COWPs localise to the oocyst wall and suture

To determine if COWP2–9 are oocyst wall proteins, CRISPR/Cas9 was used to generate reporter strains. Strains were engineered to have both a fluorescent protein and epitope tag, either mNeon-3 × HA or mScarlet-I-3 × myc (Fig 1A). Because COWPs contain an N-terminal signal peptide sequence, strains were designed as C-terminal fusions. Reporter strains were successfully generated for all cowps (S1–S4 Figs). We were unable to recover mutants where COWP3–5 were tagged at the C-terminus, therefore an alternative tagging strategy was employed where a second copy of these cowps was introduced at a non-essential locus (Fig 1B). The corresponding promoters (entire upstream intergenic region) and open reading frame for cowp3–5 were cloned into the tagging constructs (mNeon-3 × HA). CRISPR was used to insert these expression cassettes at Cpimpdh (S2 Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Cryptosporidium Oocyst Wall Protein (COWP) family are confirmed oocyst wall proteins.

A) CRISPR/Cas9 was used to produce strains of Cryptosporidium where COWP2, 6-9 are individually fused at their C-terminus to a fluorescent protein, either mNeon (yellow) or mScarlet-I (purple). B) Expression of COWP3-5 tagged at the C-terminus with mNeon were driven by their native promoter and targeted for integration at the Cpimpdh locus. C) Live microscopy of wild type Cryptosporidium parvum oocysts. D-L) Fluorescence microscopy of live oocysts confirms that COWPs localise to the oocyst wall and/or suture. Images collected, as indicated, using a Deltavision widefield epifluorescence microscope, or on a Zeiss LSM880 Airyscan microscope either in confocal or super resolution mode (Airyscan mode). Representative images shown; contrast adjusted individually for each image and not normalised. Scale bar for all images is 5 µm. Image collection parameters for each image: C) widefield epifluorescence microscope, single z-plane, exposure time 0.08 msec, laser 50%. D) confocal, projection of 29 z-planes, exposure time 0.01 msec, laser power 2%. E) super resolution, projection of 20 z-planes, exposure time 0.038 msec, laser power 2.2%. F) confocal, projection of 13 x-planes, exposure time 0.02msec, laser power 2% G) widefield epifluorescence microscope, single z-plane, exposure time 0.5 sec, laser 50%. H) widefield epifluorescence microscope, projection of 12 z-planes, exposure time 0.5 sec, laser 50%. J) super resolution, 39 z-planes, exposure time 0.043 msec, laser power 2.3%. K) confocal, projection of 10 z-planes, exposure time 0.02 msec, laser power 2%. L) confocal, projection of 16 z-planes, exposure time 0.02 msec, laser power 2%.

https://doi.org/10.1371/journal.ppat.1013561.g001

Oocysts were purified from faecal samples and examined live by fluorescence microscopy (Fig 1C). COWP5–9 localise across the oocyst wall (Fig 1D–H) in a similar pattern as previously observed for COWP1 [23]. Additionally, immunoelectron microscopy confirmed localisation of COWP6 and COWP8 to the oocyst wall (S5 Fig). Fluorescence microscopy indicates that COWP8 oocysts contain a few very small puncta in the space between sporozoites and the oocyst wall that are not associated with the residual body (Fig 1E). These were also observed by immunoelectron microscopy as globules of wall material secreted by the zygote that did not incorporate into the inner oocyst wall (S5C Fig). Fluorescent signals for COWP5, COWP7 and COWP9 were observed to be of lower intensity relative to COWP6 and COWP8 (Fig 1F–H). The expression level apparent in these images are consistent with published transcriptomics data indicating that cowp1, cowp6, and cowp8 are the most abundant members of this family (S6 Fig) [23,24]. As the fluorescence signal of COWP5-mNeon was particularly dim, a second reporter strain was generated where the promoter of cowp1 is used to drive expression of cowp5 (S7A–C Fig). Live fluorescence microscopy further confirms that COWP5 is localised to the oocyst wall (S7D Fig).

Notably, COWP2–4 localise to the oocyst wall, but more specifically their localisation pattern is consistent with the suture, in both shape and length. COWP4 is located at the suture and in small puncta, similar to the structures observed in the COWP8 strain (Fig 1J). Unlike the other COWPs that localise evenly across the oocyst wall (COWPs 1,5,6,7,8,9), COWP3 localises across the oocyst wall but is highly concentrated at the suture (Fig 1K). COWP2-mNeon localises to the suture, but also to a focus corresponding to the residual body (Fig 1L).

To test whether localisation of COWP2-mNeon to the residual body was a mislocalisation due to fusion with the mNeon protein, a second transgenic strain was generated where COWP2 is fused to the epitope tag alone, COWP2-HA (S1A Fig). COWP2-mNeon and COWP2-HA oocysts were excysted, labelled with anti-HA antibody and inner oocyst wall antibody (1E3; Kerafast) and imaged. In excysted oocysts of both tagged strains, COWP2 is localised in the residual body and present at the suture (S8 Fig). This confirms that COWP2 localisation is unlikely to be an artefact of the bulky tag but truly localises to the residual body. Further, COWP2-HA was only detected in excysted oocysts, suggesting that COWP2 is an inner wall protein. The relatively low protein abundance of the other COWPs prevented observation of labelling by immunoelectron microscopy.

Application of genetics and microscopy confirms that all members of the cowp family are true oocyst wall proteins. Further, we discovered that COWPs 2–4 specifically localise to the Cryptosporidium oocyst suture, providing the first markers for this structure.

COWP6 and COWP8 are expressed by macrogamonts exclusively and are secreted after fertilisation to form the oocyst wall

To determine which parasite life cycle stages express cowps, reporter strains were cultured in vitro or in vivo and expression of cowps was observed using fluorescence microscopy. We focussed on COWP6 and COWP8 as expression was high enough to confidently observe by fluorescence microscopy. COWP5 was not attempted; COWP7 and COWP9 were examined, but expression could not be visualised by fluorescence microscopy, likely due to low expression levels.

We lack markers for discrete life cycle stages and use shape and number of nuclei to differentiate life cycle stages. Expression of COWP6 and COWP8 was undetectable in asexual stages including single nuclei trophozoites (S9A Fig) and eight-nuclei meronts (S9B Fig). Males (microgamonts), which have up to 16 bullet-shaped nuclei, also do not express COWP6 or COWP8 (S9C Fig).

COWP6 and COWP8 expression was exclusively observed in females, called macrogamonts (Fig 2A and 2C). Macrogamonts have a single nucleus and synthesise the components of the oocyst wall as they mature [23]. These components are stored in organelles called “wall forming bodies” (WFB) which have a punctate pattern [22]. Previously, COWP1 was localised to wall forming bodies by immunoelectron microscopy [22]. Recent examination of COWP1-mNeon parasites identified a similar punctate localisation consistent in size, shape, and number with WFBs [23]. COWP6 and COWP8 reporter strains produce the same WFB-like localization pattern. Prior to fertilisation (characterised by a single large nuclei), macrogamonts parasites express COWP6 (Fig 2A) and COWP8 (Fig 2C) in numerous puncta of a similar, and small size. After fertilisation and meiosis (four nuclei observed) COWPs are secreted to form the oocyst wall (Fig 2B and 2D).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. COWPs are expressed by macrogamonts and are secreted to form the oocyst wall.

Super resolution microscopy of ileal tissue collected from IFN-γ knockout mouse infected with COWP6-mScarlet-I (A and B) or COWP8-mNeon (C and D) strains. Mice were culled at peak infection and tissue was processed for histology. Macrogamonts are identified by size (~ 5 µm) and a single nucleus; oocysts are identified similarly by size and presence of 4 nuclei. Brush border (F-actin stained with phalloidin-AlexaFluor-647, white) and nuclei (DNA stained with Hoechst, cyan). A) Prior to fertilisation, COWP6 (purple) is localised in numerous puncta in the female parasites, resembling wall forming body organelles. B) After fertilization, COWP6 is secreted to form the oocyst wall. C) COWP8 (yellow) is localised in puncta similar in size and number to COWP6, and D) after fertilisation is secreted to form the oocyst wall. Small puncta of COWP8-mNeon remain after oocyst wall formation, consistent with live microscopy of oocyst purified from faecal material (reported in Fig 1E). Images collected on a Zeiss LSM880 Airyscan microscope, Airyscan mode. Representative images shown; contrast adjusted individually for each image and not normalised. Scale bar for all images is 5 µm. Image collection parameters for each image: A) projection of 5 z-planes, exposure time mScarlet-I 0.028 msec and laser 2%, exposure time Alexa-647 0.024 msec and laser 0.5%, exposure time Hoescht 0.010 msec and laser 3%. B) projection of 10 z-planes, exposure time mScarlet-I 0.028 msec and laser 2%, exposure time Alexa-647 0.024 msec and laser 0.5%, exposure time Hoescht 0.010 msec and laser 3%.C) projection of 9 z-planes, exposure time mNeon 0.023 msec and laser 2.4%, exposure time Alexa-647 0.020 msec and laser 0.5%, exposure time Hoescht 0.022 msec and laser 3%. D) projection of 17 z-planes, exposure time mNeon 0.020 msec and laser 2.4%, exposure time Alexa-647 0.020 msec and laser 0.5%, exposure time Hoescht 0.022 msec and laser 3%.

https://doi.org/10.1371/journal.ppat.1013561.g002

Cowp8 is not essential for parasite survival, infection, or transmission

Cowp8 is a highly expressed member of the cowp family, at levels similar to cowp1. From transcriptomics data, cowp8 is in the 99% percentile of genes expressed by macrogamonts and is slightly higher in expression than cowp1 (bulk RNASeq from female parasites; S6 Fig) and is the last cowp to be expressed during female maturation (single cell RNASeq) [23,24]. When measured by proteomics, abundance of COWP8 is only slightly lower than COWP1. To understand the role of cowps in construction of the oocyst wall, infection, and parasite transmission, we used CRISPR/Cas9 to generate parasites lacking cowp8. The repair cassette was designed to replace cowp8 (cgd6_200) with an mScarlet-I fluorescent protein and the NLuc-Neo^R fusion protein, driven by a strong, constitutive promoter (Fig 3A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. COWP8 is not essential for infection or transmission.

A) Strategy to replace the full open reading frame of cowp8 (cgd6_200) with a repair cassette where mScarlet-I and NanoLuciferase-Neomycin resistance fusion protein (NLuc-Neo^R) is expressed by the constitutive Enolase promoter. The same gRNA (black arrow) that was used to target the C-terminus for fluorescent fusion and the same downstream region of 50 bp of homology (grey) was used to target the gene for deletion (S4A Fig). B) PCR with primer pairs indicated in (A) was performed using genomic DNA extracted from wild type and ∆cowp8. C) Whole genome sequencing of wild type (yellow) and ∆cowp8 (blue) confirms gene deletion. Histogram illustrates sequencing coverage mapped at the cowp8 locus. Confocal microscopy of live wild type (D and E) and ∆cowp8 (F and G) confirms mScarlet-I cytoplasmic expression of ∆cowp8 strain (magenta); scale bar 5 µm. H) Super resolution microscopy of tissue collected from interferon-gamma knockout mouse infected with ∆cowp8 culled at peak infection and processed for histology. Brush border (F-actin stained with phalloidin, white) of ileal villi (DNA stained with Hoechst, cyan) are highly infected with ∆cowp8 (mScarlet-I expressed in the parasite cytoplasm, purple). Images collected on a Zeiss LSM880 Airyscan microscope, (Airyscan mode for P). Representative images shown. J-L) Experimental design for passage of ∆cowp8 strain in interferon-gamma knockout mice. J) Initial passage where inoculum is wild type sporozoites transfected with CRISPR machinery to target cowp8 for deletion, K) second passage infected with slurry made of faecal material from first passage, L) and third passage infected with oocysts purified from the second passage. Infection level of COWP8-mNeon (white circles) and ∆cowp8 strains (black circles) is similar and follows typical acute infection pattern in M) first, N) second, and P) third passages. Infection level of mice as determined by faecal NLuc, limit of detection at 500 RLU/mg, dotted line. Average and SD of three technical replicates of one biological replicate.

https://doi.org/10.1371/journal.ppat.1013561.g003

Diagnostic PCR and whole genome sequencing confirmed that parasites lacking cowp8 were successfully generated (Fig 3B and 3C). Using confocal microscopy Δcowp8 oocysts display no difference in morphology from wild type (Fig 3K–N). Parasite burden ofΔcowp8-infected tissue was observed by fluorescence microscopy. Intestinal tissue was collected from mice at peak infection levels (determined by faecal NLuc) and imaged. High levels of infection were observed along the length of the villi (Fig 3P). Δcowp8 parasites were viable in vivo (Fig 3D–J) and parasite shedding followed a typical acute pattern, indistinguishable from the pattern observed for passage of the COWP8-mNeon reporter strain (Fig 3G and 3J). Infection was unaffected by type of inoculum; oral gavage with either a faecal sample (Fig 3E) or purified oocysts (Fig 3F) both generated robust infection.

Although cowp8 is not required for infection of immunocompromised mice under laboratory conditions, it could be required for survival in the environment. Oocysts are resistant to treatment with chlorination and can survive exposure to high levels of bleach. ∆cowp8 oocysts were assessed for their ability to survive treatment with bleach. COWP8-mNeon and ∆cowp8 oocysts were treated with 8% bleach, washed thoroughly, and used to infect mice. No difference in infection levels was observed for ∆cowp8 parasites (S10A Fig and S10B), confirming that oocysts remain resistant to chlorination despite loss of cowp8. In addition to their resistance to disinfectants, Cryptosporidium oocysts can persist and survive for considerable lengths of time. Oocysts stored below 15 °C remain infectious even after six months [25]. Parasite infectivity begins to be greatly compromised only when oocysts are stored at temperatures above 37 °C [5,26]. We tested the effect of heat treatment on oocysts lacking cowp8. Treatment at 60 °C for 5 minutes was sufficient to render both COWP8-mNeon and ∆cowp8 non-infectious (S10A Fig and S10C).

Loss of cowp8 does not alter oocyst morphology

We performed electron microscopy to visualise changes in oocyst wall morphology that may not be visible by light microscopy. General morphology was observed using scanning electron microscopy (SEM). Transmission electron microscopy (TEM) was used to visualise changes to the inner and outer layers of the wall and to measure the thickness of these layers. By SEM, wild type and ∆cowp8 cannot be distinguished (Fig 4A and 4B). When quantified, both wild type and ∆cowp8 oocysts have similar proportions of oocysts appearing ‘wrinkled’ or ‘smooth’ (Fig 4C). ‘Wrinkles’ are an artefact of the resiliency of the oocyst from sample processing for SEM. This analysis is consistent with confocal microscopy which failed to identify morphological changes (Fig 3K–N). Cross sections of wild type and ∆cowp8 oocysts across two biological replicates were prepared for TEM (Fig 4D and 4E), and the thickness of the oocyst wall was measured. Despite loss of the abundant COWP8, Δcowp8 oocyst walls were around ~12nm thicker than wild type (WT median wall width = 50.8nm and Δcowp8 = 62.9nm; Fig 4F data plotted by biological replicate in S11 Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. COWP8 does not alter robustness of oocyst wall.

Representative field of view of A) WT and B) Δcowp8 oocysts imaged using Scanning Electron Microscopy (SEM). ‘Wrinkled’ oocysts are circled in white and ‘Smooth’ oocysts are outlined with orange squares. Scale bars are 10 µm. C) Percentage of oocysts that appear wrinkled (black) or smooth (orange) quantified. n = 1787 WT oocysts (1514 wrinkly (84.7%) and 273 smooth (15.3%)). n = 593 for Δcowp8 oocysts (357 wrinkly (60.2%) and 236 smooth (39.8%)). Raw data in S5 Table. Representative Transmission Electron Microscopy was performed on D) wild type and E) Δcowp8 oocysts. Scale bars are equivalent to 1µm. F) Median width of oocyst wall. Error bars, IQR; ***p < 0.001 (Mann Whitney test). Raw data reported in S6 Table. Statistical tests performed on data grouped by biological replicate are reported in S11 Fig.

https://doi.org/10.1371/journal.ppat.1013561.g004

It was surprising to find that Δcowp8 oocysts maintain their thick wall. To investigate if parasites can compensate for loss of COWP8, for example by upregulating other members of the COWP family, oocysts were analysed by quantitative mass spectrometry. Wild type and ∆cowp8 oocysts were excysted, proteins were extracted, samples were labelled with tandem mass tags (TMTs) and then analysed using quantitative mass spectrometry. The amount of the other COWPs is not significantly changed upon loss of COWP8 (S12 Fig and S7 Table).

Cowp8 mutants retain mechanical strength

To determine if there are changes to oocyst wall strength upon loss of COWP8, we measured ∆cowp8 oocyst walls using a biomechanical approach called nanoindentation. This approach is commonly used to measure the hardness and elastic modulus of biological materials [27]. Oocysts were immobilised on a petri dish and probed under a microscope. The probe approaches the sample at a constant speed, generating a force-displacement curve as the probe indents the sample with a defined indentation depth (5 µm). Young’s Modulus, which is the mechanical property measuring stiffness of a material by defining the relationship between stress and strain, can be obtained by fitting force-displacement curves to a corresponding mathematical model.

Oocysts were immobilised on a dish and subjected to nanoindentation using a Chiaro nanoindentor. From raw force displacement curves (S13A Fig), force indentation plots were created using average force measurements obtained from the first 100 nm of indentation (S13B Fig). Young’s Modulus was calculated by fitting the force indentation plots (Fig 5A) to the Hertz mathematical model for both wild type and ∆cowp8 oocysts. No difference in stiffness was observed and wild type and ∆cowp8 oocysts were determined to have an approximate average Young’s Modulus of 1.2 MegaPascals (MPa; Fig 5B). This is in range with previous measurements of the strength of Toxoplasma gondii oocyst walls at 10⁶ - 10⁷ MegaPascals (both sporulated and unsporulated oocysts measured) [28].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Oocysts lacking COWP8 maintain mechanical strength.

A) Force indentation plot in range of 0-100 nm of indentation. Two technical replicates performed for wild type (purple and lilac; data from third wild type replicate was determined to be an outlier, see S13 Fig) and ∆cowp8 (blue, orange, and green). B) Young’s modulus was obtained from fitting the indentation plot to the Hertz mathematical model, producing box and whisker plots (10-90 percentile). Statistical analysis carried out on obtained data (Welch’s t-test) concluded that there is no significant difference in mechanical strength between wild type and ∆cowp8 oocysts (ns meaning p > 0.05), with an average Young’s Modulus of 1.2 MegaPascals (MPa) being calculated for both wild type and ∆cowp8 oocysts.

https://doi.org/10.1371/journal.ppat.1013561.g005

Animal ethics

Mice randomly assigned to cages at the time of weaning. Both sexes were used for experiments described here. As required animals were sex and aged matched. All animal studies were ethically reviewed and carried out in accordance with Animals (Scientific Procedures) Act 1986 and the Dundee University Policy on the Care, Welfare, and Treatment of Animals. Regulated procedures on living animals were approved by the University’s Ethical Review Committee and carried out under the authority of project and personal licenses issued by the Home Office under the Animals (Scientific Procedures) Act 1986, as amended in 2012 (and in compliance with EU Directive EU/2010/63). The ERC has a general remit to develop and oversee policy on all aspects of the use of animals on University premises and is a subcommittee of the University Court, its highest governing body.

Materials

Wild type Cryptosporidium parvum oocysts (Iowa strain) were purchased from Bunchgrass Farms (Deary, Idaho, USA) and stored at 4°C in PBS with penicillin/streptomycin. Oligonucleotides produced by either the University of Dundee Oligonucleotide Synthesis Service or Sigma-Aldrich (S1 Table). mScarlet-I described throughout is codon optimised for C. parvum (GeneScript; sequences available in S2 Table).

Molecular cloning and DNA sequencing

PCR performed using high-fidelity DNA Polymerase (Primestar Max, Takara) and cloning with NEBuilder HiFi DNA Assembly (NEB). DNA Sequences confirmed via sequencing by Genewiz, University of Dundee MRC PPU Reagents and Services, or Plasmidsaurus. All plasmids created for this work listed in S2 Table with links available for DNA sequences.

Construct used for C-terminal tagging with mNeonGreen-3xHA previously described, referred to as mNeon throughout [29]. A second construct was created with mScarlet-I fused to a 3 × c-myc epitope tag followed by an NLuc-Neo^R fusion under the control of constitutively expressed CpEnolase promoter to generate “pLIC-mScarlet-I-3xc-myc_aldo 3´utr_ENNE”. Construct for deletion of Cpcowp8 contains mScarlet-I, 2A linker, and an NLuc-Neo^R fusion, all of which is under the control of the CpEnolase promoter to generate “EnoP_mScarlet-I-ENNE”.

Due to the location of the gRNA at the C-terminus, the last 129 bp of COWP4 was cloned into a tagging construct fused directly to the mNeon tagging construct. The codon for cysteine at amino acid position 845 was mutated from “TGC” to “TGT” to prevent further cutting by Cas9 to generate “COWP4 C-term-mNeon-3×HA_ENNE”. Similarly, the last 114 bp of COWP9 was cloned into a tagging construct fused directly to an mNeon-3 × HA. The codon for lysine at amino acid position 1180 was mutated from “AAG” to “AAA” to prevent further cutting by Cas9 to generate “COWP9 C-term-mNeon-3×HA_ENNE”.

To express COWP3, COWP4, COWP5 under their endogenous promoter from an ectopic location, the entire intergenic region upstream of the corresponding cowp gene and its full open reading frame (except the stop codon) was amplified using PCR from C. parvum genomic DNA. This was cloned to generate “Cowp3P_COWP3ORF-mNeon-3×HA_ENNE,” “Cowp4P_COWP4ORF-mNeon-3×HA_ENNE,” and “Cowp5P_COWP5ORF-mNeon-3xHA_ENNE”.

Guide oligonucleotides were designed for COWPs 2–9 (S1 Table) and ligated into a plasmid containing Cas9/U6 promoter containing plasmid [32] using restriction enzyme cloning as previously described [33]. To generate the Δimpdh::COWP3-mNeon, Δimpdh::COW4-mNeon, and Δimpdh::COWP5-mNeon, a guide for the inosine monophosphate dehydrogenase (IMPDH, cgd6_20) locus was used [34].

CRISPR design

Linear repair cassettes were generated via PCR using PrimeSTAR MAX DNA polymerase (Takara). Fifty bp of homology, typically directly flanking the site targeted by CRISPR, was incorporated on primer.

To generate COWP2-mNeon, cowp2 (cgd7_1800) was targeted by a gRNA located 38 bp upstream of the stop codon. Regions of homology consisting of the 68 bp directly upstream of the stop codon, and 50 bp located 116 bp downstream of the stop codon. A synonymous mutation of serine at position 1373 encoding the PAM (AGG to AGA) was engineered to prevent further targeting of Cas9 upon integration of the repair cassette into the genome.

To attempt to generate COWP3-mNeon and COWP3-HA, cowp3 (cgd4_670) was targeted by a gRNA located 16 bp after the stop codon. Regions of homology consisting of the 50 bp directly upstream of the stop codon, and 50 bp located 177 bp downstream of the stop codon.

To attempt to generate COWP4-mNeon, cowp4 (cgd8_3350) was targeted by a gRNA located 124 bp after the stop codon. Regions of homology consisting of the 179 bp upstream of the stop codon, and 50 bp located 153 bp downstream of the stop codon.

To attempt to generate COWP5-mNeon and COWP5-HA, cowp5 (cgd7_5150) was targeted by a gRNA located 17 bp upstream of the stop codon. Regions of homology consisting of the 71 bp upstream of the stop codon, and 50 bp located 130 bp downstream of the stop codon. A synonymous mutation of threonine at position 623 encoding the PAM (CGG to CTG) was engineered to prevent further targeting of Cas9 upon integration of the repair cassette into the genome.

To generate Δimpdh::COWP3-mNeon, Δimpdh::COWP4-mNeon, Δimpdh::COWP5-mNeon, and Δimpdh::cowp1P-COWP5-mNeon the CpIMPDH locus was targeted using CRISPR as previously described [34].

To generate COWP6-mScarlet-I, cowp6 (cgd4_3090) was targeted by a gRNA located 13 bp downstream of the stop codon. Regions of homology consisting of the 57 bp directly upstream of the stop codon, and 50 bp located 54 bp downstream of the stop codon. A synonymous mutation of serine at amino acid position 536 encoding the PAM (AGC to AGT) was engineered to prevent further targeting of Cas9 upon integration of the repair cassette into the genome.

To generate COWP7-mNeon, cowp7 (cgd4_500) was targeted with a guide RNA located 19 bp upstream of the stop codon. Homology regions of 92 bp located directly upstream from the stop codon, and 50 bp located 56 bp downstream of the stop codon. The PAM was mutated such that proline at position 827 was replaced with alanine.

To generate COWP8-mNeon, cowp8 (cgd6_200) was targeted with a guide RNA found 39 bp upstream of the stop codon. Homology regions of 105 bp located directly upstream of the stop codon, and 50 bp located 286 bp downstream of the stop codon within the 3´ UTR. The PAM was mutated such that glycine at position 452 was replaced with alanine. The same guide RNA and downstream homology was utilised in the knockout strategy. New upstream homology was selected, consisting of 50 bp located 271 bp upstream of the cowp8 gene start codon.

The same gRNA and downstream homology for CpCOWP8 C-terminal tagging was used to generate Δcowp8. New upstream homology was selected, consisting of 50 bp located 271 bp upstream of the CpCOWP8 start codon.

To generate COWP9-mScarlet-I, cowp9 (cgd6_210) was targeted with a guide RNA located 31 bp upstream of the stop codon. Upstream homology was already cloned into the tagging construct. Downstream homology of 50 bp located 266 bp after the stop codon.

Excystation

Wild type C. parvum oocysts were incubated in 4% bleach on ice for 5 minutes and then washed three times with PBS. Treatment with bleach serves to kill residual bacteria and remove faecal contaminants; this also disrupts the “outer veil” which may be observed otherwise without bleach treatment [11]. Oocysts were incubated for 1 hour at 37°C in either 0.2 mM sodium deoxytaurocholate in PBS [33], 0.5% (w/v) taurodeoxycholic acid in 0.25% trypsin in PBS [35], or 0.75% sodium taurocholate (Merck) in RPMI-1640 medium with 1% foetal bovine serum [36] to induce excystation.

Propagation of transgenic strains in immunocompromised mice

IFN-gamma KO mice (Jackson Laboratory #B6.129S7-IfngtmlTS/J, JAX 002287 bred at the University of Dundee; at least 8 weeks old) were given antibiotics in the drinking water (1 mg/ml ampicillin, 1 mg/ml streptomycin, 0.5 mg/ml vancomycin final concentrations) for 1 week prior to infection. Excysted sporozoites were transfected using a 4D Amaxa Nucleofector as previously described [33]. cowp1P_COWP5-mNeon strain transfected using a 4D Amaxa with modifications as described [36]. Mice were gavaged with saturated sodium bicarbonate 5 min prior to infection by gavage with transfected sporozoites. This serves to neutralise stomach acid before gavage with sporozoites and enhance parasite survival and transit to the intestines. Both sexes used to generate new strains of transgenic Cryptosporidium (2 or 4 mice). We have found that two mice are sufficient to generate new strains of Cryptosporidium. This reduction in animals was specifically utilised to produce transgenic COWP2-HA and Δimpdh::COWP3-mNeon, Δimpdh::COW4-mNeon, and Δimpdh::COWP5-mNeon. Paromomycin (16 g/L) was added to the drinking water for the duration of the infection, or as indicated.

To passage transgenic strains, mice were gavaged with faecal slurry or purified oocysts. Mice were treated with selection agents as indicated. Faecal samples were collected post infection as indicated and were pooled from the cage of mice.

Faecal sample analysis

NanoLuciferase assay was performed on 20 mg of homogenised faecal sample (in 1mL of lysis buffer). 25 µL of NanoLuciferase substrate-lysis buffer mixture was added to 100 µL of faecal lysate (GloMax plate reader, Promega). Positive threshold for NanoLuciferase signal is RLU/mg > 500. Oocysts were purified from faecal samples using sucrose and Caesium chloride flotations as previously described [33].

DNA was extracted from either ~100 mg faecal material or purified oocysts. Samples were subjected to minimum of five freeze thaw cycles and then DNA was extracted using either a Zymo Quick-DNA Faecal/Soil Microbe DNA Miniprep Kit (faecal sample) or a Qiagen DNeasy Blood & Tissue Kit (oocysts). DNA was amplified by PCR using primers specific to the 5´ and 3´ regions of the repair cassette integration site. For knockout mutants, a third set of primers designed to amplify within the targeted genes ORF were also used. α-tubulin served as the positive control.

Whole genome sequencing and analysis

In the case of Δcowp8, DNA was extracted from 10 million purified oocysts and sequenced and analysed by University of Glasgow Polyomics. Briefly, library was prepared using NEBNext UItra II FS DNA kit, which uses enzymatic fragmentation. Library was then sequenced pair-ended reads of 100 bp to an average of >5 million reads on an Illumina NextSeq2000. The FASTQ files forward and reverse paired-end reads of Δcowp8 sample were aligned to the reference genome v68 of C. parvum clone IOWAII downloaded from CryptoDB [37] using Bowtie2 version 2.3.5 [38], with the ‘very-sensitive-local’ pre-set alignment option. The alignments were converted to BAM format, reference sorted and indexed with SAMtools version 1.9 [39]. The genome coverage of the aligned reads was extracted from the BAM files using bedtools version v2.29.0 [40] with the -pc option for pair end data and -bg option to output a bedGraph file. The GFF annotation file for C. parvum clone IOWAII was downloaded from CryptoDB and converted to GTF with gffread (https://github.com/gpertea/gffread). The bedGraph file and GTF file were visualized with the svist4get python package version 1.2.24 [41].

Fluorescence microscopy

SEM and TEM microscopy

For routine transmission electron microscopy and immunostained electron microscopy (immunoEM), four million oocysts (wild type or COWP8-mNeon) were bleached and washed three times in PBS (excluding Δcowp8 oocysts). All oocyst samples were fixed overnight at 4 °C in EM-grade 4% paraformaldehyde and either 2.5% (for standard TEM) or 0.2% (for immunogold TEM) glutaraldehyde in PBS. Fixed parasites were sent for processing and imaging at the Cellular Analysis Facility, University of Glasgow.

For routine TEM, oocysts were washed in 0.1 M phosphate buffer, before being post-fixed in 1% OsO₄ and 1.25% Potassium ferrocyanide (vol:vol) for 1 hour on ice in darkness. Oocysts were then washed in distilled water following contrast en bloc with 0.5% aqueous uranyl acetate for 1 hour at room temperature in darkness. Samples were then dehydrated in an ascending series of acetone and embedded in epoxy resin. Ultra-thin sections (60 nm) were cut and collected onto formvar-coated copper grids and contrasted with 4% Uranyl acetate prior imaging using a JEOL 1200 EX transmission electron microscope, operating at 80kV. Δcowp8 oocysts were pooled from passages two, three, five, eight, nine and ten for the first TEM sample preparation. For the second repeat, Δcowp8 oocysts were purified from passage 10 in NOD-SCID mice. Wild type oocysts were used from the same isolation from Bunchgrass Farms in January 2024. Two independent sample preparations were analysed and compared.

For immunoEM, after several washes in the cacodylate 0.1M buffer, the samples were dehydrated in ascending ethanol series and embedded in LR White resin (Agar Scientific). Ultrathin sections (60 nm thick) were obtained using an ultramicrotome (Leica Microsystems). The sections were collected on formvar-coated nickel grids and then blocked in PBS containing 3% bovine serum albumin for 1 hour. After this time, they were incubated in the presence of primary antibody (either a monoclonal mouse anti-c-myc (ProteinTech) or a monoclonal rat anti-HA (SigmaAldrich)). Then they were washed several times in blocking buffer and incubated with 15 nm gold-conjugated Protein A (Aurion). The grids were washed several times in the blocking buffer, dried and contrasted with 4% Uranyl acetate, and observed using a JEOL 1200 EX transmission electron microscope operating at 80kV.

For scanning electron microscopy two million oocysts were bleached and washed three times in PBS. Oocysts were fixed at 4 °C overnight in EM-grade 4% paraformaldehyde and 2.5% glutaraldehyde in PBS. Fixed oocysts were washed in 0.1 M phosphate buffer, then left on top of poly-l-lysine coated glass slides for 20 min to allow oocysts to adhere. Adhered oocysts were washed in 0.1 M phosphate buffer, before being dehydrated in an ascending series of ethanol concentrations and finally reaching critical point dehydration with a Autosamdri-815 machine (Tousimis). Dried coverslips were coated with gold/palladium (10 nm thick layer) and imaged in a JEOL IT-100 scanning electron microscope. A minimum of 10 fields of view for each sample were used for quantitation.

TEM analysis

Levels of grey images were processed with a gaussian blur filter prior to thresholding by eye (Otsu method) to create a binary image of the oocyst wall. Where sporozoite material was attached to the oocyst wall this was manually broken using the pencil tool. The oocyst wall ROI was used to create a mask. A centreline was added prior to running the Voronoi width tool using the MorphoLibJ-plugin [42] and width_profile_tools toolset [2] (available on github; see software section of Methods). This generated the width-profile of the oocyst wall calculated as Voronoi distance from line-segments perpendicular to the centreline of the wall. Line segments were measured every pixel (equivalent to a range of between 58–344 nm dependent on the scale of the image analysed).

Images were excluded from analysis where the whole width of the oocyst wall was not captured completely at the edges, or where accurate thresholding of the image was prevented, either by artefacts or by the oocyst wall touching sporozoite material extensively. For some images, the oocyst wall was split into multiple masks to avoid artefacts. Values across all masks generated from one oocyst were combined for analysis. Where more than one oocyst was captured in an image, width measurements were analysed separately.

In biological replicate one 51 images of wild type oocysts were collected. Of these, 14 whole oocysts were analysed (three images excluded) and 34 zoomed in images were collected and analysed. 23 images of Δcowp8 were collected. 5 whole oocysts and 16 zoomed in images were analysed (three images excluded). In biological replicate two, 10 images of WT and 10 images of Δcowp8 were collected. Three Δcowp8 images were excluded. 10 whole WT oocyst walls and 10 whole Δcowp8 oocyst walls were measured.

In vivo phenotyping experiments

25,000 oocysts were resuspended in PBS and exposed to 60°C in an Eppendorf ThermoMixer. A temperature of 60 °C was chosen, as this temperature is at the lower end of reported temperatures known to heat inactivate Cryptosporidium [43–45]. Samples were then stored at 4 °C overnight.

50,000 oocysts were resuspended in either 0% or 8% undiluted bleach solution and incubated on ice for 5 minutes. Oocysts were washed in PBS three times and then stored at 4 °C overnight.

Age and sex matched interferon-gamma knockout mice (4 mice were cage) were pre-treated prior to infection with 5,000 oocysts per strain by oral gavage.

Nanoindentation

Petri dishes (35 mm × 10 mm, Corning cat no. 430588) were treated with Corning Cell-Tak cell and tissue adhesive (cat no. CLS354240) the day before oocyst attachment. Cell-Tak was added to 0.1 M Sodium bicarbonate solution, applied to 35 mm dish, and incubated for a minimum of 20 min at room temperature. Cell-Tak solution was aspirated and dish washed twice with sterile water. Dishes were left to air dry and stored in at 4 °C.

One million Δcowp8 and WT oocysts in PBS were applied to a dish and incubated overnight at 4 °C. Dishes were gently washed once with sterile PBS and stored in sterile PBS to prevent oocysts desiccation. Nanoindentation measurements and analysis were conducted at the Advanced Research Centre (ARC), University of Glasgow. Nanoindentation measurements were performed using a Chiaro nanoindenter (Optics 11) mounted on top of an inverted phase contrast microscope (Evos XL Core, Thermofisher), following a previously described approach [46]. Measurements were performed at room temperature in PBS. Each sample was indented once with a spherical tip of 3 μm radius at a speed of 2 μm/s over a vertical range of 5 μm. Measurements were performed on oocysts adhered to three separates petri dishes, sampling a minimum of ten oocysts per plate (n = 60–67) and the experiment was repeated twice (N = 2).

The selected cantilever had a stiffness of 4.18 N/m. The collected curves were analysed using a custom Python code [47]. Curves were first aligned using a baseline detection method based on the histogram of the force signal [48] and the corresponding indentation was calculated for each curve. The analysis was performed using the Hertz model for a spherical indenter to fit the curves obtained.

Sample preparation and protein extraction

Samples of wild type oocysts and Δcowp8 oocysts were prepared as two biological repeats (with two technical repeats each) and analyzed using quantitative proteomics. 40 million oocysts per technical repeat were bleached, washed three times in PBS, and excysted at 37 °C in 0.2 mM Sodium taurochlorate for 2 hours. Excysted material was resuspended in 25 µL lysis buffer (4% SDS, 10 mM DTT, 1x protease inhibitor cocktail, all prepared in water). Samples were subjected to five cycles of freeze thaw to excyst remaining oocysts. Samples were incubated at 56 °C for 1 hour while shaking (800 rpm) to solubilise proteins. Samples were centrifuged at 20,000xg for 40 minutes at 4 °C. Supernatant was removed and TCEP (tris(2-carboxyethyl)phosphine) added to a final concentration of 25 mM and incubated at 37 ºC for 10 minutes. Added IAA (iodoacetamide) to a final concentration of 25 mM and incubated at room temperature in the dark for 1 hour. Added 13% volume ice-cold TCA and store at -20 C. Sample was centrifuged at 16,000 xg for 4 °C for 5 minutes. Supernatant was removed and pellet was washed with 0.5 ml cold acetone. Acetone wash repeated a total of five times and then the sample was air dried. Protein extract was resuspended in 400 µl TEAB, sonicated to resuspend the material. Sample was digested with Trypsin/LysC (Promega) overnight at 37 ºC with shaking and then dried.

TMT labelling and high pH reverse phase fractionation

Tryptic peptides (11.6 µg, from each sample) were dissolved in 100 µl of 150 mM TEAB. TMT labelling was performed according to the manufacturer’s instructions (Thermo-Fisher Scientific). The different TMT-10 plex labels (0.8 mg) (Thermo Fisher Scientific) were dissolved in 41 µl of anhydrous acetonitrile, and each label is added to a different sample. The mixture was incubated for 1 hour at room temperature, an equivalent of 0.75 µg of peptides from each sample were mixed with 20 µl 1% formic acid and used to check labelling efficiency. The remaining samples were kept at –80 °C, until the labelling efficiency was checked. Samples were pooled, desalted, and dried in a speed-vac at 30 °C. Samples were re-dissolved in 200 µl Ammonium formate (10 mM, pH 9.5) and peptides were fractionated using High pH RP Chromatography. A C18 Column from Waters (XBridge peptide BEH, 130 Å, 3.5 µm 2.1 X 150 mm, Waters, Ireland) with a guard column (XBridge, C18, 3.5 µm, 2.1 X 10mm, Waters) were used on an Ultimate 3000 HPLC (Thermo-Scientific). Buffers A and B used for fractionation consist, respectively, of (A) 10 mM ammonium formate in milliQ water pH 9.5 and (B) 10 mM Ammonium formate, pH 9.5 in 90% acetonitrile. Fractions were collected using a WPS-3000FC auto-sampler (Thermo-Scientific) at 1minute intervals. Column and guard column were equilibrated with 2% Buffer B for twenty minutes at a constant flow rate of 0.2 ml/min. Fractionation of TMT labelled peptides was performed as follows; 190 µl aliquot were injected onto the column, and the separation gradient was started 1 minute after the sample was loaded onto the column. Peptides were eluted from the column with a gradient of 2% Buffer B to 20% Buffer B in 6 minutes, then from 20% Buffer B to 45% Buffer B in 51 minutes and finally from 45% buffer B to 100% Buffer B within 1 min. The column was washed for 15 minutes in 100% Buffer B. The fraction collection started 1 minute after injection and stopped after 80 minutes (total 80 fractions, 200 µl each). Formic acid (30 µl of 10% stock) was added to each fraction and concatenated in groups of 20 fractions.

LC-MS analysis

Analysis of peptides was performed on a Q-exactive-HF (Thermo Scientific) mass spectrometer coupled with a Dionex Ultimate 3000 RS (Thermo Scientific). LC buffers were the following: buffer A (0.1% formic acid in Milli-Q water (v/v), buffer B (80% acetonitrile and 0.1% formic acid in Milli-Q water (v/v) and loading buffer (0.1% TFA).

Peptides from each fraction were resuspended in 50 µl 1% formic acid and aliquots of 5 μl were loaded at 10 μL/min onto a trap column (100 μm × 2 cm, PepMap nanoViper C18 column, 5 μm, 100 Å, Thermo Scientific) equilibrated in 0.1% TFA. The trap column was washed for 5 min at the same flow rate with 0.1% TFA and then switched in-line with a Thermo Scientific, resolving C18 column (75 μm × 50 cm, PepMap RSLC C18 column, 2 μm, 100 Å) equilibrated in 5% buffer B for 17 min. The peptides were eluted from the column at a constant flow rate of 300 nl/min with a linear gradient from 5% buffer B (for Fractions 1–10, 7% for Fractions 11–20) to 35% buffer B in 125 min, and then from 35% buffer B to 98% buffer B in 2 min. The column was then washed with 98% buffer B for 20 min and re-equilibrated in 5% or 7% buffer B for 17 min. The column was maintained at a constant temperature of 50 ^°C.

Q-exactive HF was operated in data dependent positive ionisation mode. The source voltage was set to 2.85 Kv and the capillary temperature was 250 ^°C. A scan cycle comprised MS1 scan (m/z range from 335-1600, with a maximum ion injection time of 50 ms, a resolution of 120 000 and automatic gain control (AGC) value of 3x10⁶) followed by 15 sequential dependant MS2 scans (resolution 60000) of the most intense ions fulfilling predefined selection criteria (AGC 1x10⁵, maximum ion injection time 200 ms, isolation window of 0.7 m/z, fixed first mass of 100 m/z, spectrum data type: centroid, intensity threshold 5 x 10⁴, exclusion of unassigned, singly and >6 charged precursors, peptide match preferred, exclude isotopes on, dynamic exclusion time of 45 s). The HCD collision energy was set to 32% of the normalised collision energy. Mass accuracy is checked before the start of samples analysis.

Quantitative proteomics analysis of transgenic Cryptosporidium

Proteomic data were processed through MaxQuant software (version 2.4.10.0), leveraging its integrated Andromeda search engine [49]. The search database was specifically constructed for Cryptosporidium parvum Iowa II, with annotated protein sequences obtained from CryptoDB [37], release 61. This was augmented with sequences for the reporter proteins mScarlet-I and NLuc-Neo^R. Additionally, a murine protein database, retrieved from Uniprot [50] on January 20, 2023, was concatenated with the C. parvum database to account for potential host protein contamination. The analysis encompassed both TMT 6-plex and TMT 10-plex labeling, processed in parallel within a single MaxQuant instance. Each TMT experiment was treated as a distinct analytical batch, with no normalization applied between them. Trypsin was designated as the proteolytic enzyme, with a specification for up to two missed cleavages per peptide allowed. Carbamidomethylation of cysteine residues was set as a constant post-translational modification (PTM), whereas N-terminal acetylation of proteins and methionine oxidation were configured as variable PTMs. Default settings were retained for all other parameters within MaxQuant, except for TMT label correction factors. These were adjusted according to the manufacturer’s instructions (Thermo Fisher Scientific) and are detailed within the MaxQuant parameter files. These files, alongside the raw data, have been submitted to the PRIDE database. Data is available under Project accession: PXD050089.

Pre-normalization, we filtered the MaxQuant output to exclude potential confounders. Data were refined by removing entries solely identified by peptide modification sites (only identified by site), entries marked as reverse database matches (reverse), and proteins classified as potential contaminants (potential contaminant). For our analysis of Cryptosporidium parvum COWP8 across wild type (WT) and knockout (KO) specimens, we used a robust normalization technique tailored specifically for TMT proteomic data. Our experimental set consisted of one biological replicate with two technical replicates integrated within a TMT 6-plex configuration, along with an additional biological replicate paired with two technical replicates in a TMT 10-plex setup. To normalize the signal of the two TMT batches, we adopted a modified version of the Internal Reference Standard (IRS) normalization method, as initially delineated by Plubell et al. [51] The IRS normalization was applied by taking the raw mean intensity of each plex to serve as the reference channel. We computed the sum of the intensities for each TMT channel row-wise and then calculated the geometric mean of these sums to establish a stable reference point. This method was chosen to replace the reference channel utilized in [51], which was not generated from our datasets. The scaling factors for normalization were ascertained by dividing this geometric mean by the sum intensities of each individual experiment. Subsequently, each plex underwent individual scaling to ensure uniformity in signal intensities. Upon normalizing the data, we employed Principal Component Analysis (PCA) to verify the efficacy of the IRS process between the TMT 6-plex and 10-plex setups. The differential expression analysis was performed with the limma package [52] using the WT samples versus the KO samples, with log2 values. FDR values were computed with the toptable function in limma. The output table of the analysis is available as attached csv file (S7 Table).

Statistics and software

Graphs and all statistical analyses were prepared using GraphPad Prism version 10.2.1. Mann-Whitney U Test was used to measure differences between median oocyst wall width values between two parasite strains (Fig 4F). Kruskal Wallis multiple comparisons test applied to compare median oocyst wall width values between two biological replicates of the two parasite strains (S11 Fig). Standard t-test was used to measure differences in Young’s modulus between two parasite strains, and Welch’s correction was applied due to differences in sample size (Fig 5B). Oocyst wall width measured using “Width Profile Tools” available at: https://github.com/MontpellierRessourcesImagerie/imagej_macros_and_scripts/wiki/Width-Profile-Tools. Curves generated for biomechanics measurements were analysed using a custom Python code as described above. Proteomic data were processed through MaxQuant software (version 2.4.10.0). The differential expression analysis was performed with the limma package as described above. These files, alongside the raw data, have been submitted to the PRIDE database; data is available under Project accession: PXD050089.