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%.
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%.
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-NeoR) 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.
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.
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.
Fig 6.
Model of COWPs in oocyst wall structure.
A) 3D rendering of the Cryptosporidium oocyst structure, where 4 sporozoites (olive green) and a residual body (red) are tightly packed inside the slightly oval shell of the oocyst wall. Inner and outer walls are sealed closed by the suture (dark green). B) Upon excystation the suture opens, allowing sporozoites to escape often leaving behind the residual body inside the excysted oocyst wall. COWPs 1, 5, 6, 7, 8, 9 localise to the oocyst wall only. COWP3 and COWP4 localise to both the oocyst wall and the suture, while COWP2 is localised to the residual body and the suture. Illustrations by Konstantinos Alexandrou.