2021; Rodrigues et al., 2025; Ao et al., 2025). Interestingly, several of these proteins contain aspartyl protease cleavage motifs that are required for their export to the host cell during intracellular growth (Dumaine et al., 2021; Rodrigues et al., 2025; Ao et al., 2025). It is thus likely that, similar to Plasmodium, targeting ASPs in Cryptosporidium is a viable avenue for drug development and one that potentially could benefit significantly from repurposing (Dogga et al., 2017; Nasamu et al., 2017; Richardson et al., 2022).

The Cryptosporidium genome encodes five putative ASPs, cgd1_2240, cgd1_3690, cgd4_2190, cgd6_660, and cgd6_3820 (Shea et al., 2007; Dogga et al., 2017; Šnebergerová et al., 2021) and phylogenetic analyses suggest these Cryptosporidium ASPs to be homologous to Toxoplasma and Plasmodium ASPs. Here, we focus on CpASP2, which we find the parasite to express close to the completion of its intracellular development. We engineer a conditional mutant and loss of CpASP2 blocks parasite growth in vitro and in vivo and we use this mutant to define the essential role this enzyme plays in the development of this parasite.

RESULTS

CpASP1 and CpASP2 are candidate therapeutic targets that are expressed in multiple lifecycle stages

The recent experience of the effort to develop new antimalarials suggests that targeting functions required at multiple stages of the parasite lifecycle enhances the potency of candidate treatments (Forte et al., 2021). We therefore were most interested in Cryptosporidium ASPs that are expressed in both asexual and sexual stages of the parasite lifecycle. The C. parvum genome encodes five presumptive ASP that we will refer to as CpASP1-5 (cgd1_2240; CpASP1, cgd1_3690; CpASP2, cgd4_2190; CpASP3, cgd6_660; CpASP4, and cgd6_3820; CpASP5). To understand the timing of their expression, we queried the C. parvum single-cell mRNA-sequencing atlas (Walzer et al., 2024). Pseudotime analysis for each gene across the asexual and sexual phases is shown in Figure 1A. CpASP1 and CpASP2 displayed transcriptional peaks in both asexual and sexual stages (Figure 1B). We initially focused on the highly expressed CpASP2.

FIGURE 1:

CpASP2 is expressed in both mature asexual and male stages. (A) Pseudotime analysis of CpASP1-5 gene expression throughout the C. parvum in vivo lifecycle, divided into asexual, female, and male gene clusters. (B) Uniform manifold approximation and projection (UMAP) of CpASP1 and CpASP2 transcript abundance from single-cell RNA-sequencing data (Walzer et al., 2024). Transcriptional start of the lifecycle, asexual region, and branching male and female arms are indicated. (C) The percentage of HA-positive <8N, 8N, female (F), or male (M) parasites quantified from micrographs across two experiments, each data point represents counts from one field of view. ****p<0.0001, one-way ANOVA. Data are shown as mean ± SD. (D) Representative confocal micrographs of CpASP2-HA–tagged 8N meronts and male gamonts. VVL in red, Hoechst in blue, HA in green; scale bar, 2 µm. (E) 1N and 8N CpASP2-HA–tagged parasites visualized by expansion microscopy. Hoechst in blue, NHS-ester in gray, HA in green; scale bar, 5 µm (after expansion).

To establish whether transcript abundance is an accurate prediction of protein expression, we modified the CpASP2 gene by Cas9-mediated homologous recombination to append a C-terminal hemagglutinin (HA) epitope tag onto the protein. Immunofluorescence labeling was performed using the resulting CpASP2-HA expressing parasites at 24 and 53 h postinfection. At 24 h, parasites are in the asexual phase, in which intracellular development and multiple mitosis result in stages with 1, 2, 4, and finally 8 nuclei (N) as they mature. We observed staining in all 8N meronts; in earlier stages, staining was weaker or absent (Figure 1C). Quantification at this timepoint found 96.9% of 8N parasites expressed the HA tag compared with 28.6% <8N parasites, demonstrating a significantly higher percentage in mature than immature asexual stages (p<0.0001). At 53 h, the culture is largely dominated by male or female sexual stages (Tandel et al., 2019). Male gamonts are identified by the presence of 16 bullet-shaped nuclei. Female gametes harbor a single, large nucleus with dilute Hoechst staining, and they are uniquely recognized by an antibody to the female-specific protein DMC1 (Jumani et al., 2019). CpASP2-HA was readily detectable in male stages (94.3% of males displayed HA staining) at 53 h but not in females (Figure 1C). We conclude that protein abundance is consistent with transcript abundance (Figure 1, A and B) and that the protein is abundant in mature meronts and mature male gamonts.

By confocal microscopy, CpASP2-HA staining appeared as punctate dots proximal to the nucleus, as shown in Figure 1D for both 8N meronts and male gamonts. To better define this intracellular localization, we used ultra-expansion microscopy (U-ExM). The sample is anchored to a polymer hydrogel and physically expanded by osmotic force to enhance resolution (Gambarotto et al., 2021), and we utilized N-hydroxysuccinimide (NHS)-ester that reacts with amide groups to label protein-dense regions. CpASP2 localized to a distinct structure denoted by NHS-ester labeling (Figure 1E), and in early meronts, the HA-tag was visible as a single patch. In mature meronts, the protein consistently labeled the apical face of the nucleus of each individual merozoite. Based on previous electron microscopy (Vetterling et al., 1971), we believe this structure to be the parasite Golgi apparatus; however, we currently lack a molecular marker for this organelle for Cryptosporidium to confirm this assignment.

CpASP2, but not CpASP1, is refractory to gene deletion

CpASP1 and CpASP2 both displayed multistage expression (Figure 1A), and we next tested whether they are required for parasite infection and growth. To do this, we attempted to ablate each gene individually. A nanoluciferase/neomycin resistance cassette (Nluc/Neo^R) was introduced into the respective gene loci to replace the coding sequence using CRISPR Cas9-directed homology repair (Supplemental Figure S1A). Ifnγ^−/− mice were infected with electroporated parasites, and mice were monitored for the emergence of stable transgenic parasites by measuring fecal luminesce over the course of the infection. We detected a signal for ∆CpASP1 from parasites from day 9 that persisted above a threshold of 1000 relative luminescence units until day 19 of infection (Figure 2A). Oocysts were purified from feces and correct integration of the construct into the CpASP1 locus was confirmed by PCR mapping (Supplemental Figure S1B). We conclude that CpASP1 is not essential to parasite infection. In contrast, luminescence readings of mice infected with parasites transfected to yield ∆CpASP2 mutants remained below the threshold of detection (Figure 2A). Attempts to target CpASP2 in this manner were repeated four times using two different guide RNA constructs, but none yielded successful transfection and measurable infection under paromomycin selection. Our inability to disrupt the CpASP2 locus suggests that this gene is required for parasite growth.

FIGURE 2:

Ablation of CpASP1 and conditional deletion of CpASP2 using rapamycin-inducible DiCre. (A) Fecal luminescence measurements from mice infected with parasites transfected with ∆CpASP1- and ∆CpASP2-targeting constructs. (B) Genetic strategy to replace a region of the CpASP2 gene that encodes the catalytic domain region with a recodonized sequence, including an HA epitope tag and flanking loxP sites (black triangles). The internal loxP site was placed within an intron. Nanoluciferase (Nluc) and neomycin resistance (Neo^R) markers were inserted downstream of the gene. The resulting CpASP2_loxP transgenic was crossed with a DiCre parasite (Shaw et al., 2024), and dual drug selection was used to isolate CpASP2_loxP DiCre line. (C) PCR on WT and CpASP2_loxP genomic DNA. Control (+) primers amplified the thymidine kinase gene, 5′ and 3′ integration regions indicated in B. (D) PCR on CpASP2_loxP parental and CpASP2_loxP DiCre lines using primers (from B) that span the floxed region. Addition of rapamycin leads to excision as indicated by the diagnostic 0.4 kbp band. (E) Representative micrograph of CpASP_loxP DiCre parasites treated with vehicle or rapamycin for 24 or 48 h. Loss of HA staining (green) is apparent at both timepoints. VVL in red, Hoechst in blue; scale bar, 5 µm. (F) The percentage of HA-positive parasites quantified from E. Each data point represents counts from one field of view. ****p<0.0001, one-way ANOVA. Data are shown as mean ± SD.

As the CpASP2 gene was refractory to disruption, we next sought to ablate gene function using a conditional approach utilizing a rapamycin-inducible Cre recombinase strategy (Tandel et al., 2023; Shaw et al., 2024). We first engineered a floxed strain in which two-thirds of the CpASP2 gene, including one of its predicted catalytic domains, were replaced with a recodonized version of CpASP2 flanked by loxP sites (Figure 2B). The 5′ loxP site was inserted within an artificial intron, and the 3′ loxP was inserted downstream of the CpASP2 stop codon. A neomycin resistance cassette and an Nluc marker were also introduced to select for transgenic parasites and track parasite growth, respectively. Transgenic CpASP2_loxP parasites were recovered, purified from feces, and targeting was confirmed by PCR using parasite genomic DNA (Figure 2C). Next, we used a genetic cross to introduce a rapamycin-inducible dimerizable cre (DiCre) recombinase by coinfection of Ifnγ^−/− mice with the paromomycin-resistant CpASP2_loxP and a DiCre-expressing BRD7929-resistant line (C.p. DiCre) (Shaw et al., 2024). Recombinant parasites were selected by dual drug treatment of infected mice, and oocysts of the CpASP2_loxP DiCre progeny were isolated (Supplemental Figure S2).

In this line, we tested whether rapamycin treatment led to CpASP2 gene excision. Cultured human ileocecal adenocarcinoma (HCT-8) cells were infected with CpASP2_loxP DiCre parasites and grown in the presence of rapamycin or ethanol vehicle for 24 h. Genomic DNA was extracted from the cultures, and the floxed region was amplified by PCR. In vehicle-treated cultures, we noted the expected three Kbp floxed amplicon, while rapamycin treatment resulted in conversion to a 0.4 Kbp excised version (Figure 2D). To test when the protein is lost upon rapamycin treatment, we performed immunofluorescence analysis on vehicle- or rapamycin-treated parasites grown in culture for 24 and 48 h. The HA-tag was lost at the 24-h timepoint upon rapamycin treatment but remained detectable in vehicle-treated organisms (Figure 2, E and F). We conclude that excision is efficient and complete after 24 h of rapamycin treatment, and that leaky excision in the vehicle control is detectable but remains rare.

Loss of CpASP2 blocks progression through the parasite lifecycle

To determine whether deletion of CpASP2 influenced in vitro parasite fitness, we assessed the growth of CpASP2_loxP DiCre upon rapamycin treatment. CpASP2_loxP DiCre were grown in triplicate HCT-8 cultures in the presence of rapamycin or vehicle. At 48 h, luciferase activity was measured as a readout for parasite growth, and rapamycin treatment led to a 3.6-fold reduction when compared with vehicle (p<0.0001; Figure 3A). To control for the potential toxicity of rapamycin, the parental CpASP2_loxP line was treated in parallel. In the parent, rapamycin did not significantly affect parasite growth. Both lines were also treated with nitazoxanide (NTZ), a drug used to target Cryptosporidium (Amadi et al., 2002). NTZ treatment inhibited the growth of both lines to the same degree. We also measured CpASP2_loxP DiCre parasite growth by counting parasites from micrographs of vehicle- and rapamycin-treated cultures at 48 h. Rapamycin treatment led to a 4.2-fold reduction in parasite number (p<0.0001; Figure 3B), matching the luminescence-based measurements.

FIGURE 3:

CpASP2 deletion stalls parasite development at the 8N asexual stage. (A) Luminescence readings of CpASP2_loxP and CpASP2_loxP DiCre parasites grown in HCT-8 culture for 48 h in the presence of vehicle or rapamycin. Data points represent the average of three technical replicates, n = 3 independent experiments. ns not significant, ****p<0.0001, two-way ANOVA. Data are shown here (and in B, E, F, and H) as mean ± SD. (B) Quantification per field of view (FOV) of the number of CpASP2_loxP DiCre parasites grown in vitro for 48 h across three experiments. Data points represent one field of view. ****p<0.0001, t test. (C) UMAP lifecycle atlas from Walzer K. et al. highlighting cell populations that displayed significant differences in gene expression upon deletion of CpASP2. CpASP2_loxP DiCre was compared with CpASP2_loxP parental control at 48 h postinfection. Both samples were treated with rapamycin for 24 h before RNA extraction. Colors indicate cells with increased (up; red) or decreased (down; blue) transcript abundance in the CpASP2_loxP DiCre line. (D) Representative micrographs of CpASP2_loxP DiCre parasites at 48 h postinfection and treatment demonstrating the absence of sexual stages upon rapamycin treatment. Sexual stages denoted by DMC1 staining (red; females) and 16 nuclei (nuclear Hoechst labeling; males); scale bar 5 µm. (E) Quantification of sexual and asexual stages as the percentage of total parasites from D. n = 3, *p<0.05, t test. (F) Size of vehicle- and rapamycin-treated parasites measured by segmentation of VVL-stained cultures. Each data point represents the area of a single parasite, n = 3 independent experiments, p<0.0001, t test. (G) Representative confocal micrographs of vehicle- and rapamycin-treated cultures 24 h after infection and treatment. Nuclei (N) numbers labeled on Hoechst (blue) image. VVL in red, HA in green; scale bar 5 µm. (H) Quantification of CpASP2_loxP DiCre parasites by nuclear number as the percentage of all parasites 24 h after infection and treatment. n = 3, ****p<0.0001, two-way ANOVA. (I) U-ExM of CpASP2_loxP DiCre at 24 h postinfection and vehicle- or rapamycin-treatment. Hoechst in blue, NHS-ester in gray. C, conoid; RN, rhoptry neck; RB, rhoptry bulb; DG, dense granules; N, nucleus; scale bar, 5 µm (after expansion).

To determine at which point in their development parasites are affected by loss of CpASP2, we performed RNA sequencing to compare the transcriptome of parasites with and without the gene. CpASP2_loxP DiCre or CpASP2_loxP parasites were used to infect HCT-8 cultures in triplicate. At 24 h, rapamycin was added to all cultures, and after an additional 24 h, cultures were released from the plate, and RNA was extracted from the total cell lysate. RNA sequencing was used to measure transcript abundance that was normalized for each of the triplicate samples and aggregated into a log fold-change for all detected genes.

The CpASP2 transcript was significantly reduced (−3.5 log₂ fold change; see Supplemental Table S1 for a list of all differentially expressed genes) in the CpASP2_loxP DiCre line compared with the parental, consistent with ablation of the gene. Figure 3C shows the abundance pattern of those genes either up (red) or down (blue) modulated by loss of CpASP2, mapped onto the single-cell mRNA lifecycle atlas. Compared with the parental control, CpASP2_loxP DiCre parasites displayed a decrease in transcripts correlating to male and female clusters, and an increase in transcript in asexual clusters (Figure 3C). This suggested loss of progression to sexual stages upon CpASP2 deletion. We did not observe significant upregulation of any protease that may compensate for the loss of CpASP2. We do note a decrease in the expression of CpASP5 (Cgd6_3820; -6.7 log2 fold change); however, this enzyme is expressed in a female-exclusive manner, and its loss thus reflects the loss of sexual stages in the mutant due to developmental arrest already in the asexual phase of the lifecycle.

To validate this finding, we quantified the percentage of asexual and sexual parasites in rapamycin-treated cultures by microscopy. Males were identified as 16N parasites and females using the female stage marker DMC1 (Figure 3D), and the total number of parasites was determined by VVL staining. At 48 h postinfection, in vitro parasite cultures are typically made up half of asexual and half of sexual stages (Tandel et al., 2019). This ratio was reflected in the vehicle-treated CpASP2_loxP DiCre cultures (Figure 3E). However, rapamycin treatment leading to the deletion of CpASP2 significantly reduced the number of sexual stages (p<0.05). In these cultures, parasites were almost exclusively asexual forms. This suggested that gene ablation of CpASP2 prevents differentiation to sexual stages or alternatively, halts parasite progression in the asexual cycle, before the development of sexual stages.

We next sought to identify the point at which parasite development stalled. CpASP2_loxP DiCre parasites were treated with vehicle or rapamycin and imaged at different times. We noted the accumulation of larger parasites upon rapamycin treatment, which is shown in Figure 3F for the 24 h infection timepoint. Based on this finding, we hypothesized that rapamycin-treated samples might accumulate mature meronts. To determine the maturity of asexual parasite stages, we stained their nuclei using the DNA dye Hoechst. As parasites advance through the asexual merogony cycle, they progress through four mitotic divisions, increasing the number of nuclei from one to eight before merozoite egress and subsequent invasion of new host cells. At 24 h postinfection vehicle-treated cultures displayed a mixed population of stages ranging in size and number of nuclei (Figure 3G). However, rapamycin-treated cultures consisted largely of parasites with 8 nuclei. Quantification demonstrated that rapamycin treatment resulted in significantly more 8N stages and fewer 1N stages when compared with those treated with vehicle (Figure 3H; p<0.0001).

We also imaged cultures 24 h after invasion and rapamycin treatment by U-ExM. Upon rapamycin treatment, the abundant meronts harbor eight merozoites that appear to be fully developed, each containing the typical set of secretory organelles, including dense granules, micronemes, and a rhoptry as visualized by NHS-ester staining (Figure 3I). We conclude that loss of CpASP2 arrests the parasite development at the very end of the merogony cycle, and it is this block of asexual replication that prevents their progression to the sexual phase.

CpASP2 is required for egress of mature asexual stages

The accumulation of fully formed 8N parasites upon deletion of CpASP2 led us to hypothesize that merozoites may be unable to egress. To test this, we next followed living CpASP2_loxP DiCre parasites by time-lapse imaging. Cultures were treated with vehicle or rapamycin upon infection and imaged from 4 to 31 h postinfection. In vehicle-treated control samples, we observed two generations of meronts, with parasites experiencing growth and egress, invasion of new host cells, and a second round of growth and egress. However, upon rapamycin treatment, the parasites did not undergo this second round of egress, but instead, individual merozoites can be observed forming within the vacuole that remains intracellular, again leading to an accumulation of mature meronts (Figure 4, A and B; Supplemental Videos S1 and S2).

FIGURE 4:

Rapamycin-treated CpASP2_loxP DiCre parasites are unable to egress naturally or in response to calcium ionophore. (A) Individual frames from live video imaging and of CpASP2_loxP DiCre parasites treated with vehicle (Supplemental Video S1) or rapamycin (Supplemental Video S2). Timestamps denote the time postinfection (h:min). Closed arrows denote egressing parasites, yellow arrows the resulting extracellular merozoites; scale bar, 5 µm. Gray is differential interference contrast (DIC). (B) Quantification from A of the percentage of CpASP2_loxP DiCre parasite egress during round 1 (4–14 h) and 2 (14–28 h) of intracellular replication. Each data point represents the average of onewell, imaged across four independent movies. ns, not significant, ****p<0.0001, one-way ANOVA. Data are shown here (and in D, F, and G) as mean ± SD. (C) Frames from live imaging (Supplemental Video S3) of H2B-mNeon (green; parasite nuclei) parasites egressing upon calcium ionophore treatment. Gray is DIC. Timestamp denotes time after addition of ionophore (min:s). Closed arrows denote egressing parasite; scale bar, 5 µm. (D) Quantification of time to H2B-mNeon parasite egress after addition of calcium ionophore. Each data point represents one egress event, n = 60 egress events. Ionophore treatment was performed on four independent wells for each of two independent experiments. (E) Frames from live imaging of H2B-mNeon (Supplemental Video S4) or CpASP2_loxP DiCre (Supplemental Video S5) parasites imaged between 24 and 28 h postinfection. Both lines treated with rapamycin for the duration of the infection, and ionophore immediately before imaging. Gray is DIC, red is propidium iodide. Closed arrows denote egressing parasites. Timestamp denotes time postionophore treatment (min:s); scale bar, 10 µm. (F) The percentage of 8N H2B-mNeon or CpASP2_loxP DiCre that egress 24 h after invasion and rapamycin-treatment. Each data point represents the average of one field of view, across two independent experiments, ****p<0.0001, t test. (G) Frequency of host cell nuclei that attain PI staining over the course of the time-lapse experiment (E) upon addition of ionophore. Each data point represents the count from one field of view (FOV), across two independent experiments, ****p<0.0001, t test.

Apicomplexan egress and motility are calcium-dependent processes, and in several parasite systems, egress can be induced by facilitating calcium ion (Ca²⁺) flux across membranes and into the cytoplasm from extra- and intracellular stores using calcium ionophores (Endo et al., 1982; Glushakova et al., 2013). To establish whether an ionophore could trigger Cryptosporidium egress in a similar manner, we optimized a Cryptosporidium ionophore-induced egress assay using live imaging. In contrast with Toxoplasma that is ready to egress at essentially all times of its intracellular development (Caldas et al., 2007; Tomita et al., 2009), Cryptosporidium is only able to leave the cell at the very end of its merogony cycle. To focus our egress assay on parasites physically able to respond, we engineered and used a parasite line expressing mNeon fused to histone H2B (H2B-mNeon) that marks parasite nuclei (Wallbank et al., 2024). We infected cultures with these parasites and, after 24 h, treated with 2 µM of the calcium ionophore A23187 and followed the parasites by time-lapse microscopy imaging every 20 s (Figure 4C; Supplemental Video S3). Only 8N parasites egressed (Supplemental Figure S3), which did so 3.24±1.62 min following exposure to ionophore (Figure 4D), and 41% of all 8N meronts responded by egress within a 15 min observation window. We next measured ionophore-induced egress in rapamycin-treated CpASP2_loxP DiCre parasites. Only 0.8% of these parasites egressed upon induction (Figure 4, E and F; Supplemental Videos S4 and S5). As a control, we also assayed rapamycin-treated H2B-mNeon parasites with a wild-type (WT) ASP2 locus. 57.7% of 8N meronts of this genetic background responded to ionophore with egress. We also monitored egress by adding propidium iodide (PI) to the culture medium. Breakage of the host cell membrane allows the dye access to the host nucleus, resulting in robust fluorescence. Rapamycin-treated cultures infected with CpASP2 mutants showed significantly fewer labeled nuclei than controls (p<0.0001; Figure 4G). We conclude that in the absence of CpASP2, parasites form mature meronts; however, these do not yield a new generation of extracellular merozoites by natural egress, and they are unresponsive to ionophore treatment. This suggests CpASP2 activity to be required for a critical element of the egress cascade.

CpASP2 is required for male gamete egress

CpASP2 is transcribed and translated in male gametes (Figure 1). The development of male gamonts yields 16 male gametes that egress to fertilize female gametes. We wondered whether CpASP2, similar to its role in meronts, may also be required for the egress of male gametes. To test this, CpASP2_flox DiCre-infected cultures were again treated with vehicle or rapamycin; however, this time only after 48 h of culture to sidestep the early block in meront egress, and cultures were fixed and imaged at 90 h postinfection. The number of male gamonts decreases at later times of culture due to the egress of the gametes they contain, while female stages remain intracellular, and their number remains stable from 48 h onwards (Tandel et al., 2019; English et al., 2022). We quantified male and female stages at 90 h postinfection by immunofluorescence microscopy as a proxy assay for male egress (Figure 5A) and found similar numbers of females in cultures treated with vehicle or rapamycin (Figure 5B). However, the number of male gamonts was significantly higher in rapamycin-treated cultures (p<0.001; Figure 5C). This suggests that in contrast with vehicle-treated male gametes, those treated with rapamycin remained intracellular, and that CpASP2 function is required for male gamete egress.

FIGURE 5:

CpASP2_loxP DiCre parasites treated with rapamycin at 48 h from males that are unable to egress. (A) Representative micrographs of CpASP2_loxP DiCre parasites treated with rapamycin at 48 h postinfection and imaged at 90 h. DMC1 (red) denotes females, the presence of 16 parasite nuclei (Hoechst; blue), and closed arrows denote males (M). VVL (green) stains all parasites; Hoechst also stains host nuclei; scale bar, 5 µm. (B and C) Quantification of micrographs from A demonstrating frequency of females (B) and males (C) per field of view (FOV) across two experiments performed in triplicate. ns, not significant, **p<0.01, t test. Data are shown as mean ± SD.

Ablation of CpASP2 in vivo cures infection in mice

To test the importance of CpASP2 in vivo, ifnγ^−/− mice were infected with CpASP2_loxP DiCre parasites and treated with vehicle or rapamycin by oral gavage for 5 d, starting one day before infection. Parasite burden was tracked by measuring fecal luminescence daily for the course of infection. Vehicle-treated mice showed robust, self-limiting infection typical for this model. In contrast, parasite infection did not establish in rapamycin-treated mice, where fecal luminescence activity remained below the limit of detection (Figure 6, A and B). Next, we tested the effect of the removal of CpASP2 on an already established parasite infection. Ifnγ^−/− mice were infected with CpASP2_loxP DiCre parasites and beginning on day 4 were treated with vehicle or rapamycin daily for three days. Fecal luminescence readings were collected at 0, 12, 24, 48, and 72 h following the onset of treatment. All fecal samples from vehicle-treated mice showed luminescence that increased over time (Figure 6C). In contrast, the fecal luminescence of rapamycin-treated mice peaked 12 h after treatment, followed by a rapid decrease. The difference to vehicle-treated mice was highly significant after 48 h (p<0.0001), and at 72 h, infection was no longer detectable in rapamycin-treated animals. We also measured parasite burden directly in the ileal tissue by assessing parasite luciferase activity of gut biopsies from mice killed after 12, 24, 48, and 72 h of treatment. After 48 h of treatment, tissue luminescence was ∼300-fold lower in the rapamycin- compared with vehicle-treated samples (Figure 6D). Finally, we measured parasite burden by microscopy using sections of the ileum (displayed as percentage of cells that are infected; Figure 6, E and F; Supplemental Figure S4). Again, we find a significant reduction within 48 h of the onset of treatment. We conclude that parasites fail to thrive and are rapidly cleared from the intestine upon ablation of CpASP2. Ablation of the CpASP2 gene in vitro led to inhibition of egress of both asexual and male stages. To test whether this is also seen in vivo, we counted Hoechst-labeled parasite nuclei from intestinal sections of treated mice. By 24 h of rapamycin treatment, ileal sections of mice displayed significantly higher proportions of both mature meronts and male stages that for technical reasons were counted collectively as ≥8N stages (p<0.001; Figure 6, E and G). We conclude that CpASP2 gene ablation in vivo blocks egress as previously observed in culture, and that this is followed by a rapid clearance of parasites from the intestinal epithelium.

FIGURE 6:

In vivo Nluc and immunofluorescent assays demonstrate essentiality of CpASP2 for in vivo parasite establishment and growth. (A) Fecal luminescence activity of vehicle- or rapamycin-treated ifnγ^−/− mice infected with CpASP2_loxP DiCre parasites. Each bar represents average of pooled feces from 3 to 4 mice/group. One representative from three independent experiments. (B) Area under curve (AUC) of compiled relative luciferase values. Each data point represents average of one experiment, n = 3 independent experiments. ***p<0.001, t test. Data are shown here (and in C, D, F, and G) as mean ± SD. (C) Fecal luminescence from ifnγ^−/− mice infected with CpASP2_loxP DiCre parasites. Mice treated with vehicle or rapamycin 4 d postinfection. All luminescence values were normalized to 0 h vehicle- or rapamycin-treatment timepoint. Average of pooled feces from 2 to 5 mice/group, n = 2 independent experiments. **p<0.01, ****p<0.0001, two-way ANOVA. (D) Luminescence activity in 5-mm intestinal tissue biopsies from ifnγ^−/− mice infected with CpASP2_loxP DiCre parasites. Mice treated with vehicle or rapamycin 4 d postinfection. All luminescence values normalized to 12 h vehicle- or rapamycin-treatment timepoint. Two mice per group, n = 2 independent experiments. *p<0.05, ****p<0.0001, two-way ANOVA. (E) Representative intestinal sections from CpASP2_loxP DiCre infected ifnγ^−/− mice treated with vehicle or rapamycin at day 4 postinfection and imaged 24 h after treatment. Hoechst in blue, phalloidin (F-actin) in magenta, Sporo-Glo (parasite) in green. Arrows denote parasites, nucleus (N) numbers annotated; scale bar, 10 µm. (F) The percentage of infected cells quantified from stained tissue sections (E) as parasite per host nucleus. Twelve to 72 h postvehicle- or rapamycin-treatment starting 4 d after infection. 2 to 3 mice/group from two independent experiments **p<0.01, ****p<0.001, two-way ANOVA. (G) Quantification of CpASP2_loxP DiCre parasites by nuclear number as the percentage of all parasites 24 h after vehicle- or rapamycin-treatment. We counted 16N male gamonts, 8N male parasites, and 8N meronts as one ≥8N group as those cells are difficult to distinguish rigorously in tissue sections. 2 mice/group, n = 4 independent experiments, *p<0.05, ****p<0.0001, two-way ANOVA.

DISCUSSION

Cryptosporidium is a highly adapted intracellular parasite with a small, streamlined genome. When compared with other related eukaryotes, this parasite has lost most of the genes that typically encode the enzymes required for anabolic metabolism and relies on the host cell to supply the vast majority of the metabolites it needs to grow. Getting in and out of host cells and remodeling them to efficiently extract metabolites are thus critical steps to establish and maintain a successful infection. Significant efforts have been made to understand the process of Cryptosporidium host cell invasion (Smith et al., 2005; Wetzel et al., 2005; Guérin et al., 2023; 2021); in contrast, egress from the host cell has not been studied as well. Although early time-lapse microscopy studies suggested that apicomplexans egressed through simple cell rupture after exceeding the capacity of their host cell (Lund et al., 1961), we now know this to be an active process, driven by complex parasite signaling cascades. These signals center around activation of the cyclic guanosine monophosphate-dependent protein kinase (PKG) and subsequent increase in intracellular Ca²⁺ through release from intracellular stores and uptake (Brochet et al., 2014; Borges-Pereira et al., 2015; Sidik et al., 2016; Vella et al., 2020; Balestra et al., 2021). Ca²⁺ acts on multiple proteins, including calcium dependent protein kinases whose targets mediate egress and motility (Dvorin et al., 2010; Garrison et al., 2012; Lourido et al., 2012; McCoy et al., 2012; Gaji et al., 2015), and the secretion of a perforin-like protein from the micronemes that causes rupture of the membranes of the parasitophorous vacuole (PV) and the host cell plasma membrane (Kafsack et al., 2009; Garg et al., 2013; Roiko and Carruthers, 2013; Roiko et al., 2014). Cryptosporidium encodes a PKG homologue, the silencing of which led to a defect in merozoite egress (Nava et al., 2020). Cryptosporidium also encodes multiple CDPKs at different points in the lifecycle (Etzold et al., 2014; Walzer et al., 2024), with one, CpCDPK5, recently being implicated in male gamete egress (Nava et al., 2024). In this study, we demonstrate Cryptosporidium to be sensitive to Ca²⁺ ionophore, resulting in egress. This response to ionophore, and the presence of both PKG and CDPKs in the genome, suggest that Ca²⁺ likely plays an important role as a second messenger in Cryptosporidium egress, at multiple stages in the lifecycle.

Although Cryptosporidium thus shares biology with other well-studied apicomplexans, there are significant differences. Most obvious among those is its peculiar localization within the host cell; intracellular stages are restricted to the apical brush border in a compartment that extrudes into the intestinal lumen. Several structures unique to Cryptosporidium form at the interface between host and parasites that hold the parasite in place and facilitate metabolite extraction (Guérin et al., 2021). A second differentiating feature is its significantly sped-up lifecycle (Tandel et al., 2019; English et al., 2022). Intracellular development from invasion, through replication to egress takes a mere 12 h (Guérin et al., 2021), and the full developmental cycle of three rounds of asexual replication, gametogenesis, sex and formation of the spore-like oocysts takes little more than 2 d (English et al., 2022), all likely the result of adaptation to a rapid turnover of their intestinal host environment (Haskins et al., 2024; Wallbank et al., 2024). When using time-lapse imaging, we find that consistent with its adhesion to a rigid schizogony cycle, Cryptosporidium is only capable of egress following the maturation of 8 daughter merozoites. This differs from T. gondii tachyzoites that respond to ionophore and egress at any point of intracellular development (Endo et al., 1982) and resembles Plasmodium that similarly replicates by schizogony. U-ExM imaging of intracellular stages clearly visualized the apical secretory machinery and other intracellular organelles within each mature merozoite, which are not present in immature presegmentation 8N meronts. CpASP2 mutants develop into mature, fully segmented merozoites, but do not become motile or egress upon addition of Ca²⁺ ionophore (Figure 4; Supplemental Video S5). We conclude that in the absence of CpASP2, intracellular replication is arrested after cytokinesis and merozoite assembly but before the initiation of motility. This may reflect a failure to assemble or activate the glideosome, an actin myosin-based adhesion and motility complex specific to the apicomplexans that is initiated during egress (Meissner et al., 2002; Keeley and Soldati, 2004; Frénal et al., 2017). In T. gondii, motility and egress require microneme secretion. Among the micronemal proteins that have been linked to egress are Toxolysin 4, a metalloproteinase (Huynh et al., 2021), and PLP1, a pore-forming protein thought to rupture the PV and host cell membrane (Kafsack et al., 2009; Roiko and Carruthers, 2013; Roiko et al., 2014). ASPs frequently act as preprotein maturases, and CpASP2 could act directly or indirectly to activate proteins required for glideosome activation or microneme secretion; however, the molecular targets of the enzyme are still currently unknown. In T. gondii, ASP targets have been identified biochemically using terminal amine isotopic labeling, comparing the N-terminome of WT and ASP KO parasites (Dogga et al., 2017; Coffey et al., 2018). Similar approaches may in the future yield further insight in Cryptosporidium.

Our results suggest CpASP2 to be required for male gamete egress. Merozoites and male gametes are transcriptionally and morphologically distinct (Tandel et al., 2019; Walzer et al., 2024). Merozoites possess apical secretory organelles, similar to those found in the invasive stages of other apicomplexans, and express conserved components of a canonical glideosome (Tandel et al., 2019; Guérin et al., 2023; Walzer et al., 2024). Male gametes, in contrast, possess neither flagellum nor glideosome. Their mechanism of egress and motility remains thus far elusive but appears to require calcium signaling as the male-specific calcium-dependent kinase CpCDPK5 was implicated in gamete egress (Nava et al., 2024). We found that similar to meronts, male gamonts develop to maturity in the absence of CpASP2 but appear unable to egress. Mutant male gametes remained intracellular at a timepoint at which WT males had long egressed. We found visualizing male egress through live imaging technically challenging, but recent work by Nava et al. suggests that male gamonts may respond to ionophore. This highlights the importance of a calcium-induced component of male egress and could be helpful as an approach to further validate our findings. The shared requirement for CpASP2 further suggests commonality in the egress mechanism of merozoites and gametes despite their many differences. Male gametogenesis and fertilization are essential to Cryptosporidium lifecycle progression; this is true not only for transmission but also for continued infection. Remarkably, blocking sex leads to rapid cure of infected animals (Tandel et al., 2019; Walzer et al., 2024). Pharmacological interference with CpASP2 thus would deliver a double-hit by targeting and blocking asexual and sexual replication, both of which are essential.

Significant advances have been made in recent years to develop potent ASP inhibitors to treat infections with different apicomplexan parasites (Dogga et al., 2017; Nasamu et al., 2017; Pino et al., 2017; Hodder et al., 2022; Lowe et al., 2022; Kovada et al., 2023). Based on protein sequence alignments and phylogenetic analysis CpASP2 appears most similar to Plasmodium PMV and T. gondii TgASP5 (Dogga et al., 2017), enzymes responsible for the maturation of proteins translocated by intracellular stages across the PV into the host cell (Boddey et al., 2010; Russo et al., 2010; Sleebs et al., 2014; Hammoudi et al., 2015; Coffey et al., 2018). Cryptosporidium similarly translocates proteins into the host cell, and these proteins are proteolytically processed (Dumaine et al., 2021; Rodrigues et al., 2025; Ao et al, 2025). However, the timing of CpASP2 expression at the very end of intracellular development does not match that of the exported proteins and makes it unlikely that CpASP2 functions as the critical maturase of these exported effectors. Instead, the function of CpASP2 appears more similar to that of P. falciparum PMIX and PMX and T. gondii ASP3. The mechanism of PMX has been well described, and pharmacological and genetic interference with PMX-targeting leads to an arrest similar to the one we observed for the CpASP2 mutant (Nasamu et al., 2017; Pino et al., 2017; Favuzza et al., 2020). PMX localizes to the exonemes (Nasamu et al., 2017), Plasmodium-specific secretory organelles, that release their content into the PV upon egress (Yeoh et al., 2007). At the exonemes, PMX processes the prodomain of the subtilisin-like protease SUB1, transforming it into an active form (Nasamu et al., 2017; Mukherjee et al., 2023; Withers-Martinez et al., 2024). After release into the PV, this catalytically active SUB1 cleaves additional substrates, including other proteases such as SERA6, initiating a coordinated cascade that leads to PV membrane rupture, followed by erythrocyte membrane permeabilization, and merozoite egress (Thomas et al., 2018). It is possible that CpASP2 acts through a similar mechanism, potentially activating substrates at the Golgi, but where exactly its activity unfolds is still unknown in Cryptosporidium. Cryptosporidium encodes two subtilase family enzymes (cgd6_4840 and cgd2_3660) that show 36% and 31% amino acid identity with Plasmodium SUB1, respectively. It also expresses several Plasmodium SERA6 homologues, with one, cgd7_2850, displaying a similar expression pattern to CpASP2, with greatest transcript abundance in late-stage asexuals and males (Walzer et al., 2024), making it a plausible candidate target. The role of both CpASP2 and PMX in egress and the presence of elements of the PMX signaling cascade in Cryptosporidium suggest that compounds developed to target PMX might prove effective against Cryptosporidium as well. Repurposing of antimalarial compound libraries and leads for the development of anti-cryptosporidiosis therapeutics has already been shown to be a highly productive strategy (Baragaña et al., 2019; Vinayak et al., 2020; Manjunatha et al., 2024; 2017). Cryptosporidium ASPs provide extensive opportunity for future repurposing, building on the extensive chemical matter and experience gathered in targeting ASPs in P. falciparum. We demonstrate CpASP2 to be essential for both asexual and sexual development in vitro and in vivo, making it a promising target. Other aspects of Cryptosporidium are likely to require the activity of additional CpASPs expressed at different times in the lifecycle. This provides a rich field for future research to develop additional enzymes and compounds to advance much-needed treatments for this devastating disease.

MATERIALS AND METHODS

Mice

Ifnγ^−/− mice (stock no: 002287; RRID:IMSR_JAX:002287) were purchased from Jackson Laboratory and maintained as a breeding colony at the University of Pennsylvania. Mice ranged from 4 to 10 wk and were age and sex matched within each experiment. All animal experimentation was approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania (protocols #806292 and #805045).

Plasmid and repair template construction

All primers used in this study can be found in Supplemental Table S2. For CpASP1 and CpASP2 gene replacement Nluc-Neo^R repair templates were amplified using primers with homology to the CpASP1 and CpASP2 loci. To generate the CpASP2_loxP plasmid, a recodonized gene fragment was synthesized (IDT) from base pair 904 to the 3′ end of the CpASP2 gene, excluding the STOP codon. The 5′ loxP-intron and 3′ loxP fragments were amplified from the intron-containing plasmid (Tandel et al., 2023). Due to the sequence similarity of loxP sites, plasmids containing each loxP were generated separately and ligated together using PacI and BlpI (New England Biolabs) restriction enzyme cloning. The CpASP2_loxP repair template was amplified from the resulting plasmid using primers with homology to the CpASP2 locus. To generate pCas9-guide plasmids, primers coding guide sequences were annealed with the Cas9 plasmid (Vinayak et al., 2015).

Parasite transgenesis, crossing, and gene deletion

WT C. parvum oocysts were purchased from Bunchgrass Farms. Transgenic parasites were derived as previously described (Vinayak et al., 2015). For selection of neomycin-resistant transgenics, mice were treated with 16 mg/ml paromomycin in the drinking water from day 1 after infection. For dual neomycin- and BRD7929-resistant parasite selection, mice were treated with paromomycin in the drinking water and with 10 mg/kg/day BRD7929 by oral gavage for 7 d, from day 4 postinfection (as per Shaw et al., 2024). Transgenics were propagated in ifnγ^−/− mice and oocysts purified from feces using sucrose flotation and a subsequent cesium chloride gradient as described previously (Vinayak et al., 2015; Sateriale et al., 2020).

To delete CpASP2 in vivo, mice were treated with 10 mg/kg rapamycin resuspended in ethanol, or an equivalent volume of ethanol for vehicle control, by oral gavage. Mice were treated either for 5 d, starting 1 day before infection, or on day 4, 5, and 6 of infection, depending on experiment type.

Parasite genotyping and rapamycin-induced excision PCR

To test for correct integration of transgenic constructs, parasite genomic DNA (gDNA) was extracted from 100 mg of fecal material from infected mice using the Quick-DNA Fecal/Soil Microbe Kit (Zymo Research) and used for PCR genotyping.

To test for CpASP2 gene excision in cultured parasites, 200,000 oocysts were excysted and used to infect HCT-8 cells in 24-well plates and incubated in the presence of ethanol or 100 nM rapamycin for 24 or 48 h. gDNA was extracted using the QIAamp DNA Mini Kit (Qiagen). To test for in vivo excision, gDNA was extracted from fecal material as described. PCR was performed on these extracted gDNA samples.

Measuring parasite burden from cultured parasites, feces, and tissue biopsies by luciferase assay

For measuring parasite burden in tissue culture, 5000 oocysts were excysted and used to infect HCT-8 cells in 96-well plates in quadruplet with 100 nM rapamycin, or an equivalent volume of ethanol vehicle. At 48 h, wells were scraped and incubated in Nluc luciferase lysis buffer and mixed in a 1:1 ratio with Nluc luciferase substrate/Nluc luciferase assay buffer (1:50) in a white-bottom 96-well plate. Relative luminescence was read using a Promega GloMax Plate Reader.

For fecal Nluc assays, 20 mg of mouse feces was suspended in Nluc luciferase lysis buffer and luminescence was measured as described for in vitro samples. Feces for each cage (group) were pooled, as previous work showed that mice within the same cage show similar level of infection (Manjunatha et al., 2017). For intestinal biopsy readings, tissue biopsies were taken using a 5 mm biopsy punch from the most distal region of the ileum of killed mice and incubated with 1 ml Nluc luciferase lysis buffer for 10 min, shaking at 2000 rpm. Luciferase assays were performed on this lysate as for previous samples.

Histological analyses

To prepare intestinal sections, mice were killed and the ileum removed and “Swiss-rolled” as described previously (Wallbank et al., 2024). Tissue was fixed overnight at 4°C in 10% neutral buffered formalin (Thermo Fisher Scientific), followed by 30% sucrose overnight. Tissue was frozen in OCT solution (Sakura Finetek) and cryo-sectioned. For immunostaining, sections were thawed and incubated in PBS Triton 0.1% BSA 1% (hydration buffer), followed by blocking in PBS Triton 0.1% BSA 10% for 45 mins each. Antibody staining was carried out in hydration buffer. Slides were mounted with Fluoro-Gel mounting medium (Thermo Fisher Scientific).

U-ExM

U-ExM was performed based on (Guérin et al., 2023). A total of 500,000 oocysts were excysted and used to infect HCT-8 cells grown on coverslips. After 24 h, coverslips were fixed in 4% paraformaldehyde (Electron Microscopy Sciences) before overnight incubation in 1.4% formaldehyde (Sigma-Aldrich)/2% acrylamide (Sigma-Aldrich) at 37°C. Gelation of cells was performed by adding a mix of 90% monomer solution (18% sodium acrylate (Sigma-Aldrich), 10% acrylamide (Sigma-Aldrich), 0.1% N,N’-methylenbisacrylamide (Sigma-Aldrich) in PBS, 5% tetramethylethylendiamine (10%), and 5% Ammonium persulfate (10%) to the coverslip and incubating on ice for 5 min followed by 1 h incubation at 37°C in a humid chamber. Gels were released from coverslips and subsequently incubated at 95°C for 90 min, in denaturation buffer (200 mM SDS, 200 mM NaCl, 50 mM Tris in water, pH 9). Expansion was achieved with subsequent rounds of incubation in dH₂O, followed by shrinking in PBS, blocking in 3% BSA, and staining in primary antibody overnight. Gels were then incubated in secondary antibody for 2.5 h before re-expansion and imaging by adhering to a poly-D-lysine (Life Technologies) coated, 35 mm glass-bottom imaging dish (MatTek).

Immunofluorescence staining and imaging

For visualization of the HA-tag in cultured parasites, infected HCT-8 cells grown on coverslips were fixed with 4% paraformaldehyde (Electron Microscopy Sciences), followed by methanol for 3 min and subsequent permeabilization in 0.25% triton. For all other staining, the methanol step was omitted. Primary antibodies used were rat anti-HA (Sigma-Aldrich) for cultured cells and U-ExM at 1:500, rabbit anti-HA (Cell Signaling Technology) for intestinal sections at 1:500, and mouse anti-DMC1 (courtesy of Dr. Christopher Houston (Jumani et al., 2019)) at 1:10. Secondary antibodies used were goat anti-rat or goat anti-mouse Alexa 488, 594, or 647 (Thermo Fisher Scientific) at 1:1000 for coverslips, 1:250 for tissue sections, and 1:500 for U-ExM. Additional staining reagents used were the parasite-specific lectin VVL-fluorescein (Vector Laboratories) at 1:1000, Sporo-Glo-fluorescein 20x (Waterborne), phalloidin conjugated to Alexa-Fluor 647 (Thermo Fisher Scientific) at 1:500 and Hoechst 33342 (Thermo Fisher Scientific) at 1:5000 for coverslips, 1:2000 for tissue sections, and 1:1000 for U-ExM. NHS-ester Atto 647N (Sigma-Aldrich) was used for U-ExM at 1:500. Coverslips were mounted using Vectashield (Vector Laboratories) or Fluoro-Gel (Thermo Fisher Scientific).

Images of fixed samples were taken using a Leica DM6000B Widefield Fluorescence Microscope (coverslips) or a Leica STELLARIS confocal microscope (coverslips, tissue sections, and U-ExM gels). Image deconvolution was performed using Leica LAS X software. Image segmentation was performed on VVL signal using Imaris and ImageJ for quantification of parasite number and parasite area. Parasite nuclear number and sexual stage frequencies were counted using Hoechst and DMC1 signal. All other image processing and visualization were performed using Imaris and ImageJ.

Time-lapse imaging and ionophore-induced egress assays

For natural and ionophore-induced egress time-lapse imaging, 150,000 oocysts were excysted and used to infect an 18-well glass-bottom imaging µ-slide (Ibidi) seeded with HCT-8 cells. For visualizing natural egress, imaging was performed 4 to 32 h after infection in phenol red-free, Leibowitz's L-15 medium (Life Technologies) 1% calf serum (Thermo Fisher Scientific) on an OMX SR Delta Vision microscope with automatic stage and environmental chamber using the DIC channel. Z-stacks were taken every 15 min. Samples were maintained at 37°C and 5% CO₂ throughout the imaging process. Images were deconvoluted and aligned using OMX Delta Vision software. To perform ionophore-induced egress, media were replaced with warm Ringer's buffer (NaCl 155 mM, KCl 3 mM, MgCl₂ 1 mM, Na₂HPO₄ 3 mM, HEPES 10 mM, CaCl₂ 1 mM, Glucose 10 mM) supplemented with 2 µM A23187 ionophore and 1 µg/ml propidium iodide (PI). Egressing parasites were visualized on a Leica STELLARIS confocal microscope, taking images every 10 to 20 s for a total of 15 min after addition of ionophore and PI. Movies were generated, and parasite egress events and PI⁺ cells were counted using Imaris and ImageJ.

RNA-sequencing and transcriptional analysis

A total of 500,000 CpASP2_loxP or CpASP2_loxP DiCre parasites were excysted in triplicate and used to infect 3-wells per strain of a 24-well plate seeded with HCT-8 cells. Parasites were incubated for 24 h before the addition of 100 nM rapamycin to all wells. RNA was extracted 48 h after infection (24 h after addition of rapamycin). Samples were lysed in RLT lysis buffer (Qiagen), 1% β-mercaptoethanol, and passed through a QIAshredder column (Qiagen) and RNA-extracted using the RNAeasy kit (Qiagen).

The sequencing library was prepared with Illumina Stranded mRNA-seq kit. Final QCed library was sequenced on Illumina, a P2 100-cycle flowcell with the NextSeq 2000 platform.

Raw reads were processed through the nf-core/rnaseq pipeline (version 3.17.0). Briefly, total reads across all samples have their strand orientation inferred with RSeQC (version 5.0.2; Wang et al., 2012) and Salmon (version 1.10.3; Patro et al., 2017). Reads are then quality plus adapter trimmed, and length filtered with Trim Galore (version 0.6.5). Human host reads are removed with BBSplit (version 39.1 using hg38 reference genome. Ribosomal RNA is filtered out with SortMeRNA (version 4.3.7; Kopylova et al., 2012). Transcript mapping and quantifications are done with STAR (version 2.7.11b; Dobin et al., 2012) and Salmon (version 1.10.1; Patro, 2017) using C. parvum Iowa II reference (CryptoDB, release 65). In addition, we used T2T C. parvum Iowa II BGF reference (Baptista et al., 2025) to incorporate the updated noncoding RNA transcript. Picard (version 3.1.1; broadinstitute.github.io/picard/) is used to remove optical duplicated reads, and SAMtools (version 1.21; Danecek et al., 2021) is used to keep uniquely aligned reads.

Filtered quantification table from nf-core/rnaseq pipeline is used for pair-wise differential expression analysis through the nf-core/differentialabundance pipeline (version 1.5.0; Ewels et al., 2020). PCA (Principal component analysis) and MAD (median absolute deviation) analysis are performed with R (version 4.1.3, www.R-project.org) to detect sample outliers. DESeq2 is used for differential expression analysis (version 1.34.0; Love et al., 2014).

Differentially expressed gene lists from DESeq2 are used for input into single cell RNA analysis with previously published C. parvum single cell atalas with Seurat (version 5.3.0; Hao et al., 2023). Briefly, normalized RNA-assay values for the up- and down-regulated gene lists were extracted from the Seurat object. Each gene's expression profile was multiplied by its corresponding log₂ fold-change coefficient, and these weighted values were summed across all genes to produce a per-cell score. The sum of log₂FCs was then used to normalize each cell's weighted expression, yielding two metadata vectors, UpWeightedExpr and DownWeightedExpr, which were initialized to NA for cells without overlapping genes. Finally, UMAP embeddings were generated with Seurat, using red (aggregated up-regulated) and blue (aggregated down-regulated) to highlight cells enriched for corresponding gene signatures.

Code and data availability

All scripts and commands used for analysis are available through github repo https://github.com/ruicatxiao/CpASP2_KO_Seq/tree/main. RNAseq data are available through BioProject PRJNA1274990.

Supporting information

Supporting Video 1. Movie S1.

Video 1 and 2: Rapamycin-treated CpASP2_loxP DiCre parasites are unable to egress. Live imaging of CpASP2_loxP DiCre parasites treated with vehicle (Video 1) or rapamycin (Video 2). Timestamps denote the time post infection (h:min). Gray is DIC.

Supporting Video 2. Movie S2.

Video 1 and 2: Rapamycin-treated CpASP2_loxP DiCre parasites are unable to egress. Live imaging of CpASP2_loxP DiCre parasites treated with vehicle (Video 1) or rapamycin (Video 2). Timestamps denote the time post infection (h:min). Gray is DIC.

Supporting Video 3. Movie S3.

Video 3: Calcium ionophore induces Cryptosporidium merozoite egress. Live imaging of H2B-mNeon (green; parasite nuclei) parasites egressing upon calcium ionophore treatment. Gray is DIC. Timestamp denotes time after addition of ionophore (min:s).

Supporting Video 4. Movie S4.

Video 4 and 5: Rapamycin-treated CpASP2_loxP DiCre parasites are unable to egress in response to calcium ionophore. Live imaging of H2B-mNeon (Video 4) or CpASP2_loxP DiCre (Video 5) parasites imaged between 24 and 28 h post-infection. Both lines treated with rapamycin for the duration of the infection, and ionophore immediately prior to imaging. Gray is DIC, red is propidium iodide. Timestamp denotes time post ionophore treatment (min:s).

Supporting Video 5. Movie S5.

Video 4 and 5: Rapamycin-treated CpASP2_loxP DiCre parasites are unable to egress in response to calcium ionophore. Live imaging of H2B-mNeon (Video 4) or CpASP2_loxP DiCre (Video 5) parasites imaged between 24 and 28 h post-infection. Both lines treated with rapamycin for the duration of the infection, and ionophore immediately prior to imaging. Gray is DIC, red is propidium iodide. Timestamp denotes time post ionophore treatment (min:s).

Abbreviations used:

ASP

aspartyl protease

Ca²⁺

calcium ion

DiCre

Dimerizable Cre

DMC1

DNA meiotic recombinase 1

H2B

histone 2B

HA

hemagglutinin

HCT-8

human ileocecal adenocarcinoma

Ifnγ

interferon γ

N

nucleus

Neo^R

neomycin resistance

NHS

N-Hydroxysuccinimide

Nluc

nanoluciferase

NTZ

nitazoxanide

PI

propidium iodide

PKG

cyclic guanosine monophosphate–dependent protein kinase

PM

plasmepsin

PV

parasitophorous vacuole

U-ExM

ultra-expansion microscopy.