Targeted CRISPR Screens Reveal Genes Essential forCryptosporidiumSurvival in the Host Intestine

33 34 The Cryptosporidium parasite is one of the leading causes of diarrheal morbidity and 35 mortality in children, and adolescent infections are associated with chronic malnutrition. 36 There are no vaccines available for protection and only one drug approved for treatment 37 that has limited eKicacy. A major barrier to developing new therapeutics is a lack of 38 foundational knowledge of Cryptosporidium biology, including which parasite genes are 39 essential for survival and virulence. Here, we iteratively improve the tools for genetically 40 manipulating Cryptosporidium and develop a targeted CRISPR-based screening method 41 to rapidly assess how the loss of individual parasite genes influence survival in vivo. Using 42 this method we examine the parasite’s pyrimidine salvage pathway and a set of leading 43 Cryptosporidium vaccine candidates. From this latter group we determined the parasite 44 gene known as Cp23 to be essential for survival, which was confirmed through inducible 45 knockout in vitro and in vivo. Parasites deficient in Cp23 were able to replicate within and 46 emerge from infected epithelial cells, yet unable to initiate gliding motility which is 47 essential for the reinfection of neighbouring cells. The targeted screening method 48 presented here is highly versatile and will enable researchers to more rapidly expand the 49 knowledge base for Cryptosporidium infection biology , paving the way for new 50 therapeutics. 51 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 3

52 53 Diarrhoeal related infections are a major cause of morbidity and mortality in children 54 around the world (Troeger et al., 2017) , (Troeger et al., 2018) . Cryptosporidiosis is 55 consistently found to be one of the leading causes of a moderate -to-severe diarrhoeal 56 disease in infants (KotloK et al., 2013) , (KotloK et al., 2019) . Unlike other diarrheal 57 diseases that are attributed with high incidence rates, such as rotavirus or Shigella, there 58 are no eKective drugs or vaccines for Cryptosporidium. Nitazoxanide, the only Food and 59 Drug Administration approved drug for treatment, is not eKective in 60 immunocompromised individuals, only partially eKective in adults, and not approved for 61 use in children, the patient population that needs intervention the most (Abubakar et al., 62 2007), (Amadi et al., 2009). One reason for this scarcity of therapeutics is a historic lack 63 of eKective systems to study the parasite. While there have been recent improvements in 64 genetic manipulation (Vinayak et al., 2015) , drug target identification (Caldwell et al., 65 2024), (Manjunatha et al., 2024) , (Ajiboye et al., 2024) , and animal models of infection 66 (Sateriale et al., 2019) , our comprehension of Cryptosporidium biology remains very 67 basic. Specifically, there is a limited understanding of the Cryptosporidium genes that 68 contribute to parasite fitness and survival, genes that would be the most suitable targets 69 for therapeutic intervention. 70 71 Reverse genetic approaches have been instrumental in identifying parasite genes that 72 influence survival and convey fitness in other Apicomplexan parasites (Sidik et al., 2016), 73 (Young et al., 2019), (Bushell et al., 2017), (Butterworth et al., 2023), (Smith et al., 2022). 74 Toxoplasma, which has historically served as a facile model for Apicomplexa, has 75 benefited from a high transfection eKiciency coupled with non-homologous end joining 76 (NHEJ) to be at the forefront of Apicomplexa CRISPR screening (Sidik et al., 2016) . Like 77 Cryptosporidium, the Plasmodium parasite lacks NHEJ pathways, so CRISPR -Cas9 78 driven double stranded breaks can only be fixed by homologous repair. This lack of NHEJ 79 can be leveraged to implement precise CRISPR screening, yet to date, no such screening 80

exists in Cryptosporidium. Here, we overcome the current technical barriers for 81 Cryptosporidium genetic manipulation to develop a reproducible method for pooled in 82 vivo CRISPR screens. 83 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 4 84 Cryptosporidium has a highly streamline genome and lacks many basic metabolic 85 pathways. Because of this, it is thought to be heavily reliant on its host cell for nutrients. 86 This is particularly evident in the nucleotide salvage pathway, where Cryptosporidium is 87 believed to be capable of scavenging both nucleotides and nucleotide precursors from 88 the host enterocyte (Pawlowic et al., 2019). Despite a compact genome, both purine and 89 pyrimidine salvage pathways show evidence of redundancy. The purine salvage pathway 90 has been well studied in Cryptosporidium, revealing that many genes in the pathway can 91 be independently eliminated, resulting in little or no deficit in parasite growth (Pawlowic 92 et al., 2019) . Surprisingly, this includes inosine monophosphate dehydrogenase 93 (IMPDH), a gene once considered to be a crucial drug target based on metabolic mapping 94 (Striepen et al., 2004), (JeKeries et al., 2015). The absence of a growth defect in parasites 95 that lack IMPDH suggests that Cryptosporidium can not only synthesise purine 96 nucleotides, but also take them up directly from the infected host cell. Less is known 97 about the pyrimidine salvage pathway in Cryptosporidium, however, there is still 98 evidence of redundancy. Thymidine kinase (TK) and dihydrofolate reductase -thymidine 99 synthase (DHFR -TS) both convert their respective substrates (deoxythymidine or 100 deoxyuridine monophosphate) to deoxythymidine monophosphate. For this reason, TK 101 and DHFR -TS are both independently non -essential for DNA synthesis and parasite 102 growth (Vinayak et al., 2015) , (Pawlowic et al., 2019) . Using our in vivo pooled CRISPR 103 screen, we discovered that most of the pyrimidine salvage genes significantly contribute 104 to parasite fitness within the intestine. 105 106 It is widely accepted that protective immunity can be acquired to Cryptosporidium, as 107 incidences of Cryptosporidium infections decrease with age (Haque et al., 2009), (KotloK 108 et al., 2013) , (Sow et al., 2016) , and experimental infection with attenuated parasites 109 leads to protection in calves and mice (Jenkins et al., 2004) , (Sateriale et al., 2019) . 110 Considering the success of the vaccination campaign for rotavirus (Yen et al., 2014) , 111 (O'Ryan, 2017), a diarrheal pathogen with a similar patient population and pathogenesis, 112 a Cryptosporidium vaccine is predicted to greatly reduce morbidity and mortality in 113 children. Numerous protein antigens have been associated with protection in humans 114 and several prominent surface proteins have been suggested as vaccine candidates 115 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 5 (Gilchrist et al., 2023), (Manque Patricio et al., 2011), (Askari et al., 2016). Yet, if and how 116 many of these proteins contribute to parasite fitness is unclear. Here, we assess the 117 relative fitness contributions during infection for a panel of proposed Cryptosporidium 118 vaccine candidates (collected from the literature). As Cp23 (also known as the 119 Cryptosporidium immunodominant antigen 23) is one of the leading vaccine candidates, 120 and strongly correlated with protection in humans (Gilchrist et al., 2023), we investigated 121 how this protein specifically contributes to parasite virulence and fitness. 122 123

124 125 Iterative improvements to improve Cryptosporidium genetic manipulation. To 126 develop CRISPR screening, we first sought to improve Cryptosporidium transfection 127 eKiciency. To transfect Cryptosporidium, dormant oocysts are treated with bile salts and 128 warmed to body temperature to simulate the conditions of a human intestine. Under 129 these conditions, the environmentally hardy oocysts ‘excyst’ , releasing four motile and 130 transfectable sporozoites that readily invade intestinal epithelial cells. To optimise 131 transfection, we transfected Cryptosporidium parvum parasites with a vector that 132 contained both luminescent ( NanoLuciferase) and fluorescent (mCherry) reporters 133 under constitutive expression and allowed the transfected parasites to infect an 134 intestinal cell (HCT8) monolayer. Numerous electroporation programs were trialled using 135 transient expression, with FL115 demonstrating the highest luminescen ce (Sup. 1a). 136 Furthermore, various combinations of bile salts and incubation media were tested for 137 their eKect on parasite transfection. Notably, incubation in sodium taurocholate led to 138 higher transfection eKiciency compared to the deoxy form, sodium taurodeoxycholate 139 (Sup. 1b). Combined, these adjustments resulted in more than a 50-fold improvement in 140 transfection eKiciency (Sup. 1c & d). 141 142 As Cryptosporidium parasites lack NHEJ, CRISPR driven injury of the genome is required 143 to drive homologous recombination for genetic manipulation. Although this represents a 144 significant barrier to high -throughput screening, the parasite’s lack of NHEJ very likely 145 contributes to its uniquely high level of genetic editing specificity. Indeed, an ‘oK -target’ 146 genomic insertion has yet to be reported by Cryptosporidium researchers. Further, the 147 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 6 parasite predominantly exists in a haploid state during its life cycle and has a minimal 148 requirement of homology for eKicient editing. To determine the smallest reliable 149 allowance for eKicient repair, we tested the eKiciency of editing using diKering length 150 homology arms to integrate a nanoluciferase gene into the dispensable TK locus. A 151 positive correlation between the length of the homology arms and the eKiciency of repair 152 was observed, with 50bp (the longest length tested) being the most eKicient, while no 153 integration was observed in the absence of CRISPR driven injury of TK (Sup. 1e). 154 155 To allow for pooled genetic knockout screens, we reasoned that a one-plasmid approach 156 would be required, where a single plasmid delivers the Cas9 -expression vector, a target 157 specific gRNA, and a segment of DNA to be inserted at the target genomic locus (repair 158 DNA). The repair DNA contains homologous flanks surrounding an expression cassette 159 with the selectable marker and a genetic barcode for identification. Further, to enable 160 high-throughput creation of targeted libraries, we developed a two -step Golden Gate 161 assembly method, where first a 300bp segment containing all gene -specific (unique) 162 DNA is integrated into the Cas9-expression vector. In the second step of the assembly, a 163 non-unique expression cassette is inserted that will integrate into the genome, enabling 164 the use of virtually any combination of selection markers or reporters (Fig. 1a). To fit all 165 unique DNA into a 300bp oligo nucleotide, we used a 50bp gRNA which simultaneously 166 serves as one of the homology sites for genomic integration. To test the feasibility of this 167 approach, we generated a vector targeting TK, a gene that is known to be dispensable for 168 parasite growth. The TK targeting vector was transfected into Cryptosporidium parvum 169 (C. parvum) sporozoites that were used to infect genetically immunocompromised 170 (interferon gamma deficient Ifng-/-) mice and luminescence was measured in mouse 171 faecal material 7 days post transfection, comparable to what is observed when using a 172 separate DNA repair template and a Cas9 -expression vector (Fig. 1b). As we redefined 173 the method to generate Cryptosporidium transgenics, we confirmed the approach’s 174 specificity via whole genome sequencing, and as expected, the only genome alteration 175 observed was at the targeted site within the coding region of the TK gene (Fig. 1c). 176 177 CRISPR KO screen of the pyrimidine salvage pathway. To assess the feasibility and 178 reproducibility of a recombination -based in vivo screen, we designed a pilot screen 179 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 7 targeting 11 genes in the parasite’s pyrimidine salvage pathway (Table. 1). We ran parallel 180 screens employing either 1 or 2 targeting vectors per gene, hence 11 or 22 gRNAs, 181 respectively. Targeted KO vectors were transfected into C. parvum sporozoites, which 182 were propagated in Ifng-/- mice under paromomycin selection (Fig. 1d). Luminescence 183 from the integrated nanoluciferase reporter was detected in mouse faecal material 8 184 days post transfection, in both the 1 and 2 KO vector per gene pyrimidine salvage screens 185 (Fig. 1e) . From the output, faecal material from the infected mice, p arasites were 186 purified, their DNA extracted, and the barcodes amplified via high fidelity PCR. From the 187 input, transfected parasites used to infect the mice, DNA was extracted and the barcodes 188 amplified via high fidelity PCR. Both the output and input barcodes were sequenced and 189 used to calculate fold enrichment scores for each gene. Fold enrichment scores serve as 190 a measure of relative fitness , genes whose barcodes show a negative enrichment are 191 presumed to be important to parasite survival and therefore highly fitness conferring (Fig. 192 1f). Importantly, fold enrichment scores between the 1 and 2 KO vector screens were 193 strongly correlated, achieving an R 2 of 0.842 (Fig. 1g) . Of the 11 genes included in the 194 pyrimidine salvage screen, only 3 were low fitness conferring: dihydrofolate reductase – 195 thymidylate synthase ( DHFR-TS) (cgd4_4460), thymidine kinase ( TK) (cgd5_4440) and 196 dCMP deaminase (cgd2_2780). As mentioned previously, TK and DHFR -TS play 197 redundant roles in synthesis of Cryptosporidium dTMP , making them independently non-198 essential for DNA synthesis and parasite growth. Thus, the results of our pilot screen 199 eKectively mirror those of previous experiments attempting individual knockouts of these 200 genes in vivo (Vinayak et al., 2015), (Pawlowic et al., 2019). 201 202 Refining diCRE-mediated genetic editing in Cryptosporidium. To validate our results 203 from the pyrimidine salvage screen, we sought to use an inducible Cre -recombinase 204 system to remove genes at their genetic locus. This method has been previously adapted 205 for use in C. parvum, employing the conditional split recombinase (diCRE), which 206 dimerises upon rapamycin addition and excises DNA between loxP recombination sites, 207 one of which is embedded within a n artificial intron introduced into the target gene’s 208 coding sequence (Tandel et al., 2023), (Shaw et al., 2024). This system demonstrated high 209 eKiciency, but low-to-moderate activity in the absence of the rapamycin induction (i.e. 210 leakiness). To refine this system, we tested a validated intron from the male gamete 211 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 8 fusion factor HAP2 (cgd8_2220). Insertion of this intron within the nanoluciferase 212 reporter with and without the loxP recombination site did not aKect the measured 213 luminescence during in vitro infection (Sup. 2a ). Using this validated intron/loxP 214 combination, we generated a stable C. parvum diCRE parasite line targeting TK with the 215 diCRE subunits, FRB -Cre60 and FKBP -Cre59, expressed under the same promoter, 216 separated via a self-cleaving T2A skip peptide (TK-T2A-diCRE) (Sup. 2b). However, when 217 investigating the excision dynamics of this C. parvum TK-T2A-diCRE line, by infecting 218 HCT8 monolayers in the presence or absence of rapamycin, we noted a high level of 219 excision in the absence of rapamycin, similar to what has been reported previously (Sup. 220 2c). We reasoned that the T2A skip peptide may not be functioning properly and causing 221

induction in Cryptosporidium. Consequently, we generated a stable 222 transgenic parasite line targeting TK where the diCRE segments were under independent 223 aldolase and tubulin promoters (TK-diCRE) (Sup. 2d). When investigating the excision 224 dynamics of this TK -diCRE line, we observed complete excision of the loxP -flanked TK 225 segment at 24 hours post rapamycin induction, and no measurable excision in the non-226 induced controls (Sup. 2e). Using these TK-diCRE parasites (Sup. 3a), we performed a 227 time course to measure the dynamics of knockout at both the DNA and protein level (via 228 a C -terminal HA tag). DNA excision had not occurred by 8 hours post rapamycin 229 treatment but was complete by 24 hours, and no excision was detected in the non -230 induced control (Sup. 3b). Likewise, protein levels started to decrease by 12 hours post 231 rapamycin treatment and were approximately 95% reduced compared to the non-232 induced controls by 24 hours. In contrast, the protein level s remained high throughout 233 the time course in the non-induced controls (Sup. 3c & d). 234 235 Ribonucleotide reductase is required for parasite DNA replication and survival. 236 Ribonucleotide reductases (RNRs) catalyse the conversion of nucleotides to 237 deoxynucleotides, an essential step for DNA synthesis in all organisms. In our pyrimidine 238 salvage screen, RNR showed the lowest fold enrichment score, indicating a high eKect 239 on parasite fitness. Our first attempt to create RNR -diCRE parasites was unsuccessful, 240 likely due to the addition of a C -terminal HA epitope tag. It has been suggested that the 241 C-terminus of RNR is required for its function in other organisms and this may hold true 242 for Cryptosporidium (Cohen et al., 1986) , (Dutia et al., 1986) . The second attempt to 243 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 9 generate RNR-diCRE parasites, without the C -terminal HA epitope tag, was successful 244 (Sup. 3e). In vitro, rapamycin induced DNA excision of both TK- and RNR-diCRE parasites 245 appears to follow similar dynamics, although we note some partial excision in the RNR -246 diCRE parasites at later timepoints (Fig. 2a & b ). Further, functional deletion of both TK 247 and RNR was confirmed using 5 -ethynyl2´-deoxyuridine (EdU), a thymidine analogue. 248 Thymidine kinase is required for EdU phosphorylation and incorporation, and it has been 249 shown before that loss of the TK gene in Cryptosporidium leads to a failure to incorporate 250 EdU (Vinayak et al., 2015). In contrast, RNR deletion will not block EdU phosphorylation, 251 but rather incorporation, as DNA synthesis cannot occur without other 252 deoxynucleotides. We observed a complete lack of EdU incorporation in rapamycin 253 treated TK and RNR-diCRE parasites, while EdU incorporation was observed in the non -254 induced and wildtype controls (Sup. 3f). In vitro, deletion of TK led to no growth defect, 255 whereas deletion of RNR strongly attenuated parasite growth by 24 hours post infection 256 (Fig. 2c) . Similarly, rapamycin treatment of TK -diCRE infected mice lead to no growth 257 defect, but rapamycin treatment of RNR -diCRE infected mice completely inhibited 258 parasite growth (Fig. 2d) , demonstrating that RNR is indeed essential for parasite 259 survival, further underlining the reliability of the in vivo CRISPR screening method. 260 261 CRISPR KO screen of Cryptosporidium vaccine candidates. Both the 1 and 2 KO vector 262 per gene pyrimidine salvage screens were reproducible, but to circumvent potential low 263 eKiciency gRNAs, we chose to screen the leading Cryptosporidium vaccine candidates 264 with 2 KO vectors per gene. In total, 11 vaccine candidates (22 KO vectors) were chosen 265 due to their reported surface location and implication of immunogenicity and/or immune 266 protection in the literature (Table. 2). Luminescence from the nanoluciferase reporter 267 was detected in mouse faecal material around 7 days post transfection in both screens 268 (Fig. 3a). As before, parasites were purified, DNA extracted, and barcodes were amplified 269 via high fidelity PCR. The sequenced barcode counts were used to calculate fold 270 enrichment scores for each gene (Fig. 3b) , and importantly these scores were 271 comparable, achieving an R2 of 0.611 (Fig. 3c). Of the 11 genes in the vaccine candidate 272 screen, only 2 were low fitness conferring: cgd6_1660 (thrombospondin repeat protein 273 11 (TSP11)) and cgd6_32 (apical glycoprotein 1 (AGP1)). The remaining 9 genes appeared 274 to confer some level of fitness. Akey et al., recently demonstrated that AGP1 was 275 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 10 dispensable for parasite survival, while apical glycoprotein 2 (AGP2) (cgd7_4330) was 276 likely essential, both phenotypes that our pooled in vivo CRISPR screen recapitulated 277 (Akey et al., 2023). 278 279 To gain more insight into growth and competition of pooled KO parasites, we conducted 280 an additional 1 KO vector per gene vaccine candidate screen, where instead of amplifying 281 DNA barcodes from parasites purified around the peak of infection, we amplified 282 barcodes directly from individual faecal collections (Fig. 3d ). This alternative method 283 oKered greater temporal and spatial resolution of infection where fold enrichment scores 284 could be easily and non -invasively be monitored throughout the experiment from either 285 pooled collections or individual mice (Fig. 3e & f ). Individual mice demonstrated 286 considerable heterogeneity at day 8 of infection, which decreased over time and became 287 more uniform by day 12. Again, this screen found that AGP1 and TSP11 were low fitness 288 conferring relative to the other genes within the cohort. 289 290 Immunodominant antigen 23 is required for host cell invasion. Immunodominant 291 antigen 23 (Cp23) (cgd4_3620) is one of the leading Cryptosporidium vaccine candidates, 292 having been originally discovered in 1986 (Ungar & Nash, 1986). Despite nearly 40 years 293 of research, there is very little known about Cp23’s role and function during 294 Cryptosporidium infection. In our vaccine candidate screen, Cp23 displayed a low fold 295 enrichment score, indicating a moderate-to-high eKect on parasite fitness. To investigate 296 the function of Cp23, we generated Cp23 -diCRE parasites (Sup. 4a) . In vitro , the 297 rapamycin induced DNA excision had started by 6 hours and was complete by 24 hours, 298 with no excision observed in the uninduced controls (Fig. 4a & b) . Loss of Cp23 was 299 confirmed at the protein level using a commercially available monoclonal antibody. In 300 vitro, rapamycin treatment caused a significant decrease in parasite growth after 24 301 hours (Fig. 4c ) and rapamycin treatment of Cp23 -diCRE infected mice completely 302 blocked parasite growth (Fig. 4d) , demonstrating that Cp23 is indeed essential for 303 parasite survival. 304 305 Cp23 has been reported to be on the surface of the sporozoite (Mead et al., 1988), in the 306 sporozoite trails (Arrowood et al., 1991), and is predicted to localise to the micronemes 307 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 11 (Guérin et al., 2023). To localise Cp23 at high resolution, we used a commercial antibody 308 that we genetically validated with our Cp23 -diCRE parasites (Fig. 4e) , coupled with 309 expansion and super -resolution microscopy. This revealed Cp23 was expressed at the 310 parasite’s pellicle throughout the life cycle (Fig. 4f) . To resolve the location further, we 311 carried out transmission electron microscopy (TEM) with the immunogold labelled Cp23 312 antibody (Fig. 4g) . In excysted and unexcysted sporozoites, Cp23 again appeared to 313 primarily localise to the parasite pellicle, either at the plasma membrane or inner 314 membrane complex (IMC). The immunogold-TEM also revealed that even in unexcysted 315 sporozoites, Cp23 was not present in the parasite’s micronemes. Further, we found no 316 evidence that Cp23 was present in sporozoite trails (Sup. 4b). When permeabilised with 317 Triton X-100, sporozoites demonstrated faint staining with the Cp23 antibody (Fig. 4h). 318 Without permeabilisation, C. parvum sporozoites, surprisingly, demonstrated either a 319 complete lack of Cp23 signal or a heighted intensity of signal throughout the parasite 320 (Fig. 4h). In these high intensity parasites, we can detect the internal control antibody 321 (CpTrpB - Cryptosporidium tryptophan synthase beta) at low levels. This suggests that 1) 322 high intensity Cp23 parasites are weakly permeabilised, 2) permeabilisation with Triton 323 X-100 aKects Cp23 localisation, likely by disrupting its membrane association, and 3) 324 Cp23 is likely not exposed on the surface of the sporozoite. 325 326 Within our synchronised in vitro model of infection, C. parvum parasites egress from their 327 infected cell and re-invade nearby epithelial cells around 18 hours post infection, starting 328 another round of asexual reproduction. By 22 hours, this reinvasion event is mostly 329 complete. To determine the biological function of Cp23, we examined our Cp23 -diCRE 330 parasites in the context of this reinfection event, when DNA excision and depletion of 331 Cp23 has started. At 22 hours, Cp23-diCRE parasites without rapamycin treatment were 332 mostly newly invaded life stages (1n or 2n = 81% of parasites observed, n = 356/438). 333 When Cp23 was ablated by rapamycin, fewer parasites observed were newly invaded life 334 stages (1n or 2n = 45%, n = 77/173) (Fig. 4i & Sup. 4c & e) . In wildtype controls, both in 335 the presence and absence of rapamycin, again, most of the life stages observed were 336 newly invaded (1n or 2n = 83%, n = 176/203 and 171/213 respectively) (Fig. 4i & Sup. 4d 337 & e) . We hypothesised that the reduced percentage of newly invaded para sites when 338 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 12 Cp23 was ablated might be due to a defect in reinvasion. To examine this more closely, 339 live microscopy was carried out in the presence and absence of rapamycin (Fig. 4j & k , 340 supplementary videos S1 -4). When Cp23 was ablated, merozoites could egress, but 341 were then unable to move to another cell to initiate reinvasion. Apicomplexan parasites, 342 such as Cryptosporidium, are known to use a method of locomotion called gliding 343 motility, where parasites secrete proteins that are then bound by their own surface 344 receptors, allowing for forward propulsion using an actomyosin-based complex (referred 345 to as the glideosome) (Keeley & Soldati, 2004). Loss of Cp23 in Cryptosporidium appears 346 to specifically block this gliding motility that is essential for reinfection. 347 348

349 350 Reverse genetic screens play an important role in molecular biology and have 351 transformed the field for many pathogens. Here, we overcome the current technical 352 barriers to develop a pooled in vivo CRISPR KO method that allows for reverse genetic 353 screening in Cryptosporidium. As Cryptosporidium lacks the molecular machinery for 354 NHEJ, our method had to employ homologous recombination to create stable 355 transgenics. We leveraged two important features of Cryptosporidium genetics: 1) the 356 parasite spends nearly all of its life cycle in a haploid form, thus requiring only one 357 recombination event, and 2) there is a minimal length requirement for homologous 358 recombination, in fact 30bp of DNA flanking the Cas9 cut site appears to be suKicient in 359 our experiments. Our two-step cloning approach for generating these targeted KO vectors 360 is designed to be versatile, allowing for a n unlimited range of selection markers and 361 reporters that can be inserted into the genome, which will allow researchers to develop 362 new and innovative screens. 363 364 Within an infected host, Cryptosporidium undergoes multiple rounds of asexual 365 reproduction before diKerentiating into sexual forms (male and female) that unite to 366 allow for genetic recombination. During the sexual cycle, transgenic parasites in these 367 CRISPR screens can mate and this has the potential to influence results. This risk is 368 partially mitigated by two factors: 1) the parasite must complete at least three rounds of 369 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 13 asexual replication prior to sexual diKerentiation and potential recombination. This gives 370 ample time for any genes that are detrimental to parasite fitness to exert their eKect; 2) 371 Cryptosporidium has a relatively high rate of recombination, which should allow for 372 genes, even those in close proximity, to segregate in a nearly random fashion (Kimball et 373 al., 2024). In both the pyrimidine salvage and vaccine candidate screens, we identified 374 high and low fitness conferring genes in close proximity to each other (Sup. 5a). Despite 375 these mitigating factors, there is the potential for sublethal gene knockouts to exert 376 synergistic or antagonistic eKects and this possibility must be considered during the 377 design process and while analysing and interpreting results. 378 379 Another important factor to consider is the amplifying eKect of relative fitness screens 380 within an in vivo model of infection. Genes that have a mild eKect on parasite survival and 381 fitness can be outcompeted within a larger pool. This appears to be the case for uracil 382 phosphoribosyltransferase (UPRT, cgd1_1900), which catalyses the conversion of uracil 383 and phosphoribosylpyrophosphate to uridine monophosphate. UPRT has recently been 384 shown to be non-essential to parasite survival, yet the authors noted that their UPRT KO 385 Cryptosporidium parasites appeared to have a growth defect (Kimball et al., 2024). In our 386 in vivo screens, UPRT had a negative fold enrichment, indicating that UPRT contributes 387 highly to parasite fitness and this likely reflects the observed growth defect. In contrast 388 to UPRT, dCMP deaminase (cgd2_2780) had little eKect on parasite fitness. dCMP 389 deaminase catalyses the conversion of deoxycytidine-monophosphate ( dCMP) to 390 deoxyuridine-monophosphate (dUMP), which is then transformed into deoxythymidine -391 monophosphate (dTMP) by DHFR -TS. Thymidine kinase can also produce dTMP , likely 392 explaining why the loss of dCMP deaminase had little eKect on parasite fitness. 393 394 Most of the genes in the vaccine candidate CRISPR screen were identified as fitness 395 conferring, with the exception of apical glycoprotein 1 (AGP1) and thrombospondin 396 repeat protein 11 (TSP11). Akey et al. previous showed AGP1 was dispensable for parasite 397 survival and AGP2 was likely essential, both phenotypes that were recapitulated within 398 our screens (Akey et al., 2023) . Recently , it has been suggested that there may be 399 redundancy in the Cryptosporidium thrombospondin protein family (TSPs), as there are 400 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 14 12 members, many with similar domain structures (John et al., 2023) . Although the 401 Cryptosporidium TSPs currently have an unknown function, orthologous Apicomplexan 402 TSP genes play a role in adhesion and motility . We included two TSPs in the vaccine 403 candidate screen: 1) TSP8 ( cgd6_780, also known as MIC1) and 2) TSP11 (cgd6_1660). 404 TSP8/MIC1 is a known micronemal protein (Putignani et al., 2008) and was identified as 405 high fitness conferring. In contrast, TSP11 was dispensable, demonstrating that there is 406 at least some redundancy within the Cryptosporidium TSP protein family . We also 407 confirmed GP40 (cgd6_1080, also known as GP60) and GP900 (cgd7_4020), antigens 408 that are associated with protection from reinfection in humans, influence parasite fitness 409 in vivo (Gilchrist et al., 2023). 410 411 Cp23 is one of the leading cryptosporidiosis vaccine candidates, and here we 412 demonstrated that it is highly fitness conferring, both in vitro and in vivo. Cp23 was first 413 identified in 1986 (Ungar & Nash, 1986) , and recent studies have revealed Cp23 414 recognising IgA and IgG are correlated with protection against Cryptosporidium infection 415 in humans (Gilchrist et al., 2023). Until now, the function and precise localisation of Cp23 416 has been unclear. Here we validate a commercial Cp23 monoclonal antibody using 417 diCRE mediated excision of the target gene, then use a combination of ultrastructure 418 expansion, super -resolution, and transmission electron microscopy, to describe the 419 localisation of Cp23 throughout the parasite life cycle at high resolution. Although our 420 localisation agrees with previously published studies that detects Cp23 at the pellicle 421 (Enriquez & Riggs, 1998), our data suggests that Cp23 may not be exposed at the surface 422 of the sporozoite. Cp23 has no detectable signal peptide, transmembrane domain or GPI-423 anchor, yet was recently demonstrated to be both myristoylated and palmitoylated 424 (Haserick et al., 2017) . Myristylation is a non-reversible lipid modifications that is 425 commonly found in proteins that are anchored to internal membranes in other 426 Apicomplexan parasites (Schlott et al., 2018), (Broncel et al., 2020). There are, however, 427 some exceptions to th is rule. TgMIC7 is a micronemal protein in Toxoplasma that is 428 myristoylated prior to traKicking to the parasite surface (Broncel et al., 2020). Yet, TgMIC7 429 contains a transmembrane domain that likely facilitates entry into the secretory pathway, 430 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 15 whereas Cp23 does not . While we cannot rule out traKicking of Cp23 to the outer 431 membrane of the parasite, our data supports an internal localisation. 432 433 Despite this internal localisation, it is clear Cp23 IgA and IgG have both been correlated 434 with protection in the clinic (Gilchrist et al., 2023). One interpretation of this correlation 435 is that antibodies recognising Cp23 arise from severe Cryptosporidium infection(s) where 436 cell-based adaptive immunity against the parasite has developed. Controlled 437 experiments using a natural murine model of Cryptosporidium suggest that antibody -438 based immunity is dispensable for resolution of infection (Sateriale et al., 2019). Further, 439 we found that the genetically validated Cp23 antibody used in this study was unable to 440 block parasite attachment or invasion in vitro (Sup. 6a - b). However, it has been reported 441 that antibodies to Cp23 may have a protective eKect in vivo, thus we cannot definitively 442 rule out the possibility of Cp23 -directed protection, whether through antibody or cell -443 based immunity (Enriquez & Riggs, 1998). Certainly, more investigation is warranted given 444 we now know Cp23’s essential role in parasite motility and survival. 445 446 With the targeted in vivo CRISPR screening method developed here, we assessed the 447 fitness contributions of 22 Cryptosporidium genes. This number is near to the total 448 number of Cryptosporidium genes with a genetically verified impact on parasite virulence 449 or survival, prior to this study . As this screening technology develops and improves, we 450 anticipate greater throughput, allowing for the assessment of a variety of phenotypes on 451 a larger scale. This rapid assessment of phenotypes will allow us to expand the 452 knowledge base for Cryptosporidium and explore basic biology that is essential for 453 developing more eKective interventions. 454 455

456 457 Plasmid design and construction. Genomic sites for CRISPR directed repair were 458 predicted using a customised version of EuPaGDT that retrieved 50bp of flanking DNA 459 surrounding the predicted PAM site (Alvarez-Jarreta et al., 2023). Golden Gate Assembly 460 or Gibson Assembly was used to generate all vector s used in this work. Golden Gate 461 Assembly used, BsaI, BbsI -HF or BsmBI (New England BioLabs (NEB)) restriction 462 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 16 enzymes. Gibson Assembly used HiFi DNA Assembly (NEB). To generate KO vectors for 463 CRISPR screening, 2 consecutive Golden Gate reactions were performed. Firstly, 464 between a Cas9-U6 vector and a 300bp unique segment containing 50bp homology arms 465 (one of which contained the 50bp gRNA), tracrRNA and a DNA barcode, creating a Cas9-466 unique plasmid. Secondly, between this Cas9 -unique plasmid and a n interchangeable 467 selection cassette, in turn creating the KO vector. To generate diCRE mediated inducible 468 knockouts, a gene of interest (GOI) was recodonised and cloned into a LoxP-Nluc-NeoR-469 diCRE vector via Gibson Assembly. To generate Cas9-U6-gRNA vectors, a 20bp gRNA was 470 cloned into the Cas9 -U6-BsaI vector via Golden Gate Assembly (Vinayak et al., 2015) , 471 (Pawlowic et al., 2017). 472 473 Culturing host cells. Human ileocecal adenocarcinoma cells (HCT8) were cultured in 474 RMPI-1640 medium (Gibco) supplemented with 10% heat -inactivated foetal bovine 475 serum (Merck), 120U/mL penicillin (Life Technologies) and 0.1% amphotericin B (Gibco) 476 at 37 °C under 5% CO 2. Cells were passaged at approximately 70% confluence using 477 0.25% trypsin-EDTA (Gibco). Cells were used for experiments between passage numbers 478 5 and 25. For super resolution and ultrastructure expansion microscopy, cells were 479 seeded in 24 -well plates containing coverslips. For high throughput microscopy 480 quantifications, cells were seeded in black 96 -well clear bottom tissue culture -treated 481 plates (Corning). For transient transfections using luminescence, cells were seeded in 482 24-well tissue culture-treated plates (Corning). For live imaging, cells were seeded in µ-483 slide 8-well high chambers (Ibidi). 484 485 Oocysts and excystation. C. parvum IOWAII strain oocysts were purchased from Bunch 486 Grass Farm (Deary, ID). Oocysts were stored at 4°C and used within 3 months of the date 487 of isolation. The oocysts were excysted by incubating on ice with 1% sodium hypochlorite 488 (VWR) in H 20 for 5 minutes, followed by incubating with 0.75% sodium taurocholate 489 (Merck) in RPMI-1640 medium with 1% f oetal bovine serum at 37 °C (10 minutes for cell 490 monolayer infections without transfection and 50 minutes prior to transfection). For in 491 vitro infections, primed oocysts were used to infect HCT8 monolayers. To proceed with 492 transfections, more than 60% of oocysts had to have excysted, observed using light 493 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 17 microscopy at 50 minutes post sodium taurocholate treatment (Eclipse TS2R Nikon). For 494 Supplementary Figure 1, 0.75% sodium taurocholate was substituted with either 0.75% 495 sodium deoxycholate (Sigma) or 0.75% sodium taurodeoxycholate (Sigma), and 1% 496 RPMI-1640 was substituted with PBS. 497 498 Generation of transgenic parasites. Excysted sporozoites were suspended in Lonza SF 499 buKer, combined with the appropriate DNA and electroporated using program ‘FL115’ on 500 an AMAXA Nucleofactor 4D electroporator (Lonza) (FL115 was used for all transfections 501 unless otherwise stated). For transient transfections in vitro , 5.0 x 10 6 oocysts were 502 excysted and sporozoites were transfected in the 20 μL 16 -well Nucleocuvette Strip 503 format with 20µg of plasmid. For generating stable transgenic parasites in vivo, 2.5 x 107 504 oocysts were excysted and sporozoites were transfected in the 100 μL Nucleocuvette 505 Vessel format with 20 µg of Cas9 -U6-gRNA plasmid and 40 µg of repair cassette 506 containing the 50bp homology arms. For CRISPR screens, 5.0 x 10 6 sporozoites were 507 excysted and sporozoites were transfected with 30µg of KO vector for each gene (when 2 508 KO vectors were used, 15 µg of each KO vector was used). Transfections to knockout 509 individual genes were performed separately (when 2 KO vectors per gene were used, 510 transfections of KO vectors for the same gene were performed together). Parasites were 511 pooled post transfection in 1% RMPI-1640 (Gibco) to infect mice and 4 or 5 mice were 512 infected for each screen. 513 514 Mouse model of infection. Interferon gamma deficient ( Ifng-/-) mice were bred and 515 housed in pathogen -free conditions in the Biological Research Facility at The Francis 516 Crick Institute. Mice of both sexes were used for experiments. To increase infection 517 eKiciency, mice were pretreated with an antibiotic cocktail: 1g/L ampicillin (Merck), 1g/L 518 streptomycin (Merck) and 0.5g/L vancomycin (Cambridge Biosciences) in their drinking 519 water for 3 to 10 days prior to infection with transfected sporozoites. Before infection, 520 mice received saturated sodium bicarbonate (Thermo Fisher Scientific) via oral gavage to 521 neutralise their stomach acid. A second oral gavage was then undertaken 5 minutes 522 thereafter with transfected sporozoites. Neither oral gavage exceeded the maximum 523 volume of 0.1mL per 10g of body weight. For selection of transgenic parasites, 16mg/mL 524 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 18 paromomycin (BioServ) was administered in the mice’s drinking water. To induce excision 525 when diCRE-expressing transgenic parasites were used in vivo, 0.05 mg/mL rapamycin 526 (Stratech Scientific) was administered in the mice’s drinking water. No drug treatment 527 lasted more than 3 consecutive weeks. During murine infections, faecal material was 528 collected daily and stored at 4°C. All experiments involving mice was done under the care 529 and supervision of the Francis Crick Institute veterinary and Biological Research Facility 530 staK, under protocols approved under project license PP8575470. 531 532 Measuring parasite shedding by nanoluciferase. Faecal material was collected and 533 20mg was lysed in faecal lysis buKer (50mM Tris -hydrochloric acid, 10% glycerol, 1% 534 Triton X-100, 2mM dithiothreitol, 2mM ethylenediaminetetraacetic acid (EDTA)). The 535 lysate was clarified and combined with an equal volume of a 1:50 Nano -Glo Luciferase 536 Assay Substrate: Nano-Glo Luciferase Assay BuKer (Promega). Luminescence was read 537 at 200 gain on the BioTek Cytation5 (Agilent Technologies). 538 539 Isolation of oocysts from mouse faeces. Faecal collections from the peak of infection 540 were pooled, combined with cold water and filtered through a 250mm mesh. The faecal 541 suspension was then mixed 1:1 with saturated sucrose and pelleted by centrifugation at 542 1000g for 10 minutes. Oocysts, located in the supernatant, were washed with cold water 543 and pelleted by centrifugation at 100 0g for 5 minutes. A caesium gradient was used to 544 isolate 750µL of oocysts. Pure oocysts were washed with saline and stored for up to 6 545 months at 4°C in saline. 546 547 CRISPR screening barcoding . To obtain barcodes from the output, barcodes could 548 either be extracted from the peak of infection or directly from daily faecal samples. To do 549 so from the peak of infection, oocysts were purified from pooled faecal material (5 to 7 550 days surrounding the peak of infection) , excysted and gDNA was extracted using the 551 DNeasy Blood & Tissue Kit (Qiagen). To do so from daily faecal samples, DNA was 552 extracted from 100mg of faeces using the QIAamp PowerFecal Pro DNA Kit (Qiagen). 553 Barcodes were obtained from the input material (100µL of the pool of transfected 554 parasites used to infect the mice) using the QIAquick PCR Purification Kit (Qiagen). 555 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 19 Barcodes from the input and output were amplified with high fidelity KAPA polymerase 556 (Roche) (2 5 cycles, annealing 68°C and extension 15 seconds), and amplicon 557 sequencing flanks were added with high fidelity KAPA polymerase (Roche) (10 cycles, 558 68°C annealing and 15 seconds extension). The success of the nested PCR was 559 confirmed via gel electrophoresis and barcodes were submitted for amplicon 560 sequencing to The Genomics Science Technology Platform at The Francis Crick Institute. 561 Illumina MiSeq platform with a paired -end 250bp run configuration on a Nano flow cell 562 was used. 563 564 CRISPR screening analysis. The 7bp barcode between the barcode primer binding sites 565 was bioinformatically extracted from both input and output samples. Barcodes for each 566 gene were counted and the percentage of the barcode in the pool calculated for both the 567 input and output. Using the % barcode in the pool for the input and output, fold changes 568 of all genes were calculated: 569 𝑙𝑜𝑔! 𝑓𝑜𝑙𝑑 𝑐ℎ𝑎𝑛𝑔𝑒 = 𝑙𝑜𝑔! -% 𝑏𝑎𝑟𝑐𝑜𝑑𝑒 𝑖𝑛 𝑜𝑢𝑡𝑝𝑢𝑡 % 𝑏𝑎𝑟𝑐𝑜𝑑𝑒 𝑖𝑛 𝑖𝑛𝑝𝑢𝑡 5 570 571 Immunofluorescence microscopy. At the required timepoint, HCT8 cell monolayers 572 were washed with 1X PBS, fixed with 4% PFA/PBS (Alfa Aesar) for 15 minutes, 573 permeabilised with 0.25% Triton X-100 (Merck) for 10 minutes and blocked with 4% 574 BSA/PBS (Merck) overnight at 4ºC. Primary antibodies were incubated in 1% BSA/PBS for 575 2 hours and then cell monolayers washed 5 times with 1X PBS. Secondary antibodies 576 were incubated for 1 hour in 1% BSA/PBS along with the fluorescein labelled 1:4000 Vicia 577 villosa lectin (VVL) (Vector Lab) or 1:1000 Helix pomatia agglutinin (HPA) (Invitrogen) that 578 stains parasites. Nuclei were stained by incubating with 1:10,000 Hoechst 33342 579 (Invitrogen) in 1X PBS for 5 minutes. For the EdU assays, 10mM EdU was incubated from 580 28 to 32 hours post infection and stained as the described in the Click -iT EdU Cell 581 Proliferation Kit for Imaging (C10340, Invitrogen). Stained monolayers were washed with 582 1X PBS, the coverslips mounted with ProLong Gold Antifade (Thermo Fisher Scientific) 583 and visualised using a VisiTech instant super resolution imaging system (VT -iSIM). 584 Alternatively, when super resolution was not required, the BioTek Cytation5 (Agilent 585 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 20 Technologies) was used to visualise the stained monolayer. See Table 6 for a full list of 586 antibodies used in this work. 587 588 Ultrastructure expansion microscopy. This protocol has been adapted from LiKner and 589 Absalon, 2021. For the expansion of sporozoites, coverslips were coated with 0.1mg/mL 590 poly-D-lysine for 1 hour and then washed twice with 1X PBS. Excysted sporozoites in 1% 591 RPMI-1640 were added and allowed to adhere for 10 minutes at 37°C. Adhered 592 sporozoites or cell monolayers to expand were fixed with 4% PFA/PBS for 15 minutes at 593 37°C. Protein crosslinking prevention was performed by adding 1.4% formaldehyde/ 2% 594 acrylamide in 1X PBS to each sample, then incubating overnight at 37°C. To perform 595 gelation, TEMED and APS were added to a monomer solution (19% w/w sodium acrylate 596 / 10% v/v acrylamide / 0.1% v/v N,N’ -methylenebisacrylamide in 1X PBS) and pipetted 597 under each coverslip in a pre -cooled humid chamber which was then incubated for 5 598 minutes on ice. To complete gelation, the humid chamber was incubated for 1 hour at 599 37°C. To separate gels from the coverslips, the coverslips were incubated with a 600 denaturation buKer (200mM SDS, 200mM NaCl, 50mM Tris in water, pH 9) for 15 minutes. 601 To complete the denaturation of the sample, the gel was incubated in a denaturation 602 buKer for 90 minutes at 95°C. To expand the sample, the gel was incubated with dH20 for 603 30 minutes, three times in total at room temperature (RT). Following the first round of 604 expansion, the sample was shrunk by incubating with 1X PBS, 2 times in total at RT. To 605 block the sample, 2% BSA/PBS was incubated for 1 hour at RT. To stain the sample, 606 primary antibodies were incubated in 2% BSA/PBS overnight at RT. Gels were then 607 washed 3 times in total with 0.5% Tween 20/PBS for 10 minutes. Directly conjugated and 608 secondary antibodies were incubated in 1X PBS for 3 hours at RT. The gel was then 609 washed 3 times in total with 0.5% Tween20/PBS for 10 minutes. A second round of 610 expansion took place by incubating the gel with dH 20 for 30 minutes, three times at RT. 611 The gel was measured to calculate the expansion factor, mounted onto a 0.1mg/mL poly-612 D-lysine coated 60mm dish and visualised on the VT-iSIM microscope. 613 614 Immunogold electron microscopy. 1x107 oocysts were excysted for 1 hour, pelleted and 615 re-suspended in fixative (8% formaldehyde in 0.4M HEPES buffer, pH 7.4) for 15 minutes 616 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 21 at RT. Samples were washed with 0.2M HEPES and a secondary fixation ( 2% 617 formaldehyde + 0.05% glutaraldehyde in 0.2M HEPES) step was carried out at 4 oC 618 overnight. Samples were dehydrated and infiltrated with LR white resin at -20oC 619 overnight. Prior to polymerisation, samples were brought back to RT for 1 hour and then 620 polymerised at 60 oC for 24 hours. The samples were sectioned using a Leica U C7 621 ultramicrotome with a 45° Diatome diamond knife, achieving sections of 70nm 622 thickness. The sections were collected on nickel grids and immunogold labelled. To do 623 so, the samples were quenched with PBS/glycine for 2 minutes three times and then 624 blocked with 1% BSA/PBS for 5 minutes at RT. A 1:10 dilution of the primary Cp23 625 antibody was incubated in 1% BSA/PBS for 1 hour and then samples were washed twice 626 with 0.1% BSA/PBS. A 1:50 dilution of the protein -A gold bound to 10nm gold particles 627 (PAG-10) was incubated with the sample in 0.1% BSA/PBS for 20 minutes and then 628 washed twice with PBS. A post fixation step was carried out with 1% glutaraldehyde for 5 629 minutes and samples were washed with Milli-Q H20 for 1 minute 6 times in total. Samples 630 were incubated in 1% uranyl acetate for 10 minutes and air dried. Transmission electron 631 microscopy was carried out on the 120 -kV JEOL JEM -1400Flash Electron Microscope 632 (JEOL Ltd., Welwyn Garden City, UK) with a JEOL Matataki Flash camera. 633 634 Live microscopy. At 18 hours post -infection, phase contrast imaging was performed 635 using an Eclipse Ts2R microscope (Nikon) with a 40X/0.55 NA Ph1 ADL objective (Nikon), 636 digital sight 10 camera (Nikon) and a TPi-TCSX (Tokai Hit) heated stage set at 37oC. Images 637 were acquired at 25fps for 1 hour to capture egressing merozoites. Images were imported 638 into ImageJ2 Version 2 and the number of gliding merozoites was manually observed. 639 Image sequences were converted to movies to acquire short videos of egressing 640 merozoites. 641 642 Permeabilisation assay. Coverslips placed in 24-well plates were coated with 0.1mg/mL 643 poly-D-lysine for 1 hour and washed twice with 1X PBS. Excysted sporozoites in serum 644 free RMPI-1640 were allowed to adhere for 10 minutes at 37oC. Adhered sporozoites were 645 fixed with 1% PFA/PBS for 20 minutes at RT. For the permeabilised condition, 0.1% Triton 646 X-100 was used to permeabilise for 10 minutes at RT. For the non -permeabilised 647 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 22 condition, 1X PBS was instead incubated for 10 minutes at RT. Both conditions were 648 blocked with 4% BSA/PBS overnight at 4°C. Primary antibodies were incubated in 1% 649 BSA/PBS for 2 hours at RT. The coverslips were washed 5 times and secondary antibodies 650 were incubated in 1% BSA/PBS for 1 hour at RT. All coverslips were mounted with ProLong 651 Gold Antifade (Thermo Fisher Scientific) and visualised using a VT-iSIM. 652 653 Attachment assay. Coverslips placed in 24 -well plates were coated with 0. 025mg/mL 654 poly-D-lysine for 1 hour and washed twice with 1X PBS. 60,000 oocysts per well were 655 excysted and sporozoites were added and spun at 80g for 1 minute to adhere. Attachment 656 was performed in Ringer’s solution (10mM Hepes pH 6.7, 10mM Glucose , 2mM CaCl 2, 657 1mM MgCl 2, 3mM KCl, 3mM NaH 2PO4 and 155mM NaCl) for 15 minutes at 37 °C. 658 Sporozoites were fixed using 8% PFA/PBS leak in resulting in 4% PFA/PBS/Ringer’s 659 fixation for 15 minutes at RT. Sporozoites were washed 3 times in total with 1X PBS and 660 blocked with 4% BSA/PBS overnight at 4 °C. Sporozoites were stained with 1:5000 Helix 661 pomatia agglutinin (HPA) in 1% BSA/PBS for 1 hour at RT . The BioTek Cytation5 (Agilent 662 Technologies) was used to visualise the attached sporozoites. Images were exported as 663 TIFs into ImageJ2 Version 2 and the below macro was used to analyse the data: 664 665 run("Subtract Background... " , "rolling=50"); 666 setAutoThreshold("Default dark"); 667 //run("Threshold... "); 668 setAutoThreshold("Triangle dark"); 669 setOption("BlackBackground" , true); 670 run("Convert to Mask"); 671 run("Median... " , "radius=2"); 672 run("Analyze Particles... " , "size=50-10000 circularity=0.00-1 display summarize overlay"); 673 674 Neutralisation assay. HCT8 cells were seeded to confluency in 96-well clear bottom 675 tissue culture -treated plates (Corning). An 8 -point 2 -fold dilution series of the Cp23 676 antibody (Stratech, LS -C137378) and the IgG isotype control (Stratech, GTX35009 + 677 0.02% Proclin 300 (Merck, 48912 -U) was prepared (0.02mg/mL to 0.0002mg/mL) in 96-678 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 23 well non -tissue culture -treated plates. 25,000 oocysts per well were excysted and 679 incubated with the antibody dilution series for 5 minutes at 37 °C. Post antibody 680 incubation, the sporozoites were added to the 96-well plate containing HCT8 cells in 1% 681 RPMI-1640. At 24 hours post-infection, cell monolayers were washed with 1X PBS, fixed 682 with 4% PFA/PBS (Alfa Aesar) for 15 minutes, permeabilised with 0.25% Triton X-100 683 (Sigma) for 10 minutes and blocked with 4% BSA/PBS (Merck) overnight at 4oC. Parasites 684 were stained with 1:4000 Vicia villosa lectin (VVL) and host nuclei were stained with 685 1:10,000 Hoechst 33342 (Invitrogen) in 1% BSA/PBS for 1 hour at RT. Cell monolayers 686 were washed and visualised on the BioTek Cytation5 (Agilent Technologies). 2x2 tiled 687 images per well were acquired with the 20X objective. The Gen5 analysis soft ware 688 (Agilent Technologies) was used to count the number of host cell nuclei and parasites for 689 parasite per nuclei calculations. 690 691

692 693 We would like to thank Pippa Hawes of The Francis Crick Electron Microscopy Scientific 694 Technology Platform for her helpful discussions and Elena Rodrigues of the 695 Cryptosporidiosis Laboratory for preparing Cryptosporidium samples for electron 696 microscopy. Further, we would like to thank Nicholas Chisholm and other members of 697 the Biological Research Facility, as well as the Genomics Scientific Technology Platform 698 at The Francis Crick Institute for their contributions to this work. We would also like to 699 thank Rodrigo Baptista of Houston Methodist for his insight and helpful conversations. 700 This work was supported by the Francis Crick Institute —which receives its core funding 701 from Cancer Research UK, the UK Medical Research Council, and the We llcome Trust 702 (CC2063) – and a UKRI grant awarded to A.S. (101042783). 703 704 Author Contributions 705 706 L.C.W & A.S, with help from K.S, N.B, L.B & N.B.M , pioneered the CRISPR screening 707 approach. M.P developed the attachment assay. L.C.S & L.C carried out TEM. D.P 708 developed the customised version of EuPaGDT used for predicting CRISPR guides. L.C.W 709 and A.S wrote the manuscript with contribution from all of the authors. 710 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 24 711

712 713 Abubakar, I., Aliyu, S. H., Arumugam, C., Usman, N. K., & Hunter, P . R. (2007, Apr). 714 Treatment of cryptosporidiosis in immunocompromised individuals: systematic 715 review and meta -analysis. Br J Clin Pharmacol, 63 (4), 387 -393. 716 https://doi.org/10.1111/j.1365-2125.2007.02873.x 717 718 Ajiboye, J., Teixeira, J. E., Gasonoo, M., Mattice, E. B., Korwin -Mihavics, B., Miller, P ., 719 Cameron, A. C., Stebbins, E., Campbell, S. D., Griggs, D. W., Spangenberg, T., 720 Meyers, M. J., & Huston, C. D. (2024, Sep 27). Identification of potent and orally 721 eKicacious phosphodiesterase inhibitors in Cryptosporidium parvum -infected 722 immunocompromised male mice. Nat Commun, 15 (1), 8272. 723 https://doi.org/10.1038/s41467-024-52658-y 724 725 Akey, M. E., Xu, R., Ravindran, S., Funkhouser -Jones, L., & Sibley, L. D. (2023, Feb 28). 726 Apical Secretory Glycoprotein Complex Contributes to Cell Attachment and Entry 727 by Cryptosporidium parvum. mBio, 14 (1), e0306422. 728 https://doi.org/10.1128/mbio.03064-22 729 730 Alvarez-Jarreta, J., Amos, B., Aurrecoechea, C., Bah, S., Barba, M., Barreto, A., Basenko, 731 E. Y ., Belnap, R., Blevins, A., Böhme, U., Brestelli, J., Brown, S., Callan, D., 732 Campbell, L. I., Christophides, G. K., Crouch, K., Davison, H. R., DeBarry, J. D., 733 Demko, R., Doherty, R., Duan, Y ., Dundore, W., Dyer, S., Falke, D., Fischer, S., 734 Gajria, B., Galdi, D., Giraldo -Calderón, G. I., Harb, O. S., Harper, E., Helb, D., 735 Howington, C., Hu, S., Humphrey, J., Iodice, J., Jones, A., Judkins, J., Kelly, S. A., 736 Kissinger, J. C., Kittur, N., Kwon, D. K., Lamoureux, K., Li, W., Lodha, D., 737 MacCallum, R. M., Maslen, G., McDowell, M. A., Myers, J., Nural, M. V ., Roos, D. 738 S., Rund, S. S. C., Shanmugasundram, A., Sitnik, V ., Spruill, D., Starns, D., Tomko, 739 S. S., Wang, H., Warrenfeltz, S., Wieck, R., Wilkinson, P . A., & Zheng, J. (2023). 740 VEuPathDB: the eukaryotic pathogen, vector and host bioinformatics resource 741 center in 2023. Nucleic Acids Research, 52 (D1), D808 -D816. 742 https://doi.org/10.1093/nar/gkad1003 743 744 Amadi, B., Mwiya, M., Sianongo, S., Payne, L., Watuka, A., Katubulushi, M., & Kelly, P . 745 (2009, Dec 2). High dose prolonged treatment with nitazoxanide is not eKective for 746 cryptosporidiosis in HIV positive Zambian children: a randomised controlled trial. 747 BMC Infect Dis, 9, 195. https://doi.org/10.1186/1471-2334-9-195 748 749 Arrowood, M. J., Sterling, C. R., & Healey, M. C. (1991, Apr). Immunofluorescent 750 microscopical visualization of trails left by gliding Cryptosporidium parvum 751 sporozoites. J Parasitol, 77(2), 315-317. 752 753 Askari, N., Shayan, P ., Mokhber -Dezfouli, M. R., Ebrahimzadeh, E., Lotfollahzadeh, S., 754 Rostami, A., Amininia, N., & Ragh, M. J. (2016). Evaluation of recombinant P23 755 protein as a vaccine for passive immunization of newborn calves against 756 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 25 Cryptosporidium parvum. Parasite Immunology, 38 (5), 282 -289. 757 https://doi.org/https://doi.org/10.1111/pim.12317 758 759 Broncel, M., Dominicus, C., Vigetti, L., Nofal, S. D., Bartlett, E. J., Touquet, B., Hunt, A., 760 Wallbank, B. A., Federico, S., Matthews, S., Young, J. C., Tate, E. W., Tardieux, I., & 761 Treeck, M. (2020, 2020/07/03). Profiling of myristoylation in Toxoplasma gondii 762 reveals an N -myristoylated protein important for host cell penetration. eLife, 9, 763 e57861. https://doi.org/10.7554/eLife.57861 764 765 Bushell, E., Gomes, A. R., Sanderson, T., Anar, B., Girling, G., Herd, C., Metcalf, T., 766 Modrzynska, K., Schwach, F ., Martin, R. E., Mather, M. W., McFadden, G. I., Parts, 767 L., Rutledge, G. G., Vaidya, A. B., Wengelnik, K., Rayner, J. C., & Billker, O. (2017, 768 Jul 13). Functional Profiling of a Plasmodium Genome Reveals an Abundance of 769 Essential Genes. Cell, 170 (2), 260 -272.e268. 770 https://doi.org/10.1016/j.cell.2017.06.030 771 772 Butterworth, S., Kordova, K., Chandrasekaran, S., Thomas, K. K., Torelli, F ., Lockyer, E. J., 773 Edwards, A., Goldstone, R., Koshy, A. A., & Treeck, M. (2023, Oct 11). High -774 throughput identification of Toxoplasma gondii eKector proteins that target host 775 cell transcription. Cell Host Microbe, 31 (10), 1748 -1762.e1748. 776 https://doi.org/10.1016/j.chom.2023.09.003 777 778 Caldwell, N., Peet, C., Miller, P ., Colon, B. L., Taylor, M. G., Cocco, M., Dawson, A., Lukac, 779 I., Teixeira, J. E., Robinson, L., Frame, L., Seizova, S., Damerow, S., Tamaki, F ., Post, 780 J., Riley, J., Mutter, N., Hanna, J. C., Ferguson, L., Hu, X., Tinti, M., Forte, B., 781 Norcross, N. R., Campbell, P . S., Svensen, N., Caldwell, F . C., Jansen, C., Postis, 782 V ., Read, K. D., Huston, C. D., Gilbert, I. H., Baragaña, B., & Pawlowic, M. C. (2024). 783 Cryptosporidium lysyl -tRNA synthetase inhibitors define the interplay 784 between solubility and permeability required to achieve eKicacy. Science 785 Translational Medicine, 16(770), eadm8631. 786 https://doi.org/doi:10.1126/scitranslmed.adm8631 787 788 Cohen, E. A., Gaudreau, P ., Brazeau, P ., & Langelier, Y . (1986, 1986/05/01). Specific 789 inhibition of herpesvirus ribonucleotide reductase by a nonapeptide derived from 790 the carboxy terminus of subunit 2. Nature, 321 (6068), 441 -443. 791 https://doi.org/10.1038/321441a0 792 793 Dutia, B. M., Frame, M. C., Subak -Sharpe, J. H., Clark, W. N., & Marsden, H. S. (1986, 794 1986/05/01). Specific inhibition of herpesvirus ribonucleotide reductase by 795 synthetic peptides. Nature, 321 (6068), 439 -441. 796 https://doi.org/10.1038/321439a0 797 798 Enriquez, F . J., & Riggs, M. W. (1998, Sep). Role of immunoglobulin A monoclonal 799 antibodies against P23 in controlling murine Cryptosporidium parvum infection. 800 Infect Immun, 66(9), 4469-4473. https://doi.org/10.1128/iai.66.9.4469-4473.1998 801 802 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 26 Gilchrist, C. A., Campo, J. J., Pablo, J. V ., Ma, J. Z., Teng, A., Oberai, A., Shandling, A. D., 803 Alam, M., Kabir, M., Faruque, A. S. G., Haque, R., & Petri, W. A., Jr. (2023, Aug 15). 804 Specific Cryptosporidium antigens associate with reinfection immunity and 805 protection from cryptosporidiosis. J Clin Invest, 133 (16). 806 https://doi.org/10.1172/jci166814 807 808 Guérin, A., Strelau, K. M., Barylyuk, K., Wallbank, B. A., Berry, L., Crook, O. M., Lilley, K. 809 S., Waller, R. F ., & Striepen, B. (2023, Apr 12). Cryptosporidium uses multiple 810 distinct secretory organelles to interact with and modify its host cell. Cell Host 811 Microbe, 31(4), 650-664.e656. https://doi.org/10.1016/j.chom.2023.03.001 812 813 Haque, R., Mondal, D., Karim, A., Molla, I. H., Rahim, A., Faruque, A. S., Ahmad, N., 814 Kirkpatrick, B. D., Houpt, E., Snider, C., & Petri, W. A., Jr. (2009, May 1). Prospective 815 case-control study of the association between common enteric protozoal 816 parasites and diarrhea in Bangladesh. Clin Infect Dis, 48 (9), 1191 -1197. 817 https://doi.org/10.1086/597580 818 819 Haserick, J. R., Klein, J. A., Costello, C. E., & Samuelson, J. (2017). Cryptosporidium 820 parvum vaccine candidates are incompletely modified with O -linked-N-821 acetylgalactosamine or contain N -terminal N -myristate and S -palmitate. PLOS 822 ONE, 12(8), e0182395. https://doi.org/10.1371/journal.pone.0182395 823 824 JeKeries, R., Yang, R., Woh, C. K., Weldt, T., Milech, N., Estcourt, A., Armstrong, T., 825 Hopkins, R., Watt, P ., Reid, S., Armson, A., & Ryan, U. M. (2015, 2015/01/01/). 826 Target validation of the inosine monophosphate dehydrogenase (IMPDH) gene in 827 Cryptosporidium using Phylomer® peptides. Experimental Parasitology, 148, 40-828 48. https://doi.org/https://doi.org/10.1016/j.exppara.2014.11.003 829 830 Jenkins, M., Higgins, J., Kniel, K., Trout, J., & Fayer, R. (2004, Oct). Protection of calves 831 against cryptosporiosis by oral inoculation with gamma -irradiated 832 Cryptosporidium parvum oocysts. J Parasitol, 90 (5), 1178 -1180. 833 https://doi.org/10.1645/ge-3333rn 834 835 John, A., S, M. B., Madiedo Soler, N., Wiradiputri, K., Tichkule, S., Smyth, S. T., Ralph, S. 836 A., Jex, A. R., Scott, N. E., T onkin, C. J., & Goddard -Borger, E. D. (2023, Mar). 837 Conservation, abundance, glycosylation profile, and localization of the TSP 838 protein family in Cryptosporidium parvum. J Biol Chem, 299 (3), 103006. 839 https://doi.org/10.1016/j.jbc.2023.103006 840 841 Keeley, A., & Soldati, D. (2004, 2004/10/01/). The glideosome: a molecular machine 842 powering motility and host -cell invasion by Apicomplexa. Trends in Cell Biology, 843 14(10), 528-532. https://doi.org/https://doi.org/10.1016/j.tcb.2004.08.002 844 845 Kimball, A., Funkhouser-Jones, L., Huang, W., Xu, R., Witola, W. H., & Sibley, L. D. (2024). 846 Mendelian segregation and high recombination rates facilitate genetic analyses in 847 Cryptosporidium parvum. PLOS Genetics, 20 (6), e1011162. 848 https://doi.org/10.1371/journal.pgen.1011162 849 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 27 850 KotloK, K. L., Nasrin, D., Blackwelder, W. C., Wu, Y ., Farag, T., Panchalingham, S., Sow, S. 851 O., Sur, D., Zaidi, A. K. M., Faruque, A. S. G., Saha, D., Alonso, P . L., Tamboura, B., 852 Sanogo, D., Onwuchekwa, U., Manna, B., Ramamurthy, T., Kanungo, S., Ahmed, 853 S., Qureshi, S., Quadri, F ., Hossain, A., Das, S. K., Antonio, M., Hossain, M. J., 854 Mandomando, I., Acácio, S., Biswas, K., Tennant, S. M., Verweij, J. J., Sommerfelt, 855 H., Nataro, J. P ., Robins -Browne, R. M., & Levine, M. M. (2019). The incidence, 856 aetiology, and adverse clinical consequences of less severe diarrhoeal episodes 857 among infants and children residing in low-income and middle-income countries: 858 a 12-month case-control study as a follow -on to the Global Enteric Multicenter 859 Study (GEMS). The Lancet Global Health, 7 (5), e568 -e584. 860 https://doi.org/10.1016/S2214-109X(19)30076-2 861 862 KotloK, K. L., Nataro, J. P ., Blackwelder, W. C., Nasrin, D., Farag, T. H., Panchalingam, S., 863 Wu, Y ., Sow, S. O., Sur, D., Breiman, R. F ., Faruque, A. S., Zaidi, A. K., Saha, D., 864 Alonso, P . L., Tamboura, B., Sanogo, D., Onwuchekwa, U., Manna, B., 865 Ramamurthy, T., Kanungo, S., Ochieng, J. B., Omore, R., Oundo, J. O., Hossain, A., 866 Das, S. K., Ahmed, S., Qureshi, S., Quadri, F ., Adegbola, R. A., Antonio, M., 867 Hossain, M. J., Akinsola, A., Mandomando, I., Nhampossa, T., Acácio, S., Biswas, 868 K., O'Reilly, C. E., Mintz, E. D., Berkeley, L. Y ., Muhsen, K., Sommerfelt, H., Robins-869 Browne, R. M., & Levine, M. M. (2013, Jul 20). Burden and aetiology of diarrhoeal 870 disease in infants and young children in developing countries (the Global Enteric 871 Multicenter Study, GEMS): a prospective, case-control study. Lancet, 382(9888), 872 209-222. https://doi.org/10.1016/s0140-6736(13)60844-2 873 874 Li, M., Yang, F ., Hou, T., Gong, X., Li, N., Sibley, L. D., Feng, Y ., Xiao, L., & Guo, Y . (2024, 875 2024/09/19). Variant surface protein GP60 contributes to host infectivity of 876 Cryptosporidium parvum. Communications Biology, 7 (1), 1175. 877 https://doi.org/10.1038/s42003-024-06885-0 878 879 Manjunatha, U. H., Lakshminarayana, S. B., Jumani, R. S., Chao, A. T., Young, J. M., Gable, 880 J. E., Knapp, M., Hanna, I., Galarneau, J. R., Cantwell, J., Kulkarni, U., Turner, M., 881 Lu, P ., Darrell, K. H., Watson, L. C., Chan, K., Patra, D., Mamo, M., Luu, C., Cuellar, 882 C., Shaul, J., Xiao, L., Chen, Y . B., Carney, S. K., Lakshman, J., Osborne, C. S., 883 Zambriski, J. A., Aziz, N., Sarko, C., & Diagana, T. T. (2024, Nov). Cryptosporidium 884 PI(4)K inhibitor EDI048 is a gut -restricted parasiticidal agent to treat paediatric 885 enteric cryptosporidiosis. Nat Microbiol, 9 (11), 2817 -2835. 886 https://doi.org/10.1038/s41564-024-01810-x 887 888 Manque Patricio, A., Tenjo, F ., Woehlbier, U., Lara Ana, M., Serrano Myrna, G., Xu, P ., Alves 889 João, M., Smeltz Ronald, B., Conrad Daniel, H., & Buck Gregory, A. (2011, 890 2011/11/01). Identification and Immunological Characterization of Three 891 Potential Vaccinogens against Cryptosporidium Species. Clinical and Vaccine 892 Immunology, 18(11), 1796-1802. https://doi.org/10.1128/CVI.05197-11 893 894 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 28 Mead, J. R., Arrowood, M. J., & Sterling, C. R. (1988, Feb). Antigens of Cryptosporidium 895 sporozoites recognized by immune sera of infected animals and humans. J 896 Parasitol, 74(1), 135-143. 897 898 O'Ryan, M. (2017). Rotavirus Vaccines: a story of success with challenges ahead. 899 F1000Res, 6, 1517. https://doi.org/10.12688/f1000research.11912.1 900 901 Pawlowic, M. C., Somepalli, M., Sateriale, A., Herbert, G. T., Gibson, A. R., Cuny, G. D., 902 Hedstrom, L., & Striepen, B. (2019). Genetic ablation of purine salvage in 903 Cryptosporidium parvum reveals nucleotide uptake from the host cell. 904 Proceedings of the National Academy of Sciences, 116 (42), 21160 -21165. 905 https://doi.org/doi:10.1073/pnas.1908239116 906 907 Pawlowic, M. C., Vinayak, S., Sateriale, A., Brooks, C. F ., & Striepen, B. (2017, Aug 11). 908 Generating and Maintaining Transgenic Cryptosporidium parvum Parasites. Curr 909 Protoc Microbiol, 46, 20b.22.21-20b.22.32. https://doi.org/10.1002/cpmc.33 910 911 Putignani, L., Possenti, A., Cherchi, S., Pozio, E., Crisanti, A., & Spano, F . (2008, Jan). The 912 thrombospondin-related protein CpMIC1 (CpTSP8) belongs to the repertoire of 913 micronemal proteins of Cryptosporidium parvum. Mol Biochem Parasitol, 157(1), 914 98-101. https://doi.org/10.1016/j.molbiopara.2007.09.004 915 916 Sateriale, A., Šlapeta, J., Baptista, R., Engiles, J. B., Gullicksrud, J. A., Herbert, G. T., 917 Brooks, C. F ., Kugler, E. M., Kissinger, J. C., Hunter, C. A., & Striepen, B. (2019, Jul 918 10). A Genetically Tractable, Natural Mouse Model of Cryptosporidiosis OKers 919 Insights into Host Protective Immunity. Cell Host Microbe, 26 (1), 135-146.e135. 920 https://doi.org/10.1016/j.chom.2019.05.006 921 922 Schlott, A. C., Holder, A. A., & Tate, E. W. (2018, 2018/04/13). N-Myristoylation as a Drug 923 Target in Malaria: Exploring the Role of N -Myristoyltransferase Substrates in the 924 Inhibitor Mode of Action. ACS Infectious Diseases, 4 (4), 449 -457. 925 https://doi.org/10.1021/acsinfecdis.7b00203 926 927 Shaw, S., Cohn, I. S., Baptista, R. P ., Xia, G., Melillo, B., Agyabeng-Dadzie, F ., Kissinger, J. 928 C., & Striepen, B. (2024). Genetic crosses within and between species of 929 Cryptosporidium. Proceedings of the National Academy of Sciences, 121 (1), 930 e2313210120. https://doi.org/doi:10.1073/pnas.2313210120 931 932 Sidik, S. M., Huet, D., Ganesan, S. M., Huynh, M. H., Wang, T., Nasamu, A. S., Thiru, P ., 933 Saeij, J. P . J., Carruthers, V . B., Niles, J. C., & Lourido, S. (2016, Sep 8). A Genome-934 wide CRISPR Screen in Toxoplasma Identifies Essential Apicomplexan Genes. 935 Cell, 166(6), 1423-1435.e1412. https://doi.org/10.1016/j.cell.2016.08.019 936 937 Smith, T. A., Lopez -Perez, G. S., Herneisen, A. L., Shortt, E., & Lourido, S. (2022, 938 2022/06/01). Screening the Toxoplasma kinome with high -throughput tagging 939 identifies a regulator of invasion and egress. Nature Microbiology, 7(6), 868-881. 940 https://doi.org/10.1038/s41564-022-01104-0 941 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 29 942 Sow, S. O., Muhsen, K., Nasrin, D., Blackwelder, W. C., Wu, Y ., Farag, T. H., Panchalingam, 943 S., Sur, D., Zaidi, A. K., Faruque, A. S., Saha, D., Adegbola, R., Alonso, P . L., 944 Breiman, R. F ., Bassat, Q., Tamboura, B., Sanogo, D., Onwuchekwa, U., Manna, B., 945 Ramamurthy, T., Kanungo, S., Ahmed, S., Qureshi, S., Quadri, F ., Hossain, A., Das, 946 S. K., Antonio, M., Hossain, M. J., Mandomando, I., Nhampossa, T., Acácio, S., 947 Omore, R., Oundo, J. O., Ochieng, J. B., Mintz, E. D., O'Reilly, C. E., Berkeley, L. Y ., 948 Livio, S., Tennant, S. M., Sommerfelt, H., Nataro, J. P ., Ziv -Baran, T., Robins -949 Browne, R. M., Mishcherkin, V ., Zhang, J., Liu, J., Houpt, E. R., KotloK, K. L., & 950 Levine, M. M. (2016, May). The Burden of Cryptosporidium Diarrheal Disease 951 among Children < 24 Months of Age in Moderate/High Mortality Regions of Sub -952 Saharan Africa and South Asia, Utilizing Data from the Global Enteric Multicenter 953 Study (GEMS). PLoS Negl Trop Dis, 10 (5), e0004729. 954 https://doi.org/10.1371/journal.pntd.0004729 955 956 Striepen, B., Pruijssers, A. J., Huang, J., Li, C., Gubbels, M. J., Umejiego, N. N., Hedstrom, 957 L., & Kissinger, J. C. (2004, Mar 2). Gene transfer in the evolution of parasite 958 nucleotide biosynthesis. Proc Natl Acad Sci U S A, 101 (9), 3154 -3159. 959 https://doi.org/10.1073/pnas.0304686101 960 961 Tandel, J., Walzer, K. A., Byerly, J. H., Pinkston, B., Beiting, D. P ., & Striepen, B. (2023, Apr 962 25). Genetic Ablation of a Female -Specific Apetala 2 Transcription Factor Blocks 963 Oocyst Shedding in Cryptosporidium parvum. mBio, 14 (2), e0326122. 964 https://doi.org/10.1128/mbio.03261-22 965 966 Troeger, C., Blacker, B. F ., Khalil, I. A., Rao, P . C., Cao, S., Zimsen, S. R. M., Albertson, S. 967 B., Stanaway, J. D., Deshpande, A., Abebe, Z., Alvis -Guzman, N., Amare, A. T., 968 Asgedom, S. W., Anteneh, Z. A., Antonio, C. A. T., Aremu, O., Asfaw, E. T., Atey, T. 969 M., Atique, S., Avokpaho, E. F . G. A., Awasthi, A., Ayele, H. T., Barac, A., Barreto, M. 970 L., Bassat, Q., Belay, S. A., Bensenor, I. M., Bhutta, Z. A., Bijani, A., Bizuneh, H., 971 Castañeda-Orjuela, C. A., Dadi, A. F ., Dandona, L., Dandona, R., Do, H. P ., Dubey, 972 M., Dubljanin, E., Edessa, D., Endries, A. Y ., Eshrati, B., Farag, T., Feyissa, G. T., 973 Foreman, K. J., Forouzanfar, M. H., Fullman, N., Gething, P . W., Gishu, M. D., 974 Godwin, W. W., Gugnani, H. C., Gupta, R., Hailu, G. B., Hassen, H. Y ., Hibstu, D. T., 975 Ilesanmi, O. S., Jonas, J. B., Kahsay, A., Kang, G., Kasaeian, A., Khader, Y . S., Khalil, 976 I. A., Khan, E. A., Khan, M. A., Khang, Y .-H., Kissoon, N., Kochhar, S., KotloK, K. L., 977 Koyanagi, A., Kumar, G. A., Magdy Abd El Razek, H., Malekzadeh, R., Malta, D. C., 978 Mehata, S., Mendoza, W., Mengistu, D. T., Menota, B. G., Mezgebe, H. B., Mlashu, 979 F . W., Murthy, S., Naik, G. A., Nguyen, C. T., Nguyen, T. H., Ningrum, D. N. A., Ogbo, 980 F. A . , O l a g u n j u , A . T. , Pa u d e l , D. , Pl a t t s-Mills, J. A., Qorbani, M., Rafay, A., Rai, R. 981 K., Rana, S. M., Ranabhat, C. L., Rasella, D., Ray, S. E., Reis, C., Renzaho, A. M. N., 982 Rezai, M. S., Ruhago, G. M., Safiri, S., Salomon, J. A., Sanabria, J. R., Sartorius, B., 983 Sawhney, M., Sepanlou, S. G., Shigematsu, M., Sisay, M., Somayaji, R., 984 Sreeramareddy, C. T., Sykes, B. L., TaKere, G. R., Topor-Madry, R., Tran, B. X., Tuem, 985 K. B., Ukwaja, K. N., Vollset, S. E., Walson, J. L., Weaver, M. R., Weldegwergs, K. G., 986 Werdecker, A., Workicho, A., Yenesew, M., Yirsaw, B. D., Yonemoto, N., El Sayed 987 Zaki, M., Vos, T., Lim, S. S., Naghavi, M., Murray, C. J. L., Mokdad, A. H., Hay, S. I., 988 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 30 & Reiner, R. C., Jr. (2018). Estimates of the global, regional, and national morbidity, 989 mortality, and aetiologies of diarrhoea in 195 countries: a systematic analysis for 990 the Global Burden of Disease Study 2016. The Lancet Infectious Diseases, 18(11), 991 1211-1228. https://doi.org/10.1016/S1473-3099(18)30362-1 992 993 Troeger, C., Forouzanfar, M., Rao, P . C., Khalil, I., Brown, A., Reiner, R. C., Fullman, N., 994 Thompson, R. L., Abajobir, A., Ahmed, M., Alemayohu, M. A., Alvis -Guzman, N., 995 Amare, A. T., Antonio, C. A., Asayesh, H., Avokpaho, E., Awasthi, A., Bacha, U., 996 Barac, A., Betsue, B. D., Beyene, A. S., Boneya, D. J., Malta, D. C., Dandona, L., 997 Dandona, R., Dubey, M., Eshrati, B., Fitchett, J. R. A., Gebrehiwot, T. T., Hailu, G. 998 B., Horino, M., Hotez, P . J., Jibat, T., Jonas, J. B., Kasaeian, A., Kissoon, N., KotloK, 999 K., Koyanagi, A., Kumar, G. A., Rai, R. K., Lal, A., El Razek, H. M. A., Mengistie, M. 1000 A., Moe, C., Patton, G., Platts -Mills, J. A., Qorbani, M., Ram, U., Roba, H. S., 1001 Sanabria, J., Sartorius, B., Sawhney, M., Shigematsu, M., Sreeramareddy, C., 1002 Swaminathan, S., Tedla, B. A., Jagiellonian, R. T. -M., Ukwaja, K., Werdecker, A., 1003 Widdowson, M.-A., Yonemoto, N., El Sayed Zaki, M., Lim, S. S., Naghavi, M., Vos, 1004 T., Hay, S. I., Murray, C. J. L., & Mokdad, A. H. (2017, 2017/09/01/). Estimates of 1005 global, regional, and national morbidity, mortality, and aetiologies of diarrhoeal 1006 diseases: a systematic analysis for the Global Burden of Disease Study 2015. The 1007 Lancet Infectious Diseases, 17 (9), 909 -948. 1008 https://doi.org/https://doi.org/10.1016/S1473-3099(17)30276-1 1009 1010 Ungar, B. L., & Nash, T. E. (1986, Jul). Quantification of specific antibody response to 1011 Cryptosporidium antigens by laser densitometry. Infect Immun, 53 (1), 124 -128. 1012 https://doi.org/10.1128/iai.53.1.124-128.1986 1013 1014 Vinayak, S., Pawlowic, M. C., Sateriale, A., Brooks, C. F ., Studstill, C. J., Bar -Peled, Y., 1015 Cipriano, M. J., & Striepen, B. (2015, Jul 23). Genetic modification of the diarrhoeal 1016 pathogen Cryptosporidium parvum. Nature, 523 (7561), 477 -480. 1017 https://doi.org/10.1038/nature14651 1018 1019 Yen, C., Tate, J. E., Hyde, T. B., Cortese, M. M., Lopman, B. A., Jiang, B., Glass, R. I., & 1020 Parashar, U. D. (2014). Rotavirus vaccines: current status and future 1021 considerations. Hum Vaccin Immunother, 10 (6), 1436 -1448. 1022 https://doi.org/10.4161/hv.28857 1023 1024 Young, J., Dominicus, C., Wagener, J., Butterworth, S., Ye, X., Kelly, G., Ordan, M., 1025 Saunders, B., Instrell, R., Howell, M., Stewart, A., & Treeck, M. (2019, 2019/09/03). 1026 A CRISPR platform for targeted in vivo screens identifies Toxoplasma gondii 1027 virulence factors in mice. Nature Communications, 10 (1), 3963. 1028 https://doi.org/10.1038/s41467-019-11855-w 1029 1030 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 31 1031 1032 Figure 1. In vivo CRISPR Screen of the Cryptosporidium Pyrimidine Salvage Pathway . a. Schematic illustrates how a 1033 knockout vector is generated. The first Golden Gate reaction occurs between a Cas9 expression plasmid and a 300bp unique 1034 segment (containing the two 50bp homology arms (one of which serves dual function as the gRNA), a unique DNA barcode, 1035 BsmBI restriction enzyme sites and the tracrRNA). The second Golden Gate reaction, using the BsmBI restriction enzyme sites, 1036 inserts a variable selection/reporter cassette to generate a complete knockout vector. The knockout vector then contains all the 1037 machinery to disrupt a gene of interest by inserting a variable selection cassette and barcode at the genomic locus. b. A knockout 1038 vector targeting thymidine kinase (cgd5_4440) was generated and transfected into C. parvum sporozoites that were used to 1039 infect Ifng-/- mice under paromomycin selection . Faecal samples were collected and the luminescence in faecal material was 1040 monitored. Data shows the mean of a biological replicate ± SEM of the 2 technical replicates , n = 4 mice. c. Alignment of reads 1041 from whole genome sequencing of the thymidine kinase knockout strain to the C. parvum IOWAII genome at the site of insertion. 1042 Note the complete lack of alignment to the PAM site, which is removed by the homologous repair event. d. Overview of CRISPR 1043 screening method. Following construction of KO vectors (detailed in 1a), sporozoites are transfected with gene specific KO 1044 vectors and used to infect mice. Specific barcodes are then amplified via high fidelity PCR and used to calculate fold enrichment 1045 (log2[%barcode(output) / %barcodes(input)]) which measures the relative fitness contribution of each gene. e. Mouse faecal 1046

was collected, and luminescence was monitored in the pooled sample from the pyrimidine salvage pathway CRISPR 1047 screens at 1 and 2 KO vectors per gene. Data shown is the mean of the biological replicate ± SEM of the 2 technical replicates, 1048 n = 4 mice per screen. f. Rank ordered fold enrichment scores from the 1 and 2 KO vectors per gene pyrimidine salvage CRISPR 1049 screens. The colour indicates the relative fitness contribution, with dark purple showing high fitness conferring and dark green 1050 showing low fitness conferring. g. Comparison of the fold enrichment scores between the 1 and 2 KO vectors per gene pyrimidine 1051 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 32 salvage CRISPR screens. Confidence refers to the inverse of the 95% confidence interval when comparing the log2 fold change 1052 scores from each screen (see methods). 1053 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 33 1054 Figure 2. Validating Screen Results with diCRE Mediated Excision. a-b. TK-diCRE (shown in green) and ribonucleotide 1055 reductase (RNR) -diCRE (shown in purple) parasites were generated and used to infect an HCT8 monolayer in the presence or 1056 absence of rapamycin. Genomic DNA was extracted at 6, 24 and 48 hours and diagnostic PCRs confirmed the level of excision 1057 at the given time points. c. HCT8 monolayers infected with T K-diCRE and RNR-diCRE parasites in the presence or absence of 1058 rapamycin. At 6, 24 and 48 hours the monolayers were fixed, stained and the number of parasites per host nuclei was quantified. 1059 Representative images are shown, with nuclei in blue (Hoechst), and parasites in green (Vicia villosa lectin ). Data shown is 1060 representative of 2 biological replicates ± sd of 4 technical replicates. Significance is determined using a two -tailed unpaired t -1061 test, ns = not significant, ** = p ≤ 0.01. d. TK-diCRE or RNR-diCRE parasites were used to infect Ifng-/- mice, which were either 1062 treated with rapamycin or vehicle ( DMSO) in their drinking water starting at day 2 post infection. Data shown i s a biological 1063 replicate ± SEM of the 2 technical replicates, n = 2 mice per condition. 1064 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 34 1065 1066 Figure 3. In vivo CRISPR Screen of Cryptosporidium Vaccine Candidates. a. Mouse faecal material was collected and 1067 luminescence was monitored during infection. Each replicate was conducted with 2 KO vectors per gene. Data shown is the 1068 mean of the biological replicate ± SEM of the 2 technical replicates , n = 5 Ifng-/- mice for repeat 1 and n = 3 Ifng-/- mice for repeat 1069 2. b. Rank ordered fold enrichment scores from the vaccine candidate CRISPR screens. The colour indicates the relative fitness 1070 contribution of a gene, with dark purple being high fitness conferring and dark green being low fitness conferring. c. Comparison 1071 of the fold enrichment scores between the replicate CRISPR screens for each gene. Confidence refers to the inverse of the 95% 1072 confidence interval when comparing the log2 fold change scores from each screen (see methods). d. Mouse faecal material was 1073 collected and luminescence was monitored during infection. Data shows a biological replicate ± SEM of the 2 technical replicates, 1074 n = 5 Ifng-/- mice. e-f. Barcodes from KO parasites could be easily monitored over time (e) and within individual mice (f). 1075 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 35 Figure 4. Immunodominant Antigen 23 is Essential and Required for Reinvasion of Host Cells. a-b. HCT8 cell monolayers 1076 infected with Cp23-diCRE in the presence or absence of rapamycin . At 6, 24 and 48 hours, gDNA was extracted and diagnostic 1077 PCRs confirmed the level of excision at the given timepoint. c. HCT8 cell monolayers infected with Cp23-diCRE in the presence 1078 or absence of rapamycin . At 6, 24 and 48 hours, the monolayer was fixed and stained and the parasite per host nuclei was 1079 quantified. Representative images are shown, with nuclei in blue (Hoechst), and parasites in green (Vicia villosa lectin ). Data 1080 shown is representative of two biological replicates ± sd of the 4 technical replicates. Significance is determined using a two -1081 tailed unpaired t-test, ns = not significant, * = p ≤ 0.05, ** = p ≤ 0.01. d. Ifng-/- mice were infected with 50,000 Cp23-diCRE parasites 1082 and treated with rapamycin or DMSO control in their drinking water at 2 days post infection. Data shown is the mean of a biological 1083 replicate ± SEM of the 2 technical replicates, n = 2 mice per condition. e. Super resolution microscopy at 24 hours post infection 1084 in vitro with Cp23-diCRE parasites in the presence or absence of rapamycin; green ( helix pomatia agglutinin (HPA)), parasite; 1085 magenta, (sytox), nuclei; red (Cp23 Ab), Cp23. Scale = 2 µm. f. Expansion microscopy of the C. parvum asexual stages: 1086 sporozoite, merozoite, trophozoite and meront, and the sexual stages: macrogamont (female) and microgamete (male); green 1087 (helix pomatia agglutinin (HPA)); blue (N -hydroxysuccinimide (NHS ester); magenta (sytox); red ( aCp23). Scale bar = 5 µm. 1088 Median expansion factor of 4 .5. g. Transmission electron microscopy of an excysted and unexcysted C. parvum sporozoite 1089 coupled with immunogold labelling with Cp23 antibody. Scale = 300nm. h. Permeabilisation assay of C. parvum sporozoites. The 1090 permeabilised condition used 0.1% Triton X -100 and the non -permeabilised condition used PBS . Grey (tryptophan synthase 1091 (aTrpB); green (Vicia villosa lectin); red (aCp23); blue (Hoechst). All images were taken using the same settings and exposure. 1092 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 36 Scale bar = 5µm. i. HCT8 cell monolayers were infected with either Cp23-diCRE or wildtype parasites in the presence or absence 1093 of rapamycin, and at 22 hours post infection life stage s were quantified. Data shown is 2 biological replicates, n = 438 Cp23 -1094 diCRE - RAP, n = 173 Cp23-diCRE + RAP, n = 213 wildtype - RAP, n = 203 wildtype + RAP. Significance is determined using a 1095 two-tailed unpaired t -test, ns = not significant , **** = p ≤ 0.0001. j-k. Live imaging of the reinvasion event was carried out at 18 1096 hours post infection of an HCT8 monolayer with Cp23-diCRE parasites in the presence or absence of rapamycin. Movement of 1097 merozoites was tracked using Fiji software. Representative images are shown in k. with videos in supplementary data. All live 1098 microscopy data shown is from 2 biological replicate s. Significance is determined using a two -tailed unpaired t -test, **** = p ≤ 1099 0.0001. 1100 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 37 1101 Supplementary 1. Advancement to CRISPR Screening . a-b. Relative luminescence units (RLU) from a HCT8 cell monolayer 1102 infected for 24 hours with C. parvum parasites that were transiently transfected with a nanoluciferase-mCherry expression vector, 1103 trialling multiple electroporation programs ( a) and excystation reagents used prior to transfection ( b). TDC (taurodeoxycholate), 1104 NaCIO (sodium hypochlorite), TC (taurocholate), RMPI (media) , DC (deoxycholate) PBS (phosphate -buffered saline) . c. 1105 Quantification of the improved transfection efficiency measured by RLU. In black, the standard method, in yellow, the revised 1106 method. d. Visualisation of the increased transfection efficiency of the revised method (right panel) compared to the standard 1107

(left panel) using the mCherry-expression vector. Blue (Hoechst), nuclei; green (Vicia villosa lectin (VVL)), parasites; red 1108 (mCherry Ab), mCherry/transfected parasites. All data shows the mean ± standard deviation from 2 biological replicates. e. RLU 1109 from a HCT8 cell monolayer infected for 24 hours with C. parvum parasites transfected with a repair cassette designed to replace 1110 the thymidine kinase gene, using different lengths of homology repair. Data shows the mean ± sd from three technical replicates. 1111 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 38 1112 Supplementary 2. Refining diCRE Mediated Excision. a. Schematics of disruption of a nanoluciferase (Nluc) gene mid-codon 1113 with a HAP2 (cgd8_2220) intron (Nluc int) or with a HAP2 intron containing a loxP sequence (Nluc loxPint). Vectors were 1114 transiently transfected into C. parvum sporozoites that were allowed to infect an HCT8 monolayer for 24 hours. Data shown is 1115 the mean of 2 biological replicates ± sd of 4 technical replicates . Significance is determined using a one-way ANOVA, ns = not 1116 significant b. Schematic of the TK -T2A-diCRE parasite line and the corresponding luminescence from mouse faecal material 1117 when generating these parasites. Data shown is a biological replicate ± SEM, n = 4 Ifng-/- mice. c. Diagnostic PCR to determine 1118 the level of excision occurring in the TK-T2A-diCRE parasite line after 24 hours infection of a HCT8 cell monolayer during 1119 treatment with rapamycin . d. Schematic of the TK -diCRE parasite line and the corresponding luminescence from mouse faecal 1120

when generating the parasites. Data shown is a biological replicate ± SEM, n = 4 Ifng-/- mice. e. Diagnostic PCR to 1121 determine the level of excision occurring in the TK-diCRE parasite line after 24 hours infection of a HCT8 cell monolayer during 1122 treatment with rapamycin. 1123 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 39 1124 Supplementary 3. diCRE Mediated Excision of TK and RNR . a. Schematic and genotyping PCR of TK-diCRE parasites. b. 1125 TK-diCRE parasites were allowed to infect a HCT8 monolayer and g enomic DNA was extracted at 2, 4, 8, 24, 32 and 48 hours 1126 post infection in the presence and absence of rapamycin. Diagnostic PCRs confirmed the level of excision that had occurred at 1127 the respective timepoint. c-d. TK-diCRE parasites were allowed to infect a HCT8 monolayer in the presence and absence of 1128 rapamycin, and the monolayer was fixed and stained at 4, 8, 12, 16, 24, 32 and 48 hours post infection. The HA tag per parasite 1129 was quantified by automated imaging , representation images are shown in d; blue (Hoechst); green (Vicia villosa lectin (VVL)); 1130 red ( aHA). Data shown is representative of 2 biological replicates, each with 2 technical replicates ± sd. e. Schematic and 1131 genotyping PCR of RNR-diCRE parasites. f. EdU assay with quantifications showing newly synthesised DNA between 28 and 1132 32 hours in wildtype, TK- and RNR-diCRE parasites in the presence and absence of rapamycin . Blue (Hoechst); green, ( Vicia 1133 villosa lectin (VVL)); magenta, (EdU). Data shown is two biological replicates with the number of parasites quantified indicated. 1134 Scale bar = 10µm. 1135 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 40 1136 Supplementary 4. Immunodominant Antigen 23 is Essential and Required for Reinvasion of Host Cells. a. Schematic and 1137 genotyping PCR of Cp23-diCRE parasites. b. Visualising gliding C. parvum sporozoites using super-resolution microscopy. Green 1138 (Vicia villosa lectin (VVL) which marks both parasite and trail ); red (aCp23); blue (Hoechst). Scale bar = 5 µm. c-e. Cp23-diCRE 1139 and wildtype parasites were allowed to infect an HCT8 monolayer in the presence or absence of rapamycin . At 22 hours post 1140 infection the life cycle stage was quantified. Data shows 2 biological replicates . Significance is determined using a two -tailed 1141 unpaired t-test, ns = not significant, **** = p ≤ 0.0001. Representative images are shown in e. Blue (Hoechst); green (Vicia villosa 1142 lectin (VVL)); magenta (phallodin - actin); red (aCp23). Scale bar = 10 µm. 1143 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 41 1144 Supplementary 5. Chromosome Locations of the Genes Knocked Out During CRISPR Screens. a-b. Locations of the genes 1145 from the pyrimidine salvage (a) and vaccine candidate (b) CRISPR screens. The colour indicates the relative fitness of the gene 1146 knocked out with dark purple being a highly fitness conferring gene and dark green being a low fitness conferring gene. 1147 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 42 1148 Supplementary 6. Immunodominant Antigen 23 Antibody Does Not Neutralise Infection In Vitro. a. Quantifying the number 1149 of attached sporozoites on a poly-D treated surface when incubated with the Cp23 antibody or isotype control antibody dilution 1150 series. Data shows the mean ± sd of 4 technical replicates. The black line shows Cp23 antibody incubations, the grey line shows 1151 isotype control antibody incubations, and the dotted line shows no antibody control incubations. b. Quantification of invasion 1152 (parasite per host nuclei) when cell monolayers are infected in the presence of the Cp23 antibody or isotype control . The data 1153 shows the mean and ± sd of the 4 technical replicates. The back line shows Cp23 antibody incubations, the grey line shows 1154 isotype control antibody incubations, and the dotted line shows no antibody control incubations. 1155 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 43 Supplementary videos. Live imaging of the reinvasion event was carried out at 18 hours post infection of an HCT8 monolayer 1156 with Cp23-diCRE parasites in the presence of rapamycin or vehicle (DMSO) control. 1157 Video S1. DMSO treatment 1 1158 Video S2. DMSO treatment 2 1159 Video S3. Rapamycin treatment 1 1160 Video S4. Rapamycin treatment 2 1161 1162 1163 Table 1. Genes in the Pyrimidine Salvage CRISPR Screen 1164 Gene ID Name Gene Fitness Contribution cgd1_1900 Fur1p like uracil phosphoribosyltransferase (UPRT) Dispensable but fitness conferring (Kimball et al., 2024) cgd1_3140 Adenylate kinase/UMP-CMP kinase cgd2_1630 Cytidine and deoxycytidylate deaminase family cgd2_2780 dCMP deaminase cgd4_4460 Bifunctional dihydrofolate reductase/thymidylate synthase (DHFR) Dispensable (Pawlowic et al., 2019) cgd5_1710 CTP synthase cgd5_3630 Thymidylate kinase cgd5_4440 Thymidine kinase (TK) Dispensable (Vinayak et al., 2015) cgd6_1950 Ribonucleotide reductase (RNR) cgd7_1470 Cytidine and deoxycytidylate deaminase zinc -binding domain containing protein cgd8_2810 Phosphoribulokinase/uridine kinase/Uracil phosphoribosyltransferase 1165 1166 Table 2. Genes in the Vaccine Candidate CRISPR Screen 1167 Gene ID Name Gene Fitness Contributions cgd3_3370 Uncharacterized protein cgd4_32 Apical glycoprotein 1 (AGP1) Dispensable (Akey et al., 2023) cgd4_3620 Immunodominant antigen 23393226 (Cp23) cgd6_1080 Glycoprotein GP40 (GP60) Dispensable, but fitness conferring (Li et al., 2024) cgd6_1660 Uncharacterized protein with Thrombospondin type -1 (TSP1) repeat (TSP11) cgd6_2330 Uncharacterized protein cgd6_780 Thrombospondin type -1 (TSP1) repeat/EGF -like domain containing protein (TSP8/MIC1) cgd7_1960 WD40/YVTN repeat-like+signal peptide-containing protein cgd7_4020 Cryptopsporidial mucin (GP900) cgd7_4330 Apical glycoprotein 2 (AGP2) Likely Essential (Akey et al., 2023) cgd7_5520 Hemogen 1168 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 44 Table 3. Homology Arms/gRNAs used in Pyrimidine Salvage 1 KO Vector CRISPR Screen 1169 KO Vector Target Hom1/gRNA Hom2 cgd1_1900_319 CTAATAGGATCTGGTGAAGCGATGGAGAATGCGCTTAGATTTGTATGCAG CAAATTCGTTTTCAGAAGAGTTATTTAATAACACCTTCCCTATTCTACAT cgd1_3140_313 TCAGTATCTGGAATAATCTTATACCAACCATCTAAATTATTTTGGTTACG AGATGCAAGAGTATGGTTGGAATGATAAATATTTTTTGATTGATGGTTTT cgd2_1630_243 AACGTAAGCTATAGAAACTATAAATTAAGTAAATAAACATATATTACTCA TTAACTTAACTATTAATTTTTCTAAAGCAACAATTTCACAGTGTCTAGTT cgd2_2780_524 CTCTTCAAAATCTTGAACCAATATTGAGGCAAGTTCTTTTAACTTTCCTG GAATGCAAATTTTGAAGAAGTTGAAAAATCTATTGAGGTATTAAAAAATT cgd4_4460_156 TAATTGTGACTCGAATAAGAAGAATGCACTAATTATGGGAAGAAAAACAT GGAAATAACGACAATTTTTCTATTTTTAAGAGGTCTTCTTCCAATTGAAT cgd5_1710_235 ATAGATCCTTATTTGAATATTGACGCAGGGACTATGTCTCCCTTTGAGCA TTCCCAAATCCAAATCGACTTCCTCTCCATCATCTAATACAAAAACCTCA cgd5_3630_140 ATATGAAATTGGGGGTAAGAAGAGGGGGATATGCTGGCAGAAAGAGAGAA GAATCCCTCTCCTCCATTAGTAATTTTGACTAACTTTAGATAAAGGAATT cgd5_4440_332 AGAGAATTATGCATTGTTGTTGATAAGCTAAATATTCCAGTACTATGCTA ATAAGTATTTACTTCCTTCAAATAAATTTCCCTTAAAGTCTGTTCTCAAA cgd6_1950_240 TGTGTAGATTAGAAGTCGATATCCGAGCAGCGAGTTTAGAGAAATCATTG ATCTGAATTGGACGAACTTGCTTCGCAAACTTGTGCATATATGGCAGCTA cgd7_1470_400 GATAAAGGCCAAATTTTTGAAAGTTGGGAAAACTTTTCCCTAGTTAATGG AAGAAAACCATATTGATAAAATTTGGAATGTAATAGAGGTTCCAAAGAAT cgd8_2810_291 TTCTGTGGACTTTGAACTTCTATACAATGTTTTACTAAGTTTGAAAAACG TTCCAATCTCTTATGTTGTTTAAAACAGTAGTTAGGGATGTGAACCCCTT 1170 1171 Table 4. Homology Arms/gRNAs used in Pyrimidine Salvage 2 KO Vector CRISPR Screen 1172 KO Vector Target Hom1/gRNA Hom2 cgd1_1900_232 CTTCCTTATGATTATAAGGAAATTAAAACCCCAAATGGAATCGAAGTCAA CACCAGATCCTATTAGGCTCACGCCGCAGATGGGAGTGTTAAATGCAATT cgd1_3140_162 AATGTCTAGAAAGGATGAAACCAGCGAGTTAATTGACAGTTATATCAGAG CATCTTCTTCTTTAACAAACCAACAGTAATCTCAACAGGAACAATTAATC cgd2_1630_37 TATAACAATGAAGAGTTAGAGATGTTTATGAAAAGAGCGATTGAACTAGT TCTGACTAAAAATAATAAATTATTACAGCTTTACAAAGATATTAGCTTTA cgd2_2780_222 ATGTGGCAAAAAAGCAATAAGACAAATTTGCTTTTGCGGAAGCGATTCAG TATCATGAAAGCTTTCTTTGCTCTGAACCTAATACAATCATCATAAGATT cgd4_4460_9 TGTGAATTTCAGAACTTTTAAAATGAGTAAAAAGAACGTTTCAATTGTTG CCAAGGTAATTGTCCGTTAATTCCTATTCCTCTACTCAAAACAGAAGCTG cgd5_1710_160 CTAGGTAAAGGAATAGCTATAAGCTCGCTTGGCTTATGCCTTAAAAGCAG CTGCGTCAATATTCAAATAAGGATCTATTTTTATCGCTGTTACATTATAT cgd5_3630_73 GTATTAGAAGGAACAGATAGGTAAGTTTGTTAAAATTTAAATATTAAGAG CAGCATATCCCCCTCTTCTTACCCCCAATTTCATATTTATTTCTTCTCCA cgd5_4440_223 ATTTTTTCCTTATTATTTTCTTGATTAATAATATCAAGTAGTTTTAAATC GCTCAATTTGCTCTAGAATTGGTCTCTCAGAAAAAGCACACACATTCACT cgd6_1950_124 CTATATAGACCATTTATCACTGCTTGAGTAACCCTCGCTGGGTCAACAAG CATTTGACCAGATTCTTAGCAGAATCACTAAATTATCATATGGACTTCAT cgd7_1470_202 AATTCCAAGTCATTCTCTTTAATAACTACTTTACGCATTCTTTTTATGTG CTAATGAATTTTCAAGATTTGTTAATAATGACCAGTTTACAAAGAGATTC cgd8_2810_203 TTACGGTAATTGAGACTGATAGTTTTTATAAAACTCCTGTCTTAGAAGAG TCAAAGTCCACAGAATTAGGATGATCAAAGTTGTAGTCTGCCATAGTTTG * Table 3 KO used in addition to Table 4 KO vectors 1173 1174 1175 Table 5. Homology Arms/gRNAs used in Vaccine Candidate 2 KO Vector CRISPR Screen 1176 KO Vector Target Hom1/gRNA Hom2 cgd7-4020_67_revco ACTCGATTTAAATGCAAGTGAAAAAAGTGGGTTCATAATAACAGCCACAA GGTGAAGTCAAAAATCATGGTGAACATTAAAGTGAGCTCATCGGCAATAG cgd7-4020_188_revco TCTTGCCAGTAGAATCAATAAGCAAGAAAGTTGTTGGGTCTAAATCTGAA ATTGAATCATCTGGTGCAGTTTCAAATGAAAAATTTGTAATCCCATCTCT cgd4-32_141_revcom GATTGCCACGTCAAGTTTAAATCTATTGCATTACTTGCCATTAATTCCTC GTGGCAAACTTATTGACTCAATAAACGAAAATTATGATAGATTTTCATAT cgd4-32_293 CTGAACTCTTCAACAAGAGAAGAATTAAAAACAAACTGGAGTTACACTAC TACCTTCTAATGCATTGATTGTGCACGAAACAGAACTTAGATTTTTGACT cgd4-3620_84_revcom CTTTGTTTGTTAGCATCTGCTGCAGATTTATTTTCAGCAACTTTAGTTTC ATTTTTAGTTTTATTATTCAATATTAAAAATGGGTTGTTCATCATCAAAG cgd4- 3620_183_revcom GACTTCTTTGGTTCTTCTGGCTTTTGTTGAGCTTGGTTGCTGATTGGAGC AAAGAGAATTAGCTGAAAAGAAGGCTCAATTAGCCAAGGCTGTAAAGAAT cgd6- 1080_106_revcom AACATCCTTTAAAGTTCCTCTGAGTGGAACGGCTGGGGCTGAGAATACAG AAGGAAGATGAGATTGTCGCTCATTATCGTATTACTCTCCGTTATAGTCT cgd6-1080_632 CAGGATTTCAGCACTCTCTCTGCTAATTCAAGTAGTCCAACTGAAAATGG TTTCCTCTGAGAGTGATCTTCTTGATCTTGATGAAGCCTGACCCGCAGAT cgd6-780_211_revcom AGAATACCAAGGTAGCTTATTTGAACTAAATGTATATACTAACCCTTCAC AAATTCTTTACCACCAAGTTTCAGTTGGACAAAAGCATGGAAAGATATTA cgd6-780_347_revcom TAGATTCCACATCTACTGTTTGTGCTGCAAAATTTATCATCAACTTACAG GCAGAAAATGATAAAGAAACTTTGGTAACAATTCAGAATGGTGATTTATA .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint 45 cgd7-4330_329 ACTCCATCAATATCAGGATATAATCGATGCGCTGTAATTATTGGTCTGAG TTCCTAAAGAGATGTTTTTGGAGCCAAAAGGACTAACATGAATTCTATCT cgd7- 4330_517_revcom GTTATCAACGTTAGTAATTCCCTGTGATATTGGCGTAGAAACAACTTGAG CGTACCAGAATATGCCAGATACGGGGCTATTCATTCTGGGTACAGTCCTT cgd3-3370_100 ATTCCTAGCAGGGTTGTTTCTCAACACCCTTCTGGGATGCGCTTCTGTTT ACTTTCTCTTGGAACAGATGCAGCTGGAGGAATTTCGGTTTCTGATGATG cgd3-3370_1241 GGTTCTGCAGTTGGTAGATACAAATGTAGAGTTGGAGAAACGTCTTGTTA TTTCTACTTTTTGACCGGGAGATAATGGGTTTTGTAGTCTAAATTCAGAA cgd6-2330_164 GCTAGTTGTTTGAATGCTTCGTACTCATTCATTGAAGGTGGCATGACACC ATTCTAACGATGCAGATGGTAAAAGAACGACAGAAACACGGATAGCATCG cgd6- 2330_316_revcom ACCATTCTTGTAAAGTGAATATGCTCTTAATGAAGCTATGCCACTTTCAA TTCTTCATTCCAGAATATTGCAACATTAAGATTCGGTGTGGACTCATGTT cgd6-1660_124 ATGCTTGGGGCCAGAGCATGTAGAATTTCATTTGGTCAACGGAAATAAAG TTTTCTATTTTCAACTTCGAACCTGCCATTTCCTTCTATTTGATTTAAAT cgd6-1660_998 GTAATAATTTTGAGCGAATCCCGAAATGAAATAAGGCCATACGGCTATAG CGTTGTTTAAGTCAATATTTTCGATAAATTGAGTCAGTATCATAGGGCTT cgd7-1960_160 AGCTTCAGAAGCAAAGGTAGAAACAGAAAACCCATTGCTATCAAGTACAG CATGGAAAGCAACTTATCAGAACAACAGGCACATGGATTGTTAGAACAAT cgd7-1960_317 TTAAAGAATCTATTTTGCACTTTAGCATGTAACGAGGAGTCATCAGACAT AACATCCTGACTCGCATACAGACAAGACCATAATTTCATTTGACAATGTT cgd7- 5520_1216_revco CGCTAAGTCTTTCTCAGAACCAAAAGTCATTTTATTCTCAGCGCGATTAA TTTCTATACAAAAACATGTTTTGAAAAGAATGAAGCACATTGTCTTAAAC cgd7- 5520_1115_revco AAACACAAGCGATTGAATAACTTACAGCTTCAAGAGCGTGTTTCCTATCA TTTGAGCTTGGCTTGATAAATGGTAGTTGGCTCGGAGGTGATATTTTTAT 1177 1178 Table 6. List of Antibodies Used in this Work 1179 Antibody Dilution factor (IFA or Expansion*) Source Catalogue mCherry Monoclonal Antibody (16D7) 1:1000 Invitrogen M11217 Anti-Cryptosporidium Immunodominant Antigen Cp23 1:1000 or 1:250* Stratech LS-C137378 Lot #234986 Vicia villosa lectin (VVL) 1:4000 2B Scientific FL-1231-2 Lectin HPA AF488 1:5000 or 1:1000* Invitrogen L11271 HPA-Alexa 647 1:5000 Thermo Fischer L32454 Hoechst 33342 1:10,000 Invitrogen H3570 Goat anti-Rabbit AF 647 1:1000 Invitrogen A-21245 Goat anti-Mouse AF546 1:1000 or 1:250* Invitrogen A-11030 Goat anti-Rat AF647 1:1000 Invitrogen A-21247 Alexa Fluor™ 405 NHS Ester 1:250* Thermo Scientific A30000 SYTOX Deep Red Nuclei Acid Stain 1:1000* Invitrogen S11381 Anti-TrpB 1:1000 Kind gift from lab of Boris Striepen Generated towards TrpB protein Click-IT Plus EdU Cell Proliferation Kit, Alexa Fluor 647 10mM Thermo Fischer C10340 1180 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted November 22, 2024. ; https://doi.org/10.1101/2024.11.22.624643doi: bioRxiv preprint